Assessment of insulin pen injection practices and the effectiveness of pharmacist interventions in improving injection technique among patients with diabetes mellitus

Abstract

Background

Insulin therapy is crucial for Type 1 and advanced Type 2 diabetes management. Despite the convenience of insulin pens, improper use persists due to limited patient education. Pharmacists' expertise in counseling and medication safety uniquely positions them to address these gaps through individualized guidance on use, adherence, and monitoring.

Objectives

This study aimed to evaluate the impact of pharmacist-led interventions on improving insulin injection practices among diabetes patients.

Methods

A pre–post interventional study was conducted at Dhulikhel Hospital (April–September 2018) among 81 consecutively enrolled insulin pen users (≥12 years; on pens ≥2 weeks) from inpatient and outpatient wards. Patients underwent baseline assessment of 16 injection technique elements, received one-to-one pharmacist-led training with demonstration, and were reassessed immediately and after 2 weeks. The primary outcome was the total technique score (0–16), analyzed using Friedman's test and Wilcoxon signed-rank test.

Results

Cold chain maintenance was suboptimal, with 51.9 % of patients transporting insulin cartridges without an icepack. Unsafe needle disposal was common, with 40.7 % discarding needles in municipal vehicles, 13.6 % in bushes, and 4.9 % in rivers. Pharmacist-led interventions significantly improved all practices (p < 0.05). Proper cartridge storage increased from 77.8 % pre-intervention to 91.4 % post-intervention and 88.6 % at two weeks. Correct room temperature pen storage rose from 49.4 % to 95.1 % post-intervention, declining slightly to 70.8 % at two weeks. Correct insulin mixing surged from 7.8 % to 100 % post-intervention, remaining at 81.1 % at two weeks. Injection technique scores significantly improved (median pre = 10, post = 15) and were largely sustained (median two-weeks = 14).

Conclusion

Pharmacist-led education significantly improved insulin injection techniques. Integrating pharmacists into routine diabetes care and reinforcing education is essential to sustain these improvements. These findings underscore that clinical pharmacy interventions make a tangible difference in improving healthcare outcomes.

1. Introduction

Diabetes mellitus (DM) is a chronic metabolic disorder characterized by persistent hyperglycemia, resulting from defects in insulin secretion, insulin action, or both.¹ The global prevalence of diabetes is rising steadily, with an estimated 589 million adults currently affected—over 81 % of whom reside in low- and middle-income countries, including Nepal. This number is projected to escalate to 853 million by 2050.² In Nepal, the prevalence of diabetes has been rising sharply, with an estimated national prevalence of 8.5 % as per a meta-analysis of 2020, where only 52.7 % were aware of their diabetic status.³ Insulin, a vital component in the management of Type 1 and advanced Type 2 diabetes, is also the primary treatment option when lifestyle measures fail to adequately control gestational diabetes, and is required in certain cases of malnutrition-related Type 5 diabetes, making it essential for maintaining blood glucose control and preventing complications.4, 5, 6 Despite its importance, insulin therapy requires precise and consistent administration techniques to achieve optimal outcomes and minimize complications, and this must be balanced with regular glucose monitoring, ranging from fingertip blood glucose testing with strips to advanced systems such as subcutaneous insulin pumps integrated with continuous glucose monitoring and mobile applications.

Inadequate insulin injection practices are a common problem among diabetic patients worldwide. Insulin pens have significantly improved the convenience, accuracy, and ease of insulin administration compared to traditional vial and syringe methods, leading to better adherence and quality of life.⁷ They are the most preferred delivery device among patients and clinicians due to their safety, simplicity, and precision, though correct administration techniques remain essential.⁸ A robust body of evidence consistently demonstrates that errors in insulin injection technique are alarmingly prevalent among patients using pens, encompassing critical steps such as skipping priming, selecting an inappropriate needle length, or failing to rotate injection sites, insufficient injection duration (dwell time), improper needle reuse, and injection into lipohypertrophic tissue,⁹ which has led to suboptimal diabetes control and increased complications.⁸ Poor injection practices, such as failure to rotate injection sites, incorrect handling of insulin pens, and reusing needles can contribute to complications such as lipohypertrophy, injection site infections, and even insulin resistance.¹⁰ Additionally, incorrect dosing, improper site selection, and lack of cold chain maintenance during insulin transport can further compromise treatment efficacy.¹¹ Ultimately, this translates into higher risks of diabetes complications, increased healthcare resource utilization, and diminished quality of life.¹²

Despite the recognized importance of proper technique, numerous studies have documented widespread deficiencies in insulin injection practices. In a global survey, more than 67 % of patients failed to rotate injection sites properly, and about 25 % exhibited lipohypertrophy due to repeated injections at the same site.⁹ Lipohypertrophy is one of the most frequent complications of poor injection technique, estimated to affect at least half of all insulin-treated patients.⁹ This condition impairs insulin absorption and has been directly associated with increased glycemic variability and reduced treatment efficacy.¹³ A study by Poudel et al. (2017) in Nepal further highlighted these gaps, showing that 58.1 % of insulin pen users transported insulin without maintaining the cold-chain, and most disposed of used needles inappropriately.¹⁴ These findings underscore the global challenge of improving insulin administration techniques, even with the availability of advanced insulin delivery systems.

Given the prevalence of these errors, targeted educational interventions are crucial. Although insulin instruction is often initiated by physicians or nurses, studies show that guidance from other healthcare professionals, such as nurses, on insulin pen use is often inadequate.¹⁵ Due to heavy workloads and knowledge and skill deficiencies of other health professionals, many patients do not receive structured or repeated training. In this context, pharmacists, as highly accessible medication experts embedded within communities and healthcare systems, are uniquely positioned to address this gap. Their frequent patient interactions, particularly during medication dispensing and counseling, provide natural opportunities for assessment and intervention regarding injection technique.¹⁶ Studies have shown that pharmacist-led interventions, including structured education, demonstration, observation of patient technique, and regular follow-up, can improve various aspects of diabetes management, including medication adherence and glycemic control.¹⁷ Similarly, a randomized controlled trial in Pakistan showed that patients receiving regular pharmacist-led counseling demonstrated better injection techniques, fewer complications such as lipohypertrophy, and greater reductions in HbA1c compared to those receiving standard care.¹⁸ However, a critical literature gap persists. While the prevalence of injection errors with pens is well-documented, and the potential of pharmacists is acknowledged, there is a distinct paucity of robust, controlled studies specifically evaluating the effectiveness of structured, pharmacist-delivered interventions targeted solely at correcting insulin pen injection technique and measuring their direct impact on objectively assessed technique mastery. Many existing studies focus on syringe use, combine multiple interventions, making technique-specific effects unclear, or lack rigorous pre-post or controlled designs specifically isolating the pharmacist's role in technique improvement.

