Assessment of self-medication practices in the context of Covid-19 outbreak in Togo

Background: To date, there is no effective treatment for Covid-19 which is a pandemic due to a new coronavirus called SARS-CoV-2. In Togo, where four in ve people practice self-medication, the absence of a cure for Covid-19 and its constant progression requires an assessment of self-medication patterns in the context of the pandemic. This study aimed at estimating the prevalence of self-medication to prevent Covid-19 and its associated factors in Lomé, Togo. Methods: A cross-sectional study was conducted in Lomé, capital city of Togo, from April 23 rd to May 8 th , 2020 in a sample of participants from ve sectors: healthcare, air transport, police, road transport and informal. Participants were invited to give information about their self-medication practices to prevent Covid-19 within the two weeks preceding the survey. Results: A total of 955 participants (71.6% of men) with a median age of 36 (IQR 32–43) were included. Around 22.1% were in the air transport sector, 20.5% in police, and 38.7% in the health sector. The overall prevalence of self-medication to prevent Covid-19 was 34.2% (95CI%: 31.2-37.3%). Most commonly used products were vitamin C (27.6%) and traditional medicine (10.2%). Only 2.0% of participants reported using Chloroquine/Hydroxychloroquine. Female sex (aOR=1.90; p<0.001), working in the health sector (aOR=1.89; p= 0.001), secondary education level (aOR= 2.28; p= 0.043) and university education level (aOR= 5.11; p<0.001) were associated with self-medication. Conclusion: One third of high-risk populations in Lomé have self-medication. Intensifying ght misinformation social


Background
On January 30 th 2020, the World Health Organization (WHO) declared a public health emergency of international concern due to the advent in China of a disease called Covid-19 caused by a new coronavirus, SARS-CoV-2 and its rapid spread [1]. About six months later, almost 20 million cases and about 700,000 deaths have been reported worldwide [2].
To date, there is still no treatment or vaccine for this pandemic. Several studies have evaluated the e cacy of Hydroxychloroquine-based treatment with or without azithromycin [3][4][5]. However, the e cacy of these molecules has not been proven for a curative treatment of the disease.
Chloroquine/Hydroxychloroquine were also evaluated for prophylaxis against Covid-19 in clinical trials among close contacts or health care workers. Although pre-clinical results are promising, there currently is no evidence of the effectiveness of Chloroquine/Hydroxychloroquine in the prevention of Covid-19 [6].
The Covid-19 epidemic created widespread psychosis and anxiety among the population in sub-Saharan Africa [7]. This could be linked, on one hand, to the high mortality observed in some countries such as Italy and Spain, and on the other hand, to the lack of an adequate technical platform in sub-Saharan Africa. Regarding the Africa continent, the WHO indicated that it fears the worst as the more resourced health-care systems of developed countries have faced enormous di culties in dealing with the epidemic [8]. Faced with this situation and the various information circulating on social media, many plants and substances are being proposed without the minimum requirements of e cacy and tolerance [9]. The use of these substances without medical advice is considered as self-medication which is de ned as taking medicines, herbs or home remedies on one's own initiative or on the advice of another person, without consulting a medical doctor [10]. In the context of the Covid-19 pandemic, cases of poisoning and death have been reported in the USA and Nigeria when self-medicating with Chloroquine [11,12].
Health literacy plays an important role in self-medication behaviour [13]. Concerning the covid-19 pandemic and other coronaviruses, the level of knowledge is globally low according to a meta-analysis of 70 scienti c articles. Indeed, the proportion of people with a low level of knowledge ranged from 4.3 to 57.9% among health professionals and from 4.0 to 82.5% in the rest of the population [14].
In Togo, although the dispensing of psychoactive drugs is regulated by law and requires a prescription, this is not the case for other drugs, including antibiotics, which can be sold without a prescription [15,16].
Despite the efforts of the Togolese National Order of Pharmacists to curb the phenomenon and limit access to antibiotics without a prescription, these drugs are still widely consumed as self-medication [17].
Togo reported its rst case of Covid-19 on March 5 th 2020 and, the number of cases was multiplied by ten in three months, with 98 cases for 6 deaths on April 26 th [18] against 908 cases for 18 deaths on July 31 th 2020 [2]. The absence of a recognised treatment for the disease and its constant progression requires a re-evaluation of self-medication practices in our country where 80% of people resort to selfmedication [19] and 60% to traditional pharmacopoeia [20]. Thus, this study was conducted to estimate the prevalence of preventive self-medication and its associated factors in an epidemic context where there is no preventive or curative treatment.

