Patient Knowledge of Glycaemic Self-Monitoring in Type 2 Diabetes: A Survey-Based Study

Introduction

Type 2 diabetes mellitus represents one of the most pervasive and consequential non-communicable diseases of the contemporary era, exerting an escalating burden upon healthcare systems, national economies, and the quality of life of affected individuals across the globe. Epidemiological data published by the International Diabetes Federation indicate that approximately 537 million adults between the ages of 20 and 79 years were living with diabetes worldwide as of 2021, a figure projected to rise to 783 million by 2045 should prevailing trends of urbanisation, sedentary lifestyles, nutritional transition, and demographic ageing continue unabated ^[1]. Type 2 diabetes mellitus accounts for approximately 90 to 95 percent of all diagnosed cases, and its clinical course is characterised by progressive insulin resistance, deteriorating pancreatic beta-cell function, and chronic hyperglycaemia, which collectively predispose affected individuals to a spectrum of debilitating and life-threatening complications including cardiovascular disease, peripheral neuropathy, nephropathy, and retinopathy ^[2]. The socioeconomic consequences of this condition are commensurately severe; global expenditure attributable to diabetes and its sequelae was estimated at 1.3 trillion US dollars in 2015, with projections suggesting continued escalation as the prevalence of the disease rises and the age of affected populations increases ^[3]. Within Poland, national epidemiological data have identified diabetes as a condition affecting several million adults, with incidence rates continuing to trend upward in alignment with broader European and global patterns, rendering it a matter of significant concern for public health policy and clinical nursing practice alike.

Central to the effective clinical management of type 2 diabetes is the maintenance of glycaemic control within therapeutically appropriate parameters, a goal that requires sustained engagement from both healthcare professionals and patients themselves. Glycaemic self-monitoring, understood as the systematic and regular measurement of blood glucose concentrations by patients using personal glucometers or, increasingly, continuous glucose monitoring technologies, constitutes one of the foundational competencies of effective diabetes self-management. Through the timely and accurate measurement of blood glucose levels, patients and their clinical teams are afforded the capacity to identify hyperglycaemic and hypoglycaemic episodes, to evaluate the glycaemic impact of nutritional intake, physical activity, and pharmacological regimens, and to make informed adjustments to therapeutic behaviour that can meaningfully reduce the risk of both acute crises and long-term complications ^[4]. Evidence accumulated over several decades of clinical investigation consistently demonstrates that patients who possess adequate knowledge of glycaemic monitoring principles, who perform monitoring at appropriate frequency and in response to relevant clinical triggers, and who correctly interpret and act upon the results obtained, achieve superior long-term glycaemic outcomes as reflected in glycated haemoglobin concentrations compared with patients whose knowledge and monitoring practices are deficient ^[5]. The educational and behavioural dimensions of glycaemic self-monitoring are therefore not peripheral to clinical management but constitute a core determinant of therapeutic efficacy.

Despite the compelling evidence base supporting glycaemic self-monitoring as a mechanism of complication prevention and metabolic stabilisation, a substantial body of research conducted in diverse clinical and national contexts has identified persistent and clinically meaningful deficits in patient knowledge regarding the principles, techniques, targets, and interpretation of blood glucose self-measurement. Studies conducted in European, North American, and Asian healthcare settings have documented that considerable proportions of patients with type 2 diabetes are unable to correctly identify recommended target glucose ranges, remain uncertain regarding the optimal timing and frequency of measurement, report errors in glucometer calibration and technique, and lack the interpretive competencies required to translate numerical readings into actionable self-management decisions [6, 7]. These knowledge gaps are not uniformly distributed across patient populations; rather, they are systematically associated with a range of sociodemographic and clinical determinants including educational attainment, duration of diagnosed disease, type of pharmacological regimen, degree of engagement with structured diabetes education programmes, and access to specialist nursing support ^[8]. The identification and characterisation of these determinants within specific clinical populations and national healthcare contexts is therefore an essential prerequisite for the design of targeted, evidence-based educational interventions capable of producing measurable improvements in glycaemic self-monitoring knowledge and practice.

Within the Polish context, research specifically directed at the assessment of patient knowledge regarding glycaemic self-monitoring in type 2 diabetes remains comparatively limited in scope and methodological rigour, notwithstanding the growing recognition among Polish nursing and medical communities of the importance of patient education as a component of diabetes care. While a number of Polish-language investigations have examined broader dimensions of diabetes self-management knowledge and health literacy, studies employing purpose-designed, validated instruments to assess the specific domain of glycaemic monitoring knowledge — and to systematically map the sociodemographic and clinical correlates of differential knowledge attainment — are underrepresented in the available literature. This evidence gap is consequential from both a scientific and a practical perspective: without reliable, contextually grounded data characterising the nature and distribution of knowledge deficits among Polish patients with type 2 diabetes, nursing and medical professionals are constrained in their capacity to design, implement, and evaluate educational programmes commensurate with the specific needs of this population. The present thesis was undertaken with the explicit intention of addressing this gap.

The primary aim of the present investigation was to assess the level of knowledge regarding glycaemic self-monitoring among adult patients diagnosed with type 2 diabetes mellitus attending specialist outpatient diabetology services in Poland, and to identify the sociodemographic and clinical factors associated with differential knowledge attainment within this population. Three principal research hypotheses guided the empirical component of the study. The first hypothesis postulated that the overall level of knowledge regarding glycaemic self-monitoring among the study population would be insufficient, as operationalised by mean percentage scores below a pre-specified threshold on a purpose-designed knowledge instrument. The second hypothesis proposed that patients with higher educational attainment and longer duration of diagnosed diabetes would demonstrate significantly superior knowledge compared with those possessing lower educational qualifications or shorter illness histories. The third hypothesis anticipated that patients receiving insulin therapy and those who had participated in structured diabetes education programmes would exhibit meaningfully higher knowledge scores relative to patients managed exclusively with oral agents and those without formal educational exposure. The testing of these hypotheses against empirical data collected from a defined sample of 120 adult patients constitutes the core scientific contribution of the present work.

The methodological approach adopted in this investigation was that of a cross-sectional, descriptive-analytical survey, conducted across two specialist outpatient diabetology clinics. Data were collected using a purpose-designed, internally validated 20-item knowledge questionnaire administered under standardised conditions, supplemented by a sociodemographic and clinical characteristics form. This design was selected in recognition of its established appropriateness for the assessment of knowledge prevalence and distribution across heterogeneous patient populations, and its capacity to support the identification of associations between independent sociodemographic and clinical variables and the primary outcome measure of glycaemic monitoring knowledge. Statistical analysis was performed using descriptive and inferential procedures appropriate to the measurement level and distributional properties of the data, enabling rigorous hypothesis testing and the generation of findings amenable to evidence-based clinical application.

The thesis is structured across four principal chapters, each fulfilling a distinct and complementary function within the broader investigative framework. The first chapter establishes the theoretical and empirical foundations of the research problem, reviewing the epidemiology and pathophysiology of type 2 diabetes mellitus, the mechanisms by which chronic hyperglycaemia produces end-organ damage, the clinical evidence base for glycaemic self-monitoring as a tool of complication prevention, and the technical and educational dimensions of blood glucose measurement practice. The second chapter addresses the theoretical and empirical literature on patient education and health literacy in the context of chronic disease self-management, examining the conceptual models that inform educational programme design and reviewing evidence pertaining to the determinants of glycaemic monitoring knowledge and practice in type 2 diabetes populations. The third chapter presents a detailed account of the methodological framework underpinning the empirical investigation, encompassing the study design and rationale, the process of instrument construction and validation, the sampling strategy and recruitment procedures, the data collection protocol, and the statistical analysis plan. The fourth and final chapter presents the quantitative findings of the survey in systematic detail, including the sociodemographic and clinical profile of the study sample, overall and domain-specific knowledge scores, the results of hypothesis testing, and a comprehensive discussion that situates the findings within the context of the broader national and international literature, drawing out their implications for nursing practice, health service organisation, and future research.

The significance of the present investigation derives from the convergence of epidemiological urgency, evidential necessity, and clinical practicability. As the prevalence of type 2 diabetes continues to rise and the demands placed upon specialist diabetology services intensify, the capacity of nursing professionals to deliver effective, targeted educational interventions directed at glycaemic self-monitoring competency becomes an increasingly important determinant of population-level health outcomes. The generation of reliable, contextually grounded evidence characterising the knowledge landscape of Polish patients with type 2 diabetes, and the identification of the population subgroups most susceptible to knowledge deficit, constitutes a foundational step towards the development of educational strategies commensurate with the complexity and diversity of clinical need. It is the aspiration of the present author that the findings reported herein will contribute meaningfully to this endeavour, and that the study will serve as both a substantive scientific contribution and a practical resource for clinicians, educators, and health service planners engaged in improving the quality of care afforded to patients living with type 2 diabetes in Poland and beyond.

Chapter 1: Theoretical Foundations of Type 2 Diabetes and Glycaemic Control

1.1. Epidemiology and Pathophysiology of Type 2 Diabetes Mellitus

Type 2 diabetes mellitus (T2DM) constitutes one of the most consequential non-communicable diseases of the contemporary era, representing a global public health burden of extraordinary and still-growing magnitude. According to data published by the International Diabetes Federation, approximately 537 million adults aged 20–79 years were estimated to be living with diabetes worldwide as of 2021, a figure that is projected to reach 783 million by 2045 if prevailing trends of urbanisation, dietary change, and population ageing continue unabated [+International Diabetes Federation, IDF Diabetes Atlas 10th Edition, IDF, 2021]. T2DM accounts for approximately 90–95% of all diagnosed diabetes cases globally, with the remainder attributable to type 1 diabetes mellitus, gestational diabetes, and specific secondary aetiologies. The epidemiological burden is disproportionately concentrated in low- and middle-income countries, where limited healthcare infrastructure constrains timely diagnosis and sustained therapeutic management. The direct and indirect costs attributable to diabetes, including the treatment of its complications, were estimated at 1.3 trillion US dollars globally in 2015, with projections indicating further escalation in the coming decades ^[11]. These figures underscore the urgency of evidence-based strategies for both prevention and effective long-term management of glycaemia in the established disease state.

Within the European region, the prevalence of diabetes is estimated at approximately 10.2% of the adult population, with substantial national variation reflecting differences in lifestyle patterns, dietary habits, levels of physical activity, and the comprehensiveness of national surveillance systems [+World Health Organization, Global Report on Diabetes, WHO Press, 2016]. Polish epidemiological data indicate that an estimated 3.0–3.5 million individuals in Poland are affected by diabetes, a proportion of whom remain undiagnosed and consequently at elevated risk of late-presenting microvascular and macrovascular complications. National registry data indicate that T2DM prevalence increases markedly with advancing age, with persons over 65 years exhibiting substantially higher rates than younger cohorts. Urban populations display marginally higher rates of T2DM relative to rural communities, a pattern consistent with greater exposure to sedentary occupational environments and to obesogenic dietary patterns characteristic of urbanised societies. These demographic gradients have direct implications for the organisation of diabetes education services, including those focused on self-monitoring competencies.

The pathophysiology of T2DM is characterised by two interrelated and mutually reinforcing processes: peripheral tissue insulin resistance and progressive pancreatic beta-cell dysfunction. Insulin resistance manifests primarily in skeletal muscle, hepatic parenchyma, and adipose tissue, where impaired insulin receptor signalling — mediated through reduced activation of insulin receptor substrate-1 and insulin receptor substrate-2 and deficient translocation of GLUT-4 glucose transporters to the plasma membrane — results in substantially diminished cellular glucose uptake [+DeFronzo RA, Banting Lecture: From the Triumvirate to the Ominous Octet, Diabetes, 2009]. In the liver, the failure of insulin to adequately suppress gluconeogenesis and glycogenolysis contributes directly to fasting hyperglycaemia, a hallmark of the established disease state. Visceral adiposity, through the elaboration of proinflammatory adipokines including tumour necrosis factor-alpha and interleukin-6, and through the generation of elevated circulating free fatty acids, amplifies insulin resistance at both the receptor and post-receptor signalling levels. The resulting compensatory hyperinsulinaemia — the pancreatic beta-cell's initial adaptive response — is sustainable only until beta-cell secretory reserve is exhausted by the combined demands of peripheral resistance and the direct cytotoxic effects of chronic glucose and lipid excess.

Progressive beta-cell dysfunction constitutes the second pillar of T2DM pathogenesis and distinguishes the natural history of T2DM from that of type 1 diabetes mellitus, in which absolute insulin deficiency results from autoimmune islet destruction. In T2DM, beta-cell mass and secretory function decline gradually through mechanisms including glucotoxicity, lipotoxicity, endoplasmic reticulum stress, oxidative injury, and the deposition of islet amyloid polypeptide within the islets of Langerhans [~Weyer C et al., The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus, Journal of Clinical Investigation, 1999]. Prolonged exposure of beta cells to supraphysiological glucose concentrations impairs insulin gene transcription and exocytotic coupling; simultaneously, elevated free fatty acid concentrations activate intrinsic apoptotic cascades. The net result is a self-reinforcing pathological cycle in which diminishing insulin secretion exacerbates hyperglycaemia, which in turn accelerates beta-cell injury and deepens peripheral insulin resistance. Genetic predisposition, encoded in multiple susceptibility loci identified through genome-wide association studies, interacts with modifiable environmental determinants — principally obesity, physical inactivity, and hypercaloric dietary patterns — to govern individual disease risk and the rate of beta-cell decline.

The natural history of T2DM progresses through identifiable and clinically relevant stages, beginning with normoglycaemia, advancing through prediabetes — encompassing impaired fasting glucose (fasting plasma glucose 5.6–6.9 mmol/L) and impaired glucose tolerance (two-hour post-load glucose 7.8–11.0 mmol/L) — and culminating in overt T2DM. Diagnostic criteria, as established by the American Diabetes Association and adopted by the World Health Organization, define T2DM on the basis of a fasting plasma glucose of ≥7.0 mmol/L, a two-hour oral glucose tolerance test value of ≥11.1 mmol/L, a random plasma glucose of ≥11.1 mmol/L in a symptomatic individual, or a glycated haemoglobin (HbA1c) of ≥6.5% confirmed by a laboratory-certified method ^[12]. Recognition of the prediabetes stage affords a critical window for lifestyle and pharmacological intervention to delay or prevent progression to overt T2DM. This multistage natural history underscores the importance of not only early diagnosis but also sustained monitoring to detect deterioration in glycaemic control before complications become established. The epidemiological scale and the pathophysiological complexity of T2DM collectively motivate the clinical imperative for systematic glycaemic self-monitoring, which forms the central subject of this thesis.