Despite this growing body of international evidence, there remains a significant literature gap in low-resource settings such as Nepal, where pharmacy services in the country are still in their nascent stages, and patients often receive inadequate counseling on insulin administration techniques.^19,20 In developing countries like Nepal, pharmacies are expected to be operated by licensed pharmacists; however, many are run by unqualified personnel, resulting in substandard counseling and poor guidance on proper medicine use, including specialized devices like insulin pens.¹⁹ A recent study in a tertiary care hospital in eastern Nepal revealed that patients were either not counseled or poorly counseled on the use of such devices, leading to incorrect usage and an increase in drug therapy-related problems.²⁰ Pharmaceutical care practice is suboptimal,²¹ and clinical pharmacy practice in Nepal is largely limited to medication dispensing, with minimal involvement in patient counseling or diabetes education.^22,23 As a result, patients often lack access to accurate and practical guidance on insulin administration. According to our literature search, only two studies have assessed insulin injection practices in Nepal, including among health professionals like Nurses, and these studies have revealed serious deficiencies in technique and disposal practices.^14,15 To date, no published research has evaluated the impact of structured pharmacist-led interventions on insulin injection practices in Nepal.

To fulfil the above literature gap, our study aimed to evaluate the impact of pharmacist-led interventions on insulin injection practices among patients with diabetes mellitus in a tertiary care hospital of central Nepal, with the goal of improving administration techniques and supporting better patient outcomes.

2. Methods

2.1. Study design and study site

This study employed a pre-post interventional design to assess the impact of pharmacist-led interventions on insulin injection practices among patients with diabetes mellitus. The pre-post design allowed for the comparison of insulin injection practices before and after the intervention to evaluate the effectiveness of pharmacist-led education and training.

The study was conducted from April 4 to September 30, 2018, at Dhulikhel Hospital, Kathmandu University Hospital, a tertiary care center serving the Kavre district and surrounding regions. The hospital offers comprehensive services, including specialized diabetes care and insulin therapy.

2.2. Study population and criteria

The study population included diabetes patients who had been using insulin pens for at least two weeks. A total of 81 patients were recruited based on the inclusion criteria. These patients were consecutively enrolled during the study period from the Internal Medicine ward and the Outpatient Department (OPD) of Dhulikhel Hospital.

2.3. Inclusion criteria

• •Only patients who had already been using an insulin pen for ≥2 weeks prior to hospitalization were enrolled, ensuring that they were established users.
• •Patients aged 12 years or older.
• •Patients who were willing to participate in the study and provide informed consent.

2.4. Exclusion criteria

• •Diabetic patients who were using insulin delivery methods other than insulin pens (e.g., syringes or pumps).
• •Newly diagnosed diabetes patients who initiated insulin pen use during hospitalization
• •Patients with psychiatric or neurological conditions that could impair comprehension or adherence to study instructions.

2.5. Sample size and sampling techniques

Consecutive sampling was employed to recruit participants who met the inclusion criteria. Patients admitted to the internal medicine ward or visiting the hospital for routine care and already on insulin therapy were considered for the study. Eligible patients were individually assessed and invited to participate.

2.6. Data collection tools

Data collection was carried out using a semi-structured questionnaire and observational checklists. The semi-structured questionnaire gathered demographic information, such as age, sex, education level, and the type of diabetes (Supplementary File S1). The checklist (Supplementary File S2) was designed to assess insulin injection techniques, focusing on aspects like needle storage, site rotation, dosage accuracy, and needle disposal.

In addition to the questionnaire and checklist, a placebo device (insulin pen on a sandbag) was used to observe and assess the patients' technique in real-time. This device allowed for a standardized approach to the assessment of insulin injection practices without the use of actual insulin, ensuring no risk of medication errors.

2.7. Study procedure

Upon enrollment, patients were first assessed for their baseline insulin injection practices using the checklist. The initial assessment took place on the first visit, before any intervention. If patients demonstrated incorrect insulin injection techniques, they were provided with one-on-one education and training by a qualified clinical pharmacist. The training included physical and verbal demonstrations of proper insulin injection techniques, including:

• •Proper mixing of insulin (if applicable).
• •Correct site rotation to avoid lipohypertrophy.
• •Correct needle disposal practices.
• •Cold chain maintenance for insulin storage and transportation.
• •Priming of insulin pens.

Patients and their caregivers were encouraged to ask questions to clarify any doubts regarding the insulin injection process.

After the intervention, patients' techniques were reassessed immediately using the same observational checklist to gauge improvements. Follow-up assessments were conducted two weeks after the intervention to evaluate the sustainability of the improvement in insulin injection practices. In the case of minors (12–18 years), educational interventions were provided simultaneously to both the patients and their caregivers.

2.8. Insulin pen administration practice and its scoring

Insulin pen administration practice comprised 16 questions (Supplementary File S3) evaluating insulin pen administration practices. The questionnaire focused on practical aspects such as: storage of unopened insulin cartridges, storage method for insulin pens, handwashing prior to insulin injection, cleaning of the injection site, mixing of insulin before injection, the technique used for mixing, priming of the insulin pen, identifying appropriate injection sites, systematic rotation of sites, formation of a skin fold during injection, injection angle, needle retention time post-injection, massaging the site after injection, time interval between injection and meal, needle length, and frequency of needle reuse.

