Self-Medication with Antibiotics: Prevalence, Practices and Related Factors among the Pakistani Public

Self-medication with antibiotics (SMA) has become considerably common in developing countries, which is a critical factor for driving antibiotic resistance. Individuals involved in SMA generally do not have adequate knowledge regarding the appropriate use, indications and dosage of these drugs. The objective of the present study was to investigate population SMA practices, knowledge and sociodemographic factors associated with SMA in Islamabad, Pakistan. The study adopted a cross-sectional methodology and data collection was performed through an anonymous, structured and pilot-tested questionnaire, which was interview-administered. Inferential statistics and multivariate logistic regression were performed. Out of 480 participants, 55.6% (n = 267) were male with a mean age of 37.1 ± 10.1 years; the total prevalence of SMA was 32.5%. Ciprofloxacin (42.9%) was the most commonly used antibiotic to treat coughs or colds, a runny nose, flu or sore throat, diarrhea or fevers, which were relevant reasons for SMA. Findings from multivariate logistic regression showed that predictors of SMA were: male gender (95% CI: 0.383–1.005), age (95% CI: 0.317–0.953) and highest level of education (95% CI: 0.961–0.649). Despite reasonable access to healthcare facilities, people are still obtaining antibiotics without prescription, bypassing diagnostic and consultative healthcare services. Thus, the government must implement strict healthcare policies to restrict the sale of antibiotics without prescriptions, while at the same time, targeted public awareness campaigns about the proper use of antibiotics are also required.


Introduction
Self-medication with antibiotics (SMA) is a common practice in both developing and developed countries [1][2][3][4]. The public mostly does not have adequate knowledge about how and when to use antibiotics properly, and due to a lack of professional healthcare supervision-by bypassing diagnostic and consultative healthcare services-they generally misuse prescription antibiotics [5]. Based on the Special Eurobarometer (EBM) Reports of the European Commission, and the World Health Organization (WHO) "Antibiotic resistance: multi-country public awareness survey", there is a clear association between the knowledge level of the individuals regarding infectious ailments and antimicrobials and the consumption (both via prescriptions and SMA) of antibiotics [6,7]. Studies conducted in Poland [8], Kosovo [9], Ghana [10], Nepal [11] and Pakistan [4] have all reported that non-prescription antibiotics were mainly utilized for inappropriate indications, or for

Sociodemographic Characteristics of Study Participants
Overall, a total of n = 480 study participants were included in this study; the demographic characteristics of the respondents are summarized in Table 1. In the study population, males were in the majority (55.6%; n = 267), and the mean age of the participants was 37.1 ± 10.1 years (median 36; range . The majority of the participants were in the age group of 31-40 years (44.2%; n = 212), and 76.3% (n = 366) of them were married. Most of the participants were self-employed (29.8%; n = 143) and had an undergraduate or bachelor's degree (35.4%; n = 170). At the same time, 89% (n = 427) of participants had an income over PKR 20,000.
Among participants, 28.5% (n = 137) indicated that they had used antibiotics in the last month, 26.9% (n = 129) of participants stated that they had used antibiotics at least once in the previous six months, 14.4% (n = 69) used antibiotics in the previous 12 months, and 6% (n = 29) of the participants used antibiotics more than a year ago. At the same time, 22.9% (n = 110) of participants said that they do not remember when they used antibiotics the last time. In contrast, 75.8% (n = 364) of participants claimed that they had used antibiotics at least once at the time of data collection. Furthermore, 23.5% (n = 113) of participants claimed that they received advice from a doctor, pharmacist, or nurse on how to take these medicines, 40.5% (n = 194) indicated that they did not receive any advice from any healthcare professional, and 36.0% (n = 173) had no recollection on the topic. Internal consistency (Cronbach's α value) for the different domains of the instrument was determined: for the "Practices towards SMA" domain, the Cronbach's α value was 0.817, while for the "Public-reported outcomes" domain, the α value was 0.658, respectively.

Test-Retest Reliability
Results obtained from the test-retest analysis indicate satisfactory reliability and stability: for the "Practices towards SMA" domain and the "Public-reported outcomes" domain, the reliability coefficients were 0.972 and 0.861, respectively (p < 0.05).

