Prevalence and Predictors of Antibiotic Self-Medication in Sudan: A Descriptive Cross-Sectional Study

Background: Self-medication with antibiotics (SMA) is one of the common factors which precipitate antimicrobial resistance, yet if effective implementations are amended it can be effortlessly controlled. The present study aimed to estimate the prevalence and predictors of SMA in Sudan. Methods: The study adopted a cross-sectional study design conducted in all Sudan states between June and December 2021. Multi-stage stratified cluster sampling was used. A semi-structured questionnaire was used for data collection. Descriptive statistics were used to present the data. Binary logistic regression was computed to investigate the possible factors which associated with SMA. Results: Out of 1492 participants surveyed, 71.3% utilize antibiotics as self-medication. The derived reasons for SMA were convenience (63.3%) and cost-saving (34.8%). Tonsillitis was the most common ailment behind SMA (55.5%). Log-binominal regression revealed that non-insured and low level of education participants were more likely to predict SMA. Regarding the practice, 40% changed the dose and/or antibiotics mainly owing to improvement (53.7%) or worsening of the condition (37.9%). The most commonly used antibiotic was amoxicillin/clavulanic acid (32.5%). Conclusions: Two out of three individuals in Sudan practice SMA mainly to manage upper respiratory tract ailments. Thus, the necessity of implementing an antimicrobial stewardship program throughout the country, as well as implementing effective legislation to prohibit dispensing antibiotics without prescription is urgently required.


Introduction
The behaviors of patients toward different medical conditions vary considerably, from immediately seeking medical advice, relying on self-medication (SM), or neglecting the condition [1]. The World Health Organization (WHO) described self-medication (SM) as the use of medicinal products or herbs to manage self-diagnosed disorders or symptoms; moreover, SM also comprises repeated or continued use of a prescribed drug for chronic, recurrent diseases or symptoms. This generally occurs through obtaining medicines without a prescription, sharing medicines, or using leftover medicines stored at home [2].
Out of 1492 participants surveyed in this study from all Sudan states (The distribution of the participants according to their states is provided in Supplementary Materials Table S1), 53.4% (796) were female. The majority of the participants were aged from 18 to 24 (36.3%) and 25 to 39 (38.7%). More than half (54.3%) of the participants had a low monthly income, however, more than two-thirds (69.3) were medically insured. Regarding the level of education, more than half (55.9%) of the participants completed their university studentship. The demographic characteristics of the included participants are presented in Table 1.

Prevalence, Sources, and Reasons behind SMA
More than two-thirds (71.3%) of the participants used antibiotics as SM. The Chisquare test revealed that, the participants level of education was significantly associated with antibiotic SM. The vast majority of the participants obtained antibiotics from the pharmacy (92.1%). Additionally, graduate (93.9%) and post-graduate (94.4%) participants obtained antibiotics from pharmacies in higher proportions than primary (89.2%) and secondary school (86.5%) participants, Table 2.

Common Ailments for Taking Antibiotic as SM
Tonsillitis was the most common aliment that drove participants to self-treatment. It was rated by more than half of the participants (55.5%). This was followed by cough (45%), while a small proportion of the participants used antibiotics for vomiting (10%) ( Table 3).

Factors Associated with the Use of Antibiotics as SM
At the bivariate level, medically insured (COR: 0.656; 95% CI (0.506-0.817) were more likely to use antibiotics as SM. Individuals having secondary (COR: 0.61; 95% CI (0.39-0.97), and post-graduate (COR: 0.38; 95% CI (0.23-0.63) levels of education were less likely to use antibiotics as SM, other factors were not statistically significant. Regarding the multivariate model, the model sensitivity was 70.2%, further, the model adequately fits the data since there were no differences between the observed and the predicted (Hosmer and Lemeshow test = 0.761). Medically non-insured (AOR: 0.645; 95% CI (0.487-0.855) and post-graduates (AOR: 0.27; 95% CI (0.15-0.5) were the only predictors for SMA. Complete logistic regression for the use of antibiotic as SM are presented in Table 4.

Knowledge and Adherence to Antibiotic Dosage
More than half of the participants knew the dosage of antibiotics through pharmacist consultation (56.4%) and they fully understood the instructions (59.6%). About 40% of the participants sometimes changed the dosage and/or the antibiotics deliberately during treatment, while 10% always changed the dosage. The main reasons for changing the dosage were improvement (53.7%) or worsening of the condition (37.9%). Approximately twothirds (67.1%) of the participants changed the former antibiotics if they weren't effective, on the other hand, more than half of the participants (55.1%) stopped taking the antibiotics when symptoms disappeared, and about 10% consulted the doctor or the pharmacist before stopping the antibiotics (Table 5).  Legends: a = more than one options is allowed.