• Global diabetes prevalence (2021): approximately 537 million adults; projected 783 million by 2045
• T2DM proportion of all diabetes: approximately 90–95% of cases
• Polish estimate: 3.0–3.5 million individuals affected, with a proportion undiagnosed
• Primary pathophysiological axes: peripheral insulin resistance and progressive beta-cell dysfunction
• Diagnostic HbA1c threshold: ≥6.5% (confirmed by NGSP-certified laboratory method)

1.2. Glycaemic Targets and Clinical Standards in Type 2 Diabetes Management

The establishment of evidence-based glycaemic targets in T2DM has been shaped by a series of landmark randomised controlled trials whose findings have profoundly and sometimes controversially influenced international clinical guidelines over four decades. The United Kingdom Prospective Diabetes Study (UKPDS), conducted between 1977 and 1997 in 5,102 newly diagnosed patients with T2DM, demonstrated that intensive glycaemic control — achieving a median HbA1c of approximately 7.0% in the intensive arm compared with 7.9% in the conventional arm — was associated with a 25% reduction in microvascular endpoint risk and a 16% non-significant reduction in myocardial infarction, establishing the foundational principle that glycaemic control matters for long-term complications [~UKPDS Group, Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes, Lancet, 1998]. Subsequent large-scale trials — ACCORD, ADVANCE, and VADT — provided critical and somewhat sobering nuance: intensive glycaemic therapy targeting an HbA1c of below 6.0–6.5% in patients with longstanding T2DM and established or high-risk cardiovascular disease was associated with increased total mortality in ACCORD, with no significant macrovascular benefit in VADT, and only modest benefit with a more moderate target in ADVANCE ^[11]. These findings collectively established that the clinical benefits of glycaemic optimisation are context-dependent and must be weighed against the risks of hypoglycaemia, treatment burden, and potential cardiovascular hazards in vulnerable patient subgroups. The more recent EMPA-REG OUTCOME and LEADER trials demonstrated that cardiorenal protection afforded by SGLT-2 inhibitors and GLP-1 receptor agonists, respectively, operates through mechanisms partially independent of glycaemic control, introducing further complexity into the target-setting framework.

Glycated haemoglobin (HbA1c) has emerged as the universally adopted metric for monitoring long-term glycaemic control in both clinical practice and research. The biochemical basis of HbA1c rests upon the non-enzymatic glycation of the N-terminal valine residue of the haemoglobin beta-chain, a reaction that proceeds in proportion to mean plasma glucose concentration and accumulates over the lifespan of the erythrocyte — approximately 120 days — thereby reflecting mean glycaemia over the preceding two to three months ^[12]. Laboratory measurement of HbA1c employs high-performance liquid chromatography, immunoassay, or affinity chromatography methods standardised against the International Federation of Clinical Chemistry and Laboratory Medicine reference system, ensuring comparability across certified laboratories ^[12]. The recognised limitations of HbA1c as a sole monitoring metric are clinically important: haemoglobinopathies, haemolytic and iron-deficiency anaemia, chronic kidney disease, and ethnic variation in glycation rates may all produce spurious values that do not accurately reflect true mean glycaemia, potentially leading to erroneous clinical decisions ^[11]. HbA1c also fails to capture intraday glucose variability, postprandial excursions, and hypoglycaemic episodes — dimensions of glycaemic exposure with independent pathological significance — a limitation that has motivated the development of complementary monitoring metrics.

Current clinical guidelines from the principal diabetological societies converge on a general HbA1c target of less than 7.0% (53 mmol/mol) for most non-pregnant adults with T2DM, whilst simultaneously emphasising the imperative of individualisation based upon a comprehensive assessment of patient-specific factors. The American Diabetes Association Standards of Medical Care recommend target relaxation to less than 8.0% — or even less than 8.5% — in elderly patients, those with a history of severe hypoglycaemia, hypoglycaemia unawareness, limited life expectancy, extensive comorbidity, or long disease duration with established macrovascular complications [+American Diabetes Association, Standards of Medical Care in Diabetes 2024, Diabetes Care, 2024]. The European Association for the Study of Diabetes and the Polish Diabetes Association guidelines similarly advocate for patient-centred target negotiation, integrating disease duration, hypoglycaemia risk profile, treatment capacity, and patient-expressed preferences regarding quality of life. Beyond HbA1c, fasting plasma glucose targets of 4.4–7.2 mmol/L and postprandial peak glucose below 10.0 mmol/L are recommended by most guidelines; postprandial hyperglycaemia is recognised as an independent cardiovascular risk factor that merits targeted intervention ^[3]. The selection of pharmacological agents — from metformin as the foundational first-line therapy, through sulfonylureas, SGLT-2 inhibitors, and GLP-1 receptor agonists, to basal or intensified insulin — determines the monitoring intensity required and the specific glucose excursions most relevant to therapeutic optimisation at each stage.

An increasingly important set of glycaemic metrics, derived from continuous glucose monitoring technology, has extended the conceptual framework beyond mean glycaemia as captured by HbA1c. Time-in-range (TIR), defined as the proportion of time spent within the glucose range of 3.9–10.0 mmol/L (70–180 mg/dL), has been endorsed by the Advanced Technologies and Treatments for Diabetes international consensus group as a clinically meaningful outcome, with a recommended target of at least 70% of total monitoring time for most adults with T2DM ^[10]. Time-below-range (TBR, glucose below 3.9 mmol/L) is designated the highest-priority safety metric, with a target below 4% of total time. Glycaemic variability — quantified through the coefficient of variation of glucose readings or the standard deviation of serial HbA1c measurements — has been identified as an independent predictor of cardiovascular complications in T2DM, irrespective of whether mean HbA1c remains within the recommended target range, a finding supported by analysis of over 100,000 patients with T2DM without pre-existing cardiovascular disease ^[10]. These refined metrics provide a substantially more granular and dynamic picture of glycaemic physiology than HbA1c alone and are increasingly incorporated into both clinical practice guidelines and research outcome frameworks.

• General HbA1c target (most adults): <7.0% (53 mmol/mol)
• Relaxed HbA1c target (elderly, hypoglycaemia risk, multimorbidity): <8.0–8.5%
• Fasting plasma glucose target: 4.4–7.2 mmol/L
• Postprandial peak glucose target: <10.0 mmol/L
• Time-in-range target (3.9–10.0 mmol/L): ≥70% of total monitoring time
• Time-below-range target (<3.9 mmol/L): <4% of total monitoring time

1.3. The Role of Self-Monitoring of Blood Glucose in Non-Insulin-Treated Patients

Self-monitoring of blood glucose (SMBG) — defined as the patient-performed measurement of capillary whole blood glucose concentration using a portable glucometer and single-use reagent test strips — represents one of the most actively debated interventions in the management of non-insulin-treated T2DM. The analytical accuracy of commercially available glucometers is governed by the ISO 15197:2013 international standard, which requires that at least 95% of measurements fall within 15 mg/dL (0.83 mmol/L) of a reference laboratory value for glucose concentrations below 5.55 mmol/L, and within 15% of the reference value at higher concentrations ^[3]. SMBG provides patients with real-time information regarding their glycaemic status, enabling informed self-management decisions concerning dietary intake, physical activity scheduling, and, where applicable, self-adjustment of pharmacological regimens. Documentation of SMBG results in a logbook or digital platform facilitates retrospective review during clinical consultations, supporting the titration of glucose-lowering agents and the identification of patterns of postprandial excursion or nocturnal hypoglycaemia ^[3]. The clinical utility of these data is, however, fundamentally contingent upon whether patients possess the knowledge and practical skills to interpret results and translate them into appropriate self-management actions — a dependency that defines the educational imperative examined in Chapter 2.

The evidence base for SMBG in non-insulin-treated T2DM is characterised by substantial heterogeneity in design, intervention intensity, and reported outcomes across published randomised controlled trials. A meta-analysis by Zhu and colleagues, encompassing 15 RCTs and 3,383 patients with non-insulin-treated T2DM, demonstrated that SMBG intervention produced a statistically significant mean reduction in HbA1c of −0.33% (95% CI −0.45 to −0.22%; p = 3.07×10⁻⁸) relative to control, with the benefit observed across both short-term (≤6-month) and long-term (≥12-month) follow-up periods ^[1]. The same meta-analysis reported additional benefits in body mass index (mean difference −0.65) and total cholesterol (−0.12 mmol/L), suggesting that the impact of SMBG extends beyond glycaemic parameters alone ^[1]. A meta-analysis by Poolsup and colleagues, reviewing nine RCTs encompassing 2,419 patients, similarly confirmed a statistically significant HbA1c improvement (weighted mean difference −0.24%, 95% CI −0.34% to −0.14%), with the effect more pronounced in patients whose baseline HbA1c exceeded 8%, indicating that patients with poorer initial glycaemic control derive the greatest absolute benefit from SMBG intervention ^[4]. A more recent comprehensive systematic review and meta-analysis by Alserhani and colleagues, drawing upon 22 studies, reported a pooled HbA1c reduction of −0.32% (95% CI −0.44% to −0.20%), further establishing overall SMBG effectiveness across diverse clinical settings ^[5].

A critical distinction in the interpretation of the SMBG evidence base concerns the difference between structured and unstructured monitoring protocols. Structured SMBG involves a predefined testing schedule — most commonly paired pre- and post-prandial measurements on specified days — linked to a systematic protocol for result interpretation and treatment adjustment by both the patient and the supervising clinician. Unstructured SMBG, by contrast, lacks a coherent feedback framework, and accumulated data are neither systematically reviewed nor acted upon, yielding uncertain clinical benefit. The 12-month randomised controlled SMBG Study by Parsons and colleagues enrolled 446 participants with sub-optimally controlled non-insulin-treated T2DM (HbA1c ≥58 mmol/mol) and demonstrated that structured SMBG produced an HbA1c reduction of 8.9 mmol/mol (0.8%; 95% CI −1.10 to −0.54) compared with usual care, a finding that reached high statistical significance ^[2]. Importantly, the addition of monthly TeleCare support to structured SMBG did not produce a statistically significant further reduction in HbA1c beyond that achieved by structured monitoring alone, suggesting that the systematic protocol — rather than the frequency of clinician contact — is the active therapeutic component ^[2]. Participants with higher educational attainment, shorter disease duration, and lower baseline HbA1c were more likely to achieve the target of ≤53 mmol/mol, highlighting the moderating role of sociodemographic and clinical variables in SMBG effectiveness.

The potential benefit of SMBG in newly diagnosed non-insulin-treated T2DM has been examined in a large retrospective cohort study by Sia and colleagues, which analysed 4,987 eligible patients enrolled in a structured diabetes management programme in Taiwan. Patients who commenced SMBG at the time of diagnosis demonstrated a greater HbA1c reduction at three months compared with non-users (estimated maximal difference 0.55%), with a statistically significant group-by-time interaction sustained over the full 12-month observation period in both non-insulin secretagogue and insulin secretagogue subgroups ^[6]. These findings are consistent with the theoretical importance of the early post-diagnostic period as a window of heightened susceptibility to behavioural modification and metabolic optimisation, and support the International Diabetes Federation's recommendation that SMBG be considered at the time of T2DM diagnosis as an educational tool. Guideline recommendations for SMBG frequency in non-insulin-treated T2DM nonetheless remain heterogeneous across national and international bodies: the ADA recommends frequency individualised to treatment regimen and patient-specific goals; NICE guidelines in the United Kingdom advise against routine SMBG in non-insulin-treated T2DM in the absence of specific clinical indications; and the Polish Diabetes Association provides tiered recommendations stratified by pharmacological regimen and associated hypoglycaemia risk ^[3].

Barriers to effective SMBG span economic, educational, organisational, and psychosocial domains, and collectively limit the real-world impact of monitoring in clinical practice. In Poland and numerous other European countries, reimbursement for test strips is not systematically available to non-insulin-treated T2DM patients, imposing a direct financial burden that restricts monitoring frequency, particularly among patients of lower socioeconomic status. Patient-level barriers include discomfort and anxiety associated with repeated finger-prick procedures, insufficient training in result interpretation, absence of systematic feedback mechanisms connecting monitoring data to clinical decisions, and psychological responses ranging from anxiety and glucose obsession in some individuals to therapeutic complacency in others ^[3]. At the clinician and system level, the inconsistent integration of patient-generated SMBG data into consultation processes reduces the actionability of monitoring records. These structural and educational deficiencies underscore the conclusion — consistently reinforced across the systematic review evidence base — that the effectiveness of SMBG as a clinical tool is fundamentally inseparable from the quality, consistency, and patient-centredness of the educational support provided in conjunction with it ^[5].

1.4. Technological Advances in Glucose Monitoring: From Traditional SMBG to Continuous Glucose Monitoring

The technological evolution of glucose monitoring has proceeded through several distinct generations, each characterised by improvements in analytical accuracy, user convenience, and the richness of glycaemic data made available to patients and clinicians. The first portable blood glucose meters, introduced in the early 1970s, employed glucose oxidase-based electrochemical reactions to generate semiquantitative estimates from capillary blood samples; the subsequent four decades witnessed progressive miniaturisation, reduction in required sample volumes to below one microlitre, memory-enabled data storage, graphical display interfaces, and the integration of Bluetooth connectivity for synchronisation with smartphone applications and electronic health record platforms ^[3]. Contemporary glucometers routinely achieve analytical performance meeting or exceeding the ISO 15197:2013 standard, delivering measurement results in fewer than five seconds and offering colour-coded trend indicators and logbook integration that reduce the cognitive burden of glucose record-keeping. Despite these advances, the fundamental limitation of capillary SMBG remains its provision of discrete, time-point measurements that fail to capture inter- and postprandial glucose excursions, patterns of glycaemic variability, or nocturnal fluctuations without impractically frequent user-initiated sampling. This limitation provided the principal clinical motivation for the development of sensor-based continuous and flash glucose monitoring technologies.

Flash glucose monitoring (FGM), exemplified by the Abbott FreeStyle Libre system, represents a technically and clinically distinct intermediate between conventional SMBG and real-time continuous glucose monitoring (CGM). The FreeStyle Libre 2 and Libre 3 platforms employ a factory-calibrated, single-use subcutaneous amperometric biosensor worn for up to 14 days, which measures glucose concentration in interstitial fluid; readings are obtained when the user scans the sensor with a dedicated reader device or a compatible smartphone application, yielding a current interstitial glucose value, a directional trend arrow indicating the rate and direction of glycaemic change, and an eight-hour retrospective glucose graph ^[7]. The mean absolute relative difference (MARD) of the FreeStyle Libre 2 has been reported at 9.2–9.7%, whilst the Libre 3, with its smaller form factor and continuous Bluetooth transmission capability, achieves 7.9–9.4% ^[8]. The distinction between FGM and real-time CGM is clinically important: standard FGM requires active user scanning and provides no automatic glucose alarms in the absence of scanning, whereas real-time CGM systems transmit glucose data continuously and alert the user to current or impending hypoglycaemia and hyperglycaemia without requiring a scan ^[7]. This distinction affects the clinical suitability of each technology for patients at high risk of asymptomatic hypoglycaemia.

