Adherence to COVID-19 preventive measures and associated factors in Oromia regional state of Ethiopia

Background Adherence to preventive measures of Coronavirus disease 2019 (COVID-19) was among the means to tackle the transmission of the virus. However, reluctance to implement the recommended preventive measures has been reported to be a major problem everywhere including Oromia Regional State. Purpose This research was aimed to assess the level of adherence to COVID-19 preventive measures and associated factors in the study area. Participants and methods Community based cross-sectional study was conducted. Sample of 2751 adults aged ≥ 18 years were used for the quantitative study. Also, 20 FGDs and 30 KIIs were conducted in the qualitative approach. The collected data were entered into Epi info version 7.2.0.1 and analyzed using STATA 15. The qualitative data were entered into NVivo version 12 for its organization. Bivariate and multivariable binary logistic regression analyses were conducted to determine the association between the study variables. Odds Ratio with its 95%CI was calculated and P- Value < 0.05 was used as a cut off points to declare the significance. Results The level of adherence to COVID-19 preventive measure was 8.3. Age [AOR, 4.00; 95% CI: 1.50, 10.45], Illiterate AOR, 0.38; 95% CI: 0.15, 0.93], read and write [AOR, 0.26; 95% CI: 0.10, 0.72], attended primary [AOR, 0.30; 95% CI: 0.13, 0.70], occupation (AOR; 95% CI: 0.29, 0.96] and knowledge [AOR, 0.20; 95% CI: 0.01, 0.11] were factors associated with level of adherence to COVID-19 preventive measures. Political context, unemployment, livelihoods, and social events were mentioned as reasons for the poor adherence to COVID-19 preventive measures. Conclusions The overall level of adherence to COVID-19 preventive measures in the study area was low. Age, level of education, occupation, and knowledge were factors associated with level of adherence to COVID-19 preventive measures. Activities to increase the adherence to COVID-19 preventive measures should be implemented by the concerned bodies.

General guidance is provided below.
Consult the submission guidelines for The Ethical issues was handled by the regional health Bureau and no human subjects were involved detailed instructions. Make sure that all information entered here is included in the Methods section of the manuscript.  Considering its pandemicity and absence of effective treatment, authorities across the globe have designed various mitigation strategies to combat the spread of COVID-19 7 . Accordingly, to limit the transmission, the World Health Organization (WHO) recommends minimizing contact between infected and non-infected persons, early detection and isolation of cases, and general personal and collective hygiene measures 6,8,9 . As part of these measures, the use of face masks, hand washing, physical distancing, cough etiquette and avoidance of crowded places are recommended 9 .
Although adherence towards preventive measures is the only means to tackle the virus, reluctance to do so has been reported to be a major problem everywhere 7 . Also, community's risk perception and poor adherence towards COVID-19 mitigation measures remains a major problem. A significant proportion of communities did not perceive the virus as a risk for health 10 .
People also think that it originated from a laboratory, and mostly causes mild symptoms, and affects the elderly 10 Thus, this study was designed to determine the level of adherence to COVID-19 preventive measures and associated factors in Oromia regional state by using a mixed method approach.
6 | P a g e

Study Area
This study was conducted in Oromia Regional State of ten selected zones and towns. Oromia is one of the largest and most populous regions in Ethiopia with an estimated population of 39,074, 846.

Study design and Period
A community-based cross-sectional design was conducted using a mixed method of quantitative and qualitative approach from September 2020 to March 2021.

Population
All adults living in the Oromia Regional State during the study period were the source population, while all adults living in the selected households of Oromia region during the study period were the study population.

Inclusion and Exclusion Criteria
Men and women aged 18 years and above who have been residing in the area for at least six months were included in to the study, while those who were critically sick, hearing difficulties and unable to communicate during the data collection time were excluded from the study.

Quantitative
The sample is required to compare the adherence of COVID-19 preventive measures among urban population in comparison to rural population is calculated using formula to compare and test difference between two population proportions in comparative study designs. In using this formula, the following assumptions were considered: the proportion of adherence to COVID-19 preventive measures among urban population is 50% in the absence of previous study. Under With these assumptions, considering the scenario where the alternate hypothesis is true and the proportions are significantly different and the general formulae is given as follows.
8 | P a g e By replacing the corresponding values for the symbols in the formulae and having design effect of two and adding for the possible non response rate, a total of 2851 respondents were obtained where 1426 from urban areas and another 1426 respondents from rural areas (Woredas) were selected.

Qualitative
The intention of the study was to explore the perception, knowledge, attitudes and practices of people towards the COVID-19 preventive measures. The required information for this purpose not only acquired through survey but also qualitative methods. This approach was applied based on the assumption that it allows triangulate the method and data. That is, in addition to the collection of quantifiable information using survey method, the qualitative method helps to explore the lived experiences of the study participants in the context of COVID-19. In this regard, the qualitative method supplements the quantitative findings with evidence generation.
More specifically, the qualitative method mainly aimed at addressing the "why" people in the study area adhere/not adhere to the preventive practices of COVID-19 Hence, Key Informant Interview (KIIs) and Focused Group Discussion (FGDs) were used as methods of qualitative data collection.