Each correct practice was awarded a score of “1” and each incorrect or inappropriate practice received a score of “0.” This binary scoring system allowed calculation of a total practice score for each respondent, with a maximum possible score of 16 and a minimum of 0. The highest, lowest, and median scores obtained by the participants were recorded. The median score was used as a benchmark for comparative analysis between different subgroups (e.g., based on experience, education level, or work setting), offering a robust measure less influenced by outliers.

2.9. Statistical analysis

Statistical analysis was performed using IBM SPSS Statistics Version 20. Descriptive statistics were used to summarize the demographic characteristics and insulin injection techniques. Non-parametric tests were employed for statistical analysis, as the data were not normally distributed. The Friedman's Test was employed to compare the insulin injection technique scores at three different time points: baseline, immediately after the intervention, and two weeks post-intervention. To compare the median scores before and immediately after the intervention, the Wilcoxon Signed Rank Test was used. The Mann–Whitney U Test was applied to compare the baseline scores across different demographic groups, such as age, type of diabetes, and education level. To explore the relationship between demographic factors (e.g., age and type of diabetes) and the accuracy of insulin injection techniques, Spearman's Correlation was utilized. Finally, the Kruskal–Wallis Test was conducted to assess the differences in insulin injection technique scores based on categorical variables like residential area and the source of instruction. A p-value of <0.05 was considered statistically significant for all tests.

2.10. Ethical consideration

Ethical clearance for the study was obtained from the Institutional Review Committee (IRC) of Kathmandu University, School of Medical Sciences. The protocol number for approval was 19/18. Informed consent was obtained from all participants prior to their inclusion in the study. Participants were assured of the confidentiality of their data and their right to withdraw from the study at any time without any consequences.

3. Results

3.1. Sociodemographic characteristics of patients

Of the 81 participants, 69.1 % were male and 88.9 % were from hilly regions. Likewise, 44.4 % lived in urban areas. The predominant age group was 66–75 years (24.7 %). In terms of educational status, 32.1 % of participants were uneducated, while 21 % had completed secondary education, and 19.8 % had received only primary education. A small proportion held a bachelor's degree (7.4 %) or a master's degree and above (4.9 %). Regarding employment status, 53.1 % of participants were unemployed, while 46.9 % were employed. Detailed information is presented in Table 1.

Table 1.

Sociodemographic characteristics of patients and their baseline assessment of scores (N = 81).

Variables	Category	Frequency	Percentage (%)	Below Median f (%)	Median (10) f (%)	Above median f (%)
Age group (Years)	12–21	2	2.5	0 (0)	0 (0)	2 (2.5)
	21–30	6	7.4	0 (0)	0 (0)	6 (7.4)
	30–39	9	11.1	1 (1.2)	3 (3.7)	5 (6.2)
	39–48	6	7.4	2 (2.5)	1 (1.2)	3 (8.8)
	48–57	17	21	11 (13.6)	4 (4.9)	2 (2.5)
	57–66	17	21	9 (11.1)	3 (3.7)	5 (6.2)
	66–75	20	24.7	8 (9.9)	3 (3.7)	9 (11.1)
	75–84	4	4.9	1 (1.2)	1 (1.2)	2 (2.5)
Sex	Male	56	69.1	21 (25.9)	13 (16)	22 (27.2)
	Female	25	30.9	11 (13.6)	2 (2.5)	12 (14.8)
Geographical Distribution	Hilly	72	88.9	29 (35.8)	13 (16)	30 (37)
	Terai	9	11.1	3 (3.7)	2 (2.5)	4 (4.9)
Residential area	Urban	36	44.4	18 (22.2)	3 (3.7)	15 (18.5)
	Semi-rural	24	29.6	5 (6.2)	9 (11.1)	10 (12.3)
	Rural area	21	25.9	9 (11.1)	3 (3.7)	9 (11.1)
Educational status	Uneducated	26	32.1	12 (14.8)	4 (4.9)	10 (12.3)
	Primary education	16	19.8	7 (8.6)	1 (1.2)	8 (9.9)
	Secondary level	17	21	7 (8.6)	3 (3.7)	7 (8.6)
	School leaving certificate	6	7.4	2 (2.5)	1 (1.2)	3 (3.7)
	Intermediate	6	7.4	1 (1.2)	3 (3.7)	2 (2.5)
	Bachelors	6	7.4	2 (2.5)	2 (2.5)	2 (2.5)
	Master's and above	4	4.9	1 (1.2)	1 (1.2)	2 (2.5)
Occupation	Unemployed	43	53.1	17 (21)	6 (7.4)	20 (24.7)
	Employed	38	46.9	15 (18.5)	9 (11.1)	14 (17.3)

Note: Below (<10), Median(10) and Above (>10) classification based on median score; f = Frequency; % = Percentage.

3.2. Baseline clinical characteristics, complications, and needle disposal

Of the 81 patients, 84 % had Type 2 diabetes and 86.4 % self-administered insulin, mostly twice daily (76.5 %). The most commonly used insulin was biphasic human insulin 30/70 (48.1 %). While 63 % were confident in injection technique, 22.2 % were unsure, and 14.8 % lacked confidence. Nurses were the primary source of instruction (70.4 %), and 51.9 % transported insulin without an icepack.

Pain was the most frequently reported complication (23.5 %), followed by bleeding (17.3 %), lipohypertrophy (4.9 %), and burning or inflammation (3.7 %). A notable proportion reported combined symptoms such as pain with bleeding (11.1 %), lipohypertrophy with pain and bleeding (6.2 %), and other mixed presentations like swelling, itching, or infection (1.2 % each).

Needle disposal practices varied. The most reported method was discarding needles into dustbins and transferring to municipal vehicles (40.7 %), followed by disposal in household waste (34.6 %). However, unsafe disposal was also reported, including dumping into bushes (13.6 %) and rivers (4.9 %). Only a few followed safer methods like returning to hospital waste systems (4.9 %) or incineration (1.2 %). Detailed information is presented in Table 2.

Table 2.

Baseline clinical characteristics of patients and their score category (N = 81).