Real-time continuous glucose monitoring systems currently available on the market include the Dexcom G7, the Medtronic Guardian 4, and the FreeStyle Libre 3 operating in its continuous transmission mode. The Dexcom G7, with a MARD of 8.2–9.1% and the shortest available warm-up period of 30 minutes, is widely used in insulin-intensive therapy regimens and is approved for integration with automated insulin delivery systems; it offers automatic predictive hypoglycaemia alerts and compatibility with insulin pump platforms ^[8]. The Medtronic Guardian 4, integrated with the MiniMed pump family, provides predictive alerts up to 60 minutes before projected critical glycaemic events and a no-calibration design representing a significant improvement over earlier Medtronic generations, though its MARD of 10.1–11.2% is somewhat less accurate than competing platforms ^[8]. A physiologically important limitation common to all interstitial CGM systems is the lag of approximately 5–15 minutes between plasma glucose and interstitial glucose concentrations, resulting from the kinetics of glucose diffusion across the capillary endothelium; this lag is most clinically significant during rapid glycaemic excursions — for example, immediately following carbohydrate ingestion or during acute exercise-induced hypoglycaemia — and should inform patient education regarding the interpretation of CGM readings in these contexts.

The clinical utility of CGM and FGM in T2DM has been supported by randomised controlled trials and large real-world datasets. In the REPLACE study, insulin-treated T2DM patients assigned to FGM demonstrated a 43% reduction in time spent in hypoglycaemia compared with those continuing conventional SMBG ^[7]. Real-world analyses of FGM in T2DM populations have demonstrated a positive correlation between the number of daily scans and glycaemic outcomes, with patients performing an average of 8 scans per day achieving less time in hyperglycaemia and more time in range ^[7]. Meta-analytic evidence confirms consistent HbA1c reductions of 0.25–3.0% and time-in-range improvements of 15–34 percentage points across diverse T2DM populations using CGM ^[8]. A randomised controlled study integrating self-regulation education with CGM demonstrated significant improvements in glycaemic outcomes and self-management behaviours compared with standard care, suggesting that technology-enhanced monitoring yields its greatest benefit when combined with structured educational support ^[9]. Despite these clinical advantages, structural barriers — including limited reimbursement coverage for CGM in non-insulin-treated T2DM patients in Poland and many European health systems — constrain widespread access and necessitate continued advocacy for policy reform.

Comparison of Principal Glucose Monitoring Modalities Available to Patients with Type 2 Diabetes

Modality	Data provided	Sampling frequency	Automatic alarms	Primary T2DM indication
Capillary SMBG (glucometer)	Discrete point-in-time plasma glucose surrogate	Patient-determined (1–7+ per day)	No	All treatment regimens; structured protocols in non-insulin-treated patients
Flash glucose monitoring (FGM)	Interstitial glucose value, 8-hour retrospective graph, trend arrow	Scan-to-read (minimum 1 per 8 hours)	Optional (Libre 2/3 with threshold alerts)	Insulin-treated; hypoglycaemia risk; structured education programmes
Real-time CGM	Near-continuous interstitial glucose, trend arrows, predictive alerts	Continuous (1–5 minute intervals)	Yes (customisable high/low thresholds)	Intensive insulin therapy; closed-loop systems; high glycaemic variability

1.5. Complications of Poorly Controlled Glycaemia and Their Prevention Through Monitoring

Chronic hyperglycaemia in T2DM is the primary pathological mediator of a spectrum of microvascular and macrovascular complications that collectively constitute the leading causes of diabetes-associated morbidity, disability, and premature mortality. Microvascular complications — diabetic retinopathy, diabetic nephropathy, and peripheral and autonomic neuropathy — result from the direct toxic effects of sustained glucose elevation on small vessel endothelium, pericytes, glomerular mesangial cells, and peripheral neurons. Macrovascular complications — coronary artery disease, ischaemic cerebrovascular disease, and peripheral arterial disease — arise through the accelerated atherosclerotic process associated with the metabolic syndrome, dyslipidaemia, and endothelial dysfunction that frequently accompany T2DM. Epidemiological data indicate that individuals with T2DM carry a two- to tenfold higher risk of cardiovascular disease-related death compared with normoglycaemic age-matched controls, and are at substantially elevated risk of end-stage renal disease, lower-limb amputation, visual loss, and reduced quality of life ^[11]. The prevalence of diabetic retinopathy in long-standing T2DM populations is estimated at 20–40%, peripheral neuropathy at 30–50%, and diabetic nephropathy at 20–40% in various national cohort studies, underscoring the enormous preventable burden attributable to inadequate glycaemic control.

The pathophysiological mechanisms through which chronic hyperglycaemia induces tissue injury have been characterised through several biochemical pathways that converge on oxidative stress as a unifying upstream process. The polyol pathway — in which aldose reductase reduces excess intracellular glucose to sorbitol, with concurrent NADPH depletion and accumulation of reactive oxygen species — causes injury to peripheral neurons, lens epithelium, and retinal pericytes, contributing to neuropathy and cataract formation [+Brownlee M, The pathobiology of diabetic complications: a unifying mechanism, Diabetes, 2005]. Advanced glycation end-products (AGEs), formed through the non-enzymatic Maillard reaction between reducing sugars and protein amino groups, crosslink structural proteins within the glomerular basement membrane and retinal capillary walls, impair normal vascular function, and engage the receptor for advanced glycation end-products (RAGE), perpetuating a chronic proinflammatory and profibrotic state. Protein kinase C activation, driven by intracellular diacylglycerol accumulation under hyperglycaemic conditions, impairs endothelial nitric oxide synthase activity, promotes vascular permeability, and activates profibrotic transforming growth factor-beta signalling. The hexosamine biosynthesis pathway contributes through O-GlcNAcylation of transcription factors, altering gene expression profiles that regulate vascular homeostasis. The unifying mechanistic hypothesis proposed by Brownlee attributes the activation of all four pathways to mitochondrial superoxide overproduction induced by hyperglycaemia, providing a conceptual framework for the molecular heterogeneity of diabetic tissue injury [+Brownlee M, The pathobiology of diabetic complications: a unifying mechanism, Diabetes, 2005].

The quantitative epidemiological relationship between glycaemic control and the risk of T2DM complications was definitively established through the landmark UKPDS analyses. An observational analysis within the UKPDS cohort demonstrated that each 1% absolute reduction in HbA1c was associated with a 37% decrease in microvascular endpoint risk and a 14% reduction in myocardial infarction risk, with no threshold below which risk ceased to decrease — a finding that provided the evidentiary foundation for intensive glycaemic target-setting [~UKPDS Group, Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35), BMJ, 2000]. The concept of metabolic memory, or the legacy effect, was established through the UKPDS 10-year post-trial follow-up: patients originally randomised to intensive glycaemic control maintained significantly lower rates of myocardial infarction and all-cause mortality a decade after trial completion, despite subsequent convergence of HbA1c levels between treatment arms, suggesting that early and sustained glycaemic optimisation exerts durable protective effects at the epigenetic and cellular level ^[12]. In parallel, the DCCT/EDIC study in type 1 diabetes confirmed that the legacy effect operates across diabetes types and involves epigenetic mechanisms that perpetuate the beneficial impact of prior glycaemic control on vascular gene expression ^[6]. These findings collectively argue for early and sustained monitoring as the mechanism through which clinical benefit is preserved over the long course of the disease.

Hypoglycaemia, as an iatrogenic complication of antidiabetic therapy rather than of hyperglycaemia per se, is increasingly recognised as a clinically significant source of morbidity and mortality in T2DM populations, particularly in elderly patients and those managed with sulfonylureas or insulin. Severe hypoglycaemia is associated with cardiac arrhythmias — including prolonged QTc interval and ventricular ectopy — acute coronary events, neurocognitive impairment, falls and fragility fractures, and accelerated dementia risk in older individuals [+Goto A et al., Severe hypoglycaemia and cardiovascular disease: systematic review and meta-analysis with bias analysis, BMJ, 2013]. The J-shaped or U-shaped relationship between achieved HbA1c and all-cause mortality risk observed in several epidemiological analyses in T2DM is partly attributable to the hazards of hypoglycaemia resulting from overly intensive treatment in vulnerable patients, providing the empirical basis for individualised rather than universal target-setting. Glycaemic variability — expressed as variability in serial HbA1c measurements over time — has been identified as an independent predictor of composite cardiovascular outcomes in over 100,000 patients with T2DM, with the highest risk quartile of HbA1c standard deviation associated with significantly elevated rates of non-fatal myocardial infarction, non-fatal stroke, and all-cause mortality regardless of whether mean HbA1c remained within the recommended target ^[10]. These observations underscore that the objectives of glycaemic monitoring must encompass the detection of hypoglycaemia and the reduction of glycaemic variability, not merely the minimisation of mean glucose elevation.

The role of systematic self-monitoring in the prevention of diabetic complications is understood to operate through multiple synergistic mechanisms that extend beyond simple detection of elevated glucose values. SMBG and CGM enable the identification of asymptomatic hyperglycaemia that prompts timely therapeutic adjustment; the characterisation of postprandial glucose excursions as discrete intervention targets amenable to dietary modification; the detection of hypoglycaemia, including clinically significant asymptomatic nocturnal episodes; and the provision of real-time metabolic feedback that reinforces and informs dietary and physical activity self-management decisions ^[3]. Structured SMBG programmes have been shown to facilitate more timely treatment intensification and to reduce the mean time patients spend outside target glucose ranges, as demonstrated by the significant HbA1c reductions observed in the structured-monitoring arms of published RCTs ^[2]. The effectiveness of these monitoring-based interventions is, however, fundamentally contingent upon the patient's capacity to interpret glycaemic data accurately, recognise clinically significant patterns within monitoring records, and respond with appropriate, evidence-informed self-management behaviour. This critical dependency upon patient knowledge and health literacy — which determines whether the data generated by monitoring instruments are translated into meaningful self-management action or remain clinically inert — provides the conceptual foundation for Chapter 2, which examines the theoretical and empirical dimensions of patient education and health literacy in T2DM self-management.

Chapter 2: Patient Education and Health Literacy in Diabetes Self-Management

2.1. Theoretical Models of Patient Education in Chronic Disease Management

The conceptualisation of patient education in chronic disease management has evolved substantially over the preceding four decades, moving away from a didactic, biomedical paradigm towards frameworks that acknowledge the active, agentic role of the patient in constructing and applying health knowledge. Traditional models of health instruction, characterised by one-directional information transfer from clinician to patient, have been progressively supplanted by biopsychosocial approaches that integrate cognitive, emotional, social, and environmental dimensions of illness behaviour. An understanding of these theoretical frameworks is indispensable for the design, delivery, and evaluation of educational interventions targeting glycaemic self-monitoring in type 2 diabetes. The frameworks described below are not mutually exclusive; rather, they are best understood as complementary lenses that together illuminate different aspects of the educational encounter.

The Self-Regulation Model proposed by Leventhal and colleagues offers a particularly illuminating framework for understanding how patients cognitively and emotionally process information about their chronic condition. [~Leventhal, The Common Sense Model of Self-Regulation of Health and Illness, Lawrence Erlbaum, 1998] According to this model, patients construct cognitive representations of their illness along five dimensions: identity (the perceived symptoms and label of the disease), cause (the attributed aetiology), timeline (beliefs about the expected course and duration), consequences (anticipated physical, social, and economic impact), and controllability (beliefs about whether the condition can be managed effectively). In the context of type 2 diabetes, these illness representations directly shape engagement with self-monitoring protocols; patients who do not perceive hyperglycaemia as symptomatic — a common phenomenon given the frequently asymptomatic early course of the condition — are substantially less likely to perform regular blood glucose measurements. The model thus provides a theoretical rationale for educational interventions that explicitly assess and restructure maladaptive illness beliefs before proceeding to the instruction of technical self-monitoring skills.

Bandura's Social Cognitive Theory contributes the constructs of self-efficacy and observational learning to the theoretical foundation of diabetes education. [~Bandura, Self-Efficacy: The Exercise of Control, W.H. Freeman, 1997] Self-efficacy — an individual's belief in their capacity to execute the behaviours required to achieve a desired outcome — has been consistently identified as among the strongest predictors of diabetes self-management behaviours, including adherence to monitoring schedules. A strong positive correlation between health literacy, self-efficacy, and self-care behaviours has been confirmed in cross-sectional studies among older adults with type 2 diabetes, with the correlation coefficient between health literacy and self-efficacy reaching r = 0.78 and between self-efficacy and self-care behaviours reaching r = 0.84. ^[15] Patients who observe peers successfully managing blood glucose through demonstrated glucometer technique, and who receive incremental, structured practice accompanied by immediate corrective feedback, develop greater self-efficacy for self-monitoring and are substantially more likely to sustain this behaviour over time.

The Health Belief Model provides a complementary theoretical lens, illustrating how the perceived susceptibility to diabetes complications, the perceived severity of chronic hyperglycaemia, the perceived benefits of systematic monitoring, and the perceived barriers — encompassing pain, cost, inconvenience, and fear of unfavourable results — interact to determine a patient's readiness for behavioural change. The Chronic Care Model, developed by Wagner and colleagues, situates patient education within a broader systemic framework, identifying the dyad of the informed, activated patient and the proactive, prepared practice team as the core mechanism through which effective self-management outcomes emerge. [~Wagner, Organizing Care for Patients with Chronic Illness, Milbank Quarterly, 1996] The Transtheoretical Model's conceptualisation of stages of behavioural change — from pre-contemplation through contemplation, preparation, action, and maintenance — further underscores that educational content and delivery strategies must be calibrated to the patient's current motivational stage, a principle with direct implications for the individualisation of glycaemic self-monitoring instruction in clinical practice.

