Assessing the prevalence of non-prescribed medicines at home and associated factors: a community-based study – BMC Public Health

Educational settings, planning, and timing

A cross-sectional community study was conducted in the community of Gondar, located in the Amhara Region of northwestern Ethiopia. Gondar City is located 728 km from Addis Ababa, the capital of Ethiopia. This city is among the oldest and most populated cities in the country. According to the Central Statistical Agency (CSA) report in 2021, the total population of Gondar city was 443,156. [25]. The city of Gondar is characterized by rapid population growth, driven by natural increase and migration from rural to urban areas. At the time of the study, the city administration reported a population of 454,445 for the financial year 2021/2022. The city has six sub-towns and 36 Kebele administrative centers [26]. The study was conducted from July 30 to August 30, 2023. A total of 847 families were selected, which corresponds to the total number of participants included in the study.

Citizens

The population source for this study was all residents of Gondar city while the study population consisted of residents living in selected households.

Inclusion and exclusion criteria

People above 18 years of age, of both sexes, who were the main persons responsible for obtaining unused medicines in the selected households were included. Participants who refused to allow observation of unused medication package labels or medications themselves were excluded from the study.

Sample size determination and sampling technique

A sample size of 847 was calculated using the design for one section of the population, based on an earlier study in Arba Minch town where 49.4% of households had unused medicines. [22]. This calculation included a 95% confidence interval, a 5% margin of error, and an additional 10% for the non-response rate. In addition, the effect of plan 2 was added to the final sample size by multiplying the result. Due to the similarity of the population, the researchers selected nine Kebeles (Bilajig, Arebegnoch, Teda, Fasiledes, Kerkos, Ayra, Abajale, Angereb and Hidase) from 36 Kebeles to be included in the study, using a lottery. method. A total sample size of 847 was allocated to each of the nine selected Kebeles through stratified sampling. Within each Kebele, samples were selected according to population, with the number of households serving as the sampling frame. A systematic random sampling method was used to select households from nine Kebeles. The sampling period for each Kebele was established by dividing the total number of households in that Kebele by the size of the allocated sample. Within each Kebele, only one street was designated, and the family with the lowest address on that street was the starting point. After that, K^th the street family was interviewed, ​​following a fixed order (Figure 1).

Data Collection Tool and Data Collection Method

The data collection instrument used in this study was adapted from previous research [15, 21, 22, 27]. The questionnaire consisted of two parts – Part 1 covered social and demographic information, while Part 2 assessed the presence of non-use drugs, family health status, and information about stored medicines. A single yes/no question was used to determine whether there were many medications not being used, similar to previous studies. [21, 22]. The first questionnaire was developed in English and then translated into Amharika, the local language. After editing, including deleting, translating words, and adding instructions, a final questionnaire was created. For electronic data collection, the questionnaire was programmed into the KoboToolbox desktop (https://www.kobotoolbox.org/) for use on mobile devices and tablets during data collection. Data were recorded using structured data collection forms. In cases where the person responsible for the acquisition of unused medicines was not present during the data collection, any family member over 18 years of age and an older member were selected to participate in the study. If no family member was available at the time of data collection, the next family in the sampling sequence was selected. The unused medicines stored in each household were noted, and the different types of unused medicines were systematically recorded.

Data collectors and training

In this study, the data collectors were selected based on their prior knowledge related to the topic of interest, which can understand the value of the topic and focus on collecting insightful data. Four individuals with a Bachelor of Pharmacy degree were selected for data collection. All had a mobile smartphone that supported the application for data collection. Data collection training was given to all employees for two days, with the supervisor present. The training included the objectives of the study, an introduction to the data collection software (KoboToolbox), and guidance on recording the responses of the respondents to each question. The training also includes information on how to troubleshoot and deal with technical problems that may arise with the collection software.

Educational changes

Dependent modification

Prevalence of unused medicines.

Independent types

Social and demographic data such as gender, age and educational background, as well as baseline data, history of chronic diseases, self-medication, and medication use during the past six months.

Measurement method

Survey questions included participants’ age, gender, income, education level, and presence of children under 12. Follow-up questions included: (1) Do you have any medications not used at home? (2) Reasons for having unused medicines at home. (3) Diseases/symptoms in households with unused medicines. (4) How to get medicine. (5) Medicines kept at home. (6) Groups of medicines stored in homes.

The participants were asked if they had unused medicines in their homes. Those who answered “yes” were considered to have medicines that are not used at home. The type of medicine not used was known by examining the package or the medicine itself. The unused drugs were then classified according to their pharmacological properties. Dosage forms of non-prescription drugs, such as tablets, capsules, liquids, injections, and semisolid formulations, were also identified and categorized.

Information Quality Control

The questionnaire was initially in English, translated into the local Amharic language, and finally translated into English to ensure consistency. The questionnaire was checked to ensure the inclusion of meaningful statements and the inclusion of important information. The first survey was conducted on 5% of households from different Kebeles in Gondar city, different from Kebeles selected randomly. Necessary corrections were then made to address any potential problems, unexpected comments, and cultural objections to the questionnaire.

The data collectors were trained on the purpose of the study, the questionnaire, and how to conduct the interviews and record the responses. Finally, before data entry, the collected data was analyzed and checked by the researchers for completeness. Data were password protected on the computer and shared only with the research team.

Data Analysis

Data were imported into SPSS version 27 for analysis after cleaning in Excel. Descriptive statistics, including means, standard deviations, and frequencies, were calculated to analyze the data. Binary logistic regression analysis was performed to identify associated variables and dependent variables. We performed a purposeful selection of covariates based on the Wald test from bivariate logistic regression, using p– threshold value of 0.25. Changes with a p-values ​​below 0.25 were entered into the multiple regression model. Finally, an odds ratio (AOR) at the 95% confidence level and an ap value of less than 0.05 was considered significant.