Variables	Category	Frequency	Percentage (%)	Below Median f (%)	Median (10) f (%)	Above median f (%)
Types of diabetes	Type 1 diabetes	13	16	1 (1.2)	0 (0)	12 (14.8)
	Type 2 diabetes	68	84	31 (38.3)	15 (18.5)	22 (27.2)
Administration by	Self	70	86.4	29 (35.8)	13 (16)	28 (34.6)
	Caretaker mediated	11	13.6	3 (3.7)	2 (2.5)	6 (7.4)
Frequency of administration	1	16	19.8	5 (6.2)	5 (6.2)	6 (7.4)
	2	62	76.5	26 (32.1)	9 (11.1)	27 (33.3)
	3	2	2.5	1 (1.2)	1 (1.2)	0 (0)
	4	1	1.2	0 (0)	0 (0)	1 (1.2)
Types of insulin used	Biphasic human insulin 30/70	39	48.1	16 (19.8)	7 (8.6)	16 (19.8)
	Glargine	16	19.8	3 (3.7)	6 (7.4)	7 (8.6)
	Biphasic insulin aspart 30/70	16	19.8	9 (11.1)	1 (1.2)	6 (7.4)
	Biphasic insulin lispro 25	5	6.2	0 (0)	1 (1.2)	4 (4.9)
	Biphasic insulin lispro 50/50	3	3.7	3 (3.7)	0 (0)	0 (0)
	Insulin aspart & insulin degludec	1	1.2	1 (1.2)	0 (0)	0 (0)
	Insulin aspart & Glargine	1	1.2	0 (0)	0 (0)	1 (1.2)
Confident on techniques	No	12	14.8	4 (4.9)	3 (3.7)	5 (6.2)
	Not sure	18	22.2	7 (8.6)	2 (2.5)	9 (11.1)
	Yes	51	63	21 (25.9)	10 (12.3)	20 (24.7)
Source of Instruction	Nurses	57	70.4	25 (30.9)	11 (13.6)	21 (25.9)
	Doctor	17	21	6 (7.4)	3 (3.7)	8 (9.9)
	Pharmacy personnel	1	1.2	0 (0)	1 (1.2)	0 (0)
	Nonprofessionals/others	6	7.4	1 (1.2)	0 (0)	5 (6.2)
Method of transportation	Icepack box	39	48.1	14 (17.3)	7 (8.6)	18 (22.2)
	Without icepack box	42	51.9	18 (22.2)	8 (9.9)	16 (19.8)
Complications associated with insulin injection	Pain	19	23.5	6 (7.4)	6 (7.4)	7 (8.6)
	Bleeding	14	17.3	6 (7.4)	3 (3.7)	5 (6.2)
	Lipohypertrophy	4	4.9	1 (1.2)	0 (0)	3 (3.7)
	Burning & inflammation	3	3.7	0 (0)	1 (1.2)	2 (2.5)
	Infection	1	1.2	1 (1.2)	0 (0)	0 (0)
	Pain & bleeding	9	11.1	5 (6.2)	0 (0)	4 (4.9)
	Lipohypertrophy, pain & bleeding	5	6.2	4 (4.9)	1 (1.2)	0 (0)
	Pain & burning	3	3.7	2 (2.5)	0 (0)	1 (1.2)
	Pain, swelling and itching	1	1.2	0 (0)	0 (0)	1 (1.2)
	Burning, pain & bleeding	1	1.2	1 (1.2)	0 (0)	0 (0)
Needle disposal	Collecting in dust bin and throwing in municipality vehicle	33	40.7	15 (18.5)	4 (4.9)	14 (17.3)
	Collecting in dust bin and throwing in waste disposal	28	34.6	10 (12.3)	6 (7.4)	12 (14.8)
	Collection in dust bin and throwing in bush	11	13.6	6 (7.4)	3 (3.7)	2 (2.5)
	Collection in dust bin and throwing in river	4	4.9	1 (1.2)	2 (2.5)	1 (1.2)
	Incineration by self	1	1.2	0 (0)	0 (0)	1 (1.2)
	Returning to hospital waste disposal	4	4.9	0 (0)	0 (0)	4 (4.9)

Note: Below (<10), Median(10) and Above (>10) classification based on median score; f = Frequency; % = Percentage.

3.3. Baseline injection technique assessment

At baseline, patients demonstrated several deficiencies in insulin pen injection practices. While 77.8 % stored insulin cartridges appropriately in the refrigerator, only 49.4 % stored the insulin pen at room temperature as recommended, with another 49.4 % incorrectly refrigerating it. Critical hygienic steps were poorly practiced: only 46.9 % of participants performed hand washing, and 45.7 % cleaned the injection site before administration. Correct mixing of cloudy insulin was reported by 75.3 %, yet only 7.8 % executed the correct mixing method. Similarly, priming of pens was done by only 46.9 %. Site rotation was significantly inadequate; only 21 % of participants followed systematic rotation. Injection angle and site selection were relatively appropriate, with 95.1 % using a 90-degree angle and 88.9 % using the abdomen. However, only 8.6 % held the pen for the optimal 10 s post-injection, and a mere 34.6 % reported the most appropriate injection-to-meal gap of 20–30 min. Needle reuse was prevalent, with many reusing needles beyond the recommended single-use. The most common reuse frequency was up to >6 times, highlighting significant practice gaps. Detailed information is presented in Table 3.

Table 3.

Comparison of correct insulin injection technique: Pre-intervention (n = 81), immediately after intervention (n = 81), and two weeks after intervention (n = 65).

