Practices for glucose measurement with a glucometer in a population of diabetic patients. An observational study

Abstract

The ability to use a glucometer for self-monitoring of blood glucose is an important component of optimal blood glucose control in diabetic patients. However, the importance of this skill is still not well understood by numerous patients and in medical facilities. The aim of this study was to evaluate glucose measurement practices in people with diabetes. A Cross-sectional, descriptive study. This cross-sectional, observational study was conducted in a group of 212 patients who attended follow-up visits in diabetes clinics located in two hospitals with the same level of referral. The criteria for inclusion in the study were: age ≥ 20 years, diabetes diagnosed at least 1 year before the study, glucose monitoring using only a glucometer. The ability to measure glucose with a glucometer was assessed using a 16-item observation questionnaire. The relationship between two variables was calculated using the R Spearman’s rank correlation coefficient. The non-parametric Mann-Whitney U test was used to assess differences of one feature between two groups. The non-parametric Chi-squared test was also used in the analyses. A multiple linear regression model was constructed to examine the associations between of multiple independent variables and single dependent variable. The data showed that 49.1% of the patients were rather able to measure glucose with a glucometer. Over half of them (53.8%) had never participated in training to learn how to measure glucose levels. As regards patients included in the study, we identified 3 errors related to measurement technique in 19.3% of them, while 9 errors defining correct measurement skills were identified in 5.7% of the patients. The most common errors included failure to replace the lancet in the fingertip lancing device (81.6%) and not washing hands with warm soapy water prior to the measurement (65.1%). The number of errors reported during the measurement was significantly related to the age of the patients, education level, professional status, place of residence, self-assessment of the financial situation, type of diabetes and participation in education. Educating patients in the skill of the self-monitoring of blood glucose with a glucometer should constitute an integral part of the care of a diabetic patient in a diabetes clinic. Only one in ten patients with diabetes fully adhered to the principles of correct glucose measurement with a glucometer. When educating patients on how to take measurements, special attention should be paid to replacing lancets in the lancing device and washing hands with soap and water before collecting a blood sample. When assessing the ability to measure glucose levels using a glucometer in this group of patients, the following factors should be considered: glycated hemoglobin concentration, patient participation in education on how to take measurements, self-assessment of the financial situation and the duration of diabetes.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-42065-2.

Introduction

Self-monitoring of blood glucose (SMBG) is a common test used to monitor glucose levels in people with diabetes. It is estimated that in some high-income countries, about 82% of diabetics regularly use SMBG. In 2023, SMBG devices accounted for about 66%, while continuous glucose measurement (CGM) systems accounted for 34%¹. In samples including Polish patients with type 1 and type 2 diabetes, the use of a glucometer was confirmed by 76.8% to 90% of the respondents^2–4.

Despite the widespread use of modern glucose monitoring systems, such as the intermittently scanned continuous glucose measurement (isCGM) and CGM systems which measure and display glucose levels continuously in real time (real-time CGM; rtCGM), numerous patients still use glucometers for monitoring. According to the American Diabetes Association (ADA) guidelines, diabetics should be equipped with glucose monitoring devices. They should also be permanently accessible for persons using CGM devices⁵. According to a report issued by the WHO, guidelines for the self-monitoring of blood glucose are needed, especially in low- and middle-income countries. Such guidelines should stay in line with the recommendations for home use or in use in medical facilities. Education is necessary to make optimal use of medical technologies for SMBG and treatment of diabetes in general, because even if they are available free of charge, poor results are still observed in patients⁶.

It was estimated that in 2023, 3.26 million people suffered from diabetes in Poland, 2.11 million people with diabetes used test strips, and the cost of reimbursement was PLN 459.43 million⁷. In adults with nonintensively managed type 2 diabetes, using SMBG seemed to be less expensive compared to CGM⁸. In patients with insulin-treated type 2 diabetes and poor glycemic control, the use of rtCGM may be cost-effective compared to glucometer use⁹.

According to current Polish guidelines, multiple monitoring of blood glucose levels is recommended to diabetic patients receiving at least 3 insulin injections per day¹⁰. Other diabetic patients require no routine glucose monitoring (except during pregnancy). However, monitoring may be informative if the results are inconsistent with individual therapeutic goals, or in the context of changes in drug therapy or behavioral changes (diet, physical activity). Data on the evaluation and comparison of the perceived value, recommended frequency, and usefulness of SMBG in patients with non-insulin-treated type 2 diabetes (NITT2DM) by primary care providers (PCP) and endocrinologists showed that they found SMBG practice valuable in the NITT2DM group and believed it promoted behavioral changes in their patients. Endocrinologists tend to recommend SMBGs more frequently and find SMBG values more useful when making treatment decisions, while primary care physicians tend to view glycated hemoglobin results as more valuable than SMBGs in adjusting drugs in therapy¹¹. A meta-analysis showed that the recommendation of structured self-monitoring of blood glucose by medical staff and the use of the results to modify therapy brought greater benefits in terms of improving glycated hemoglobin (HbA1c) outcomes¹². Notably, the provision of care that corresponds with the diabetic patients’ preferences, values and needs constitutes the basis for making clinical decisions comprising a personalized and patient-centered approach¹³.

The frequency of measurements and avoiding measurement-related errors may be important for the patient and medical staff when making therapeutic decisions and introducing lifestyle changes, likely demonstrating potential to improve clinical outcomes^12,14,15. According to the recommendations of the Polish Diabetes Association (PDA), the American Diabetes Association and the Polish Federation of Education in Diabetes (PFED), the recommended frequency of measurements depends on the type of therapy implemented and the individual needs of the patient^5,10,16. Research showed that higher SMBG frequencies were associated with lower HbA1c scores in diabetics^12,14,17,18 and might be helpful in modifying b, ehaviors related to nutrition, physical activity and the use of drugs⁵. Some authors suggested that about two-thirds of adult patients monitored their glucose levels less frequently than recommended^19,20. The importance of SMBG was less recognized by patients experiencing pain during the measurement and more recognized in medical facilities, where doctors regularly checked the patients’ reported glucose results during the visit to provide them with feedback²¹.

Due to the design and software solutions implemented in glucometers, the accuracy of the glucose measurement result is influenced by numerous factors. These may include the preparation of the fingertip for collecting a blood sample, the technique of collecting a blood sample, the way how test strips are stored^13,22–27. Failure to comply with current recommendations related to the measurement may result in obtaining a false result that may contribute to making incorrect therapeutic decisions by the patient at home.