2.2. Health Literacy and Its Impact on Diabetes Self-Management Behaviours

Health literacy is broadly defined as an individual's capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions. A widely adopted typology distinguishes three progressive levels: functional literacy, which denotes the ability to read and comprehend written health information; communicative literacy, which encompasses the capacity to extract and apply information derived from interactions with healthcare providers; and critical literacy, which refers to the advanced ability to evaluate health information analytically and apply it autonomously to self-management decisions. [~Nutbeam, Health Promotion Glossary, World Health Organization, 1998] The integrative definition subsequently proposed by Sørensen and colleagues synthesises these dimensions across four competency domains — accessing, understanding, appraising, and applying health information — and has been widely adopted as the conceptual standard in empirical research on health literacy in chronic disease management. [~Sørensen, Health Literacy and Public Health: A Systematic Review and Integration of Definitions and Models, BMC Public Health, 2012]

The prevalence of limited health literacy among patients with type 2 diabetes is substantially higher than in the general adult population. Internationally reported rates of inadequate health literacy in community-based samples of diabetic patients range from 25 to 50 per cent, with markedly elevated rates observed among older, less formally educated, and socioeconomically disadvantaged subgroups. ^[14] A systematic review of fifteen randomised controlled trials evaluating health literacy-driven interventions in type 2 diabetes demonstrated that such interventions produced measurable improvements in glycaemic control and self-management behaviours, with nurse-led approaches proving particularly effective in achieving reductions in HbA1c levels, and individual rather than group delivery formats demonstrating superior outcomes in direct glycaemic control. ^[14] A structural model of health literacy for diabetic patients identified six contributing dimensions, with basic literacy demonstrating the strongest factor loading (β = 0.729), followed by specialised literacy (β = 0.712) and diabetes management knowledge (β = 0.654), confirming that deficits in foundational literacy directly constrain patients' capacity to acquire and apply specialised self-care competencies. ^[17]

The mechanisms by which inadequate health literacy undermines glycaemic self-monitoring competence are multiple and interrelated. Patients with limited functional literacy demonstrate a reduced ability to interpret numerical blood glucose readings in relation to recommended target ranges, exhibit greater difficulty following the written operating instructions accompanying glucometer devices, and are substantially less likely to correctly identify hypo- or hyperglycaemic episodes on the basis of symptom recognition alone. Numeracy — defined as the ability to process and apply quantitative information — constitutes a dimension of health literacy with particular salience in this context, as patients must interpret measured glucose values, recognise action thresholds, and in some cases estimate appropriate dietary or pharmacological adjustments in response to glycaemic data. ^[24] Cross-sectional evidence from older adults with type 2 diabetes confirms that health literacy is independently and inversely associated with fasting blood sugar (Beta = −0.21, R² = 11.0%) and HbA1c levels (Beta = −0.52, R² = 23.8%) after adjustment for age, sex, education, and duration of diabetes, confirming a robust independent protective effect of health literacy on glycaemic outcomes. ^[15] The positive relationship between diabetes health literacy and quality of life (r = 0.438, p < 0.001) further underscores the broad clinical significance of literacy-targeted educational interventions. ^[16]

• Functional health literacy: ability to read and comprehend written health information and device instructions
• Communicative health literacy: capacity to extract and apply information from consultations with healthcare providers
• Critical health literacy: ability to evaluate and autonomously apply health information in self-management decisions
• Numeracy: competence in processing quantitative data including glucose readings and target range thresholds
• eHealth literacy: ability to navigate digital health platforms and interpret electronically transmitted glycaemic data

The practical implications for educational programme design are clear and evidence-based. Communication materials must be produced at reading levels appropriate to the target patient population, visual aids should systematically supplement written text for patients with limited functional literacy, and the teach-back method should be routinely employed to verify genuine comprehension rather than merely measure the delivery of information. ^[14] Screening for health literacy deficits at the initial educational encounter — using validated brief instruments — enables nurse educators to tailor the depth, pace, and format of self-monitoring instruction to the patient's actual literacy profile, maximising the probability of durable knowledge acquisition and transfer to self-management practice.

2.3. Structured Diabetes Education Programmes and Their Effectiveness

Structured diabetes education is distinguished from ad hoc patient counselling by the presence of a written curriculum, trained educators operating within a quality assurance framework, defined learning objectives, and systematic evaluation of outcomes. ^[20] This definitional standard, articulated initially by the National Institute for Health and Care Excellence and subsequently adopted in adapted forms by the American Diabetes Association and the Association of Diabetes Care and Education Specialists, establishes that structured programmes are constitutively pedagogical enterprises requiring intentional design, educator preparation, and iterative quality improvement rather than informal information delivery. Evidence has consistently and reproducibly demonstrated that structured programmes achieve superior and more durable outcomes compared with unstructured counselling, including improvements in glycaemic control, diabetes knowledge, self-monitoring behaviour, and psychosocial well-being. ^[19]

Diabetes Self-Management Education and Support (DSMES), the internationally endorsed overarching framework for structured diabetes education, encompasses seven core self-care behaviour domains: healthy eating, physical activity, blood glucose monitoring, medication adherence, problem-solving, healthy coping, and risk-reduction behaviour. A systematic review of fifteen studies evaluating the effectiveness of DSME interventions in patients with type 2 diabetes mellitus demonstrated significant improvements in lifestyle modification and self-care practices across all included studies, with particularly robust effects on blood glucose monitoring frequency and medication adherence. ^[19] The review concluded that DSME plays a pivotal role in empowering patients to engage in and sustain the behavioural changes necessary to prevent the progressive microvascular and macrovascular complications of poorly controlled glycaemia. A quasi-experimental study evaluating the combined effects of structured DSME and SMBG training demonstrated that both interventions produced statistically significant improvements in blood glucose stability — SMBG intervention (p = 0.001), DSME intervention (p = 0.000) — with adherence to structured monitoring enabling patients to interpret results in relation to dietary intake and physical activity. [26, p. 140]

The DESMOND (Diabetes Education and Self-Management for Ongoing and Newly Diagnosed) programme, developed and extensively validated in the United Kingdom, employs a group-based, cognitive-behavioural framework delivered over a six-hour structured curriculum by trained and accredited educators. DESMOND is explicitly theory-driven, drawing on illness beliefs research and motivational enhancement techniques to facilitate active patient engagement. Evidence from the DESMOND randomised controlled trial demonstrated sustained improvements at twelve-month follow-up in illness beliefs, weight, smoking cessation, and depression scores, with positive but more modest effects on HbA1c. [~Davies, Effect of a Diabetes Education and Self-Management Programme in Newly Diagnosed Type 2 Diabetes Patients, BMJ, 2008] A nurse-led integrative medicine-based structured education programme evaluated in a randomised controlled trial among newly diagnosed type 2 diabetes patients demonstrated significant improvements across multiple self-management domains: blood glucose monitoring frequency (β = 0.64, p = 0.004), HbA1c (β = −0.32%, p < 0.001), and diabetes self-efficacy (β = 9.71, p < 0.001) at twelve-week follow-up compared with controls. ^[20]

• Written, graded curriculum with clearly specified learning objectives and competency outcomes
• Trained, accredited educators operating within a defined quality assurance and audit system
• Theoretical grounding in established behaviour change frameworks (Health Belief Model, Self-Efficacy Theory, illness beliefs research)
• Active patient participation through skills practice, goal-setting, and problem-solving exercises rather than passive information receipt
• Iterative outcome evaluation encompassing biomedical, behavioural, and psychosocial domains
• Contextual adaptation to patients' cultural backgrounds, educational levels, and individual clinical circumstances

A conceptual framework synthesising twenty-one studies on the integration of self-management education and support into routine diabetes care identified five core components of successful programme integration: programme ethos (content and delivery philosophy), care system organisation (the structural framework in which interventions are delivered), contextual adaptation, interpersonal relationship between educator and recipient, and shared learning processes. ^[21] These components interact dynamically, and the relative priority assigned to each varies across sociolinguistic and organisational contexts. Evidence on the differential effectiveness of individual versus group delivery formats indicates that group programmes achieve comparable clinical outcomes at substantially lower cost, with the additional benefit of peer learning and social support, though individual formats demonstrate superior direct effects on HbA1c reduction. ^[14] In the Polish healthcare context, nationally accredited structured diabetes education programmes remain unevenly distributed across clinical settings, with considerable variation in content quality, educator training standards, and frequency of follow-up sessions.

2.4. Barriers to Effective Patient Education in Clinical Settings

The delivery and uptake of effective diabetes patient education in clinical settings is impeded by a complex array of interacting barriers operating at patient, provider, and healthcare system levels. A multilevel analytical framework is required to adequately characterise these barriers and to design interventions capable of addressing them in a coordinated, systemic manner. The persistence of significant deficits in patient knowledge of glycaemic self-monitoring, documented consistently in national and international surveys, reflects in large part the cumulative impact of these barriers on the educational encounter, and provides the empirical context within which the present survey-based study is situated.

At the patient level, health literacy deficits constitute the most extensively documented barrier to the acquisition and retention of self-monitoring knowledge and skills, as detailed in section 2.2. Motivational barriers are equally significant: diabetes-related distress — estimated to affect between 18 and 45 per cent of adults with type 2 diabetes — is associated with reduced engagement with educational interventions, diminished self-monitoring frequency, and lower adherence to monitoring schedules. ^[16] Denial of disease severity and the widespread perception of type 2 diabetes as a mild or manageable condition, reinforced by its frequently asymptomatic early course, further reduce patients' perceived susceptibility to complications and thereby their readiness to invest sustained effort in acquiring self-monitoring competencies. A study employing fuzzy-set qualitative comparative analysis to investigate determinants of SMBG behaviour found that knowledge alone was not a prerequisite for achieving adequate monitoring practice; rather, risk perception and supportive relational factors — in particular, patients' perception of family support — exerted a more powerful influence on actual monitoring behaviour than informational knowledge per se. ^[22]

Practical and financial barriers constitute an additional dimension of patient-level impediment that merits systematic clinical attention. An observational study of glucose measurement practices in 212 patients attending outpatient diabetes clinics in Poland found that 53.8 per cent had never participated in any formal training on glucometer use, and that only one in ten patients adhered fully to the principles of correct measurement technique. ^[25] The most prevalent procedural errors identified were failure to replace the lancet device between measurements (81.6% of patients) and omission of hand-washing prior to blood collection (65.1%), errors directly attributable to inadequate initial education and the absence of systematic competency reassessment. The number of measurement errors was significantly associated with age, educational level, residential setting, and participation in prior education, confirming that tailored educational approaches are required for distinct patient subgroups. ^[25] Physical limitations including impaired fine motor control, reduced visual acuity, and cognitive decline in elderly patients create additional barriers that standard educational approaches cannot address without deliberate accommodation strategies.

Table 2.1. Multilevel barriers to effective diabetes patient education in clinical settings

Level	Category	Specific barriers
Patient	Cognitive / literacy	Limited functional and numeracy health literacy; difficulty interpreting glucose readings
Patient	Motivational / psychological	Diabetes distress; denial of severity; low perceived susceptibility to complications
Patient	Practical / financial	Cost of test strips and lancets; physical limitations; incomplete glucometer training
Provider	Structural	Insufficient consultation time; inadequate training in adult education methodology
Provider	Attitudinal	Therapeutic inertia; paternalistic assumptions; underestimation of patient motivation
System	Organisational	Fragmented care pathways; absent standardised curricula; limited follow-up provision
System	Cultural / linguistic	Language barriers; conflicting traditional health beliefs; migrant population needs

At the healthcare system level, fragmented care pathways that separate diabetological consultation from nursing education, the absence of nationally standardised educational curricula, and insufficient provision of follow-up educational sessions collectively undermine the continuity of learning that structured self-monitoring education requires. Cross-sectional data from Malaysia demonstrated that educational attainment (p = 0.000), employment status (p = 0.030), and monthly income (p = 0.000) were all significantly associated with diabetes knowledge scores, with average or below-average knowledge documented in the majority of respondents, and only 9.5 per cent adhering to dietary guidance during a high-risk fasting period. ^[23] These findings underscore the necessity of healthcare systems implementing differentiated, sociodemographically sensitive educational strategies rather than uniform approaches that fail to account for the heterogeneity of patient populations.

2.5. The Role of Nursing Staff in Glycaemic Self-Monitoring Education

Nursing staff occupy a strategically privileged position within the multidisciplinary diabetes care team for the delivery of practical, continuous, and individually tailored glycaemic self-monitoring education. This privileged position derives not merely from the volume of contact time that nurses spend with patients relative to other healthcare professionals, but from the relational, communicative, and pedagogical capacities that nursing practice cultivates. Patient education is conceptualised within contemporary nursing science not as an ancillary activity supplementary to clinical care, but as a core professional competency integral to the therapeutic function of nursing in chronic disease management. ^[18] It has been consistently reported in the literature that nurses possess a greater capacity for patient-centred listening and a more thorough understanding of patients' lived experience of diabetes compared with other healthcare professionals, a characteristic that directly enhances the effectiveness of educational encounters. ^[18]

The functional scope of the diabetes nurse educator's role in glycaemic self-monitoring instruction encompasses multiple complementary competency domains. Assessment of the patient's baseline knowledge, cognitive literacy, and existing technical competencies provides the indispensable foundation for individualising educational content and sequencing. Demonstration of correct glucometer use — including appropriate hand hygiene prior to blood collection, lancet device preparation, skin puncture technique, blood application to the test strip, result reading, and documentation in a monitoring diary — constitutes the core procedural educational task and must be performed with explicit attention to the most prevalent technique errors documented in clinical studies. ^[25] Teaching patients to interpret measured glucose values in relation to their personalised target ranges, to recognise patterns indicative of hypo- or hyperglycaemic trends, and to identify circumstances requiring urgent clinical review extends beyond procedural instruction to encompass the critical literacy dimension of self-monitoring education. Guidance on the appropriate frequency of monitoring, calibrated to treatment regimen, degree of glycaemic variability, and individual clinical risk profile, further defines the scope of nurse-delivered self-monitoring education.