Prevalence of SMA among Study Participants
The total prevalence of SMA among study participants was 32.5%. Out of the 480 participants, 34.2% had used prescription antibiotics; 32.5% had self-medicated with antibiotics; and 33.3% did not remember whether they had used prescribed or non-prescribed antibiotics. The prevalence of SMA was higher in males (21.9%) compared to females (10.6%). Regarding age groups, the highest prevalence was noted among age groups of 31-40 years old (15.4%). On the other hand, the lowest prevalence of SMA was noted in the younger age group, 21-25 years old, where only 2.1% of participants had used antibiotics without a prescription. Postgraduate participants (2.9%) were less likely to use antibiotics than others. There was no significant correlation between SMA and some sociodemographic variables including age groups (p = 0.129), educational attainment (p = 0.224), income levels (p = 0.578,) and having insurance (p = 0.686). On the other hand, gender (p = 0.001), marital status (p = 0.003), occupation type (p = 0.001), and parents having children (p = 0.001) were significantly associated with SMA. All the results regarding SMA and associated demographic characteristics are shown in Table 2.

Commons Reasons for SMA and Common Sources for Obtaining Antibiotics
Results showed that most of the participants (35.4%) did not see doctors when they were sick because of lack of time. At the same time, 8.1% said that they do not implicitly trust doctors, and 15.2% of participants said that they do not have enough money for a doctor's visit. In addition to this, 2.9% of participants stated that they face difficulties in communication with doctors, and 16.9% said that they face difficulties in reaching public health clinics or hospitals, or no transportation is available. The majority of participants who used antibiotics for themselves obtained their antibiotics from community pharmacies (50.6%), while 8.3% of participants obtained their antibiotics from a friend or family member. However, only (0.6%) of participants obtained their antibiotics from a stall or hawkers ( Figure 1). Legend: *** = p < 0.001, ** = p < 0.01.

Trends in Antibiotic Usage
When participants were asked about why they consumed antibiotics; the respondents in the study referred to the following reasons for SMA: a cough or a cold, runny nose, a flu or sore throat, diarrhea or fever. The most commonly used antibiotic was ciprofloxacin (42.9%), followed by metronidazole (35.2%), erythromycin (28.7%), levofloxacin (28.5%), and amoxicillin-clavulanate (25.2%). The rest of the antibiotic groups inquired were less frequently utilized by study participants, depicting the frequency of usage for each antibiotic ( Figure 2). Participants were allowed to choose more than one option, so the overall percentage is greater than 100%.

Trends in Antibiotic Usage
When participants were asked about why they consumed antibiotics; the respondents in the study referred to the following reasons for SMA: a cough or a cold, runny nose, a flu or sore throat, diarrhea or fever. The most commonly used antibiotic was ciprofloxacin (42.9%), followed by metronidazole (35.2%), erythromycin (28.7%), levofloxacin (28.5%), and amoxicillin-clavulanate (25.2%). The rest of the antibiotic groups inquired were less frequently utilized by study participants, depicting the frequency of usage for each antibiotic ( Figure 2). Participants were allowed to choose more than one option, so the overall percentage is greater than 100%.

Trends in Antibiotic Usage
When participants were asked about why they consumed antibiotics; the respondents in the study referred to the following reasons for SMA: a cough or a cold, runny nose, a flu or sore throat, diarrhea or fever. The most commonly used antibiotic was ciprofloxacin (42.9%), followed by metronidazole (35.2%), erythromycin (28.7%), levofloxacin (28.5%), and amoxicillin-clavulanate (25.2%). The rest of the antibiotic groups inquired were less frequently utilized by study participants, depicting the frequency of usage for each antibiotic ( Figure 2). Participants were allowed to choose more than one option, so the overall percentage is greater than 100%.

Knowledge, Attitude, and Practices towards SMA
Less than half (41.7%) of participants completed their course of treatment as recommended, and 43.1% stopped taking their medicine when their symptoms began to improve. At the same time, 41.3% of participants indicated that it is okay to use antibiotics given to a friend or a family member, if they were used to treat the same illness, regardless of the outcome. More than half of the study participants (62.5%) wrongly identified that antibiotics are useful to treat colds and the flu (i.e., common viral infections). In addition, 67.7% believed that antibiotics may be used to treat fever followed by sore throat (50.4%), body aches (44.0%), headaches (42.7%) and urinary tract infections (UTIs) (40.2%) ( Table 3).  Table 4.