According to clinical guidelines, SMBG skills should be assessed by a nurse at least once a year^16,28–30. To test the compliance of the ability to measure capillary blood glucose with a glucometer with the latest recommendations, we evaluated the patients’ practices related to the measurement. We focused on assessing the patients’ preparation for the measurement, the patients’ skills when performing the measurement and the patients’ behavior after the measurement. Contrary to previous research that focused mainly on the frequency of measurements and their relationship with metabolic control, this paper provides new data on the practical aspects of glucose measurement technique and the most common errors. The identification of factors that may be related to correct SMBG performance was not comprised in previous analyses.

The effectiveness of chronic disease management may be influenced by numerous facilitating factors and obstacles, which are important both for the design of educational interventions, patient-centered care planning, and planning and designing further research. Education and support in SMBG play a key role in the care of people with diabetes³¹. The results obtained in this study may be relevant to the planning of diabetes education programs, indicating areas that require special attention when training patients and potentially reassessing their skills.

Methods

Aim

The aim of this study was to evaluate glucose measurement practices in people with diabetes.

Sampling procedure

The study included 212 diabetic patients who attended follow-up visits in two diabetes clinics located in two hospitals on two randomly selected days of the week (Tuesday and Thursday). It was estimated that during the study period from September 1, 2024 to December 1, 2024, approximately 470 patients presented to both diabetes clinics on Tuesdays and Thursdays. The study included patients who met the following criteria: age ≥ 20 years, diabetes diagnosed at least 1 year before the study, glucose monitoring using only a glucometer. Women with gestational diabetes and patients using CGM were excluded.

The size of the study sample was estimated on the basis of the sample size formula for a finite population.

P – estimated proportion (0.5 for maximum variability).

z – z-value corresponding to the desired confidence level (1.96 for 95%).

N _min – required sample size.

N – population size.

e – margin of error (0.05).

Based on this formula, the final sample size was determined to be 212 patients assuming the confidence level of 95% and the maximum error margin of 5%.

Setting

It was a cross-sectional, descriptive, two-center study. The study was conducted in two diabetes clinics located in two different hospitals. The criteria for the selection of hospitals: specialist hospital, location in different voivodeships, secondary level of referral, availability of a diabetes clinic, availability of a diabetes nurse. The study was performed during the patient’s visit to the nurse’s educational office.

Measurements and data collection procedures

In order to collect the research material, the initial part of the questionnaire intended for the patient contained information on sociodemographic (7 questions) and clinical (7 questions) data. It was completed by an experienced diabetes nurse basing on an interview with the patient and medical documentation. Another part of the 12-item questionnaire concerned the content of the patient’s glucometer case and was also completed by the nurse after the patient showed her the case. To collect feedback on the patient’s glucose measurement practices using a glucometer, a 16-item follow-up questionnaire was developed. This part was completed by the nurse following the observation of the patient’s skill to measure glucose concentrations with the glucometer. Binary scoring was used in the assessment of measurement skills: 1 (yes) – if a given criterion was met, which meant correct performance; 0 (no) – if a given criterion was not met, which meant incorrect performance.

The areas of correctly performed measurement were determined by the team designing the study on the basis of a review of scientific data, PDA 2024 recommendations and the 2023 recommendations of the Polish Federation of Education in Diabetology^32,33. The developed tool was subjected to external review by two consultants in the field of diabetes nursing. The tool for assessing the ability to measure glucose concentration with a glucometer turned out to be a reliable scale. The Cronbach’s alpha coefficient for the entire tool was 0.862, and the mean correlation coefficient between scale items was 0.272. The following items were the least correlated: he/she placed the used lancet in a hard case − 0.275; he/she washed his/her hands with soap and water before taking the measurement − 0.240, and he/she applied the obtained drop of blood to the test strip − 0.085.

Errors in the measurement skills analyzed in this study included: (1) the preparation of the capillary blood collection site, (2) lancet use, (3) the selection of the capillary blood collection site, (4) the technique of obtaining a drop of blood for measurement, (5) capillary blood collection site care, (6) waste equipment handling, (7) the interpretation of the measurement result. The correct ability to measure within these categories was defined as follows: the patient washed his/her hands with warm soapy water (1), he/she dried his/her hands (2), the patient changed the lancet/set a new lancet in the drum lancet device/used a disposable lancing device (3), the patient selected the 3rd, 4th or 5th fingertip for blood collection (4), the patient pricked the lateral part of the fingertip (5), the patient did not squeeze a drop of blood (6), the patient applied a gauze pad/patch/clean tissue over the puncture site (7), the patient placed the used sharp lancet in a hard container (8), the patient interpreted the measurement result as recommended, too low or too high (9).

Ethical considerations

The research was performed in accordance with the Declaration of Helsinki. It was voluntary for the subjects to answer the questionnaire questions, and they had the right to withdraw their participation at any time. All the answers were treated as strictly confidential, and the participants were guaranteed full anonymity. The subjects provided their informed consent to participate in the study. The study was approved by the Bioethics Committee at the Medical University of Warsaw, 10/06/2024 (approval no AKBE/169/2024).

Data analysis

Statistica 10.0 software and the standard functions of Microsoft Excel spreadsheet were used to analyze the data. In the descriptive analysis, tables were used to present the number and percentage of responses to individual questionnaire questions.

The relationship between two variables was calculated using the R Spearman’s rank correlation coefficient. The non-parametric Mann-Whitney U test was used to assess differences of one feature between two groups. The non-parametric Chi-squared test was also used in the analyses. A multiple linear regression model was constructed to examine the associations between of multiple independent variables and single dependent variable. Independent variables were selected based on previous literature reports and their clinical significance for glycemic control. Initially, when building the model, we took account of all variables that were correlated significantly or were at the border of statistical significance with the number of errors in the measurement. We built several models and selected one with the highest value of multiple R2 and multiple R. In addition to a minor lack of data on the HbA1c value in a group of 1.9% (n = 4) of patients, the analysis included complete cases. As regards the HbA1c variable, analyses were performed based on data obtained from 208 patients. No adjustments were made.

Results

Characteristics of participants

The study group included 58.5% of women, 50.5% of whom were married and in civil partnerships. General and vocational secondary education was reported by 40.1% of the respondents, 54.7% of the respondents were retired, and 55.2% lived in urban areas. The financial situation was assessed as satisfactory by 40.6% of the patients (see Supplementary file Table S1). In the study group, 28.3% of patients were 76–92 years old, with the mean age of the patients being 64.4 ± 15.59 years (see Supplementary file Table S2).