Study design and sampling
This study was part of a survey aimed at describing the prevalence of SARS-CoV-2 in high-risk populations in Lomé (capital city of Togo) [21]. It was a cross-sectional study conducted from April 23 rd to May 8 th , 2020. A total of 955 participants were included, and the prevalence of SARS-CoV-2 was 0.7% according to Polymerase Chain Reaction test [21].
Participants were recruited from ve professional sectors: healthcare (doctors, nurses, pharmacy auxiliaries, hospital administrators), air transport, police, road transport (taxi and moto-taxi drivers) and the informal sector (market sellers and craftsmen). These groups were targeted because they are at high risk of contamination during epidemics [22,23]. Thus, these working professionals had a high probability of being in close contact with travelers or with Covid-19 patients. Participants were eligible to participate in the study if the following four criteria were ful lled: (i) ≥ 18 years of age; (ii) working in one of the ve sectors under study; (iii) have been regularly present at workstation in the month prior to the survey, i.e. people who had not taken any time off work in the last 30 days or had no sick leave; (iv) living in Lomé for the past 3 months.
Several sampling methods were used for participants' selection based on the expected total size of the target population and the availability of a sampling frame. First, exhaustive recruitment was performed among the police (road safety o cers) and people in air transport (International Airport Gnassingbe Eyadema, Lomé, Togo). Second, participants from the informal sector were recruited based on an open invitation. Third, random sampling (two or three stage) was performed for the recruitment of taxi and moto-taxi drivers (road transport) and health care workers. For example, for the selection of moto-taxi drivers we performed a two-stage sampling with the selection of the company, then the selection of the drivers working in the company.

Sample size
The sample size of participants was calculated using a single proportion population formula with a 95% con dence level. We hypothesized that 50% of the population would practice self-medication with a 5% margin error. The estimated minimum sample size was 384 participants. A 10% non-response rate was anticipated and the minimum number of participants was estimated at 422. With a sample size of 955, we reached a 3% margin error.

Data collection
After eligibility screening and written informed consent approval, sociodemographic characteristics, Covid-19 epidemiological data and self-medication practices were collected using a standardized questionnaire developed for this survey by a multidisciplinary team involving two medical epidemiologists, two virologists, one pharmacist and one sociologist. All of the ve sections of the questionnaire (sociodemographic characteristics; knowledge, attitudes and practices toward Covid-19; symptoms of Covid-19 and biological tests) were developed based on the literature data available since the beginning of the pandemic due to SARS-CoV-2. The questionnaire, was tested on a sample of six doctors in medicine (the highest supposed level of understanding) and six taxi drivers (the lowest assumed level of understanding), which allowed us to reformulate or remove some questions that seemed complicated or di cult to answer. Five trained medical doctors administered the questionnaire during a face-to-face interview. Participants were invited to give information on self-medication practices to prevent Covid-19 within the two weeks preceding the interview.

Measurements
Self-medication practice was the outcome variable. It was assessed by the selection and use of medicines/drugs alleged to treat or prevent Covid-19 by individuals without physician's order in the past 2 weeks.

Statistical analysis
Descriptive statistics were performed and results were presented with frequency tabulations and percentages for categorical variables. Quantitative variables were presented as medians with their interquartile range (IQR). Prevalence of self-medication was estimated with its 95% con dence interval (95%CI).
Univariable and multivariable logistic model regression were performed to assess factors associated with self-medication in the intent of preventing Covid-19. For model building, characteristics that had a p-value <0.20 in univariable analysis were considered for the full multivariable models, which were subsequently nalized using a stepwise, backward elimination approach (p-value <0.05). This allowed the estimation of adjusted odds ratios (aOR) with their 95% con dence interval. Predictor variables were selected as those found to be relevant according to the literature review. Data analyses were performed using R © version 3.4.3 software and the level of signi cance was set at 5%.

Ethical considerations
Ethical approval was obtained from the 'Comité de Bioéthique de Recherche en Santé' (Bioethics Committee for Health Research) from the Togo Ministry of Health (No. 004/2020/CBRS). Potential participants were informed about the study purpose and procedures, potential risks and protections. Those willing to participate were invited to sign a consent prior to participation.

Results
This study was an ancillary study to an investigation of the circulation of SARS-CoV-2 in a population considered to be at high risk of infection. A total of 976 people was approached and 21 people refused to be sampled for SARS-CoV-2 and were therefore excluded from the study, for a response rate of 97.8%.