Questionnaires (Check List)	Response Category	Pre-intervention f(%) (n = 81)	Immediate intervention f(%) (n = 81)	After 2 weeks interventions f(%) (n = 65)	Correct response
Q1 storage of cartridge	Refrigerator	63(77.8)	74(91.4)	55(88.6)	Refrigerator and Earthen pot
	Room temperature	8(9.9)	–	3(3.7)
	Earthen pot	5(6.2)	7(8.6)	7(10.8)
	Deep freeze in refrigerator	5(6.2)	–	–
Q2 Storage of Pen	Refrigerator	40(49.4)	4(4.9)	19(29.2)	Room temperature (20–250C)
	Room temperature	40(49.4)	77(95.1)	46(70.8)
	Earthen pot	1(1.2)	–	–
Q3 Hand washing	Yes	38 (46.9)	81(100)	60(92.3)	Wash with bathing soap or hand sanitizer liquid
	No	43(53.1)	–	5(7.7)
Q4 Site cleaning	Yes	37(45.7)	81(100)	59(90.8)	Site cleaning with spirit or boiled Lukewarm water
	No	44(54.3)	–	6(9.20
Q5 Insulin Mixing	Yes	61(75.3)	64(79.0)	53(81.5)	Insulin should be mixed for cloudy insulin
	No	20(24.7)	17(21)	12(18)
Q6 Mixing Method	Correct	5(7.8)	64(100)	43(81.1)	Insulin should be mixed gentle horizontally rolling within the palm of hands and tip up and down for 10 times
	Incorrect	59(92.2)	–	10(18.9)
Q7 Priming pen	Yes	38(46.9)	81(100)	61(93.8)	Priming pen is done by throwing 2–3 drops of insulin outside the needle by pushing the plunger
	No	43(53.1)	–	4(6.2)
Q8 Injection site	Abdomen	72(88.9)	73(90.1)	56(86.2)	Injection site depends on the nature of insulin which can be Abdomen, Thigh, buttock, arms
	Thigh	1(1.2)	1(1.2)	1(1.5)
	Abdomen and thigh	8(9.9)	7(8.6)	8(12.3)
Q9 Systematic site rotation	Yes	17(21.0)	81(100)	50(76.9)	Dividing the site into 4 quadrants and rotation of each site in clockwise direction with 1-2 cm apart from umbilical cord if abdomen.
	No	64(79.0)	–	15(23.1)
Q10 Skin lifting	Yes	71(87.7)	81(100)	65(100)	Done by stretching and loosely lifting the skin by making folds
	No	10(12.3)	–	–
Q11 Injection angle	Nearly 90	77(95.1)	81(100)	64(98.5)	Nearly 90 degrees in subcutaneous fat with skin folds
	Nearly 45	4(4.9)	–	1(1.5)
Q12 Holding time (sec)	<10	35 (43.2)	0 (0)	6 (9.2)	Minimum 10 s or counting 10 slowly, holding time can be increased as per the insulin higher unit doses
	10	7 (8.6)	17 (21)	16 (24.6)
	>10	39 (48.1)	64 (79)	43 (66.2)
		Md: 10, IQR:(20−0) =20	Md: 20, IQR:(20–15) =5	Md: 20, IQR:(20−10) = 10
Q13 Injection site massage	Yes	21(25.9)	–	6(9.2)	No for injection site massage
	No	60(74.1)	81(100)	59(90.8)
Q14 Injection to meal gap (min)	<20	52 (64.2)	40 (49.4)	34 (52.3)	Time Depends on type of insulin injected
	20–30	28 (34.6)	40 (49.4)	31 (47.7)
	>30	1 (1.2)	1 (1.2)	0 (0)
		Md: 10, IQR:(30−2) =28	Md: 20, IQR:(30–2) =28	Md: 15, IQR:(30–2) =28
Q15 needle length	4 mm	28(34.5)	28(34.6)	40(61.5)	Most preferable is 4 mm, 5 mm also can be used
	5 mm	48(59.3)	49(60.5)	22(33.9)
	6 mm	5(6.20)	4(4.9)	3(4.6)
Q16 Needle reuse frequency	<2	0 (0)	1(1.4)	0 (0)	Single use disposable needle
	2–6	29 (35.8)	63 (88.7)	46 (70.8)
	>6	52 (64.2)	7 (9.9)	19 (29.2)
		Md: 10, IQR:(14–6) =8	Md: 5, IQR:(6–4) =2	Md: 6, IQR:(8–4) =4

Note: f = Frequency, % = Percentage, M_d: Median, IQR(Q₃-Q₁) = Interquartile range.

3.4. Immediate post-intervention evaluation

Following immediate pharmacist intervention, a marked improvement was observed across nearly all parameters. Hand washing, site cleaning, priming of the pen, skin lifting, systematic site rotation, and correct injection angle all reached 100 % compliance. Appropriate mixing technique surged from 7.8 % to 100 %, indicating a high impact of the intervention. Storage of insulin cartridges improved to 91.4 % proper refrigeration, and all improper deep-freezing practices were eliminated. However, some confusion remained about pen storage, with 4.9 % storing it in a refrigerator. Regarding injection timing, 49.4 % achieved the optimal 20–30 min gap. Additionally, the percentage of patients using the correct holding time (≥10 s) increased notably, with 79 % holding for >10 s and 21 % for 10 s. Needle reuse behavior showed a shift: most reuse frequencies moved to the 2–6 range, indicating partial behavioral change. However, single-use adherence was still absent. Detailed information is presented in Table 3.

3.5. Two-week post-intervention follow-up

Among 81 patients, sixteen patients were lost to follow-up due to non-response to calls, migration, or scheduling constraints, while 65 patients were successfully reassessed at the two-week follow-up point. At two weeks, sustained improvements were evident. Correct practices such as hand washing (92.3 %), site cleaning (90.8 %), priming (93.8 %), and systematic site rotation (76.9 %) remained high, though a slight decline from immediate post-intervention levels was noted. The proportion of patients storing the insulin pen at room temperature revealed some decline to 70.8 %. Correct mixing of insulin was sustained (81.5 %), although the mixing technique dropped slightly from 100 % to 81.1 %, indicating some decline in technical retention. Injection angle and site remained consistently correct (98.5 % at 90 degrees and 86.2 % using the abdomen). Holding time ≥ 10 s was reported by 66.2 % and 24.6 % (10 s), showing retention of proper injection duration. The optimal injection-to-meal gap (20–30 min) was maintained by 47.7 %, indicating behavioral consistency. Needle reuse frequency remained high, but moderate levels (2–6 times) became more common (70.8 %), suggesting incremental behavioral change post-intervention. Detailed information is presented in Table 3.