A total of 82.1% of the patients had type 2 diabetes (Table 1). The mean duration of diabetes was 10.5 ± 8.69 years in the study group of patients (see Supplementary file Table S2). Oral antihyperglycemic agents as the only diabetes treatment were reported by 64% of the patients. The majority of the respondents declared that they were rather able to measure glucose with a glucometer – 49.1%. A total of 29.7% of the patients did not participate in education concerning the performance of glucose measurement with a glucometer; they only used the information leaflet of the glucometer to learn how to measure glucose levels. Several glucose measurements during the day were performed by 36.2% of the patients (Table 1).

Table 1.

Clinical characteristics of the patients (n = 212).

Variables		n(%)
Type of diabetes	Type 1	38 (17.9)
	Type 2	174 (82.1)
Duration of diabetes	1–10 years	139 (65.5)
	11–20 years	52 (24.6)
	21–30 years	14 (6.6)
	31–40 years	7 (3.3)
Type of diabetes therapy	oral antihyperglycemic drugs	128 (64.0)
	insulin therapy	38 (17.9)
	oral antihyperglycemic drugs and insulin therapy	46 (21.4)
Participation in educational meetings to learn how to measure blood glucose with a glucometer	yes, once	9 (4.2)
	yes, many times	30 (14.2)
	yes, individual education	34 (16.0)
	yes, group education	25 (11.8)
	no, I used the support of a person using a glucometer	51 (24.1)
	no, I used the glucometer information leaflet	63 (29.7)
Self-assessment of the ability to measure glucose with a glucometer	I definitely can	37 (17.5)
	I rather can	104 (49.1)
	Hard to assess	50 (23.6)
	I rather can’t	21 (9.9)
Number of glucose measurements performed with a glucometer	one measurement per day at any time	11 (5.2)
	one measurement per day – in the morning before a meal	18 (8.5)
	several measurements a day	77 (36.2)
	several measurements a week	66 (31.2)
	several measurements a month	36 (17.0)
	once a month or less often	4 (1.9)

The mean value of glycated hemoglobin (HbA1c) results in the study group from the last 6 months was 7.54 ± 1.23%, and the mean value of the random glucose concentration measurement during the visit was 167 ± 45.24 mg/dL (see Supplementary file Table S2).

Contents of the glucometer case

At the follow-up visit, 23.1% of the patients did not bring their own glucometer. The analysis of the contents of the glucometer case showed signs of glucometer contamination in 54.0% of the patients. The presence of used blood-stained test strips was confirmed in the cases of 20.2% of the patients (see Supplementary file Table S3).

Ability to measure glucose concentrations with a glucometer

Type and number of errors made by patients during measurements

The correct performance of measurements was defined by 9 elements. The highest rate of errors made by patients was recorded for the following items: inserting a new lancet/setting a new lancet in the drum/using a disposable lancing device – 81.6%, and washing hands with warm soapy water before the measurement – 65.1%. The lowest error rate was recorded for: pricking the lateral surface of the fingertip – 20.3% and allowing the blood to flow freely – 19.8% (Table 2).

Table 2.

Ability to measure glucose concentrations with a glucometer.

Elements to be assessed	Yes	No
	n (%)	n(%)
The patient washed his/her hands with soap and water before taking the measurement (1)*	74 (34.9)	138 (65.1)
The patient washed his/her hands before taking the measurement	35 (16.5)	177 (83.5)
The patient dried his/her hands thoroughly (2)*	82 (38.7)	130 (61.3)
The patient inserted a new lancet/set another lancet in the drum/used a disposable lancing device (3)*	39 (18.4)	173 (81.6)
The patient set the penetration depth of the fingertip in the lancing device	43 (20.3)	169 (79.7)
The patient selected a finger to prick (except the thumb and index finger) (4)*	179 (84.4)	33 (15.6)
The patient assessed the capillary blood collection site visually and by palpation	171 (80.7)	41 (19.3)
The patient performed finger massage from the base of the hand towards the fingertip of the pricked finger	131 (61.8)	81 (38.2)
The patient pricked the lateral surface of the fingertip (5)*	169 (79.7)	43 (20.3)
The patient allowed the blood to flow freely (6)*	170 (80.2)	42 (19.8)
The patient applied the obtained drop of blood to the test strip	205 (96.7)	7 (3.3)
The patient applied a gauze pad/patch/clean wipe over the puncture site (7)*	158 (74.5)	54 (25.5)
The patient replaced the lancet in the lancing device after use	45 (21.2)	167 (78.8)
The patient placed the used lancet in a hard box (8)*	109 (51.4)	103 (48.6)
The patient recorded the result of the blood glucose measurement	81 (38.2)	131 (61.8)
The patient correctly interpreted the measurement result (9)*	159 (75.0)	53 (25.0)

Note: * a measurement element defining the correct measurement skill.

The average number of errors made by patients during glucose measurement was 3.6 ± 2.38. Three errors were identified in 19.3% of patients, and four errors in 16.0% (Fig. 1).

Fig. 1.

Distribution of the number of procedural errors made by patients during self-monitoring of blood glucose (n = 212).

The figure shows the percentage of patients according to the total number of errors identified during glucose measurements, illustrating the variability in measurement performance across the study population.

Type and number of errors vs. patient characteristics

The age of the respondents remained in a statistically significant, average correlation with the results for the following items: he/she placed the used lancet in a hard box (R = -0.320; P < 0.001) and the total number of errors (R = 0.340; P < 0.001). A weak correlation was obtained with the results of other measurement elements, except “he/she allowed the blood to flow freely” (P = 0.113). The older the patients were, the less often they performed a given measurement element. The education level of the respondents remained in a statistically significant, average correlation with the number of errors (R = -0.359; P < 0.001) and in a weak correlation with the results of other measurement elements, except: “the patient placed the used lancet in a hard box” (P = 0.201). The better educated the respondents were, the more often they performed the measurement elements in a correct manner (Table 3).

Table 3.

Correlation of patients’ age and education level with the number and type of errors.