Sociodemographic characteristics
In sum, 955 people with a median age of 36 (IQR 32-43) were included in the study and 71.6% (n=684) were men. Among recruited participants, 22.1% (n=212) were enrolled in air transport sector, 20.5% (n=196) in police, 5.8% (n=55) in the informal economy sector, 38.7% (n=370) in the health sector and 12.8% (n=122) in the road transport sector. None of the participants had been previously diagnosed Covid-19 positive or hospitalized in the last 30 days before their enrolment. Among the participants, 936 (98.0%) were Togolese, around two-thirds (66.6%; n=636) were in couple and half (51.0%; n=487) of them had a university degree. The sociodemographic characteristics according to the sector of activity are summarized in Table 1.
Products used for self-medication Table 2 describes the products used for the prevention of Covid-19 according to sector of activity. Most commonly used products were Vitamin C (27.6%), and traditional medicine (10.2%). Chloroquine/Hydroxychloroquine was used by 2.0% and Azithromycin by 1.2%.

Prevalence of self-medication
The overall prevalence of using at least one product to prevent Covid-19 was 34.2% (95CI%: 31.2-37.3%). It ranged from 16.4% (95%CI= [9.8-23.0]) for participants from the road transport sector to 51.9% (95%CI= [46.8-57.0]) for those from the health sector. Table 3 reports the overall self-medication's prevalence according to sector of activity among participants.

Factors associated with self-medication used
In multivariable logistic regression model after adjustment on the other variables, three factors were positively associated with self-medication: being female (aOR=1.90; p<0.001), working in the health sector (aOR=1.89; p= 0.001) and having attained secondary level or more (aOR= 2.28; p= 0.043 for secondary level and aOR=5.11; p<0.001 for university level). Having at least one symptom related to SARS-CoV-2 was not associated with self-medication ( Figure 1).