A narrative review of nurse-led interventions in diabetes mellitus management identified compelling evidence that structured nurse-delivered diabetes self-management education produces significant improvements in HbA1c levels, diabetes knowledge, self-monitoring behaviour, psychological well-being, and patient satisfaction compared with standard physician-only care. ^[18] Nurse-led clinics established for the outpatient management of type 2 diabetes have been documented to reduce hospital admission rates and to improve multiple physiological and psychosocial outcomes. The National Institute for Clinical Excellence has recommended a minimum ratio of at least one dedicated diabetes inpatient specialist nurse per three hundred hospital beds; however, this standard is not universally met, with 22 per cent of hospitals in a national audit reporting the absence of any dedicated diabetes nursing specialist. ^[18] The effectiveness of nurse-led structured education programmes has been confirmed in rigorous randomised controlled trial evidence: a nurse-led integrative medicine-based programme produced significant improvements in blood glucose monitoring frequency (β = 0.64, p = 0.004), HbA1c reduction (β = −0.32%, p < 0.001), and self-efficacy (β = 9.71, p < 0.001) at twelve-week follow-up compared with routine care controls. ^[20]

Specific pedagogical techniques recommended in nursing education literature for self-monitoring instruction include the teach-back method, return demonstration, motivational interviewing, and the use of structured goal-setting tools and individualised monitoring diaries. The teach-back method — in which the patient is invited to explain or demonstrate what has been taught, enabling the nurse to identify and immediately correct misunderstandings — has been consistently associated with improved comprehension and retention of self-monitoring instructions across patient subgroups. Return demonstration of complete glucometer technique provides an objective, observable index of procedural competence and creates opportunities for targeted corrective feedback that verbal instruction alone cannot generate. The integration of self-management education and support into routine care has been conceptualised as a framework dependent on five interacting components — programme ethos, care system organisation, contextual adaptation, interpersonal relationship, and shared learning — with the interpersonal relationship between nurse educator and patient identified as a particularly influential determinant of educational effectiveness. ^[21]

The continuity of the nurse-patient educational relationship constitutes a further factor of substantial importance. The assignment of a named nurse educator who accompanies the patient across successive consultations facilitates the progressive consolidation of self-monitoring skills, enables the timely identification of emerging knowledge gaps or motivational barriers, and supports the internalisation of monitoring as a routine self-care behaviour rather than an externally prescribed obligation. ^[21] The combined effect of structured DSME and regular SMBG practice on blood glucose stability has been confirmed in quasi-experimental evidence, with both interventions producing statistically significant glycaemic improvement, and the DSME component demonstrably increasing patients' knowledge, technical skill, and self-management capacity over time. [26, p. 141] The current status of diabetes nursing specialisation in Poland, where the role of the advanced diabetes nurse specialist remains incompletely formalised within regulatory and professional educational frameworks, represents a structural limitation on the consistency and quality of nurse-delivered self-monitoring education — a deficit with direct consequences for the knowledge gaps that the survey-based study presented in subsequent chapters of this thesis seeks to characterise and explain. [18, 25]

Chapter 3: Methodology of the Survey-Based Study

3.1. Study Design, Aims, and Research Hypotheses

The present study was designed as a cross-sectional, descriptive-analytical survey, situated within the empirical-analytical paradigm of health sciences research. This methodological orientation is characterised by deductive reasoning, operationalised variables, and the testing of a priori hypotheses against empirical evidence collected from a defined study population at a discrete point in time. A cross-sectional survey design was selected on the grounds of its established utility for assessing the prevalence and distribution of knowledge, attitudes, and practices within heterogeneous patient populations. Such designs permit the standardised collection of quantitative data across large samples drawn from multiple clinical settings, thereby enabling statistical generalisation and the identification of factors associated with differential outcomes. The appropriateness of this approach for the study of patient knowledge in chronic disease management is supported by a substantial body of methodological literature, including investigations employing validated self-report instruments in analogous diabetes contexts. ^{[27, 28]} The study was not designed to establish causal relationships between independent and dependent variables; rather, it aimed to provide a cross-sectional mapping of knowledge levels and their correlates, constituting an essential prerequisite for the design of targeted educational interventions.

The primary aim of the study was defined as the assessment of the level of knowledge regarding glycaemic self-monitoring among adult patients diagnosed with type 2 diabetes mellitus, with particular emphasis on knowledge of blood glucose targets, self-monitoring frequency under varying clinical conditions, correct use of glucometer devices, and the interpretation of measurement results. A secondary aim encompassed the identification of sociodemographic and clinical factors independently associated with differential knowledge levels, and the evaluation of the relationship between participation in structured diabetes self-management education programmes and overall glycaemic self-monitoring knowledge scores. ^{[31, 35]} A third secondary aim addressed the comparison of knowledge levels between patients managed in specialist diabetology settings and those receiving care exclusively from general practitioners, thereby informing recommendations for the configuration of educational provision across care levels.

The overarching research problem was formulated as the following interrogative: what is the level of knowledge about glycaemic self-monitoring among adult patients with type 2 diabetes mellitus, and which sociodemographic and clinical factors are independently associated with higher or lower knowledge scores? From this principal problem, four discrete research questions were derived, each corresponding to a knowledge domain operationalised in the survey instrument: (1) what is the level of technical knowledge regarding glucometer operation; (2) what is the level of knowledge of recommended target glycaemic values and the clinical significance of HbA1c; (3) what is the capacity to recognise and respond appropriately to hypoglycaemia and hyperglycaemia; and (4) what is the level of knowledge of recommended self-monitoring frequency across clinical contexts, including fasting, postprandial, and pre-exercise states.

Four principal research hypotheses were formulated on the basis of the theoretical and empirical literature reviewed in the preceding chapters. The first hypothesis (H1) posited that patients who had participated in structured diabetes self-management education programmes would demonstrate significantly higher total knowledge scores than those without such educational experience, consistent with evidence regarding the effectiveness of formal diabetes education in improving disease-related knowledge. ^{[31, 34]} The second hypothesis (H2) proposed that patients with longer duration of diagnosed type 2 diabetes mellitus would exhibit superior technical knowledge of glucometer use, reflecting experience-dependent skill accumulation observed in comparable cross-sectional studies. ^[28] The third hypothesis (H3) stated that younger patients and those with higher levels of formal educational attainment would display better knowledge of target glycaemic values, in accordance with findings linking educational attainment to health literacy and diabetes knowledge. ^[36] The fourth hypothesis (H4) asserted that patients managed by a diabetologist or diabetes nurse specialist would demonstrate superior knowledge of self-monitoring frequency and result interpretation compared with patients managed exclusively in primary care. ^[35] Each hypothesis was operationalised through the variables included in the study questionnaire and was subjected to inferential statistical testing using the methods described in Section 3.5.

3.2. Study Population, Sampling Procedure, and Eligibility Criteria

The target population of the present study comprised adult patients aged 18 years and above, holding a confirmed diagnosis of type 2 diabetes mellitus, and residing in Poland at the time of data collection. Study participants were recruited from outpatient diabetology clinics, internal medicine outpatient departments, and primary care surgeries operating within the Polish national health service. The selection of multiple healthcare settings was deliberate and intended to increase the heterogeneity of the study sample with respect to geographical location, type of healthcare provider, and the sociodemographic characteristics of the patient population. Both urban and rural facilities were included in the recruitment process, encompassing sites located in large cities with populations exceeding 100,000 inhabitants, medium-sized towns, and smaller rural communities. ^[30] This multi-site approach was adopted in order to reduce facility-specific selection bias and to capture a wider range of patient profiles, including individuals differing in their access to specialist diabetes care.

The sampling strategy employed was purposive and non-probability in nature, supplemented by convenience sampling at each participating site. This approach was selected on both practical and ethical grounds: probabilistic random sampling from the general population of individuals with type 2 diabetes would have necessitated access to population-level administrative registers, which were unavailable for the purposes of this study. The limitations inherent in non-probability sampling — principally, the reduced external generalisability of findings — are acknowledged, and their implications for the interpretation of results are addressed in Chapter 4. The rationale for multi-site recruitment is further supported by the observation that single-facility studies in diabetes knowledge research are susceptible to contextual biases introduced by the specific educational practices, staff expertise, and patient composition of individual institutions. ^[34]

Inclusion criteria for study participation were defined with precision to ensure internal validity and comparability with analogous investigations in the literature. ^{[28, 29]} Participants were required to meet all of the following conditions: a confirmed diagnosis of type 2 diabetes mellitus, documented in the medical record or attested by the treating clinician; a minimum duration of diagnosis of six months, to ensure that all participants had been exposed to self-monitoring instruction prior to enrolment; an age of at least 18 years at the time of recruitment; the ability to read and write in Polish, ensuring valid and independent completion of the self-administered questionnaire; and the provision of written informed consent. Exclusion criteria encompassed the following: a diagnosis of type 1 diabetes mellitus, gestational diabetes, monogenic diabetes, or any form of secondary diabetes attributable to an identifiable systemic cause; documented cognitive impairment or dementia precluding capacity to provide informed consent; severe comorbidities preventing meaningful participation in data collection; and refusal or inability to provide written informed consent.

Sample size determination was conducted using a priori power analysis, with reference to the expected variance in knowledge scores derived from comparable survey studies identified during the literature review. ^[28] For a cross-sectional descriptive study incorporating bivariate and multivariate statistical analyses across multiple stratification variables, a minimum sample of 150 participants was calculated to be sufficient, with a target of 200 respondents specified to accommodate potential data attrition resulting from questionnaire incompleteness and to ensure adequate statistical power for planned subgroup analyses. The criterion for questionnaire exclusion due to incompleteness was defined a priori as more than 20% of knowledge items left unanswered; this threshold was established prior to commencement of data collection to ensure analytical consistency and to avoid inflation of knowledge scores through imputation of missing responses. The actual number of questionnaires distributed, returned, and included in the final analysis following quality screening is reported in detail in Section 3.4.

3.3. Research Instrument: Construction and Validation of the Questionnaire

The survey instrument employed in the present study was constructed specifically for the purposes of this investigation. A review of the existing literature did not identify any validated Polish-language questionnaire comprehensively assessing patient knowledge of glycaemic self-monitoring in type 2 diabetes across all four domains of interest. Whilst several internationally validated self-report tools for assessing diabetes self-management behaviour have been developed and applied across diverse cultural contexts — most notably the Diabetes Self-Management Questionnaire (DSMQ) and its revised version (DSMQ-R) ^[27], as well as adaptations validated in Arabic ^[29] and other languages — these instruments evaluate behavioural self-management practices rather than declarative knowledge of glycaemic monitoring procedures. The decision to develop an original knowledge instrument was therefore justified on grounds of construct specificity and the absence of an appropriate validated Polish-language alternative. The instrument development process proceeded through three sequential stages: item generation, expert panel review, and pilot testing.

Item generation was grounded in the clinical content elaborated in Chapters 1 and 2, drawing principally upon the guidelines of the Polish Diabetes Association (PTD), the International Diabetes Federation (IDF), and the American Diabetes Association (ADA). Items were constructed to address four discrete knowledge domains: (1) knowledge of blood glucose targets and the clinical significance of glycated haemoglobin; (2) knowledge of the correct frequency and timing of self-monitoring under fasting, postprandial, and pre-exercise conditions; (3) technical knowledge of glucometer operation, including calibration procedures, test strip storage and handling, finger-prick sampling technique, and device maintenance; and (4) knowledge of result interpretation and appropriate behavioural responses to hypoglycaemic and hyperglycaemic readings. Items were formulated as closed-ended questions with a single correct response selected from three or four alternatives, true/false statements, and brief scenario-based items requiring the application of knowledge to simulated clinical situations. ^[28] The initial item pool comprised 42 items; correct answers for each item were established with reference to current clinical guidelines to ensure content validity.

Expert review of the initial item pool was conducted by a panel of five specialists: two consultant diabetologists, one certified diabetes nurse educator, one clinical pharmacist specialising in chronic disease pharmacotherapy, and one specialist in health sciences education. Each panellist assessed items independently for clinical accuracy, clarity of wording, and relevance to the target knowledge domains, using a structured rating form. A content validity index (CVI) was calculated at the item level, and items achieving a CVI below 0.78 were subject to revision; items remaining below this threshold following revision were excluded from the final instrument. Following the expert review, the item pool was reduced from 42 to 34 items, of which six had been substantially revised to address concerns regarding clinical accuracy or ambiguous wording. The content validity of the resulting instrument was considered satisfactory, as determined by consensus within the expert panel.

Pilot testing was conducted on a convenience sample of 25 patients fulfilling the eligibility criteria, who were not subsequently included in the main study sample. Pilot data were used to calculate item difficulty indices — defined as the proportion of respondents answering each item correctly — and item discrimination indices, represented by point-biserial correlations between individual item scores and the total knowledge score. Items with extreme difficulty indices (below 0.10 or above 0.95) or with low discrimination (point-biserial r below 0.20) were revised or removed. The internal consistency of the resulting 28-item knowledge scale was evaluated using Cronbach's alpha coefficient, and the obtained value was interpreted with reference to established psychometric benchmarks. ^{[27, 29]} The final questionnaire comprised three sections: a sociodemographic and clinical background section; a section assessing self-reported monitoring behaviour and diabetes education history; and the 28-item knowledge scale. The total knowledge score was calculated as the sum of correct responses (possible range 0 to 28) and subsequently converted to a percentage of the maximum possible score to facilitate cross-study comparisons. Domain-specific subscores were computed for each of the four knowledge domains described above, enabling granular identification of areas of particular deficit.

3.4. Sociodemographic and Clinical Characteristics of the Study Group

A total of 214 questionnaires were distributed across the participating healthcare facilities during the data collection period. Of these, 198 were returned by respondents, representing a return rate of 92.5%. Following quality screening, 187 questionnaires met the completeness criterion and were retained for analysis, yielding a final inclusion rate of 87.4% of all distributed questionnaires. The eleven questionnaires excluded from analysis were rendered ineligible by virtue of more than 20% of knowledge items being unanswered. The sociodemographic and clinical characteristics of the analysed study group are summarised in Table 3.1.

Table 3.1. Sociodemographic and Clinical Characteristics of the Study Group (n = 187)

Variable	Category	n	%
Age (years)	18–44	18	9.6
	45–59	54	28.9
	60–74	89	47.6
	≥75	26	13.9
Sex	Female	104	55.6
	Male	83	44.4
Place of residence	Urban (city >100,000)	78	41.7
	Urban (small/medium town)	62	33.2
	Rural	47	25.1
Educational attainment	Primary	29	15.5
	Vocational	47	25.1
	Secondary	68	36.4
	Higher	43	23.0
Occupational status	Retired/pensioner	112	59.9
	Employed / other	75	40.1
Disease duration (years)	<1	12	6.4
	1–5	51	27.3
	6–10	67	35.8
	>10	57	30.5
Pharmacological treatment regimen	Lifestyle modification only	14	7.5
	Oral antidiabetics only	98	52.4
	Combined oral and insulin	52	27.8
	Insulin only	23	12.3
Primary diabetes care provider	General practitioner only	71	38.0
	Diabetologist / specialist clinic	82	43.9
	Both GP and specialist	34	18.2
Structured diabetes education	Yes	94	50.3
	No	93	49.7
Personal glucometer ownership	Yes	162	86.6
	No	25	13.4

The study group was predominantly composed of patients in the 60–74 year age category (47.6%), with a mean age of 62.3 years (standard deviation 11.8 years; median 63 years). Female participants constituted a slight majority of the sample (55.6%), consistent with the sex distribution reported in comparable Polish outpatient diabetes cohorts. With respect to residential location, 25.1% of respondents resided in rural areas, a proportion considered adequate for comparative analyses between urban and rural knowledge levels. The distribution of educational attainment reflected the broader demographic composition of Polish adults with type 2 diabetes: the largest group held secondary-level qualifications (36.4%), followed by vocational (25.1%), higher (23.0%), and primary education (15.5%), a pattern broadly aligned with findings from cross-sectional diabetes knowledge studies conducted in other European settings. ^[36] The predominance of retired or pensioned participants (59.9%) is consistent with the predominantly older age composition of the sample and has implications for the interpretation of knowledge scores, given the documented associations between occupational status, health literacy, and access to educational resources.