Elements to be assessed	Age							Education
	23–55 years (n = 52)	56–67 years (n = 54)	68–75 years (n = 46)	76–92 years (n = 60)	R	t(N-2)	P	primary/ junior high school (n = 20)	basic vocational/ trade school (n = 43)	general/ vocational secondary school (n = 85)	higher bachelor/ engineer (n = 32)	master’s degree (n = 32)	R	t(N-2)	P
	M ± SD	M ± SD	M ± SD	M ± SD				M ± SD	M ± SD	M ± SD	M ± SD	M ± SD
The patient washed his/her hands with soap and water before taking the measurement	0.42 ± 0.5	0.69 ± 0.47	0.78 ± 0.42	0.72 ± 0.45	-0.223	-3.323	0.001	0.85 ± 0.37	0.84 ± 0.37	0.61 ± 0.49	0.56 ± 0.50	0.47 ± 0.51	0.267	4.022	< 0.001
The patient dried his/her hands thoroughly	0.38 ± 0.49	0.67 ± 0.48	0.78 ± 0.42	0.63 ± 0.49	-0.187	-2.761	0.006	0.85 ± 0.37	0.81 ± 0.39	0.56 ± 0.5	0.53 ± 0.51	0.41 ± 0.5	0.290	4.392	< 0.001
The patient inserted a new lancet/set another lancet in the drum/used a disposable lancing device	0.65 ± 0.48	0.83 ± 0.38	0.89 ± 0.31	0.88 ± 0.32	-0.213	-3.160	0.002	0.95 ± 0.22	0.95 ± 0.21	0.76 ± 0.43	0.75 ± 0.44	0.75 ± 0.44	0.197	2.910	0.004
The patient selected a finger to prick (except the thumb and index finger)	0 ± 0	0.2 ± 0.41	0.17 ± 0.38	0.23 ± 0.43	-0.209	-3.097	0.002	0.25 ± 0.44	0.28 ± 0.45	0.14 ± 0.35	0.09 ± 0.30	0.03 ± 0.18	0.221	3.281	0.001
The patient pricked the lateral surface of the fingertip	0.10 ± 0.3	0.19 ± 0.39	0.28 ± 0.46	0.25 ± 0.44	-0.152	-2.234	0.027	0.30 ± 0.47	0.35 ± 0.48	0.19 ± 0.39	0.09 ± 0.30	0.09 ± 0.30	0.221	3.290	0.001
the patient allowed the blood to flow freely	0.08 ± 0.27	0.28 ± 0.45	0.20 ± 0.4	0.23 ± 0.43	-0.109	-1.591	0.113	0.25 ± 0.44	0.33 ± 0.47	0.19 ± 0.39	0.16 ± 0.37	0.06 ± 0.25	0.186	2.739	0.007
The patient applied a gauze pad/patch/clean wipe over the puncture site	0.08 ± 0.27	0.26 ± 0.44	0.33 ± 0.47	0.35 ± 0.48	-0.227	-3.382	0.001	0.25 ± 0.44	0.44 ± 0.50	0.25 ± 0.43	0.16 ± 0.37	0.13 ± 0.34	0.197	2.907	0.004
The patient placed the used lancet in a hard box	0.33 ± 0.47	0.39 ± 0.49	0.41 ± 0.5	0.77 ± 0.43	-0.320	-4.891	< 0.001	0.30 ± 0.47	0.56 ± 0.50	0.61 ± 0.49	0.28 ± 0.46	0.38 ± 0.49	0.088	1.282	0.201
The patient correctly interpreted the measurement result	0.10 ± 0.3	0.26 ± 0.44	0.35 ± 0.48	0.3 ± 0.46	-0.176	-2.595	0.010	0.50 ± 0.51	0.40 ± 0.49	0.19 ± 0.39	0.16 ± 0.37	0.16 ± 0.37	0.246	3.676	< 0.001
Number of measurement errors	2.13 ± 1.72	3.76 ± 2.47	4.20 ± 2.2	4.37 ± 2.4	0.340	5.233	< 0.001	4.50 ± 1.96	4.95 ± 2.39	3.51 ± 2.38	2.78 ± 2.03	2.47 ± 1.97	-0.359	-5.572	< 0.001

M – mean; SD – standard deviation; R – Spearman’s correlation coefficient; t(N-2) – t-test result with N-2 degrees of freedom; P – statistical significance level

Significant values are in bold.

Statistically significant differences occurred between groups relative to the sex, regarding the results of only one measurement element – the patient applied a gauze pad over the puncture site (Z = 2.419; P = 0.016) (see Supplementary file Table S4).

Statistically significant differences occurred between marital status groups regarding the following measurement elements: the patient washed his/her hands with warm soapy water before the measurement (P = 0.011), the patient dried his/her hands thoroughly (P = 0.043), the patient applied gauze over the puncture site (P = 0.003), and the patient placed the used lancet in a hard box (P = 0.030). The highest mean number of errors (4.26 ± 2.31) was recorded in case of widowed patients (see Supplementary file Table S5).

Statistically significant differences between the occupational status groups were demonstrated regarding the results of the majority of the assessed measurement elements (P < 0.05) with the exception of: “the patient allowed the blood to flow freely” (P = 0.065) and the total number of errors made (P = 0.107). The highest mean number of errors (5.39 ± 2.25) was recorded in the group of disability pensioners and the unemployed (4.43 ± 1.27 errors) (Table 4).

Table 4.

Differences in the number and type of errors between groups in relation to the occupational status.

Elements to be assessed	Employed (n = 71)	Retired (n = 116)	Disability pensioner (n = 18)	Unemployed (n = 7)	Chi-square	df	P
	M ± SD	M ± SD	M ± SD	M ± SD
The patient washed his/her hands with soap and water before taking the measurement	0.45 ± 0.5	0.74 ± 0.44	0.78 ± 0.43	0.86 ± 0.38	16.404	4	0.003
The patient dried his/her hands thoroughly	0.44 ± 0.5	0.71 ± 0.46	0.67 ± 0.49	0.71 ± 0.49	19.562	4	0.001
The patient inserted a new lancet/set another lancet in the drum/used a disposable lancing device	0.66 ± 0.48	0.89 ± 0.32	0.89 ± 0.32	1.0 ± 0	11.154	4	0.025
The patient selected a finger to prick (except the thumb and index finger)	0.06 ± 0.23	0.19 ± 0.39	0.39 ± 0.5	0 ± 0	11.174	4	0.025
The patient pricked the lateral surface of the fingertip	0.07 ± 0.26	0.25 ± 0.43	0.44 ± 0.51	0.14 ± 0.38	11.659	4	0.020
The patient allowed the blood to flow freely	0.10 ± 0.3	0.22 ± 0.41	0.44 ± 0.51	0.29 ± 0.49	8.847	4	0.065
The patient applied a gauze pad/patch/clean wipe over the puncture site	0.07 ± 0.26	0.31 ± 0.46	0.61 ± 0.5	0.29 ± 0.49	12.432	4	0.014
The patient placed the used lancet in a hard box	0.27 ± 0.45	0.56 ± 0.50	0.78 ± 0.43	0.71 ± 0.49	15.996	4	0.003
The patient correctly interpreted the measurement result	0.11 ± 0.32	0.30 ± 0.46	0.39 ± 0.50	0.43 ± 0.53	16.241	4	0.003
Number of measurement errors	2.23 ± 1.93	4.16 ± 2.25	5.39 ± 2.52	4.43 ± 1.27	46.821	36	0.107

M - mean; SD – standard deviation; n – sample size; chi – Chi-squared test value; df – degrees of freedom; P – statistical significance level.