Public Reported Outcomes
The last part of the study was specially designed to assess public perception after taking antibiotics. When participants of the study were asked whether they intend on self-medicating with antibiotics again if the same symptoms appear in the future, 41.8% gave a positive reply (i.e., they would do it again), while only 26.3% said that they will not self-medicate in the future; 31.9% of participants were unsure about their answer. A higher potential future SMA prevalence was noted among male participants (28.7% vs. 13.1%). The sociodemographic factors that were significantly associated with future prevalence of SMA were male gender (p ≤ 0.001), income rate (p ≤ 0.001), and occupation type (p = 0.003) ( Table 5).
Moreover, Table 5 further includes results related to public reported outcomes. When participants were asked that how they perceived their symptoms after taking antibiotics, 72.9% of participants indicated that their symptoms improved after taking antibiotics. In contrast, 4.6% of participants stated that their symptoms worsened, and for 22.5% symptoms remained the same. At the same time, 45.8% of participants indicated that this practice of treating with antibiotics to take care of their health is acceptable, while 35.0% of participants did not agree with this statement. A total of 38.3% of participants stated that treating with antibiotics proved economical; 18.3% of participants were not sure about this. Among study participants, 20.4% indicated that they saved below PKR 500 on average at a time when they self-medicated with antibiotics, 10.9% of participants stated that they saved over PKR 1500, while 31.8% of participants could not say how much money they saved at one time. Legend: *** = p < 0.001; ** = p < 0.01.