3.6. Association between socio-demographic factors and injection technique scores

Significant associations were observed between certain socio-demographic variables and insulin injection technique scores. At the pre-intervention stage, type of diabetes (p = 0.001), age (p = 0.04), duration of diabetes (p = 0.01), and duration of therapy (p = 0.01) showed statistically significant correlations with baseline scores. Participants with Type 1 diabetes demonstrated better baseline technique.

Following immediate intervention, residential area (p = 0.02), duration of therapy (p = 0.02), and transportation with icepack (p = 0.01) were significantly associated with improved scores. Urban residents showed higher immediate post-intervention improvements.

At the two-week follow-up, sustained improvement remained significantly associated with age (p = 0.04), residential area (p = 0.02), administration method (p = 0.03), and current insulin type (p = 0.01). Self-administering participants and urban residents continued to show higher technique retention, as shown in Table 4.

Table 4.

Association of socio-demographic with pre-intervention score (n = 81), immediate intervention score (n = 81), and post-intervention score after 2 weeks (n = 65).

Socio demographic variables	Category	Pre-interventional score Median (IQR)/R	P value	Post-interventional score immediately after intervention Median (IQR)/R	P value	Post-interventional score after 2 weeks of intervention Median (IQR)/R	P value
GenderU	Male	10(9–11.75)	0.95	15(15–15)	0.95	14(15–13)	0.88
	Female	10(9–11.5)		15(15–15)		14(15–13)
GeographicalU	Hilly	10(11.75–9)	0.552	15(15–15)	0.55	14(15–13)	0.55
	Terai	10(11.5–9)		15(15–15)		14(15–13)
Ageρ		R = −0.21	0.04*	R = 0.05	0.06	R = 0.06	0.04*
Residential AreaH	Urban	9.5(9–11)	0.35	15(15–15)	0.02*	14(15–12)	0.02*
	Semi urban	10 (10−12)		15(15–15)		15(15–14)
	Rural	10 (8–11)		15(15–15)		14(15–13)
Educational statusH	Uneducated	10(11.25–8)	0.95	15(15–15)	0.26	14(15–13)	0.26
	Primary	10.5(12–9)		15(15–15)		14.5(15–14)
	Secondary	10(11.5–9)		15(15–15)		14(15–13)
	SLC	10.5(11–8.75)		15(15–14.75)		14(14–12)
	Intermediate	10(12.5–9.25)		15(15–15)		13(14.25–11.75)
	Bachelors	10(11.75–9)		15(15–15)		15(15–15)
	Master's and above	10.5(11.75–8.5)		15(15–14.25)		12(12−12)
OccupationU	Unemployed	10(12–9)	0.7	15(15–15)	0.08	14(15–13)	0.17
	Employed	10(11–9)		15(15–15)		14(15–12)
Type of diabetes U	Type 1	12(12.5–11)	0.001*	15(15–15)	0.12	14.5(15–13)	0.56
	Type 2	10(11–9)		15(15–15)		14(15–13)
Duration of diabetesρ		R = -0.31	0.01*	R = -0.176	0.11	R = -0.03	0.81
Duration of therapyρ		R = 0.218	0.01*	R = -0.24	0.02*	R = -0.23	0.05
Administration U	Self	10(11–9)	0.38	15(15–15)	0.5	14(15–13)	0.03*
	Caregiver mediated	11(12–9)		15(15–15)		15(15–14)
Current insulinH	Biphasic Human Insulin 30/70	10(11–9)	0.05	15(15–15)	0.001*	14(15–12)	0.01*
	Glargine	10(12−10)		15(15–15)		14.5(15–13)
	Biphasic Insulin aspart 30/70	9(12–9)		15(15–15)		14(15–13)
	Biphasic insulin lispro 25	11(11−10)		15(15–15)		13(13−13)
	Biphasic insulin lispro 50/50	8(0–5)		14(0–14)		14(0−12)
	Insulin aspart and insulin degludec	8(8–8)		15(15–15)		14(14–14)
	Insulin aspart and glargine	11(11−11)		14(14–14)		14(14–14)
Maximum frequency of insulinρ		R = -0.058	0.65	R = 0.049	0.66	R = 0.02	0.87
Confidence on insulin inj. TechniqueH	No	10(11–9)	0.7	15(15–15)	0.433	14(15–13)	0.433
	Yes	10(11–9)		15(15–15)		14(15–13)
	Not sure	10.5(12–8.75)		15(15–15)		14(15–12)
Instruction sourceH	Doctor	10(11.5–8.5)	0.5	15(15–15)	0.93	14(14–12)	0.86
	Pharmacy personnel	10(10−10)		15(15–15)		14(14–14)
	Nurse	10(11.5–9)		15(15–15)		14(15–13)
	Others	11(12–10.5)		15(15–15)		15(15–13.75)
Anxiety and fearH	At the beginning	10(12–9)	0.22	15(15–15)	0.088	15(15–13.5)	0.08
	Sometimes	10(12–10)		15(15–15)		15(15–12)
	Always	8.5(0–8)		15(15–15)		14(14–14)
	Never	10(11–9)		15(15–15)		14(15–12)
Transportation U	Icepack box	10(12–9)	0.55	15(15–15)	0.01*	14.5(15–14)	0.08
	Without icepack box	10(11–9)		15(15–15)		14(15–12)
Complication U	No	11(11.5–9)	0.46	15(15–15)	1	14(15–12)	0.7
	Yes	10(11.75–9)		15(15–15)		14(15–12.75)

Note: IQR: Interquartile Range; R: Range; U: Mann-Whitney U test for 2 categories; H: Kruskal-Wallis Test for more than 2 categories; ρ: Spearman's correlation coefficient; *Significant at p < 0.05.

3.7. Overall effects of pharmacist intervention on insulin injection technique

Friedman's test was used to compare the median scores of insulin injection technique at three different time points: pre-intervention (baseline), immediately after the intervention, and two weeks post-intervention. A statistically significant difference in scores was observed across the three time points (Friedman's test, p < 0.001), suggesting that the pharmacist-led intervention significantly improved insulin administration practices among participants, which is shown in Table 5. Detailed information about Post-hoc Pairwise Comparisons can be found in Supplementary File S4.