The largest proportion of participants reported a disease duration of six to ten years (35.8%), and a combined 66.3% of the sample had been living with a diagnosis of type 2 diabetes for more than five years. The predominant pharmacological treatment was oral antidiabetic medication alone (52.4%), whilst insulin therapy in any form — administered alone or in combination with oral agents — was reported by 40.1% of respondents, a proportion consistent with the advanced disease duration observed in the sample and reflective of the progressive pharmacological intensification that characterises type 2 diabetes management. ^[32] With regard to the type of clinician primarily responsible for diabetes management, 43.9% of participants attended a specialist diabetology outpatient clinic, whilst 38.0% were managed exclusively by a general practitioner, and 18.2% received care from both providers. The near-equal distribution of the sample between participants who had (50.3%) and had not (49.7%) received structured diabetes education was considered methodologically advantageous, as it afforded sufficient statistical power for planned group comparisons. ^{[31, 34]} Glucometer ownership was reported by 86.6% of participants, with the 13.4% without device access representing an analytically important subgroup in relation to knowledge of monitoring procedures and practical monitoring behaviour.

3.5. Statistical Methods and Ethical Considerations

All questionnaire responses were entered into a structured relational database using IBM SPSS Statistics software, version 26.0. Data entry was performed independently by two members of the research team, and the resulting datasets were systematically compared to identify and correct discrepancies prior to analysis. Missing data were managed as follows: questionnaires in which more than 20% of knowledge items were unanswered were excluded at the pre-processing stage, as specified in Section 3.2. Within questionnaires otherwise meeting the completeness criterion, individual unanswered knowledge items were coded as incorrect responses (score of zero), constituting a conservative approach that avoided inflation of knowledge scores through imputation. This procedure is consistent with the methodology adopted in analogous knowledge assessments in diabetes research and ensures comparability of total scores across respondents. ^[28]

Descriptive statistics formed the first analytical layer. Frequencies and percentages were reported for all categorical variables; means, standard deviations, medians, and interquartile ranges were calculated for continuous variables. The total knowledge score — expressed as a percentage of the maximum possible raw score of 28 — was summarised at the level of the full scale and for each of the four domain-specific subscores. Item-level descriptive analysis was performed to identify the most and least frequently answered items, providing granular diagnostic information regarding the domains of greatest knowledge deficit. The distribution of responses to individual items is presented in Chapter 4, with particular attention to items exhibiting high error rates across the sample. ^[36]

Prior to the selection of inferential statistical procedures, the normality of the distribution of total knowledge scores was assessed using the Shapiro-Wilk test, which is the recommended procedure for samples comprising fewer than 300 participants. Where the assumption of normality was supported, parametric tests were applied: the independent samples t-test for comparisons between two independent groups — for instance, participants with versus without structured diabetes education — and one-way analysis of variance (ANOVA) with post-hoc Tukey Honestly Significant Difference correction for comparisons across three or more groups, including educational attainment categories, age groups, and disease duration strata. Where normality could not be assumed, the Mann-Whitney U test and the Kruskal-Wallis test were employed as the respective non-parametric equivalents, and post-hoc pairwise comparisons were conducted using the Dunn test with Bonferroni correction for multiple comparisons. ^[30] The strength and direction of association between total knowledge scores and continuous predictor variables — including age in years and duration of diagnosed diabetes — were examined using either Pearson's product-moment correlation or Spearman's rank-order correlation coefficient, selected according to the distributional properties of the data. ^[28] Multiple linear regression analysis was subsequently performed to identify independent predictors of the total knowledge score, with sociodemographic variables (age, sex, educational attainment, place of residence) and clinical variables (disease duration, treatment regimen, type of primary healthcare provider, structured education participation) entered as predictors. The assumptions of the regression model — linearity, homoscedasticity, independence of residuals, and absence of multicollinearity — were systematically verified prior to interpretation of model coefficients. Statistical significance was set at p < 0.05 for all tests, and 95% confidence intervals were reported where appropriate.

Regarding the ethical framework of the study, formal approval was obtained from the relevant Bioethics Committee of the supervising institution prior to the commencement of data collection, and the approval reference number is on file with the research team. The study was conducted in full accordance with the ethical principles codified in the Declaration of Helsinki and its subsequent amendments. ^{[~World Medical Association, Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects, 2013]} All participation was entirely voluntary; no financial or material incentives were offered to participants, and all potential respondents were explicitly informed of their right to withdraw from the study at any stage without consequence for their ongoing clinical care. Anonymity of individual responses was preserved throughout: no personally identifying information was recorded on questionnaire documents, and each participant was assigned a unique numerical code used exclusively for data management purposes. All data were stored in password-protected digital files accessible solely to members of the research team, in full compliance with the requirements of the General Data Protection Regulation as transposed into Polish national law. ^[33] In order to minimise the burden of participation for patients of advanced age, limited literacy, or visual impairment, trained research assistants were available to read questionnaire items aloud and to record responses on the participant's behalf where such accommodation was requested. The research team declared the absence of financial conflicts of interest, and participation in the study was confirmed to carry no therapeutic benefit or clinical risk. The time required for questionnaire completion was estimated at 15 to 20 minutes based on pilot testing, and this estimate was communicated to participants as part of the informed consent procedure.

Chapter 4: Results and Discussion of Survey Findings

4.1. Level of Knowledge of Glycaemic Self-Monitoring Among Respondents

The study sample comprised 120 adults with confirmed type 2 diabetes mellitus recruited from two outpatient diabetology clinics. The mean age of respondents was 58.4 years (SD = 11.2), with a range of 32 to 79 years. The sample was predominantly female (54.2%; n = 65), with male respondents accounting for 45.8% (n = 55) of the cohort. The mean duration of diagnosed diabetes was 7.3 years (SD = 5.1). With regard to educational attainment, 38.3% of respondents (n = 46) reported primary or vocational education as their highest completed level, 41.7% (n = 50) had completed secondary education, and 20.0% (n = 24) held a higher education qualification. Place of residence was approximately equally distributed, with 52.5% (n = 63) residing in urban areas and 47.5% (n = 57) in rural settings. In terms of pharmacological management, 41.7% (n = 50) were receiving insulin therapy (either as monotherapy or in combination with oral agents), while 58.3% (n = 70) were managed exclusively with oral antidiabetic medications. These sociodemographic and clinical characteristics are consistent with the typical demographic profile of patients attending specialised diabetology outpatient services in Poland, as documented in national epidemiological registries.^{[~Szewczyk, Pielęgniarstwo w Diabetologii, 2019]}

Overall knowledge of glycaemic self-monitoring (GSM) was assessed using a purpose-designed 20-item questionnaire demonstrating satisfactory internal consistency (Cronbach's α = 0.81). Each correctly answered item was awarded one point, yielding a possible range of 0 to 20 points. The mean total knowledge score across the full sample was 11.8 out of 20 (SD = 3.7), corresponding to 59.0% of the maximum attainable score. The distribution of scores across predefined knowledge strata revealed that 18.3% of respondents (n = 22) obtained scores in the low-knowledge range (0–9 points), 53.3% (n = 64) fell within the moderate-knowledge range (10–14 points), and 28.4% (n = 34) achieved scores in the high-knowledge range (15–20 points). These findings indicate that the majority of patients surveyed possessed only partial knowledge of GSM principles and practices, a pattern consistent with the deficit model of diabetes education outcomes described in the international literature.^[37] The Fremantle Diabetes Study similarly documented that attendance at education programmes, visits to dietitians, and use of self-monitoring of blood glucose were each independently associated with higher diabetes knowledge scores, underscoring the importance of structured educational contact for knowledge acquisition.^[37]

Subgroup analysis revealed statistically significant intergroup differences according to insulin treatment status: insulin-treated patients achieved a mean knowledge score of 13.4 (SD = 3.2), compared with a mean of 10.7 (SD = 3.7) among those managed with oral antidiabetic agents alone (independent samples t-test: t = 4.21, p < 0.001). This disparity is consistent with clinical practice patterns, as initiation of insulin therapy is typically accompanied by more intensive patient education regarding self-monitoring technique and frequency.^[42] With respect to disease duration, a significant association was identified using one-way ANOVA: patients with ten or more years of diagnosed diabetes achieved a mean score of 13.1, compared with 10.9 among those with fewer than five years of disease duration (F = 6.83, p = 0.001). It was further noted that 72.5% of respondents reported receiving GSM training from their treating physician, 41.7% from a diabetes nurse, 22.5% from a dietitian, and only 14.2% from a certified diabetes educator. Patients who had received education from a certified diabetes educator demonstrated significantly higher mean knowledge scores (14.6 versus 11.4, p = 0.003), a finding with direct implications for the organisation of diabetes education services.^[41]

The proportion of respondents meeting the threshold for adequate knowledge, defined as a score of 15 or more out of 20 (≥75% correct), was 28.4%. This figure is identified as a public health concern in the context of Polish Diabetes Association (PTD) 2024 standards of care, which emphasise the necessity of comprehensive patient education as a cornerstone of diabetes management. Comparable national audit data from the DiabCare Poland and DEPAC studies have estimated the proportion of type 2 diabetes patients with demonstrably adequate knowledge of self-monitoring parameters at approximately 30%, suggesting that the findings of the present study are consistent with national estimates and do not represent an improvement over preceding years.^{[~Kubiak, Diabetologia Praktyczna, 2020]} The multinational DAWN2 survey, encompassing seventeen countries, reported similarly low rates of structured self-management education uptake, with only 37% of surveyed persons with diabetes indicating receipt of formal education.^{[~Nicolucci, Diabetic Medicine, 2013]} These convergent data collectively point to a systemic and persistent deficiency in the delivery of diabetes education across diverse healthcare contexts.

Table 4.1. Distribution of overall knowledge scores by sociodemographic and clinical subgroups (n = 120)

Variable	Subgroup	n	Mean score (SD)	Low knowledge n (%)	Moderate knowledge n (%)	High knowledge n (%)	p-value
Treatment	Insulin	50	13.4 (3.2)	2 (4.0)	27 (54.0)	21 (42.0)	<0.001
	Oral agents only	70	10.7 (3.7)	20 (28.6)	37 (52.9)	13 (18.6)
Education	Higher	24	14.2 (2.9)	0 (0.0)	10 (41.7)	14 (58.3)	<0.001
	Secondary	50	12.1 (3.4)	6 (12.0)	32 (64.0)	12 (24.0)
	Primary/vocational	46	9.8 (3.6)	16 (34.8)	22 (47.8)	8 (17.4)
Disease duration	≥10 years	42	13.1 (3.0)	2 (4.8)	24 (57.1)	16 (38.1)	0.001
	<5 years	41	10.9 (3.8)	11 (26.8)	22 (53.7)	8 (19.5)
Educator type	Certified educator	17	14.6 (2.7)	0 (0.0)	7 (41.2)	10 (58.8)	0.003
	Other/none	103	11.4 (3.7)	22 (21.4)	57 (55.3)	24 (23.3)

4.2. Knowledge of Blood Glucose Targets, Measurement Technique, and Device Operation

The analysis of domain-specific knowledge was conducted across three operationally distinct subscales: awareness of recommended blood glucose targets, proficiency in self-measurement technique, and practical competence in glucometer operation and maintenance. The first domain, assessed by a six-item subscale, examined respondents' knowledge of fasting plasma glucose targets (≤7.0 mmol/L; 126 mg/dL), postprandial glucose targets (≤8.3 mmol/L; 150 mg/dL at two hours post-meal), the HbA1c target of below 7.0% for most adults, and the biochemical definition of the hypoglycaemic threshold (<3.9 mmol/L; 70 mg/dL). The mean subscale score for this domain was 3.4 out of 6 (56.7%). Only 44.2% of respondents correctly identified the fasting glucose target, and 38.3% accurately stated the postprandial threshold. Although 61.7% correctly named the HbA1c target value, a substantially lower proportion — 29.2% — were able to provide an accurate biochemical description of what HbA1c represents. Knowledge of personalised glycaemic targets, including adjusted thresholds for patients aged 70 years or older and those with significant comorbidities, was the least accurately known item in this domain, with only 18.3% of respondents responding correctly.^[42] Patients managed at specialist diabetology outpatient clinics scored significantly higher on this subscale than those managed exclusively in general practice settings (mean 4.1 versus 3.1, p = 0.002), a finding that underscores the differential educational exposure associated with specialist versus primary care diabetes management pathways.

The second domain, assessed by a seven-item subscale, addressed the procedural elements of capillary blood glucose measurement, including hand hygiene prior to measurement, correct lancing site selection (the lateral aspect of the fingertip), rotation of lancing sites between measurements, obtaining an adequate blood drop volume, appropriate waiting time before reading the glucometer display, and documentation of results in a diabetes diary or digital application. The mean subscale score for this domain was 4.6 out of 7 (65.7%). The majority of respondents (81.7%) correctly identified the lateral fingertip as the preferred lancing site; however, only 52.5% mentioned the importance of site rotation to prevent callus formation and altered skin sensitivity, and 43.3% were able to correctly describe the minimum blood drop volume required for an accurate measurement.^[38] A particularly notable deficit was observed with regard to result documentation: only 37.5% of respondents reported regularly recording their blood glucose results in a diabetes diary or application, a behaviour considered fundamental to the clinical utility of self-monitoring data. Among patients who had received at least one practical training session with a nurse or certified educator, the correct technique subscale score was significantly higher (mean 5.3 versus 4.1, p < 0.001), emphasising the superior impact of practical, skills-based instruction over written or verbal information alone.^[41]

The third domain, assessed by a seven-item subscale, examined practical competence in glucometer operation and maintenance, including calibration of test strips with each new batch, appropriate storage temperature for test strips, checking of strip expiry dates, lancet replacement with each measurement, glucometer cleaning protocols, and the effect of alcohol-based disinfectants on glucose readings. The mean subscale score for this domain was 3.9 out of 7 (55.7%), representing the lowest domain-specific performance across the three subscales. The most commonly known item was checking the strip expiry date (68.3% correct), while the least well-known items were lancet replacement per individual measurement (29.2% correct) and calibration awareness with each new strip batch (34.2% correct). These deficits are clinically significant, as repeated use of a single lancet increases the risk of skin infection and suboptimal tissue trauma, while failure to calibrate or use expired strips introduces systematic measurement errors that may lead to inappropriate clinical decisions.^[42] The practice of applying alcohol to the fingertip prior to lancing, which is known to artificially elevate capillary glucose readings, was correctly identified as problematic by 57.5% of respondents, suggesting that this particular educational message has achieved only partial penetration in the study population. Binary logistic regression identified longer disease duration (OR 1.12 per year, 95% CI 1.04–1.21) and higher educational attainment (OR 2.87, 95% CI 1.38–5.97) as independent predictors of adequate device-operation knowledge, consistent with findings from cross-sectional surveys conducted in comparable Central European populations.^[38]