Significant values are in bold.

Statistically significant differences between the groups in relation to the place of residence were identified for the following measurement elements: the patient washed his/her hands with warm soapy water before the measurement (P < 0.001), the patient dried his/her hands thoroughly (P = 0.002), the patient inserted a new lancet/set another lancet in a drum/used.

a disposable lancing device (P = 0.001), the patient applied a gauze pad over the puncture site (P = 0.005) and the total mean number of errors (P = 0.005) (see Supplementary file Table S6).

The financial situation of the respondents remained in a statistically significant, average correlation with the results of the total mean number of errors (R = -0.410, P < 0.001), and the following items: the patient applied a gauze pad over the puncture site (R = 0.376, P < 0.001), the patient placed the used lancet in a hard box (R = 0.413, P < 0.001), and the patient interpreted the measurement result correctly (R = 0.322, P < 0.001). The better the financial situation of the respondents, the more often they performed a given measurement element correctly (Table 5).

Table 5.

The correlation of the financial situation of the patients and the number and type of errors made during measurements.

Elements to be assessed	very poor (n = 7)	poor (n = 33)	satisfactory (n = 86)	good (n = 53)	very good (n = 33)	R	t(N-2)	P
	M ± SD	M ± SD	M ± SD	M ± SD	M ± SD
The patient washed his/her hands with soap and water before taking the measurement	0.71 ± 0.49	0.88 ± 0.33	0.62 ± 0.49	0.60 ± 0.49	0.58 ± 0.5	0.159	2.327	0.021
The patient dried his/her hands thoroughly	0.71 ± 0.49	0.70 ± 0.47	0.59 ± 0.49	0.60 ± 0.49	0.58 ± 0.5	0.065	0.950	0.343
The patient inserted a new lancet/set another lancet in the drum/used a disposable lancing device	1.0 ± 0	0.97 ± 0.17	0.83 ± 0.38	0.74 ± 0.45	0.73 ± 0.45	0.215	3.184	0.002
The patient selected a finger to prick (except the thumb and index finger)	0.43 ± 0.53	0.21 ± 0.42	0.24 ± 0.43	0.04 ± 0.19	0 ± 0	0.284	4.295	< 0.001
The patient pricked the lateral surface of the fingertip	0.43 ± 0.53	0.36 ± 0.49	0.21 ± 0.41	0.15 ± 0.36	0.06 ± 0.24	0.233	3.475	0.001
The patient allowed the blood to flow freely	0.43 ± 0.53	0.30 ± 0.47	0.22 ± 0.42	0.17 ± 0.38	0.03 ± 0.17	0.215	3.193	0.002
The patient applied a gauze pad/patch/clean wipe over the puncture site	0.57 ± 0.53	0.48 ± 0.51	0.31 ± 0.47	0.13 ± 0.34	0 ± 0	0.376	5.873	< 0.001
The patient placed the used lancet in a hard box	0.71 ± 0.49	0.67 ± 0.48	0.65 ± 0.48	0.32 ± 0.47	0.09 ± 0.29	0.413	6.577	< 0.001
The patient correctly interpreted the measurement result	0.57 ± 0.53	0.52 ± 0.51	0.26 ± 0.44	0.13 ± 0.34	0.09 ± 0.29	0.322	4.925	< 0.001
Number of measurement errors	5.57 ± 2.23	5.09 ± 1.99	3.93 ± 2.67	2.89 ± 1.75	2.15 ± 1.44	-0.410	-6.514	< 0.001

M – mean; SD – standard deviation; R – Spearman’s correlation coefficient; t(N-2) – t-test result with N-2 degrees of freedom; P – statistical significance level.

The bold values indicate statistically significant results (p 0.05).

Statistically significant intergroup differences occurred in relation to the self-assessment of the skill, regarding the results of the majority of measurement elements except: the patient applied a gauze pad over the puncture site (P = 0.283). Respondents who declared to be rather unable to perform measurements obtained the poorest results in the majority of the assessed measurement elements, while respondents who declared to be definitely able to perform the measurement obtained the best results (Table 6).

Table 6.

Differences in the number and type of errors between groups in relation to the self-assessment of the skill.

Elements to be assessed	I definitely can (n = 37)	I rather can (n = 104)	Hard to assess (n = 50)	I rather can’t (n = 21)	chi-square	df	P
	M ± SD	M ± SD	M ± SD	M ± SD
the patient washed his/her hands with soap and water before taking the measurement	0.32 ± 0.47	0.59 ± 0.49	0.90 ± 0.3	0.95 ± 0.22	41.318	3	< 0.001
the patient dried his/her hands thoroughly	0.32 ± 0.47	0.55 ± 0.50	0.82 ± 0.39	0.95 ± 0.22	34.079	3	< 0.001
the patient inserted a new lancet/set another lancet in the drum/used a disposable lancing device	0.51 ± 0.51	0.84 ± 0.37	0.94 ± 0.24	0.95 ± 0.22	30.567	3	< 0.001
the patient selected a finger to prick (except the thumb and index finger)	0.05 ± 0.23	0.16 ± 0.37	0.14 ± 0.35	0.33 ± 0.48	8.092	3	0.044
the patient pricked the lateral surface of the fingertip	0.05 ± 0.23	0.19 ± 0.40	0.20 ± 0.40	0.52 ± 0.51	18.520	3	< 0.001
the patient allowed the blood to flow freely	0.03 ± 0.16	0.22 ± 0.42	0.18 ± 0.39	0.43 ± 0.51	14.289	3	0.003
the patient applied a gauze pad/patch/clean wipe over the puncture site	0.16 ± 0.37	0.24 ± 0.43	0.34 ± 0.48	0.29 ± 0.46	3.804	3	0.283
the patient placed the used lancet in a hard box	0.30 ± 0.46	0.56 ± 0.50	0.58 ± 0.5	0.24 ± 0.44	14.349	3	0.002
the patient correctly interpreted the measurement result	0.14 ± 0.35	0.16 ± 0.37	0.36 ± 0.48	0.62 ± 0.50	25.238	3	< 0.001
number of measurement errors	1.89 ± 2.17	3.51 ± 2.34	4.46 ± 1.96	5.29 ± 1.85	84.368	27	< 0.001

M – mean; SD – standard deviation; N – sample size; chi – Chi-squared test value; df – degrees of freedom; P – statistical significance level.