Quantitative
The region was categorized into three clusters, namely; agrarian, semi-pastoralist, and pastoralists. The main reason to use this method was based on the fact that the region is heterogeneous with regard to economic, cultural, geographic and climatic conditions. From each geographic area zones and towns were randomly selected and three woredas per zones and three sub-cities per towns were also randomly selected to have the participants from the households.
After identifying households in the respective woredas and sub-cities participants were randomly selected to include in to the study. Using this method the residents have equal and independent chances of being enrolled in the study ( Figure 1).
Insert Figure 1 here Moreover, based on the aforementioned livelihood clusters, zones and towns the desired eligible sample was allocated proportionally. Using population to size proportionate methods the required sample size was determined in each study site. Then, using systematic sample by calculating interval (total HH population of the area/sample size) preferably the household heads or the available eligible were selected and included in the study (Table 1).
Insert Table 1 here Research collaborators in the six sites who are all healthcare practitioners working in the tertiary health facilities supported the recruitment of policymakers and service providers in the area. The service providers helped in the recruitment of the service users who presented for health care services during the study period ( Figure 2).
Insert Figure 2 here

Data Collection
The quantitative part of the study involves the collection of quantifiable and measurable data on the implementation of COVID-19 preventive measures endorsed by the government. In this regard, the preparation of the questionnaire was based on conceptual framework of the study and previous similar research work to answer the objectives. The questionnaire was first prepared in English, and then translated to Amharic and Afaan Oromo for data collection and back to English by different people to ensure its consistency. Health professionals having diploma and above were recruited based on their previous experiences of data collection and interest for data collection. Data were collected by face to face interview from the eligible. One participant was randomly selected from the household if there were two and above respondents to prevent intrahousehold correlation. The

| P a g e
To enhance the quality of the instruments of the data collection, pre-testing of the questionnaire was undertaken prior to data collection. In addition, three days training was given for data collectors and supervisors concerning the objective, the tools, methodology, and ethical issues.
During the data collection period, the collected data were checked for completeness and consistency by the supervisors and principal investigators. Moreover, each supervisor was given his/her own household enumerators and data collectors and reoriented them during each day before data collection. They also supervised them by cross checking the registered households and questionnaire for its completeness. Before starting data entry, unique codes were given to each questionnaire. Missing values and outliers were checked using frequency tabulations, residual plotting and managed accordingly. Data was edited and checked manually by hand for checking completeness both during collection and entering into data entry templates.

Data Management
Data was entered into Epi info version 7.2.0.1, data entry template and exported to STATA 15 software for analysis. Missing values and outliers were checked by frequency tabulations.
Randomly selected 5% of the data set was double entered to check the accuracy and similarities based on the questionnaires identification numbers. Any decision or changes used on the data set was clearly documented for further explanations of unexpected errors that may happen at the end of the day. In addition, check for item and unit-missing values, outliers for accuracy, causes of outliers were considered and determined.

Data Analysis
The quantitative data was analyzed using STATA 15 software. Descriptive statistical analysis such as frequency, percentages, proportions with 95% CI, mean and standard deviation were used. The associations between level of adherence to COVID-19 preventive measures and independent variables were modeled using binary logistic regression analysis. Simple logistic regression analysis was used to assess the existence crude relationship between independent variables and level of adherence to COVID-19 preventive measures. At this level the candidate independent variables for multiple regression analysis were selected at P-value < 0.25 significance level. Multiple logistic regressions were applied to estimate the adjusted effects of independent variables on level of adherence to COVID-19 preventive measures. The regression model was developed using forward stepwise approach. The odds of being adhered to COVID-19 preventive measures were estimated using odds ratio within 95% confidence intervals. At this level the significance of associations was declared at p-value of 0.05.
The final fitted model was assessed for assumptions like normality of continuous variables using histogram and normal curve, multicolliniarity between independent variables using Variance Inflation Factor (VIF) and goodness of fit using Hosmer and Lemishow test. Moreover, the model ability to correctly classify those subjects who experience outcome of interest and those who do not was assessed using Receiver Operating Characteristics (ROC) curve. Findings were presented on frequency tables, graphs and discussed accordingly.
The qualitative data analysis was begun with the work of transcription, translation and theme development during data collection. Initially the KII and FGD were transcribed and translated.
Then a workshop was prepared to develop themes by reading all translated data. The data was then entered into NVivo version 12 for its organization and management.

Operational Definitions and Measurements
Level of Adherence: Adherence towards prevention and control measures for COVID-19 was computed from the response category of the preventive measures endorsed by the government (hand washing, using a facemask, keeping physical distance, not travel to a crowded place, home stay, and not travel to anyplace during the pandemic) regularly practiced during 14 days before data collection time. The score was computed from those who properly practiced. Those respondents who scored 95% and above where labeled to have "Good adherence to COVID-19 preventive measures" and otherwise 7 .