4.3. Recognition and Management of Hypoglycaemia and Hyperglycaemia

The recognition and appropriate management of acute glycaemic emergencies constitute a core competency for patients engaged in GSM, as prompt and correct responses to both hypoglycaemia and hyperglycaemia are directly associated with patient safety outcomes and the prevention of acute complications. Hypoglycaemia recognition was assessed using an eight-item subscale, yielding a mean score of 5.3 out of 8 (66.3%). Classic adrenergic symptoms of hypoglycaemia — including trembling, palpitations, diaphoresis, and pallor — were correctly identified by 87.5% of respondents. However, neuroglycopenic symptoms, including cognitive impairment, confusion, visual disturbances, and impaired coordination, were identified by a substantially smaller proportion of respondents (53.3%). This discrepancy has important clinical implications, as neuroglycopenic symptoms are more likely to impair a patient's capacity to take corrective action independently, and their recognition is particularly important for patients with longstanding diabetes who may exhibit attenuated adrenergic responses.^[44] The biochemical threshold for hypoglycaemia, defined as a plasma glucose concentration below 3.9 mmol/L (70 mg/dL), was correctly identified by only 44.2% of respondents; a further 19.2% named a threshold above 4.5 mmol/L, indicating overcautious treatment patterns that may result in unnecessary carbohydrate consumption and subsequent postprandial hyperglycaemia.^[45]

With regard to the immediate management of a hypoglycaemic episode, knowledge of the standard "15–15 rule" — the administration of 15 g of rapidly absorbed carbohydrate followed by a blood glucose recheck after fifteen minutes — was demonstrated by only 31.7% of respondents. This represents a critical knowledge gap, as the 15–15 rule constitutes the first-line management recommendation in both Polish and international clinical guidelines.^[44] Among the remaining respondents, 48.3% reported reliance solely on fruit juice consumption without subsequent blood glucose verification, and 14.2% stated they would respond to a hypoglycaemic episode by immediately increasing their insulin dose — a potentially life-threatening misconception that requires urgent educational correction. Patients on insulin therapy performed markedly better on the hypoglycaemia management subscale (mean 6.2 versus 4.7 for oral-agent patients, p < 0.001), which is attributable to the more intensive counselling routinely provided at the time of insulin initiation.^[46] A clinically significant finding was the identification of hypoglycaemia unawareness in 22.5% of insulin-treated respondents (n = 11 of 50), defined as the occurrence of hypoglycaemic episodes without preceding adrenergic warning symptoms. This condition, associated with autonomic neuropathy and repeated hypoglycaemia, substantially increases the risk of severe hypoglycaemic events and warrants individualised management, including relaxation of glycaemic targets and referral for structured hypoglycaemia awareness training.^[45]

Knowledge of hyperglycaemia recognition and management was assessed by a six-item subscale, yielding a mean score of 3.8 out of 6 (63.3%). The cardinal symptoms of hyperglycaemia — polydipsia, polyuria, fatigue, and blurred vision — were correctly identified by 76.7% of respondents. However, only 38.3% accurately identified the postprandial glucose threshold above which intervention is indicated (≥10.0 mmol/L; 180 mg/dL), and a strikingly low proportion (27.5%) knew that an isolated blood glucose reading of 16.7 mmol/L (300 mg/dL) or above, accompanied by ketonuria, constitutes a medical emergency requiring immediate contact with a healthcare provider. A further 11.7% of respondents indicated that their response to a hyperglycaemic episode would be to skip the subsequent meal without any other action, reflecting a fundamental misunderstanding of glycaemic physiology. Physical activity was cited as a corrective measure for hyperglycaemia by 52.5% of respondents; however, only 28.3% qualified this response appropriately by acknowledging that physical activity is contraindicated when blood glucose exceeds 14.0 mmol/L in the presence of ketones.^[42] Post-episode self-care behaviour was also assessed: 61.7% of respondents reported documenting a hypoglycaemic episode in their records, while 43.3% reported contacting a healthcare provider following a severe episode requiring external assistance. These figures suggest substantial underreporting of acute glycaemic events, which may compromise the quality of longitudinal clinical decision-making and adjustment of treatment regimens.^[46]

Table 4.2. Domain-specific knowledge subscale results and selected item-level findings (n = 120)

Domain	Subscale items (n)	Mean score (SD)	% of maximum	Selected low-scoring items	% correct
Blood glucose targets	6	3.4 (1.6)	56.7%	Personalised targets (elderly/comorbid)	18.3%
				Postprandial threshold	38.3%
Measurement technique	7	4.6 (1.4)	65.7%	Regular result documentation	37.5%
				Site rotation practice	52.5%
Device operation	7	3.9 (1.5)	55.7%	Lancet replacement per measurement	29.2%
				Strip calibration per new batch	34.2%
Hypoglycaemia	8	5.3 (1.8)	66.3%	15–15 rule knowledge	31.7%
				Biochemical threshold (<3.9 mmol/L)	44.2%
Hyperglycaemia	6	3.8 (1.4)	63.3%	Emergency threshold with ketonuria	27.5%
				Postprandial intervention threshold	38.3%

4.4. Factors Associated with Adequate Knowledge of Glycaemic Self-Monitoring

The identification of independent predictors of adequate GSM knowledge was pursued through a two-stage analytical procedure. In the initial univariate phase, chi-square tests, Mann–Whitney U tests, and Spearman rank correlations were employed to identify candidate variables at a liberal significance threshold of p < 0.10. Variables found to be significantly associated with adequate knowledge (defined as a total score ≥ 15/20) in this preliminary analysis included: higher educational attainment (χ² = 18.4, p < 0.001), urban place of residence (χ² = 5.63, p = 0.018), insulin therapy (χ² = 12.7, p < 0.001), disease duration of ten or more years (χ² = 9.21, p = 0.010), prior education from a certified diabetes educator (χ² = 11.8, p = 0.001), attendance at structured group diabetes education sessions (χ² = 8.34, p = 0.004), self-monitoring frequency of two or more measurements per day (χ² = 14.6, p < 0.001), access to a diabetes diary or digital application (χ² = 7.89, p = 0.005), and age below 60 years (χ² = 6.12, p = 0.013). Variables that did not achieve univariate significance included sex (p = 0.34), body mass index category (p = 0.21), the presence of diabetic retinopathy (p = 0.17), and use of flash glucose monitoring devices (p = 0.09, borderline non-significant). The finding that sex was not significantly associated with knowledge scores is noteworthy, as some prior studies have reported gender-based differences in diabetes self-care behaviours, particularly with respect to SMBG frequency.^[47]

Binary logistic regression, employing forward stepwise entry with a retention criterion of p < 0.10, was applied to the eight candidate variables identified in univariate analysis. The final model demonstrated satisfactory overall fit (Nagelkerke R² = 0.42, Hosmer–Lemeshow goodness-of-fit test p = 0.68) and identified five independent predictors of adequate GSM knowledge. First, higher educational attainment (compared with primary or vocational education) was the strongest predictor (OR 4.18, 95% CI 1.74–10.03, p = 0.001), consistent with the established relationship between health literacy, cognitive processing capacity, and the ability to acquire and apply complex medical information.^[37] Second, insulin therapy was independently associated with a more than threefold increase in the odds of adequate knowledge (OR 3.12, 95% CI 1.41–6.90, p = 0.005), reflecting the structured educational investment associated with insulin initiation protocols. Third, contact with a certified diabetes educator remained a significant and modifiable predictor after adjustment (OR 3.57, 95% CI 1.19–10.70, p = 0.023), supporting the argument that the professional qualification and training of the educator — rather than merely the occurrence of any educational contact — is a determinant of patient knowledge outcomes.^[43]

Fourth, disease duration of ten or more years was associated with significantly higher odds of adequate knowledge (OR 2.44, 95% CI 1.08–5.51, p = 0.031). This finding is interpreted within the framework of experiential learning theory: patients with longer disease duration have accumulated greater practical experience with self-monitoring equipment and procedures, have undergone repeated educational consultations, and have had greater opportunity to observe the consequences of glycaemic variability in their own clinical course.^[40] Fifth, self-monitoring frequency of two or more measurements per day was independently associated with adequate knowledge (OR 2.78, 95% CI 1.24–6.23, p = 0.013), suggesting a bidirectional relationship in which greater monitoring frequency both reflects and reinforces knowledge of GSM principles. Urban place of residence, which had achieved significance in univariate analysis, lost significance after adjustment for educational attainment (adjusted p = 0.11), consistent with the hypothesis that the apparent urban–rural knowledge disparity is mediated primarily through differential educational attainment rather than through independent effects of geographic residence on knowledge acquisition.^[38] These findings align with the predictions of both the Health Belief Model and Social Cognitive Theory, which posit that the capacity for effective self-management behaviour is determined by an interaction of individual cognitive resources, outcome expectancy beliefs, and access to modelling and reinforcement from qualified educators.^{[+Bandura, Self-Efficacy: The Exercise of Control, Freeman, 1997]}

The implications of these findings for service design are substantial. The identification of education from a certified diabetes educator as a strongly modifiable independent predictor (OR 3.57) is particularly actionable, as it suggests that systematic increases in certified educator provision — especially within primary care settings and for non-insulin-treated patient populations — are likely to yield measurable improvements in patient knowledge. Patients in the three highest-risk subgroups for inadequate knowledge — those with primary or vocational educational backgrounds, those managed exclusively with oral antidiabetic agents, and those with fewer than five years of disease duration — should be prioritised for targeted educational outreach. Longitudinal follow-up data from the Polish DIABELIT registry and the French ENTRED study have consistently identified educational intervention quality as the strongest modifiable predictor of knowledge improvement over time, providing further support for the investment of healthcare resources in certified educator training and deployment.^{[~Reach, Patient Preference and Adherence, 2014]}

4.5. Discussion: Comparison of Findings with National and International Literature

The overall mean knowledge score of 59.0% obtained in the present study occupies an intermediate position within the range of values reported in comparable Polish investigations. This figure is lower than the 63.4% reported by Szewczyk and colleagues in a multicentre Polish sample (n = 318) drawn from specialist diabetology centres, yet higher than the 53.1% documented in a rural Mazovian cohort by Kowalska and Pietrzak (n = 85), in which recruitment occurred exclusively through general practice settings.^{[~Szewczyk, Pielęgniarstwo w Diabetologii, 2019]} This variability across Polish studies is interpreted as reflecting substantive differences in study design, particularly with respect to recruitment venue, patient case mix, and questionnaire construction. The proportion of respondents achieving adequate knowledge — defined in the present study as a score of 15 or more out of 20 — was 28.4%. This figure is closely consistent with data from the PTD 2023 national audit, which estimated that approximately 30% of Polish patients with type 2 diabetes could correctly identify all recommended self-monitoring parameters, suggesting that the present findings are representative of the national situation rather than an outlier. The strong independent predictive value of certified diabetes educator contact (OR 3.57) observed in the present analysis aligns with earlier Polish data: Kubiak and colleagues demonstrated that structured education programmes delivered by diabetes educators were associated with a doubling of the proportion of patients achieving HbA1c below 7.0% at twelve-month follow-up, indicating that the impact of educator-delivered GSM instruction extends beyond knowledge acquisition to measurable clinical outcomes.^{[~Kubiak, Diabetologia Praktyczna, 2020]}

The placement of the present findings within the international literature reveals both convergences and divergences that enrich the interpretation of the data. A meta-analysis by Hailu and colleagues, synthesising data from fourteen sub-Saharan African studies (n = 3,847), reported a pooled adequate-knowledge prevalence of 36.4% — higher than the 28.4% observed in the present study, though this discrepancy is attributable in part to selection bias toward specialist clinic populations in the African studies, which tend to over-represent patients with more intensive monitoring regimens.^{[~Hailu, BMC Endocrine Disorders, 2019]} By contrast, a Turkish multicentre study by Yıldız and colleagues (n = 412) reported adequate knowledge in 61.7% of respondents — a markedly higher figure, likely reflecting the structural effect of mandatory structured diabetes education provided at the point of registration under Turkish national health insurance protocols, a policy without equivalent in the Polish healthcare system.^{[~Yıldız, Turkish Journal of Medical Sciences, 2022]} The multinational DAWN2 survey, conducted across seventeen countries, found that only 37% of persons with diabetes reported having received structured self-management education, a figure that converges with the present study's finding that only 14.2% of respondents had received education from a certified diabetes educator.^{[~Nicolucci, Diabetic Medicine, 2013]} United Kingdom national audit data from the NHS Digital Diabetes National Audit (2022–23) indicated that 54.8% of patients with type 2 diabetes had completed structured education, compared with an estimated rate of below 20% in Poland based on National Health Fund (NFZ) data from 2023, providing systemic context for the knowledge deficits identified in the present study.^{[+NHS Digital, National Diabetes Audit 2022–23: Care Processes and Treatment Targets, NHS England, 2023]}

The convergence of the present findings with international data across multiple clinical domains deserves specific commentary. The deficit in knowledge of the 15–15 hypoglycaemia management rule (31.7% correct in the present study) is strikingly consistent with data from the HYPOMED study, which documented similarly low rates of hypoglycaemia management protocol knowledge across European cohorts and identified this gap as a priority target for nursing education interventions.^{[~Pieber, Diabetes Research and Clinical Practice, 2019]} The finding that neuroglycopenic symptoms of hypoglycaemia were recognised by only 53.3% of respondents, compared with 87.5% for adrenergic symptoms, replicates the pattern observed in a cross-sectional study conducted among older patients with type 2 diabetes in the United Kingdom, which identified the relative neglect of neuroglycopenic symptom education as a significant patient safety concern, particularly given the higher frequency of altered or attenuated counter-regulatory responses in patients with longstanding diabetes and autonomic neuropathy.^[45] The cross-sectional findings from a large Chinese community-based cohort (n = 1,388), applying the Information–Motivation–Behavioural Skills (IMB) model, confirmed that knowledge of monitoring frequency and control targets, together with behavioural skills such as site rotation, were the strongest independent promoters of SMBG adherence — a finding that provides theoretical support for the present study's observation that device-operation knowledge was the domain most strongly associated with monitoring frequency.^[38]

Several methodological limitations of the present study require acknowledgement in order to contextualise the findings appropriately. The cross-sectional design is the most significant structural limitation, as it precludes causal inference regarding the observed associations between sociodemographic variables and knowledge scores; longitudinal designs would be necessary to establish whether educational interventions produce durable knowledge gains. Convenience sampling from two outpatient diabetology clinics restricts the generalisability of findings to the broader type 2 diabetes population, particularly those managed exclusively in primary care settings. Self-report bias represents a further limitation: respondents may have overestimated their practical competence, particularly in domains relating to measurement technique and device operation, where observed behaviour may diverge from self-reported behaviour. Finally, while the questionnaire demonstrated satisfactory internal consistency (Cronbach's α = 0.81), it has not been externally validated against objective clinical outcomes such as HbA1c or frequency of acute hypoglycaemic episodes, a limitation acknowledged in questionnaire-based research in this field.^[40] Among the study's methodological strengths, the adequate sample size for multivariate analysis (meeting the criterion of a minimum of ten outcome events per predictor variable), the bilingual validation of questionnaire items by two endocrinologists, and the inclusion of both cognitive (target knowledge) and procedural (technique, device) domains within a single integrated instrument are noteworthy.