Significant values are in bold.

As regards the type of diabetes, statistically significant differences between groups were confirmed for the number (P < 0.001) and the type of errors. A higher mean number of errors was recorded in the group of subjects with type 2 diabetes compared to subjects with type 1 diabetes (P < 0.001) (Table 7).

Table 7.

Differences in the number and type of errors between groups in relation to the type of diabetes.

Elements to be assessed	Type 1 (n = 38)	Type 2 ( n = 174)	Chi-square	df	P
	M ± SD	M ± SD
The patient washed his/her hands with soap and water before taking the measurement	0.16 ± 0.37	0.76 ± 0.43	49.535	1	< 0.001
The patient dried his/her hands thoroughly	0.18 ± 0.39	0.71 ± 0.46	35.925	1	< 0.001
The patient inserted a new lancet/set another lancet in the drum/used a disposable lancing device	0.42 ± 0.50	0.90 ± 0.30	48.116	1	< 0.001
The patient selected a finger to prick (except the thumb and index finger)	0.03 ± 0.16	0.18 ± 0.39	5.893	1	0.015
The patient pricked the lateral surface of the fingertip	0.03 ± 0.16	0.24 ± 0.43	8.922	1	0.003
The patient allowed the blood to flow freely	0.03 ± 0.16	0.24 ± 0.43	8.602	1	0.003
The patient applied a gauze pad/patch/clean wipe over the puncture site	0.08 ± 0.27	0.29 ± 0.46	7.535	1	0.006
The patient placed the used lancet in a hard box	0.32 ± 0.47	0.52 ± 0.5	5.360	1	0.021
The patient correctly interpreted the measurement result	0 ± 0	0.30 ± 0.46	15.433	1	< 0.001
Number of measurement errors	1.24 ± 1.57	4.15 ± 2.20	89.165	9	< 0.001

M – mean; SD – standard deviation; n – sample size; chi – Chi-squared test value; df – degrees of freedom;P – statistical significance level.

Significant values are in bold.

In the multiple regression model, four independent predictors of the number of errors made by patients during blood glucose measurements were identified: glycated hemoglobin levels, participation in educational meetings, self-assessment of the financial situation, and the duration of diabetes (Table 8).

Table 8.

Univariate significance tests for the number of errors.

Effect	SS	df	MS	F	P
Intercept term	8.329	1	8.329	2.697	0.102
Age	2.104	1	2.104	0.681	0.410
Sex	0.010	1	0.010	0.003	0.955
Marital status	1.370	1	1.370	0.444	0.506
Education	1.465	1	1.465	0.474	0.492
Employment status	8.305	1	8.305	2.689	0.103
Place of residence	10.549	1	10.549	3.416	0.066
Financial status	14.591	1	14.591	4.725	0.031
Self-assessment of the ability to measure glucose with a glucometer	1.071	1	1.071	0.347	0.557
Participation in educational meetings to learn how to measure blood glucose with a glucometer	31.029	1	31.029	10.047	0.002
Duration of diabetes	14.066	1	14.066	4.555	0.034
Type of diabetes	6.629	1	6.629	2.146	0.145
Type of diabetes therapy	1.380	1	1.380	0.447	0.505
HbA1c *	92.575	1	92.575	29.977	< 0.001
Error	599.116	194	3.088

* n = 208.

SS – sum of squares; df – degrees of freedom; MS – mean square; F – statistic value; P – level of statistical significance.

Significant values are in bold.

The coefficient of multiple determination (multiple R²) indicates an explanation of about 50% of the value of the dependent variable (number of errors) (see Supplementary file Table S7). The entire model turned out to be statistically significant. The total correlation of all variables (multiple R) with the “number of errors” variable was R = 0.705 in the mean.

Discussion

The study aimed to evaluate the practices of glucose measurement with a glucometer in people with diabetes. When selecting a glucometer for a patient, it is important to ensure that the person with diabetes receives initial and ongoing education, and the continuous assessment of the ability to measure, interpret the obtained results and the ability to use them^5,10,16. Our study showed that half of the patients had never benefited from education on how to use a glucometer to perform blood glucose measurements. A significant subgroup of the respondents participated in measurement education or obtained the support of a person using a glucometer and a glucometer information leaflet. According to the PFED guidelines, a nurse should offer training in the use of a glucometer and test strips to measure glucose levels at the facility where the patient is treated¹⁶. This study confirmed that patients with low self-assessment of the ability to perform the measurement and those who had not participated in such a training made more errors during measurements. The obtained results suggest that the subjective self-assessment of the ability to perform measurements reflects the actual level of the competence of patients, which may be a useful indicator of the identification of those in need of targeted re-education. Education in SMBG skills is particularly important for patients taking insulin. After starting treatment with insulin or sulfonylurea derivatives, the patient is promptly asked to self-monitor blood glucose at home, without medical supervision. Schütt et al. demonstrated that a significant percentage of patients with diabetes did not receive systematic education in glycemic self-control, despite the applicable recommendations. The lack of the regular evaluation of SMBG technique promotes the consolidation of incorrect habits³⁴.

Notably, people with diabetes who had been trained in the use of glucose monitoring devices achieved better results³⁵. Therefore, even if patients make numerous measurement errors, they may be corrected through assessment and re-education. Our study showed that numerous errors were made during measurements, which means that many patients require re-training in the skill. Re-training allows the elimination of bad habits and provides the patient with access to the latest recommendations. It is recommended that the reassessment of measurement skills should be performed at least once a year^5,10,16. In Poland, educational consultation provided by physicians or nurses as part of coordinated care of a person with diabetes in primary health care (PHC), and advice provided by nurses in diabetes outpatient specialist care include education and re-education of patients in the area of blood glucose monitoring.