Discussion
To the best of our knowledge, this is the first public-based study conducted in Islamabad, Pakistan to evaluate knowledge about antibiotics tendencies of self-medication with these drugs. Results showed that the prevalence of SMA was 32.5%, and the result was comparable with the study performed in the United Arab Emirates (UAE) (31.7%) [31]. In comparison, previous studies conducted in two different cities of Pakistan (Karachi, 80%; Peshawar, 69%) showed a higher prevalence of SMA than the present study [4,28]. In addition to this, studies performed in other developing countries, such as Kenya (60.0%), Nigeria (82.2%), Yemen (87.1%), and Saudi Arabia (40.8%) [32][33][34][35] also showed a higher prevalence of SMA by the public. The differences in the SMA rates may be due to ethnic diversity among the different populations and different healthcare systems in each of these countries. However, SMA rate may be decreased by adopting universal health coverage (UHC). UHC has also been adopted as a key target under the sustainable development goals by the United Nations, G20, and WHO to reduce AMR by improving the health system in terms of equity, quality, efficiency, accountability, sustainability and resilience [36]. Moreover, the present study also reported that male gender was an important predictor of SMA. Studies carried out in Pakistan and the UAE also showed that the prevalence of SMA was found to be higher among men [37], which might be due to men having greater access to medical stores and pharmacies in Pakistan. In comparison, other studies also showed similar results that women tended to use more antibiotics than men. Meanwhile, females are more prone to SMA, and also they visit physicians more often [38,39].
Results from the present study also indicated that most of the participants who selfmedicated with antibiotics had a low level of education. A similar result was also achieved in studies carried out in Lebanon [29,40]. At the same time, in the present study, age was not linked to SMA, which is in contrast to other studies that have assigned either the young adult or adult population to be more prone towards practicing SMA; these studies were performed in Italy, Saudi Arabia, and United Arab Emirates [37,41,42]. Moreover, gender (p = 0.001), marital status (p = 0.003), occupation (p = 0.001), and parents having children (p = 0.001) were the factors that were significantly associated with SMA. Whereas findings from multivariate logistic analysis also found that male gender, being between 31 and 40 years of age with a low level of education are most prone towards SMA. A study performed in Malaysia and Uganda also confirmed the latter findings [43,44].
Another contributing factor is the lack of strict disciplinary regulation policies (compared to some Western countries), and due to this reason, people easily obtain antibiotics from pharmacies or medical stores without a prescription. The present study revealed that most of the participants obtained their antibiotics from pharmacies, either from a pharmacist or a pharmacy technician/assistant (50.6%). Results obtained from other studies performed in Malaysia, the UAE, and Pakistan also highlighted pharmacies as important sources of antibiotics for SMA [4,31,43]. A systematic review also provided evidence that community pharmacies remain the leading source of antibiotics for SMA in developing countries. This might be due to the profit-oriented nature of service delivery in this sector and/or inadequate supervision [22]. The present study also revealed that ciprofloxacin, followed by metronidazole and levofloxacin, were the most frequently used antibiotics for self-medication of cough, cold, flu, diarrhea, and fever. A study performed in Pakistan also showed similar results [29]. In contrast, other studies conducted in Pakistan showed that amoxicillin and metronidazole were the most commonly used antibiotics [4,45]. Interestingly, a scoping review noted that β-lactam antibiotics (more specifically, penicillin and aminopenicillin/β-lactamase inhibitors) are the most commonly used antimicrobials for SMA [46]. Other studies also indicated that antibiotics were most commonly used for fever, cold, and pain [41,47,48]. This irregular use of antibiotics for common viral infections such as cold and flu is an important factor in worsening antibiotic resistance; this serious problem was also verified by other studies [31,49,50].
In our study, it was noted that 40.2% of participants were most likely to stop taking antibiotics after they were feeling better. These results are similar to another study performed in Pakistan, where participants also stopped taking antibiotics on the third day or even earlier, if their symptoms improved [29]. This non-compliance may also lead to the spread of severely drug-resistant pathogens; according to the WHO, when any individual or patient stops taking antibiotics too early, it may benefit bacteria strains with some natural intrinsic resistance. As per the guidelines of the WHO, it is therefore advisable that individuals/patients always take the entire course of antibiotics prescribed to them, overseen by a certified healthcare professional (physicians, pharmacists, nurses or others). Moreover, the present study also revealed that the main reasons that influence the use of antibiotics for self-medication were the lack of trust in doctors and economic conditions; this was in accordance with another study [4]. The last part of the study consisted of public reported outcomes associated with SMA among our participants; 41.8% of participants indicated that they would still use antibiotics in the future if the same disease symptoms appear, and this rate is actually higher than the reported prevalence of SMA in the present population. This high future prevalence of SMA might be due to the high doctor's fees for a checkup and poor patient management, as well as under-staffed healthcare centers that cannot bear the burden of the visiting population, eventually forcing people to look for alternative solutions and SMA seems to be the most viable choice. The future prevalence of SMA is significantly associated with the participant's gender (p = 0.001), income levels (p = 0.001), and occupation type (p = 0.003). Moreover, most of the participants indicated that their symptoms improved after taking antibiotics, and 45.8% of participants said that treating with antibiotics is an acceptable practice.
This study has several limitations: as with any cross-sectional, questionnaire-based survey, there is no way to gauge the truthfulness of the answers from the participants. Furthermore, this study was conducted in a federal area (Islamabad); hence, these results cannot be generalized to the whole country.

Study Design and Settings
The present study adopted a prospective cross-sectional study design. This study was carried out in the capital city of Pakistan (Islamabad), located in the country's northwest region. The total area of Islamabad is 906.5 km 2 , being the ninth largest city according to its population. The total population of Islamabad is above 2 million people, with an annual growth of 4.9% [51]. At the same time, the majority of the population in Islamabad city is 15-64 (59.38%) years old, while the literacy rate is also highest (88%) compared to other parts of the country [51].

Study Population
The participants included in this study were adults (≥18 years) willing to participate without having any communication problems either due to illness or other reasons. This study included laypeople only (i.e., people with no expert knowledge or professional qualifications in the particular subject) [52]. Participants with a medical background, such as doctors, pharmacists, nurses, paramedical and allied health professionals, were therefore excluded from this study. The respondents or participants were selected through convenience and snowball sampling methods. Furthermore, this study only included those participants who could read and write in Urdu and English.

Sample Size Calculation
To establish the required sample size for our study, a sample size calculation was performed by using Lorenz's formula (1), described below [53]: where p is the pre-study estimates of self-medication prevalence, q is 1 − p, z is 1.96 (standard normal deviation for a 95%CI:), and d is 0.05 (tolerated sampling error or precision of 5%). At 95%CI:, the z statistic value is 1.96, and p was determined to be 0.5 from a previous study, as reported by Bilal et al. [4]. Assuming a degree of precision of 0.05, the calculated minimum sample size was set at n = 234; however, to strengthen the study results, the final sample size was increased for contingency by more than 50%, and the final sample size of n = 480 persons was set.