Commonly Used Antibiotics and Common Adverse Reactions
The most commonly used antibiotics were amoxicillin/clavulanic acid combinations (32.5%), followed by amoxicillin (26.5%), metronidazole (25.3%), and azithromycin (25.3%). About one-fifth (21.8%) of the participants experienced ADR when they took antibiotics. The primary action taken by more than half (57%) of those who experienced ADR was to stop the antibiotics (Table 6). About half of the participants thought that SMA (51.9%) was not an acceptable practice, and more than one-third (37.4%) thought that they cannot treat the infectious disease on their own. Most of the participants (61%) selected antibiotics based on its indications, and about half of the participants used antibiotics when recommended by community pharmacists (47.9%) and according to their own experience (46.8%). The type of antibiotics was the main factor considered by the majority of the participants (45.4%), while the brand of the antibiotics was only considered by less than one-fifth of the participants (16.3%) ( Table 6).

Discussion
SMA is one of the common factors which precipitate AMR, yet if effective implementations are adopted it can be easily controlled. Tracking SM behaviors of public individuals are of paramount importance since it facilitates the development of preventable measures towards this condition. Moreover, it also uncovers the weakest domains in the health system. It was on these grounds that the current nationwide survey was conducted to determine the prevalence of SMA in different Sudanese states, as well as to provide an insight into the reasons, and factors associated with SMA.
The current cross-sectional survey indicated that the prevalence of SMA among the Sudanese community was 71.3%. The reported figure lies in the middle of local and regional figures in previous studies. For instance; locally, one of the earliest and most comprehensive surveys conducted by Abdelmoneim Awad et al. in 2005 reported a prevalence of 73.1% [16], while a recent study which included Sudanese university medical students reported that antibiotics were self-medicated by 60% of the students [17]. Regionally, a meta-analysis review pooled the prevalence of SMA in Africa using 40 studies from 19 countries and the computed prevalence ranged from 50-93.8% [11]. On the other hand, another systematic review outlined the proportion of SMA in middle-eastern countries to be in the range of 12.1-93.1% [10]. Our finding points towards a higher prevalence of SMA in Sudan, unfortunately, it also emphasizes the fact that the practice of SMA in Sudan has remained consistent throughout the last 20 years.
Different reasons behind SMA were mentioned previously, such as cost saving, previous experience, and convenience which were reported repeatedly. However, less frequently, the emergence of illness, and the long delays in clinic have been reported [1,[20][21][22]. Likewise, participants in the present study cited convenience (63.3%) and cost-saving (41%) as common reasons for SMA ( Table 2). The finding is consistent with the results reported previously in Sudan, Malaysia, India, and Ethiopia [14,[23][24][25].
The present study illustrated the common ailments for SMA. Generally, SMA to manage upper respiratory tract infections (URT) such as tonsillitis (55.5%), cough (45%), runny nose (37.8%), and nasal decongestion (33.4%) were higher than other ailments which comprise fever (31.9%), pain (29.1%), diarrhea (21.9%), and wound infection (14.9%) ( Table 3). This finding is in line with previous studies conducted in Sudan, Tanzania, and India [12,21,23]. Bearing in mind that most of the URT infections are of viral origin, and antibacterial agents must be preserved only for bacterial infections which indeed requires a series of investigations and diagnoses provided by health care specialists, and considering that this pattern remains consistent in Sudan through the last 20 years with a gradual increment, health authorities in Sudan should effectively implement an antimicrobial stewardship program to optimize the utilization of antimicrobial agents.
Health services in Sudan are provided by different bodies including: government, private sectors, army, police, universities, and civil society [26]. The national health insurance fund (NHIF) is an extension of social health insurance which was introduced in 1994, the finance of NHIF is based on cost sharing (national social system based on the cooperation between the government and community) [27]. The coverage in all states is around 50% (except Khartoum = 70%), and the out of pocket share in Sudan is reported to be 70% [27,28]. Nearly one third (30.7%) of the participants from the present study were not medically insured (Table 1). Further, insurance status was significantly associated with SMA (p-value < 0.00), binary logistic regression indicated that medically non-insured participants were less likely to use antibiotics as SM in comparison to insured participants (COR: 0.656; 95% CI (0.506-0.718), (AOR: 0.645; 95% CI (0.487-0.855) ( Table 4). Similar findings have been reported previously in Pakistan [12]. Additionally, 41% declared cost saving as one of the main reasons behind SMA. On the other hand, participants with secondary school and post-graduates were less likely to take SMA compared to primary school levels of education (Table 4). This pattern is not limited to this study, and it has been observed in previous studies in Lebanon, Uganda, and Malaysia [18,29,30]. However, it contradicted studies carried out in Sudan, Eritrea, and Bangladesh [16,20,31]. Such a finding is best explained by the fact that educated individuals understand the difficulties in discriminating infectious diseases and knew the consequences of SMA, therefore, they prefer to visit doctors instead of self-medicating.
Previous studies outlined that the main focus of community pharmacists in Sudan is to efficiently prescribe medications [32,33]. A considerable proportion of the participants from the present study sought antibiotics mainly from the community pharmacies (90%), and the remaining participants obtain antibiotics from leftover medication ( Table 5). Previous researchers in Sudan reported a similar pattern [14,17]. This finding indicated that the gap between the actual role of the community pharmacist which is extended to include patient counseling and education is a promising area for mitigating SMA.
Additionally, more than half of the participants in this study (51.9%) thought that SMA is not an acceptable practice (Tables 5 and 6). Paradoxically, the practice of the participants diverges from rationality, when you come to know that 41% of the participants change the dosage of the antibiotics deliberately (Table 5). Moreover, the fact that a high percentage of the enrolled participants switch antibiotics harmonizes with the finding that only 37% stop taking antibiotics after dosage completions. The abovementioned malpractice is consistent with previous studies conducted in India (24%), Malaysia (41%), and Egypt (71%) that participants change the dosage of antibiotics during usage [23,24,34]. In Afghanistan, 33% of the participants stop taking antibiotics [35], while in Malaysia 35.3% of university students switch the dosage of antibiotics [24]. Given the above backdrop, it is not surprising that at the national level multi drug resistant and extensively drug-resistant isolates detected from clinical specimens are increasingly reported [7,36,37].
Participants in the current study cited amoxicillin/clavulanic acid as the most common antibiotic used as SM (32.5%), followed by amoxicillin (26.5%), azithromycin, and metronidazole (25.3%) ( Table 6). Similar results were observed previously in Sudan [14,16], where azithromycin (29.9%) and amoxicillin/clavulanic (26.8%) were found to be the most common antibiotics self-medicated by university students [17]. Multiple studies in Africa and the Middle-East concluded the extensive use of beta-lactam antibiotics especially amoxicillin and amoxicillin/clavulanic as SM [10,11]. It is however, worth mentioning that earlier studies in Sudan reported amoxicillin as the most common antibiotic used in comparison to the present study and a recent study in 2022 [12,16]. This shift can be explained by the fact that patients always seek the most effective antibiotics, or it might be due to extensive promotion applied by different companies to promote their antibiotics (amoxicillin/clavulanic).
Besides accelerating antimicrobial resistance, SMA can also be associated with ADR. One-fifth (21.8%) of study participants reported that they experienced ADR (Table 6). This is slightly lower than a previous study in Malaysia (28.3%) [5]. Alarmingly, a considerable amount of the participants either switched the antibiotics or continued the antibiotic with the rate of 20.8% and 11.3%, respectively. ADRs associated with antibiotics ranged from mild side effects such as GIT symptoms to life-threatening conditions such as anaphylactic shock which is associated with a large number of antibiotics impacting patients' health as well as cost [25,38].
The finding from the present study can be partially generalized to the overall Sudanese community owing to the large and diverse sample size. However, one of the limitations of this study was the recall bias since not all participants were able to exactly remember for instance the types of antibiotics. Further, the study was subjected to selection bias, since it was conducted during the daytime in public areas, it is for this reason most of the participants were aged below 39 years old. Additionally, the questionnaire used in the present study adopted close-ended limited options which made it difficult for some respondents to express their opinions.

Study Design and Setting
The study adopted a cross-sectional descriptive study design, conducted in all Sudan states (all 18 states) through the period between 1 June and 15 December 2021.

Study Population
All Sudanese adult aged above 18 years old and willing to participate in the study were considered eligible.