Table 5.

Effects of Pharmacist Intervention on Insulin Injection Technique (n = 65).

Assessment Time Point	Median (IQR) Score	P-value
Pre-intervention	10 (9–11.5)
Immediate post-intervention	15 (15–15)	p < 0.001*
2 weeks post-intervention	14 (13–15)

Note: *Significant at p < 0.05.

4. Discussion

Our study found that unsafe insulin injection practices were common at baseline, including poor hygiene, storage, and disposal. A brief pharmacist-led intervention produced large improvements in these practices. However, some errors were more frequent in older patients and those from rural areas, highlighting groups that may need additional support.

In this study, unsafe needle disposal practices among patients were found to be concerning. A large number reported discarding used needles by first collecting them in dustbins and then disposing of them through municipal waste systems, with others admitting to discarding them in bushes or even rivers. These findings are consistent with a study by Patil et al., in which the vast majority of patients disposed of needles and syringes directly into general garbage or public drainage systems.²⁴ Studies have revealed that inappropriate disposal of insulin needles is a common problem not only in developing countries but also in developed countries.¹⁰ Ideally, used needles should be collected in designated sharps containers and returned to health facilities or primary health care centers for safe disposal. Unsafe disposal of insulin needles, such as discarding them in household waste, bushes, or rivers, poses serious public health risks, including needle-stick injuries to sanitation workers, environmental contamination, and the spread of blood-borne infections such as hepatitis B, hepatitis C, and HIV, thereby underscoring the urgent need for improved patient education on biomedical waste management.²⁵

A significant share of insulin users in this study were from urban areas, likely reflecting the sedentary lifestyles and unhealthy dietary habits that are more prevalent in urban settings. This observation aligns with previous studies, such as one by Patil et al., which also reported a majority of urban insulin users.²⁴ Educational status was also found to influence insulin usage practices. Nearly one-third of patients in this study were uneducated, mirroring findings from Kapoor et al., where a large portion of patients had limited literacy.²⁶ A poor educational background likely contributes to suboptimal injection techniques and a lack of awareness about safe practices.

Type 2 diabetes predominated among the study population, as observed in other studies, including that of Frid et al..²⁷ Interestingly, despite a majority of participants expressing confidence in their insulin injection technique, actual practices were often inadequate. This disconnect between perceived and actual competency has also been documented by Mitchell et al., suggesting that patient confidence alone should not be relied upon as a measure of injection adequacy.²⁸ Despite pharmacists being well-positioned to provide expert counseling, this study found that only a negligible number of patients had received insulin education from pharmacy personnel. Instead, most were educated by nurses, indicating a missed opportunity to involve pharmacists more actively in diabetes management. This underutilization of pharmacists in patient education was similarly noted by Mitchell et al., and suggests a need to redefine roles in multidisciplinary diabetes care teams.²⁸

Cold chain maintenance during insulin transportation was also found to be poor, with more than half of the patients failing to ensure appropriate storage conditions. These findings closely mirror those of Poudel et al.¹⁴ and Patil et al.,²⁴ and highlight a widespread gap in knowledge regarding insulin stability. Poor cold chain adherence can compromise insulin potency and, ultimately, glycemic control. Injection-related complications such as pain, bleeding, and lipohypertrophy were also reported, aligning with previously documented studies. For instance, Patil et al. reported similar issues, including swelling suggestive of lipohypertrophy.²⁴ These complications are often preventable with proper injection techniques, further supporting the need for comprehensive and sustained patient education.

The pharmacist-led intervention in this study led to marked improvements in several key aspects of insulin injection practices, particularly regarding insulin storage and hygiene measures. Prior to the intervention, most patients reported storing insulin cartridges in refrigerators, and this proportion increased further after the educational session and remained high at the two-week follow-up. This reflects greater awareness compared to earlier findings by Poudel et al. (2017), which had highlighted suboptimal cold chain maintenance.¹⁴ In contrast, inappropriate practices such as storing insulin in deep freezers, which were observed initially in a small fraction of patients, were completely eliminated following the intervention. However, the improper refrigeration of in-use insulin pens remained an issue. Although the number of patients refrigerating pens declined notably over two weeks, a significant proportion still continued the practice, contrary to global recommendations, which advise against refrigeration due to the risks of air bubble formation, impaired insulin flow, and pain associated with cold insulin.^9,29 Although more patients began storing in-use pens at room temperature, this practice may be problematic in Nepal's hot climate due to insulin instability.³⁰ This gap highlights the need for localized solutions such as earthen pots or cooling bags.

In terms of technique, hygiene practices improved substantially. The number of patients performing handwashing and cleaning the injection site before administration increased markedly, exceeding levels reported by Kapoor et al. (2016), where such basic measures were frequently neglected.²⁶ Similarly, consistent site rotation became much more common after the intervention, aligning with international recommendations to prevent complications such as lipohypertrophy—a condition observed in a few patients in this study, and also reported by Patil et al. (2017).²⁴ Priming of pens, which was poorly understood initially and often only performed during cartridge or needle changes, became routine for most patients after intervention. A similar trend was observed in the mixing of premix insulin, which saw a dramatic improvement following the educational session. These enhancements echo the findings of Forough et al. ³¹ and Kapoor et al.²⁶ both of whom reported significant technique improvements post-education.

