Factors Associated with Practice of Covid-19 Precautionary Measures among Clients Visiting A Teritiary Hospital, Addis Ababa, Ethiopia

Background: Despite efforts to contain the COVID-19 virus, Addis Ababa, the country's COVID-19 epicenter, is experiencing a sharp increase in the number of cases and death rate. While poor public adherence to individual COVID-19 precautionary measures is evident, factors associated with it are not well studied. We aimed to assess the level of practice of COVID-19 precautionary measures and associated factors. Methods: This was a hospital-based cross-sectional study conducted from February 1 st to 15 th , 2021 at Saint Paul's Hospital Millennium Medical College, a tertiary hospital in Addis Ababa. We used a structured questionnaire to conduct a face-to-face exit interview with randomly selected clients visiting the hospital. Binary and multivariable logistic regressions were implemented to examine factors associated with prevention practice. Results: We analyzed data from 262 participants. The mean age of participants was 36(SD=12) years. The majority (207, 79%) of the study participants had a favorable attitude towards prevention and control measures of the disease. A little above half (116, 55.7%) of the respondents had a satisfactory level of practice of COVID-19 precautionary measures. Being a resident of the capital city [AOR: 2.25 (1.22-4.15)], and having a favorable attitude [AOR: (AOR: 4.88(2.08-11.68)] were signicantly associated with satisfactory COVID-19 precautionary practice. Conclusions: The level of COVID-19 precautionary practice was unsatisfactory. A favorable attitude and stricter enforcement of COVID-19 preventive measures were associated with adherence to precautionary practices. The ndings highlight the need for a public health education strategy targeted at improving attitudes of the community on COVID-19 preventive practice.

prevention practice in a population. Among others, individual awareness, knowledge, attitude, and perception of risk are key in the adoption of precautionary practices (6,8,9). However, there is scarce evidence as to which of these factors contribute to a practical adoption and adherence to precautionary practices. The few studies conducted previously were limited to subsets of the population, particularly health care workers (10)(11)(12). Our study aimed to investigate the level and factors associated with COVID-19 prevention practice among clients visiting a tertiary hospital in Addis Ababa.

Study design and setting
This hospital-based cross-sectional study was conducted from February 1st to 15th 2021 at Saint Paul's Hospital Millennium Medical College (SPHMMC), in Addis Ababa. This tertiary hospital is the second-largest referral and teaching hospital in the country. Currently, it has an inpatient capacity of 700 beds and an average of 1,200 emergency and outpatient clients daily. The hospital is one of the few COVID-19 centers in the capital city and currently runs the largest COVID-19 treatment center in the country.

Study population and sampling
All clients visiting the hospital were the source population, of which randomly selected clients were included as a study population. Being an adult (> 18 years) was an inclusion criterion and history of COVID-19 infection was an exclusion criterion. A single population proportion formula with a marginal error of 5% (d = 0.05); and a standard score corresponding to 95% con dence was considered to determine the sample size. We calculated separate sample sizes for each of the two study objectives (level of practice and factors associated with practice). We took the prevalence (p) of the practice of 'hand washing' from a similar study (77.3%) (13) which yielded the maximum sample size of 270. After considering a 5% non-response rate, the nal sample size for this study was 284 clients.
A systematic random sampling technique was used to identify respondents and conduct a face-to-face exit interview. The number of patients who visited the hospital's outpatient department on a daily basis three months before data collection was determined by consulting client register books. A sampling interval was determined by dividing the expected number of clients by the total sample size, which was then proportionally allocated to major departments in the hospital.

Study variables and measurement
COVID-19 precautionary measure was the dependent variable in this study, and the independent variables were divided into socio-demographic characteristics: age, gender, education, family size, marital status, and occupation; knowledge level and source of information on COVID-19 transmission, prevention, and control methods; attitude towards COVID-19 prevention, and control measures; perceived level of risk of contracting COVID-19 and presence of underlying chronic illnesses of the respondent or their family.
Bloom's cut-off point of > 80% on the aggregate score was adapted to rank respondents' overall knowledge, attitude, and practice scores (14). Practice, knowledge, and attitude were operationally de ned as follows: Satisfactory practice: a score of > = 80% on nine items with a possible response and point score of "Always = 2", "Occasionally = 1" and "Never = 0" assessing the practice of COVID-19 precautionary measures.
Adequate Knowledge: a score of > = 80% on 19 items with a possible response and point score of "Yes = 1" or "No = 0" assessing the knowledge of COVID-19 prevention measures, transmission ways, and disease symptoms. Favorable attitude: a score of > = 80% on ve, ve point-Likert scale items with a response and point score of "Strongly agree = 5", "Agree = 4", "Neutral = 3", "Disagree = 1" to "Strongly disagree = 0" assessing the attitude towards COVID-19 prevention and control measures.