Discussion
The WHO does not recommend self-medication with any medicines including antibiotics, as a prevention or cure for Covid-19. Despite advice of clinicians and governments, 34.2% of surveyed people in our study used a treatment without prescription. This include modern treatments as well as traditional medicine. This prevalence of self-medication in our study is probably related to: i) the long delay in nding an appropriate treatment for Covid-19 based on an adequate powered randomized trial [24]; ii) the in uence of social media which propose any type of products to prevent or treat Covid-19 [9]; iii) the in uence of leaders (political and religious) who have claimed the e cacy of certain products or who claim to have discovered some traditional remedies [25,26]; iv) the stigmatization of people infected with SARS-CoV-2 encouraging some people to take care of themselves at home [27].
Chloroquine/Hydroxychloroquine was used by 2.0% of population and this proportion vary between 0.8% in people working in health sector to 7.3% in people working in the informal sector. The use of Chloroquine/Hydroxychloroquine could be linked to the fact that a French non-randomized open label trial was conducted showing signi cant decrease in viral load and recovery duration in Covid-19 patients receiving Hydroxychloroquine (600 mg/ day) associated with or without azithromycin during 10 days [5].
However, there is a lot of warning about the improper use of Chloroquine/Hydroxychloroquine, outside of hospital or clinical trial settings for Covid-19. Its use may increase risk for arrhythmias or death [28,29]. Political leaders such as president Trump also claimed to use Chloroquine for Covid-19 prevention [25]. This type of declaration sharing to the community could be destructive and non-productive in regard to public health communication.
In our study, Azithromycin was used by 1.2% of the sample. Self-medication with antibiotics such as Azithromycin could cause harm to the patient and increase the risk of antimicrobial resistance [30]. The low use of Azithromycin could be explained by its relatively high cost and by the fact that, in recent years, although this is not regulated by law, the Togolese pharmacists' association insists that these products be sold only with a prescription.
If Chloroquine/Hydroxychloroquine are controlled medicines which are sold in pharmacy, this is not the case for Vitamin C. In our study, Vitamin C was used by around one third (27.6%) of participants. Several studies suggested the effectiveness of high dose of Vitamin C in the management of Covid-19 [31,32].
However, it is also important to note that high doses of Vitamin C may cause side effects, most speci cally an increased risk of kidney stones [33].
In April 2020, the promotion of a traditional medicine called Covid-Organics for prevention and treatment of Covid-19 was made in Madagascar [25]. However, the number of cases of Covid-19 in Madagascar has quadrupled from 2,214 to 10,748 in July 2020 [2,34]. Several reasons could justify the rise in the number of cases of Covid-19; but, it raises questions on the effectiveness of Covid-Organics, which has not yet been properly tested in therapeutic trials. Artemisia plant, the main component of Covid-Organics, has shown some bene cial effects, less than those of artemisinin-based combination therapies (ACTs) in the treatment of malaria [35]. The WHO, fearing a risk of developing resistance to ACTs linked to the use of this plant, does not recommend it for the treatment of malaria [35]. Furthermore, no studies have yet proven its e cacy for use in the prevention or treatment of Covid-19. In our study one out of ten (10.2%) participants declared that they used traditional medicine for Covid-19 prevention. This could be explained by the fact that the use of traditional medicine is common in the African culture and relatively less expensive [20], although the composition of these mixtures are most of the time unknown [20,36]. In regards to traditional medicine, the WHO welcomes innovations around the world including repurposing drugs, traditional medicines and developing new therapies in the search for potential treatments for Covid-19 [9]. The WHO is working with research institutions to select traditional medicine products which can be investigated for clinical e cacy and safety for Covid-19 treatment [9].
In this study, self-medication was found to be signi cantly associated with being female, working in the health sector and having a high school level or more. There are con icting data on the relationship between gender and self-medication [37,38]. Some studies conducted on self-medication reported the fact that female sex was signi cantly associated with self-medication. A study conducted among undergraduate students of a private university in Nigeria showed that 88.2% of females versus 71.1% of male reported using self-medication [39]. In Spain, the prevalence of self-medication was 16.93% (2,715) for women and 14.46% (1,469) for men (p<0.05) in a study about sex differences on self-medication [40]. Reason of association between female sex and self-medication is not clearly known, but in the context of Covid-19 outbreak, greater anxiety among women as described in Iran and Italy could not be excluded [41,42].
A 2018 systematic review and meta-analysis of observational studies conducted in Ethiopia showed that healthcare professionals and students were the main practitioners of self-medication [43]. In our study, self-medication was associated with working in the health sector. Knowledge and access to prescriptiononly medicines are potential factors for self-medication among health professionals. Other factors that are often quoted include the complaint of extensive demands on their time, issues of privacy and con dentiality [44]. According to the WHO, about 10% of all Covid-19 cases globally are among health workers. In Africa, information on health workers' infections is still limited, but preliminary data nds that they make up more than 5% of cases in 14 countries in sub-Saharan Africa alone, and in four of these, health workers make up more than 10% of all infections [45]. The higher risk of infection among health care professionals, knowledge on drugs and the ease of access to drugs may also explain a higher use of self-medication [46,47].
Self-medication was usually associated with a lower education level. A study conducted in a Lebanon adult sample about their knowledge and self-medication with antibiotics reported that self-medication was signi cantly associated with low education level (p=0.036) [48]. This was not the case for the present study conducted in a context of Covid-19 outbreak. Indeed, participants with a high school level or more were more likely to self-medicate. This could be explained by the fact that a good knowledge of diseases is known to be associated with self-medication [49][50][51]. The greater access of the educated population to the internet and their ability to understand information about treatment (often published in o cial languages) found on social networks may also explain this trend.
Surprisingly, history of clinical manifestation was not associated with self-medication in our survey, and that could con rm the fact that self-medication was more likely to be used for prevention of Covid-19 and not to treat speci c clinical manifestation of Covid-19, which are closed to malaria symptoms. This study has some limitations. We did not collect data on the doses of the drugs used and the length of time they were used. For traditional medicines, the composition of the different potions used has not been collected. It should also be noted that in the Togolese context, these traditional medicines are very often used in combination with modern medicines. Another limitation of this study is that the questionnaire used was entirely developed by our team and had never been used before. Even if this questionnaire has been pre-tested, biases cannot be excluded (primacy effect, order effect etc.). Furthermore, the study was based on declarative data, which could lead to an underestimation of the prevalence of self-medication due to a social desirability bias. Finally, according to the characteristics of the surveyed population (people with a high risk of SARS-CoV-2 infection), extrapolation of these results to the general population should be made with great caution.

Conclusion
This study was the rst to assess the prevalence of self-medication to prevent Covid-19 in Togo. Approximately one third of people reported ever having used self-medication. Vitamin C and traditional medicines were the most commonly used. Health professionals, women and people with a high level of education were the most likely to use self-medication. Close collaboration is needed with pharmacists to not sell Chloroquine/Hydroxychloroquine without any medical prescription. It is also important to ght against misinformation about supposed Covid-19 preventive products on social media by improving awareness campaigns. Psychology is also important in Covid-19's response strategies, as reducing anxiety and ambient psychosis could reduce the use of dangerous self-medication. Studies in the general population should be conducted to con rm these results. participants were told about the study purpose and procedures, potential risks and protections, and compensation. Written informed consent was obtained prior to participation.

Consent for publication
Not applicable.

Availability of data and material
The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Competing interests
The authors declare that they have no competing interests.