Blood glucose monitoring allows people with diabetes to assess their individual response to therapy. Integrating blood glucose results into diabetes treatment may constitute a useful tool to adjust medication, manage nutritional therapy and physical activity, and prevent acute complications of diabetes. The frequency and time of measurements recommended to patients should comprise the needs and individual goals of the treatment of a person with diabetes^5,10. In this study, about one-third of the patients took multiple glucose measurements during the day, and almost one-fifth took only several measurements per month. A considerable variation in the frequency of measurements may reflect not only differences in treatment regimens, but also a different understanding by patients of the purpose and importance of glycemic self-control. Insulin-treated patients using SMBG should be encouraged to monitor blood glucose levels regularly, based on the type of therapy. A study by other authors revealed that primary care physicians most often recommended SMBG to patients to check glucose levels (67%), see the result (65%), or to evaluate the effectiveness of medications (61%). Almost half of the patients noted that SMBG reduced their diabetes-related concerns and 61% experienced an improvement in their quality of life. Half of them declared that they would stop the measurements if the physician agreed. Patients using SMBG following medical advice were more likely to want to give up the measurements, while those who did so out of habit or for a better understanding of the disease were less likely to consider giving up. Therefore, physicians and nurses should discuss the motivation to use SMBG with the patients³⁶. A patient’s lack of understanding of the importance of taking measurements may constitute a factor contributing to their decision to stop taking the measurements. According to the literature, the main reasons for not performing SMBG by people with diabetes included the lack of time, forgetfulness and embarrassment, lack of a glucometer, concerns about the result, and the costs incurred^20,37,38. The determinants of the self-monitoring of blood glucose at home included the fact of owning a glucometer, being provided with advice on the self-monitoring of blood glucose, level of education, and the place of residence³⁹. Notably, it is necessary to increase support for SMBG implementation and to introduce effective changes in the behavior of patients in relation to this measurement. Wang et al. concluded that patients who knew the recommended frequency of measurements and those who had their own glucometer were better at adhering to SMBG. Adherence to the recommendations was also easier in people on oral antihyperglycemic agents compared to patients using insulin¹⁹. The study by Wang et al. also showed that only 27.5% of patients with type 2 diabetes followed SMBG recommendations according to the guidelines of the Chinese Diabetes Society. The highest percentage of adherence (36.6%) was observed in people taking oral hypoglycemic drugs. Multivariate analysis revealed that the use of oral antihyperglycemic drugs, having one’s own glucometer and a higher level of education significantly increased the chances of regularity in the self-monitoring of blood glucose¹⁹. Variation in measurement frequency observed in this study is comparable to the results published by other authors. However, the percentage of patients taking measurements sporadically was relatively high. These differences may be related to the individual approach of physicians to SMBG recommendations and different levels of patient awareness.

regarding the importance of glycemic self-control.

The present study revealed no relationship between the type of therapy and the number of errors, which may result from differences in initial education and the lack of regular verification of the measurement technique over the later stages of treatment. The number of errors made was significantly lower in people with type 1 diabetes. More frequent performance of SMBG in people treated with insulin may be indirectly associated with the better performance of the measurements, mainly through: greater practice in operating the device; better recognition and avoidance of technical measurement errors; increased patient awareness of the correct technique, especially when these measurements are regularly assessed by medical staff and the patient receives feedback on the correctness of the measurement and its clinical significance.

Another study confirmed that many people who checked their blood glucose levels at least once a day took no action when their glucose levels were too high or too low⁴⁰. The problem of the misinterpretation of SMBG results was also described by Malanda et al., whose systematic review showed that numerous patients did not take any action in response to abnormal blood glucose values⁴¹. The present study revealed that every fourth patient misinterpreted glucose measurement results. However, it was not assessed what patients would do if their glucose levels were too low or too high. People with a lower socio-economic status were more likely to make mistakes in interpreting their measurement results. A study conducted in of patients with type 2 diabetes showed that lower levels of health literacy were associated with higher HbA1c values and poorer glycemic outcomes, which might indirectly indicate problems with interpreting the results and taking appropriate therapeutic measures⁴². The lack of data interpretation skills constitutes a significant limitation in terms of SMBG effectiveness, and the misinterpretation of measurement results raises concerns about patients’ ability to make adequate therapeutic decisions, especially in groups with lower socioeconomic status. However, it should be noted that not all research projects comprise the assessment of the understanding of the measurement results, mainly focusing on the frequency of their performance.

Despite technological advances in glycemic measurement, the improper preparation of the capillary blood sampling site still leads to preanalytical errors. Residual glucose on the skin derived from products containing sugar (e.g., fruits, sweets) may cause the overestimation of measurement results^23,43. Moreover, the use of some creams and body lotions before glucose measurement may elevate the results^24,25,44. This study revealed that two-thirds of the respondents did not wash their hands (which could result in an overestimation of the results), and did not thoroughly dry their hands (which could result in an underestimation of the results) before collecting a blood sample. Other authors indicated that the result of the measurement might be influenced by repeated squeezing of blood drops, and diluting the collected blood drop with foreign fluid at the test site was associated with pseudohypoglycemia^26,44,45. In this study, one-fifth of the patients pressed their fingertip to obtain a sufficient drop of blood. The importance of this measurement step has increased due to the smaller capillary blood volume required by newer glucometers. Therefore, comprehensive patient education is still important to avoid preanalytical measurement errors. Numerous patients are unaware of the importance of the thorough cleansing of the blood drop collection site or do not pay enough attention to it. As regards the measurement technique used by the respondents in studies aimed at evaluating various blood glucose monitoring systems, the most common errors included failure to check the expiry date of the test strips before the measurement, incorrect blood application (insufficient amount of blood, pressing the fingertip to the test strip, applying blood outside the test field) and errors in operating the device^46,47. The persistence of these abnormalities, despite the development of glucometer technology, may be due to the routine nature of measurement activities and insufficient emphasis on the importance of pre-analytical stages in the patient education process.

Scientific data and instructions attached to the test strip packaging showed that the accuracy of glucose measurement results deteriorated if test strips were exposed to moisture, or stored outside the original packaging⁴⁴. One case report confirmed falsely low glucose concentrations in a patient (hypoglycemia) associated with the improper storage of test strips. This resulted in the patient making inappropriate decisions related to taking medication²². In order to provide patients with stable diabetes control, they should be advised to carefully read the instructions attached to the test strips. It should also be checked whether the person with diabetes has understood the information contained therein. The authors of the study indicated that patients’ non-compliance with the manufacturer’s instructions when performing the test was the main cause of errors⁴⁷.

The present study showed that factors related to the number of errors in measurement included the patient’s age, social status (i.e., education, occupational status, and financial situation assessment) and the type of diabetes. Other authors also emphasized the importance of an individual approach in the care of a person with diabetes, taking account of social, cultural and technological factors³¹. Literature data confirmed that the demographic characteristics of patients, such as sex and occupation, and disease-related factors, such as type 2 diabetes, the presence of diabetes complications, the frequency of visits, treatment regimen, knowledge of SMBG frequency recommendations, were associated with adherence to SMBG recommendations^19,48,49. However, no studies aimed to assess SMBG skills. The obtained results suggest that the level of education, as an indirect indicator of health competence, played an important role in the proper performance of SMBG, which indicates the need to adapt educational strategies to the capabilities of patients. A higher number of errors among patients living in rural areas may reflect limited access to organized diabetes education and specialized care.

