Study area and period
Dessie Comprehensive Specialized Hospital, located in Northeast Ethiopia, is a key health facility providing specialized medical services to the region. It serves as a referral center for various chronic conditions, including diabetes. The hospital offers regular follow-up services for patients with DM, aiming to manage their condition effectively through multidisciplinary approaches involving endocrinologists, dietitians, and diabetes educators.
The study was conducted from July 1 to 30/2023.
Study design and participant characteristics
An institution-based cross-sectional study was employed at Dessie Comprehensive Specialized Hospital, Northeast Ethiopia, 2023. All diabetes mellitus patients on regular follow-up at the diabetes clinic in Dessie Comprehensive Specialized Hospital 2023 who were available during the study period were included.
Sample size determination and sampling procedure
The sample size was determined by using the single population proportion formula by considering 36.1% of patients reported essential medicines as affordable in the Ethiopian study [13]. The sample size for the secondary objective has been calculated, and the highest final sample size was taken. The margin of error is 5% at a 95% confidence level by adding 10% of respondents.
$$\begin{aligned}\:n=&\frac{\left(\left({{Z}_{\alpha\:/2})}^{2}p\right(1-p)\right)}{{d}^{2}}=\:\frac{\left(({1.96)}^{2}\:\times\:0.361(1-0.361)\right)}{{\left(0.05\right)}^{2}}\\&=355.5\hspace{0.17em}\sim\hspace{0.17em}356\end{aligned}$$
Non-response rate: 10% = 36,
Final sample size (n) = 356 + 36 = 392.
Where;
-
n = sample size.
-
Z α/2: Level of confidence at 95%.
-
P = population proportion = 36.1%.
-
d: Margin of error = 5%.
After all, the final sample size for this study is 392.
Study Participants from the DCSH diabetes mellitus clinic were selected using systematic random sampling. The average number of diabetes mellitus patients on regular follow-up at the diabetes mellitus clinic daily was estimated at 30 patients, and two months were used for data collection (44 working days). The total number of diabetes mellitus patients in the study period was 1320. The K interval was determined by dividing the total diabetes mellitus patients on follow-up by the final sample size of N/n (1320/392), which enabled us to select participants every 4th unit. The starting number was 2 after being drawn randomly, and then participants selected every 4th interval (2, 6, 10, and 14).
Data collection procedure and quality control
The data was gained from face-to-face participants’ interviews with Amharic speakers from July 1 to 30, 2023.
The data collection tool was translated into Amharic and then back to English to ensure its consistency. The tool was pre-tested on 10% of nurses at Woldia Comprehensive Specialized Hospital for appropriateness two weeks before the actual data collection. Two BSc nurses were trained as data collectors. The questionnaire has three parts: the first is socio-economic demographic characteristics; the second contains questions about disease and treatment-related factors; and the third is the affordability of healthcare. The tool is adapted from different studies [13,14,15,16,17]. During data collection, the supervisor followed the day-to-day data collection process closely and ensured the completeness and consistency of the interview checklist each day before transferring it into computer software. Problems concerned with data collection were corrected early, and non-overlapping numerical codes were given for each question to enter in Epi-Data Manager version 4.6.
Data processing and analysis
After data collection, the data was checked and entered into a computer using Epi Data version 4.6 and exported to SPSS version 26 for data analysis. Descriptive statistics, including frequencies and percentages, were done for all variables. Initially, univariable logistic regression analysis was carried out to see the association between the outcome and each explanatory variable, and then multivariable logistic regression analysis was computed. Multicollinearity among variables was evaluated using VIF (variables found with VIF < 10), and the Hosmer and model fitness were checked using the HosmeLemeshow test, which was found insignificant (0.44). Variables with P-values < 0.05 in multivariable logistic regression analysis were used to declare a significant association.
Variables and operational definition
Affordability of health service was the outcome variable measured as affordable health service: after computing the sum of the five (0–4) Likert scale-based responses of participants, health service is said to be affordable for patients with a score of mean and below the mean score (4 ± 2.4) of computed values [18]. And non-affordable health service: a health service said to be non-affordable if the patient scores above the mean score (4 ± 2.4) of computed values [18].
Socio-demographic characteristics (age, sex, marital status, level of education, employment status, religion, residence, family size, occupation), economic-related factors ( source of finance for care, average transport cost, average expenditure of follow-up, monthly income, monthly follow-up), payment methods (CBHI, OOP, others), and disease- and treatment-related factors (classification of drugs, duration of the disease, family history of DM, types of DM, comorbidity, number of drugs prescribed, additional drugs used, nature of drugs prescription) were the independent variables that affect the outcome variable.
Essential medicines are those that satisfy the priority healthcare needs of a population. They are selected with due regard to disease prevalence, public health relevance, evidence of efficacy and safety, and comparative cost-effectiveness. They are intended to be available in functioning health systems at all times, in appropriate dosage forms, of assured quality, and at prices individuals and health systems can afford [19].
link