Data collection
Data were collected by trained data collectors using a structured tool adapted from WHO's COVID-19 knowledge, risk perceptions, precautionary behaviors survey tool (15). The tool was pre-tested on a sample of clients at a nearby Selam health center. The questionnaire was translated into Amharic (the local working language) and back to English to assess the consistency of meaning of items on the two tools. The Amharic version of the questionnaire was used for the interview.

Data analysis
The collected data were entered into Epi Info and exported to SPSS version 23 for analysis. Firstly, descriptive statistics such as frequency, proportion, mean and standard deviation were computed. Then, binary logistic regression analysis was implemented to examine factors associated with prevention practice. Variables with a p-value of 0.25 or lower on binary analysis were further analyzed in a multivariable logistic regression model. Odds ratio with 95% Con dence Interval (CI) was used to test signi cance association.

Ethical consideration
Ethical clearance was obtained from SPHMMC Institutional Review Board. The participants were informed on the purpose of the study, con dentiality, voluntary participation, anonymity, and withdrawal. Oral informed consent was obtained from all the respondents.

Socio-demographic characteristics
A total of 284 clients were initially approached for an interview. After excluding those who refused to participate and those with incomplete responses, 262 (92.2%) respondents were included in the nal analysis. The mean age of participants was 36 (± 12) years. [ Table 1] Perceived risk of COVID-19 Just above a third (101, 38.5%) of the respondents reported that they had a chronic illness; diabetes was the most commonly reported (35, 34.6%). The majority (79.8, 50.8%) of the respondents perceived their level of risk of contracting COVID-19 was "low" or "medium". [ Table 2]

Factors associated with COVID-19 prevention practice
The binary analysis outputs showed that respondents who had primary school education had greater odds of poor practice as compared to those with no formal education (Crude odds ratio (COR) 0.32 (0.14-0.76)).

Discussion
Just above half (55.7%) of the study participants had satisfactory COVID-19 prevention practices. The face mask was the most commonly utilized precautionary method. More than two-thirds of the respondents reported they always wore a face mask. This was in line with the knowledge assessment result where almost all of the respondents mentioned face masks as the primary means of prevention. This was also in agreement with previous studies that reported a positive association between COVID-19 knowledge and utilization of face masks (16). However, according to WHO, the use of a mask alone is insu cient to provide an adequate level of protection and must be combined with hand hygiene to prevent transmission of COVID-19 (17). In the present study, hand hygiene practice was low, as less than half of respondents reported washing hands with water and soap, and only a third reported using sanitizers. On top of this, hand hygiene practice levels are declining as compared to a year ago at the start of the pandemic (13,18). This nding might indicate that inadequate hand hygiene might be hindering the effort to slow the spread of the virus.
In this study, the lowest score on COVID-19 precautionary measures was for social distancing; only a quarter of the respondents avoided crowded places or social gatherings, and a mere one-tenth maintained physical distance. This nding also markedly showed a decline as compared to previous studies (19). A multitude of factors can explain the poor adherence. For one, the odds of not adhering to all social distancing rules increases if a participant does not self-identify as highly vulnerable to 19).
Congruently, in this study, less than a third of the respondents perceived their risk of infection as high, indicating a lower perceived vulnerability. Another possible explanation could be that maintaining social distance may not be always entirely possible due to living in crowded areas (19) and use of public transportation, leading to unintentional non-adherence of social distancing guidelines (20).
The KAP (Knowledge, Attitude, and Practices) theory states that for the adoption and formation of new behavior, both the acquisition of knowledge and generation of attitude is important (21). In this study, a favorable attitude towards COVID-19 preventive measures was strongly associated with good practice.
Likewise, previous studies have reported positive attitude had a signi cant and robust impact on preventive behavior (6, 19,22,23). For instance, participants who felt COVID-19 will be successfully controlled were more likely to refrain from handshaking (13). Similarly, a favorable attitude predicted the intention of handwashing, one of the key precautionary measures against COVID- 19 (22). This nding might have implications for the adoption of public health education strategies targeted at improving attitude among a segment of the community with poor practice.
In this study, as compared to being from outside the capital city, being a resident was associated with better practice. The possible explanation for this could be the relatively strict enforcement of COVID-19 Public Health and Social Measures (PHSM) in the city as compared to parts of the country (4). For instance, a mandatory face mask is a requirement to receive any public service in the city. This nding was in agreement with studies that reported implementation of a mask mandate led to widespread uptake of masks (24,25 Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing Interests
The authors declare that they have no competing interests Funding Funding was obtained from SPHMMC's research resource mobilization and project management directorate. The study's sponsors had no role in study design, the collection, analysis, interpretation of data, or manuscript preparation.

Authors' Contributions
HB conceived designed the study, interpreted data, and prepared the manuscript. SM prepared data a collection instrument, analyzed data, and AY analyzed, interpreted data, and reviewed the manuscript. All authors read and approved the nal manuscript.