Other aspects, such as the proper use of injection angle and skin-lifting technique, showed modest improvement, as the majority of patients were already demonstrating acceptable practices in these areas. However, needle reuse remained universal, a finding consistent with earlier studies by Tandon et al. (2015) ³² and Frid et al. (2016),⁹ which identified economic constraints as the main barrier to single-use compliance. Despite international consensus discouraging reuse due to safety and efficacy concerns, the absence of reimbursement policies and limited affordability likely contribute to this persistent behavior in Nepal. The intervention did, however, help reduce other harmful practices, including massaging the injection site and removing the needle too quickly after injection; improvements that align with current insulin delivery guidelines.^9,33

Several demographic factors were found to influence the effectiveness of the intervention. Age, type of diabetes, duration of diabetes, and duration of insulin therapy were significantly associated with baseline insulin injection practices. Younger patients, as well as those with a shorter duration of diabetes, tended to have better injection techniques initially, which aligns with findings from other studies.³⁴ This could reflect a greater adaptability to new instructions or fewer entrenched habits. In contrast, older patients and those with a longer duration of diabetes may face more challenges in adapting to correct techniques, possibly due to entrenched habits or cognitive impairments related to aging. The intervention was also more effective in patients from urban areas, who were more likely to have received detailed training and education from pharmacists due to the availability of many pharmacies and clinics in urban areas. This may reflect differences in healthcare access and educational resources between urban and rural populations. Similar findings have been reported in studies conducted in other regions, where urban dwellers had better access to healthcare and diabetes education.¹⁴ Furthermore, in some domains of injection practice where hands-on training and demonstrations by pharmacists were possible, such as insulin pen priming, patients showed greater improvement in technique compared to domains where only verbal instruction was provided, such as cartridge and insulin pen storage, suggesting that practical demonstrations are more effective than verbal guidance alone.

While the improvement in insulin injection techniques was significant immediately after the intervention, a slight deterioration was observed during the two-week follow-up. This indicates that while the intervention led to short-term improvements, the sustainability of these changes may be a concern. Several studies have shown that without reinforcement, patient behaviors tend to revert to previous practices over time.¹⁰ This highlights the importance of regular follow-ups and reinforcement sessions to maintain the improvements achieved. Given this, reinforcement of pharmacist-led education should ideally be incorporated into routine follow-up visits every 2–3 months. This timing is both practical within clinical workflows and clinically meaningful, as it coincides with HbA1c monitoring intervals, thereby supporting sustained improvements in insulin injection practices. Periodic reminders, either through phone calls or follow-up visits, could further reinforce proper techniques and address any emerging issues.

Our study demonstrated that improper insulin injection practices, such as unsafe disposal, poor site rotation, and needle reuse, were highly prevalent among patients. Pharmacist-led education markedly improved these practices, particularly in hygiene measures, site rotation, and insulin storage. These behavioral changes are clinically relevant because improper injection technique is known to impair insulin absorption, contribute to lipohypertrophy, and lead to unstable glycemic control.^11,35 Randomized trials have shown that structured injection training can reduce HbA1c by about 1 % and lower fasting plasma glucose within six months.¹¹ Although our study did not measure blood glucose levels, since glycemia is strongly affected by diet, exercise, and comorbidities beyond the scope of our study, it is reasonable to expect that sustained improvements in injection technique would enhance glycemic control and reduce long-term risks of complications such as retinopathy, nephropathy, and cardiovascular disease, as supported by landmark studies linking HbA1c reduction to fewer microvascular events.^11,35 In low-resource settings like Nepal, where access to advanced therapies and regular monitoring is limited, improving basic practices such as injection technique offers a practical and cost-effective strategy to strengthen diabetes care. Thus, integrating pharmacist-led interventions into routine diabetes education could play an important role in improving both patient safety and long-term outcomes.

5. Strengths and limitations

This study has several strengths, including its practical pharmacist-led intervention in a real-world clinical setting, comprehensive assessment of insulin injection techniques, and rigorous statistical analysis. The focus on pharmacist-led education fills a gap in diabetes care in Nepal, where insulin therapy is often suboptimal. However, the study has limitations, such as being conducted at a single center, which limits generalizability, and a short two-week follow-up period that may not reflect long-term outcomes. Self-reported data on cold chain maintenance and needle disposal may introduce bias, and the sample size could be larger to improve statistical power. The absence of a control group limits causal inferences, and unmeasured confounding factors may have influenced the results. Similarly, patients who demonstrated correct injection techniques at baseline did not receive full demonstration-based training, which may pose a risk of technique deterioration over time and could not be fully accounted for in this study. Despite these limitations, the study provides valuable evidence supporting the role of pharmacists in improving insulin injection practices and highlights the need for further research with larger sample sizes, longer follow-ups, and multi-center designs.

6. Conclusion

This study demonstrates that pharmacist-led education significantly improves insulin injection practices among patients with diabetes mellitus. The intervention led to notable improvements in key areas such as insulin mixing, site rotation, needle disposal, and cold chain maintenance. However, while short-term improvements were achieved, the sustainability of these changes over time remains a challenge, highlighting the need for continued reinforcement through follow-up education and periodic training. The findings emphasize the critical role of pharmacists in diabetes management, particularly in settings where insulin therapy is underutilized, and proper injection techniques are lacking. Future research with larger sample sizes, longer follow-up periods, and multi-center designs is necessary to confirm the long-term impact of pharmacist interventions and to explore strategies for sustaining these improvements in insulin injection practices. Integrating pharmacist-led education into standard care may significantly improve outcomes and reduce diabetes-related complications, and improve the overall quality of life for patients with diabetes.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Availability of data and materials

The data that support the findings of this study are available on reasonable request from the corresponding author.

Clinical trial

Not applicable.

Consent for publication

Not applicable.

Ethics approval and consent to participate

Ethical approval was obtained from the Institutional Review Committee (IRC) of Kathmandu University, School of Medical Sciences (Ref no: 19/18). Institutional permission was also granted by the hospital administration of Dhulikhel Hospital before commencing the study. All participants provided written informed consent prior to data collection, with participation being voluntary. Confidentiality and patient rights were assured throughout the study.

CRediT authorship contribution statement

Bidur Sharma: Writing – review & editing, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Devindra Kumar Neupane: Writing – review & editing, Visualization, Validation, Methodology, Investigation, Formal analysis, Conceptualization. Rahi Bikram Thapa: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis. Rajani Shakya: Visualization, Validation, Supervision, Methodology, Investigation, Formal analysis. Rojeena Koju Shrestha: Visualization, Validation, Supervision, Methodology, Investigation, Formal analysis. Pooja Rimal: Visualization, Validation, Methodology, Investigation.

Declaration of competing interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Appendix A. Supplementary data

Supplementary Material 1

Supplementary Material 2

Supplementary Material 3

Supplementary Material 4