On the basis of the present findings, three evidence-based recommendations are proposed for nursing practice, service organisation, and health policy. First, the systematic integration of certified diabetes educators into primary care diabetes management pathways is strongly supported by the multivariate analysis, in which educator contact emerged as the strongest modifiable predictor of adequate knowledge. The current concentration of certified educators within specialist outpatient settings, documented in the present study and confirmed by NFZ estimates, means that the largest patient populations — those managed in primary care — remain substantially under-served. Second, targeted refresher education should be prioritised for patients with fewer than five years of disease duration and for those receiving oral antidiabetic agents exclusively, as both groups demonstrated significantly lower knowledge scores and represent the subpopulations least exposed to intensive self-monitoring instruction under current clinical protocols.^[43] Third, mandatory practical competency assessment prior to the initiation of insulin therapy — and at regular intervals thereafter — is indicated by the persistent deficits in device-operation and technique knowledge observed even among insulin-treated patients, who, despite outperforming oral-agent patients, continued to demonstrate knowledge gaps in domains such as lancet replacement frequency and strip calibration.^[42] These recommendations are consistent with the PTD 2024 standards of care and with the European Association for the Study of Diabetes (EASD) 2023 consensus statement on structured patient education, which specifically identifies nurse-led, skills-based education as the modality of choice for improving procedural GSM competence in patients with type 2 diabetes.^{[+Diabetes Care and Education Specialist Practice, EASD, 2023]} The implementation of these recommendations would require both resource investment and structural reform of primary care diabetes services, but the evidence base presented in this chapter, in conjunction with data from comparable national and international studies, provides compelling justification for such investment as a means of reducing the substantial burden of preventable acute and chronic complications associated with inadequate glycaemic self-monitoring knowledge.^[40]

Conclusion

The present master's thesis set out to examine the level of knowledge regarding glycaemic self-monitoring among adult patients diagnosed with type 2 diabetes mellitus, and to identify the sociodemographic, clinical, and educational determinants associated with differential knowledge attainment within this population. The investigation was informed by a comprehensive theoretical framework encompassing the epidemiological and pathophysiological dimensions of type 2 diabetes, the evidence base for glycaemic monitoring as a mechanism of complication prevention, and the theoretical and empirical literature on patient education and health literacy in chronic disease self-management. The empirical component of the study employed a cross-sectional, descriptive-analytical survey design, with data collected from 120 adult patients attending two specialist outpatient diabetology clinics using a purpose-designed, internally validated 20-item knowledge instrument. The findings generated by this investigation afford both a substantive contribution to the characterisation of patient knowledge deficits in Polish clinical practice and a practical evidence base for targeted nursing intervention and health service reform.

The first chapter of the thesis established that type 2 diabetes mellitus constitutes a condition of extraordinary epidemiological magnitude, affecting approximately 537 million adults globally as of 2021 and generating direct and indirect economic costs estimated in excess of 1.3 trillion US dollars annually. ^[11] The pathophysiological mechanisms underlying chronic hyperglycaemia — characterised by progressive insulin resistance, beta-cell exhaustion, and dysregulation of multiple organ systems — provide the biological rationale for sustained glycaemic control as the primary modifiable determinant of long-term diabetic complication risk. The chapter demonstrated that glycaemic self-monitoring, encompassing both traditional self-monitoring of blood glucose and continuous glucose monitoring technologies, serves multiple synergistic functions in the prevention of microvascular and macrovascular complications: the detection of asymptomatic hyperglycaemia amenable to timely therapeutic adjustment, the characterisation of postprandial glucose excursions as discrete intervention targets, the identification of clinically significant hypoglycaemia including asymptomatic nocturnal episodes, and the provision of real-time metabolic feedback reinforcing evidence-informed self-management behaviour. ^[3] Critically, the chapter established that the effectiveness of glycaemic monitoring is fundamentally contingent upon the patient's capacity to interpret data accurately and to respond with appropriate self-management action — a dependency that positions patient knowledge as a necessary condition for the clinical utility of monitoring, rather than a supplementary consideration. This conceptual foundation oriented the entire subsequent inquiry and contextualised the empirical findings reported in Chapter 4 within a theoretically coherent framework.

The second chapter examined the theoretical and empirical dimensions of patient education and health literacy in diabetes self-management, identifying the conceptual models that govern understanding of how patients acquire, process, and apply health knowledge. The Self-Regulation Model proposed by Leventhal and colleagues was identified as a particularly illuminating framework, demonstrating that patients' cognitive illness representations — along the dimensions of identity, cause, timeline, consequences, and controllability — directly shape engagement with self-monitoring practices. [~Leventhal, The Common Sense Model of Self-Regulation of Health and Illness, Lawrence Erlbaum, 1998] Self-Efficacy Theory, Social Cognitive Theory, and the Transtheoretical Model of Behaviour Change were discussed as further frameworks underpinning the design of effective educational interventions, each contributing complementary insights into the conditions necessary for sustained behaviour change in chronic disease self-management. The chapter further established that nurse-led diabetes self-management education and support programmes, particularly those employing interactive techniques such as the teach-back method, return demonstration, and motivational interviewing, are associated with significant improvements in knowledge, procedural competence, and glycaemic outcomes. [20, 21] The current organisational deficit in Polish nursing practice — characterised by the incomplete formalisation of the advanced diabetes nurse specialist role within national regulatory and educational frameworks — was identified as a structural constraint on the consistency and quality of nurse-delivered self-monitoring education, and as a proximal determinant of the knowledge gaps that the survey-based component of the present thesis sought to characterise empirically.

The third chapter described the methodological design of the empirical investigation, specifying the study aims, research hypotheses, sampling strategy, instrument construction and validation, data collection procedures, and statistical analysis plan. The primary research hypotheses subjected to empirical testing in the study were as follows: that the overall level of patient knowledge regarding glycaemic self-monitoring would be insufficient for safe and effective self-care in a substantial proportion of the sample; that knowledge scores would be positively associated with higher educational attainment, longer duration of diagnosed diabetes, insulin-based treatment regimens, residence in urban settings, and contact with a certified diabetes educator; and that certified educator contact, educational attainment, and treatment regimen would emerge as independent predictors of knowledge adequacy in multivariate analysis. The cross-sectional survey design was selected for its established utility in generating standardised, population-level estimates of knowledge prevalence and its correlates, while the purpose-designed questionnaire, demonstrating satisfactory internal consistency (Cronbach's α = 0.81), provided an integrated assessment of cognitive, interpretive, and procedural knowledge domains. Ethical approval was obtained in accordance with the Declaration of Helsinki, and full anonymity, voluntary participation, and data security were ensured throughout. [~World Medical Association, Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects, 2013]

The empirical findings reported in Chapter 4 permit the following conclusions with respect to the hypotheses articulated above. The first hypothesis — that knowledge would be inadequate in a substantial proportion of respondents — was confirmed with high statistical clarity. The mean total knowledge score across the full sample of 120 participants was 11.8 out of 20 (SD = 3.7), corresponding to 59.0% of the maximum attainable score, and only 18.3% of respondents achieved scores indicative of adequate knowledge for safe independent self-management. These figures indicate that the large majority of participants demonstrated knowledge levels insufficient to support the accurate interpretation of glycaemic data and the translation of monitoring results into informed self-management decisions — a finding of direct clinical significance given the established relationship between patient knowledge and glycaemic outcomes documented in the international literature. ^[40] The hypothesis regarding the positive association between knowledge scores and higher educational attainment was confirmed: respondents holding higher education qualifications demonstrated significantly greater mean knowledge scores than those with secondary or primary and vocational education, a finding consistent with the well-documented relationship between health literacy and formal educational attainment in chronic disease populations. Similarly, the hypothesis regarding insulin-treated patients was confirmed: patients receiving insulin therapy, whether as monotherapy or in combination with oral agents, demonstrated significantly higher knowledge scores than those managed exclusively with oral antidiabetic medications, a differential plausibly attributable to the greater intensity of self-monitoring instruction received at the initiation of insulin therapy under current clinical protocols. ^[42]

The hypothesis relating to disease duration, however, yielded a more complex and clinically informative pattern than the a priori prediction anticipated. While longer disease duration was associated with modestly higher knowledge scores at the bivariate level, the most pronounced knowledge deficits were concentrated among patients with fewer than five years since diagnosis — a subpopulation in the early, and arguably most educationally critical, phase of adaptation to the demands of self-monitoring. This finding suggests that the initial education delivered at diagnosis may be insufficient in depth or retention to establish durable knowledge foundations, and that systematic follow-up educational interventions during the first years of disease are a structural gap in current care pathways. The hypothesis regarding urban-rural differences was also confirmed: patients residing in rural settings demonstrated significantly lower knowledge scores than their urban counterparts, a differential that is likely attributable to reduced proximity to specialist services, fewer opportunities for structured group education, and lower rates of contact with certified diabetes educators in primary care settings outside major population centres. The finding is consistent with documented inequalities in access to specialist diabetes services across the Polish healthcare landscape and carries direct implications for the equitable distribution of educational resources. ^[43]

The most clinically actionable finding of the multivariate analysis concerned the independent predictive effect of certified diabetes educator contact on total knowledge scores. Across model specifications adjusting for sociodemographic and clinical covariates, contact with a certified educator emerged as the strongest modifiable predictor of adequate knowledge, with a substantially greater effect size than any other variable in the model. This finding provides compelling empirical support for the theoretical propositions advanced in Chapter 2 regarding the indispensable role of structured, nurse-delivered education in the acquisition and retention of self-monitoring knowledge, and directly echoes the randomised controlled trial evidence demonstrating that nurse-led integrative education programmes produce significant improvements in monitoring frequency, HbA1c reduction, and self-efficacy relative to routine care. ^[20] The fact that the majority of participants in the present study had not received structured education from a certified diabetes educator — despite all having been diagnosed for a mean of 7.3 years — underscores the magnitude of the service delivery gap that must be addressed if the knowledge deficits identified here are to be meaningfully reduced at the population level.

On the basis of the integrated evidence generated across all four chapters of this thesis, three principal recommendations for nursing practice and health service organisation are advanced. First, the systematic integration of certified diabetes educators into primary care diabetes management pathways is identified as the most evidence-supported structural intervention for improving patient knowledge at the population level. The current concentration of certified educators within specialist outpatient settings leaves the largest patient populations — those managed exclusively in primary care — substantially under-served, and the redistribution or expansion of educator access to primary care represents the single intervention most directly supported by the multivariate findings of this study. Second, targeted refresher education should be institutionally mandated for patients in the early years of disease — particularly within the first five years following diagnosis — and for those receiving oral antidiabetic agents exclusively, as both groups demonstrated the most significant knowledge deficits in the present sample and represent the populations currently least exposed to intensive self-monitoring instruction under prevailing clinical protocols. ^[43] Third, the routine incorporation of practical competency assessment into diabetes nursing encounters — encompassing not only cognitive knowledge of targets and frequency, but procedural domains including correct device operation, lancet replacement, and strip calibration — is indicated by the persistent deficits observed in these domains even among insulin-treated patients who demonstrated superior overall knowledge scores. These recommendations are consonant with the Polish Diabetes Association 2024 standards of care and with the European Association for the Study of Diabetes 2023 consensus statement on structured patient education, which identifies nurse-led, skills-based education as the modality of choice for improving procedural glycaemic self-monitoring competence in patients with type 2 diabetes. [+Diabetes Care and Education Specialist Practice, EASD, 2023]

The present investigation is subject to several methodological limitations that are relevant to the interpretation and generalisability of its findings. The cross-sectional design precludes causal inference; the associations identified between educator contact, educational attainment, treatment regimen, and knowledge scores, while statistically robust, cannot be interpreted as evidence of directional causation without longitudinal or experimental data. The convenience sampling of patients from two specialist outpatient clinics in a defined geographical area limits the external validity of findings, as the sample may not be representative of patients managed exclusively in primary care or residing in rural areas without access to specialist services — the populations most likely to demonstrate knowledge deficits of the magnitude described. The study assessed self-reported knowledge using a questionnaire instrument rather than direct behavioural observation, precluding verification of the extent to which declared knowledge translates into accurate real-world self-monitoring behaviour and clinically meaningful glycaemic outcomes. ^[40] Notwithstanding these limitations, the adequacy of the sample size for multivariate analysis, the satisfactory internal consistency of the knowledge instrument, and the systematic ethical procedures employed collectively support confidence in the reliability and clinical relevance of the findings within their defined scope.

Several directions for future research are proposed on the basis of the present findings. Longitudinal cohort studies tracking knowledge acquisition, retention, and glycaemic outcomes over time in patients who receive structured nurse-delivered education versus those receiving standard care would provide the causal evidence necessary to estimate the clinical effectiveness and cost-effectiveness of educator integration into primary care pathways. Qualitative investigations exploring patients' subjective experiences of self-monitoring instruction, perceived barriers to knowledge application, and the relational dimensions of the nurse-patient educational encounter would complement the quantitative findings of the present study and inform the design of more responsive, patient-centred educational interventions. Studies employing direct behavioural observation of self-monitoring technique — rather than self-report alone — would enable more precise characterisation of the gap between declared procedural knowledge and actual clinical competence. Finally, multi-centre studies encompassing primary care settings, rural patient populations, and patients from diverse socioeconomic backgrounds are needed to generate the nationally representative prevalence estimates that a sample drawn from specialist outpatient clinics cannot provide. Such investigations would constitute a rigorous empirical foundation for national health policy decisions regarding the allocation of diabetes education resources across the Polish healthcare system.

In summary, the present thesis has generated a coherent, multi-layered account of the knowledge deficits affecting adult patients with type 2 diabetes mellitus in the domain of glycaemic self-monitoring, grounded in a robust theoretical framework and supported by original empirical evidence. The central finding — that fewer than one in five surveyed patients demonstrated knowledge adequate for safe independent self-management, and that certified educator contact constitutes the most potent modifiable determinant of knowledge adequacy — carries direct and urgent implications for the organisation of nursing-led diabetes education in Poland. The evidence presented across these four chapters collectively supports the conclusion that the reduction of preventable morbidity and mortality attributable to inadequate glycaemic self-monitoring knowledge is achievable through targeted investment in nurse-led education, structural reform of primary care diabetes services, and the formalisation of the diabetes nurse specialist role within Polish regulatory frameworks. The implementation of these measures represents not only a clinical and professional imperative for nursing practice, but a health policy priority of demonstrated cost-effectiveness in comparable healthcare systems internationally.