Sample Size and Sampling Technique
According to the last census, the total population of Sudan is around 46,000,000. Using the formula below: n = Z 2 p (1 − p)/w 2 where n: sample size, Z: the critical vale (using confidence interval of 99% (Z = 2.326)), p: proportion of the target population estimated to have a particular characteristic (since there were no previous nationwide study the frequency of occurrence was assumed to be, p = 50%), (1 − p): (frequency of not occurrence of an event), w: desired margin of error tolerated (degree of precision, w = 4%). Thus, the calculated sample size was 1041 participants. We collected data from 1492 accounting for missing data. A multistage stratified sampling technique was applied to the participants. Sudan was divided into 18 states. Each state was considered a stratum, and then within each stratum, participants were selected randomly using a convenience sampling technique. Samples were collected from public places such as markets, parks, and bus stations.

Operational Definitions and Study Variables
SMA (dependent variables) was defined as the selection and use of antibiotics by participants, within the last 12 months, to manage at least one self-recognized illness or symptom without professional prescription and supervision regarding indication, dosage, and duration of treatment. Independent variables (predictors of SMA) were carefully selected based on previous studies, including: participants' gender, age, monthly income, insurance status, and educational status (all were categorical variables).

Data Collection
The current study used semi-structured questionnaires for data collection (participants who found difficulties in writing were interviewed by the trained data collectors based on the questionnaire). A comprehensive search of the literature for potential studies reporting SMA was carried out through different databases to get guidance in designing the questionnaire [20,23,24,36]. The questionnaire consisted of 25 items (provided in Supplementary Materials Table S2), which can be broadly divided into two main sections; the first section gives information regarding the demographical characteristics of the study participants, which includes gender, age, economic status, insurance status, and levels of education, while the second section starts with a main question which seeks information about any previous use of antibiotics without prescription in the last 12 months through a closed-end format (yes/no). Participants whose answer is 'yes' in the previous question were further asked to explain the main reasons and major ailment that led participants to self-medicate (multiple choice questions). It also emphasizes the practice of the participants through enquiring about the sources, selection, and adherence to antibiotic regimens (closed and close multiple choice questions). Furthermore, commonly used antibiotics and adverse drug reaction histories were also reported (closed and close multiple choice questions). For the purpose of validation, two experts in pharmacy practice were asked to highlight the main weakness of the developed questionnaire, and their comments were considered in the final version. Additionally, a pilot study was distributed to 20 individuals to confirm the clarity of the questions, the questionnaire was further validated through Cronbach alpha (α = 0.78). Responses from the pilot study were excluded from the study. To ensure the quality of the data; data was collected only through trained fifth-year pharmacy students who were taught courses in research methodology and given a comprehensive presentation on the research topics. Furthermore, data collectors were asked to check the completeness of each questionnaire.

Data Analysis
Data were entered into a Microsoft Excel spreadsheet, coded, and exported to the statistical software package SPSS (version 25.0). Both descriptive and inferential statistics were used to analyze the data. The main parts of the questionnaire were expressed in terms of frequency and percentage. A Chi-square test was employed to study the relationship between socio-demographic factors and other variables. A binary logistic regression model was used to assess the association between prevalence SM and explanatory variables. Regardless of their p-value in the unadjusted analysis, all variables were included in the final multiple regression model, and the model appropriateness was tested using Hosmer and Lemeshow test. Both crude odds ratio (COR) and adjusted odds ratios (AOR) were reported with a 95% confidence interval (95% CI). Finally, a p-value less than 0.05 was considered significant.

Ethical Consideration
Ethical approval was obtained from the ethical committee at the University of Khartoum, Faculty of Pharmacy (FPEC- . Before conducting the study, all participants signed written informed consent after a clear explanation of the research objectives, and each participant had the right to withdraw at any time from the study. To ensure confidentiality, all questionnaires were coded and personal identifiers remained anonymous throughout the study.

Conclusions
Two out of three individuals in Sudan SMA mainly to manage URT ailments, this mal-practice was explained by most of the participants by it is convenience and cost-saving. Amoxicillin/clavulanic were the most commonly used antibiotics. SMA was associated with participant's level of education and insurance status. The findings from the present study indicate the necessity of activating antimicrobial stewardship programs throughout the country, as well as implementing effective legislation to prohibit dispensing antibiotics without prescription.
Supplementary Materials: The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/antibiotics12030612/s1, Table S1: Stratification of the sample based on states; Table S2: questionnaire for data collection.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
Most of the relevant data are available in the main text, further data are available from the corresponding author upon reasonable request.