Study Instrument
The research instrument utilized in the present study was a questionnaire containing both open-ended and close-ended (multiple choice) questions. Our research group developed the questionnaire, based on literature screening and screening the references of a scoping review conducted previously on the topic [46]. The questionnaire comprised 43 questions with a total of 71 items. The items in the questionnaire were broadly grouped into the following four categories: (i) sociodemographic characteristics of the participants (gender, age, marital status, education level, employment status, healthcare expense coverage); (ii) personal information on health and medicine use; (iii) practices towards SMA and factors contributing SMA; and (iv) public-reported outcomes. For the purpose of adaptation and translation of the questionnaire, a set criterion was followed, developed by Beaton et al. [54]. Before conducting the main study, pilot testing (involving n = 100 participants) of the translated questionnaires was carried out to ensure internal consistency (Cronbach alpha), reliability (test-retest reliability), face validity, and content validity. At the same time, pilot testing was necessary to determine if the questionnaire would be comprehensive, understandable, and appropriate among respondents [55]. The results obtained from the pilot testing of the questionnaire allowed quality improvement in view of specific cultural considerations, and thus, certain modifications were made in the wording of the questionnaire to produce the final instrument.

Data Collection
The data were collected in public places, including public parks, outside of big hospitals, shopping malls, bus stations, supermarkets, and metro stations. The final version of the questionnaire was used to collect data through the interview-administered method from February and July 2020. A written signed informed consent form was obtained from all participants who agreed to take part in this study. At the same time, an introductory letter was given to all respondents, which included a brief description of the study objectives and its importance. The data were collected through an interview-administered questionnaire by the principal investigator with the help of trained interviewers. The confidentiality and anonymity of the participants were protected throughout the study.

Statistical Analysis
The descriptive and inferential statistical data analyses were processed with the assistance of Statistical Package for the Social Sciences (SPSS) version 24 (SPSS Inc., Chicago, IL, USA). Descriptive statistics, including cross tables and frequencies, were used for variables, and a χ 2 test was performed to determine variables associated with dependent variables. Multivariate logistic regression analysis was used to determine the predictors of SMA within the study population, and in this analysis, the prevalence of SMA was considered as the outcome. Participants' sociodemographics were considered exposure variables (gender, age, income, education, job type, and health insurance availability). Finally, univariate and multivariate logistic regression at 95%CI: were computed. All factors that achieved a p-value of less than 0.05 at the univariate level were considered fitting to be included in the multivariable model to determine the predictors of SMA. Statistical significance for all analyses was set at p < 0.05.

Ethical Considerations
Before the commencement of this study, ethical approval for this study was obtained from the ethical committee at International Islamic University Malaysia (IREC 2019-004) and Hamdard University Pakistan (HU-ERC-19-408).

Conclusions
In conclusion, SMA is common among the public of Pakistan; males were highlighted among the population to have higher intent for such use of these drugs. Underlying factors of why the public self-medicates with antibiotics were: mainly lack of trust towards doctors, economic considerations and easy availability of antibiotics as OTC from pharmacies. The lack of strict legislation/laws over the availability of antibiotics without prescription and limited availability of health insurance further complicates the issue. At the same time, healthcare professionals should play an important role in patient education, counseling and guide proper pharmaceutical use. Adequate competence and communication on the part of the healthcare professionals may increase patient trust towards them, which will-in turn-lead to higher rates of appropriate medicine use. The government must introduce educational and awareness programs specifically targeting the public so that people without sufficient knowledge are made aware of the dangers of SMA. Lastly, it is recommended to bring in legislation on managing antimicrobial prescriptions.
Author Contributions: A.A., S.I.A., P.P. and M.G. collected and analyzed the data, and wrote the manuscript. C.S.Z., S.J. and N.S.A.R. conceptualized and designed the study, guided data analysis and interpretation, and critically reviewed the manuscript. All authors have read and agreed to the published version of the manuscript. Informed Consent Statement: Written informed consent was obtained from the respondents before participating in this study.

Data Availability Statement:
The anonymized datasets used and/or analyzed during the present study are available from the corresponding author on reasonable request.