The study revealed that as the age increased, patients made more errors in the measurement technique and performed individual elements of the measurement less often. This may be related to deteriorating manual dexterity, cognitive functions and the coexistence of chronic diseases that hinder the proper performance of measurements. Some authors emphasized the importance of the involvement of caregivers, family members, especially in a situation in which a person with diabetes has additional difficulties related to older age, visual impairment, or concomitant diseases. It was demonstrated that self-care, including glycemic control, was poorer in the older population of diabetics with cognitive deficits^50,51. Kimura et al. highlighted the possibility of using the devices and functions of glucometers in education and in everyday practice to support the patient’s own measurements⁵². The increase in the number of errors with the age of patients indicates the need for additional support for this group, including the involvement of caregivers and the use of educational functions of modern glucometers.

The present study confirmed that fewer errors were made during the measurement when performed by respondents with higher education. This result may not be compared to research outcomes of other authors analyzing the correctness of the measurement because no such analyses have been performed so far. Nevertheless, research conducted by other authors in patients with type 1 and type 2 diabetes showed that people with a higher level of education monitored their blood glucose levels more frequently compared to those with lower education levels^19,53.

This study confirmed that the number and type of errors in measurement were affected by the professional and financial situation of patients. Other authors confirmed the relationship between the patient’s participation in the cost of test strips and the continuation of SMBG. Decreasing costs may reduce the barrier to continuing measurements⁵⁴. In Poland, lancets and lancing devices are not reimbursed for people with diabetes, and people with type 2 diabetes incur a higher cost of test strips compared to those with type 1 diabetes despite the reimbursement. According to the literature, low-income groups bore the greatest financial burden of diabetes⁵⁵. Differences between tests may result from different reimbursement systems and the availability of diabetic equipment. The data obtained indicate that financial barriers may indirectly affect the quality of SMBG through limited access to appropriate equipment and regular measurement practice, thus increasing the risk of errors^56,57.

A systematic review of research in patients with diabetes showed that the factors that might affect self-care included: the patient’s knowledge about diabetes, lack or difficulty in accessing health services, social support received, denial of the disease, costs incurred, everyday life difficulties⁵⁸. Research by other authors showed there was a critical need for more interdisciplinary and collective services, especially in rural areas⁵⁹. This study confirmed that patients living in rural areas made more measurement errors. The report on diabetes care in Poland revealed that the lack of formal diabetes education in Poland remained a significant problem, especially in relation to people with diabetes type 2 diabetes. For most patients with type 2 diabetes, education provided by medical staff after diagnosis was very limited⁶⁰.

To make SMBG, which is an important tool for maintaining self-care, effective in the treatment of diabetes, it is important to support patient and family involvement by emphasizing patient-centered and personalized care and education, i.e., taking account of preferences, needs, values, characteristics and comorbidities in people with diabetes⁵².

Although this study was conducted in diabetes clinics, the results obtained may also be applicable to primary care and other healthcare systems. The errors observed concerned the basic, routine stages of the measurement of glycemia and were related to sociodemographic factors of patients, such as the age, level of education, financial situation, professional status or the type of diabetes. These factors are common in the population of patients treated in primary care facilities, where diabetes care often focuses on patients with type 2 diabetes. This indicates the need for regular, practical assessment of the technique of glycemic measurement and targeted education provided by doctors and nurses, especially in older, less educated patients with a poorer financial situation and with type 2 diabetes. Because the study assessed the patients’ practical skills rather than systemic solutions, its results may be applied to different healthcare systems. Glucometers, measurement principles and basic educational recommendations are largely standardized internationally, which makes it possible to extrapolate conclusions to other countries and models of diabetic care.

Study limitations

This study has several limitations. The patients were recruited in two diabetes clinics and their experiences may vary from facility to facility as protocols for patient education may differ. Patients under the care of specialist clinics may show a higher level of diabetic knowledge, a greater experience in SMBG and more frequent contact with medical staff compared to patients treated in primary care. Therefore, the frequency and nature of glucose measurement errors observed in this study may differ from those in other healthcare settings. Future studies including primary care patients and diverse diabetes care systems might allow a more complete assessment of the scale and nature of measurement errors.

The method of sampling used and the small size of the study group limit the possibility of generalizing the obtained results. The study was non-interventional and observational in nature, and it was not designed to assess the values of glucose concentrations obtained by patients.

The study did not include a question concerning comorbidities that could hinder independent glucose measurements. Diabetes is often concomitant with other diseases, which may affect the number of errors made during measurements. In addition, the rotation of capillary blood collection sites for glucose measurement was not evaluated in the study. Given that 90% of the patients made at least one error among nine measurement elements assessed, we did not expect a change in our main assumption that additional therapeutic education, or even a fundamental change in the educational approach to SMBG, was necessary.

Conclusions

Educating patients how to measure blood glucose levels with a glucometer should constitute an integral part of the care of a diabetic patient in a diabetes clinic. Only one in ten patients with diabetes fully adheres to the principles of correct glucose measurement with a glucometer. When educating patients on how to take measurements, special attention should be paid to replacing lancets in the lancing device and washing hands with soap and water before taking a blood sample. When assessing the ability to measure glucose levels using a glucometer in diabetics, the following factors should be considered: glycated hemoglobin concentration, patient participation in education concerning measurement skills, the self-assessment of the financial situation and the duration of diabetes. Solutions should be constantly sought to help overcome barriers related to the correct measurement of glucose levels with a glucometer, such as a difficult access to education and a low social status of patients. Providing evidence-based recommendations for SMBG may significantly support the proper performance of glucose measurements in people with diabetes. However, the fact that patients continue to make mistakes indicates the need for not only initial but also continuous education. Regular assessment of glucose measurement skills by nurses may also play an important role in implementing these key recommendations.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

EK – Conceptualization, Methodology, Formal analysis, Investigation, Project administration, Resources, Supervision, Writing – original draft, Writing – review and editingEKZ – Data collection, Investigation, Writing – review and editingAŁ – Conceptualization, Data Collection, Methodology, Formal analysis, Writing – original draftBH – Investigation, Resources, Writing – review and editingAll authors read and approved of the final manuscript.

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval

This study has been reviewed and approved by the Bioethics Committee at the Medical University of Warsaw (approval no AKBE/169/2024) and performed in accordance with the Declaration of Helsinki. Informed consent to participate was obtained from all participants prior to data collection. All methods were performed in accordance with the relevant guidelines and regulations..