Ethics Consideration
Ethical approval was obtained from the ethical review board of Oromia Regional State Health Bureau. Permission letters were secured from Regional and Zonal Health Offices and shared with the randomly selected health care facilities and community administrators. Assent for less than 18 years and verbal consent was obtained from participants.

| P a g e
Approval and permission was sought from the concerned bodies for the study. The ethical review was undertaken by all project and investigators ensured standard processes (dignity, autonomy, informed consent, confidentiality, anonymity, ability to adhere to protocol) and data security are maintained. Voluntary and informed participation, confidentiality and safety of participants constituted key principles of researcher respondent interaction. Informed verbal and written consents were obtained from residents, service users, service providers and policymakers prior to their enrolment in the study. The study was conducted according to the Helsinki declarations on ethical principles for medical research involving human subjects. Finally, the collected data was stored in a separate computer and kept confidentially. On completion of the study, both the quantitative and transcribed data were stored in password-protected computers/laptops and only the core research team has access to the data. ETB to 650,000 ETM with the median (+IQR) of 10,000 (+649,000) ETB (Table 2).

Socio-demographic Characteristics of the Study Participants
Insert Table 2 here

Knowledge about COVID-19
The majority of the respondents, 2525 (91.6%) have heard about COVID-19, but only 61.3% believe the existence of COVID-19 in their area. Moreover, less than one in ten, 258 (9.36%) of the respondents believes as COVID-19 is a killer disease ( Figure 4).
Insert Figure 4 here The qualitative data also shows that people have information about the disease. from their close friends ( Figure 5).
Insert Figure 6 here Insert Figure 8 here The study participants were also asked about the symptoms experienced by a person infected with COVID-19 when sick. Accordingly, about eight in ten (79.4%) and (75.93%) mentioned cough and fever, respectively, as the main symptom of COVID-19 when a person get sick.
Whereas, nearly 1% of the participants responded as there is no symptom from the COVID-19 infected person when sick (Figure 9).
Insert Figure 9 here  Also about 6% perceived as COVID-19 is not a killer disease and no need for the frequent use of preventive measures (Figure 10).

Attitudes towards COVID-19 preventive measures
Insert Figure 10 here Participants were asked about COVID-19 preventive measures whether it was practiced within their family members. Accordingly, the vast majority (93.76%) have practiced in covering their mouth and nose during coughing and sneezing. Moreover, about quarter (24.83%) of the family members have practiced at least any one of the preventive measures ( Figure 11).
Insert Figure 11 here The difference between awareness, knowledge, attitude and adherence to

COVID-19
Even though the awareness level of people was extremely high decreasing trend is seen across the knowledge, attitude and adherence to COVID-19 preventive measures ( Figure 12).

Factors associated with Adherence to COVID-19 Preventive measures
Insert Table 5 here The qualitative method also explored hindering factors for applying COVID-19 prevention mechanisms. Accordingly, socio-economic problems, lack of COVID-19 confirmed cases, low enforcement mechanisms and low level of perceiving risk were the main reasons for not practicing the prevention methods. For instance, one of the KII in Dinsho woreda described; "economic problem, politics and culture can be the reasons for not practice COVID-19 prevention method. Massive meeting and rallies conducted in different place that we observed affected our community to decrease the practice of COVID-19 prevention methods." A 40 years male As mentioned in the above quotation there were political events (in support of or against the existing political system) such as rally (public meeting) and violence that brought many people together created conducive environment for the spread of the disease. These circumstances made people to be careless and avoid using prevention methods. A key informant in Bale Zone Health Office pointed out one of the incident as follows: "The possible challenges not to use the preventive methods were the mass grievance and violence after the death of artist Hacalu Hundessa that the community said no disease but the political actors are the virus by themselves." A 34 years male The discussants and informants also described that there was lack of or loose law enforcement to re-enforce people in the use of the prevention methods in their area. The informant said that,

"In the beginning, law enforcement by the government had helped for proper utilization of COVID-19 prevention methods. Religious and cultural leaders are also played, major role in helping the community to proper use of COVID-19 prevention method. Later on, this law enforcement from the government declined. The people start to stop utilization of COVID-19 prevention methods. Currently, public gathering is underway without proper care in our area. Keeping social distancing and personal hygiene is not properly practiced in our zone." A 51 years male
Furthermore, lack of commitment from the side of the government itself made the rules of the prevention measures to be over sighted. Example, the key informant from Dinsho woreda described that the government itself did not adhere to the rules. He said, "We advised on different preventive methods and we practiced as much as we can after attentively follow. As a political concern we observed that still meeting of many people by the government during the time of

corona." A 28 years Female
The study discussants and informant also mentioned that absence of COVID

Limitations of the study
Firstly, due to the cross-sectional nature of the study design, it might be difficult to ascertain the cause effect relationship between the study variables. Secondly, social desirability bias might be introduced despite their poor actual implementation. Thirdly, the tool used in this study was developed by the research team based on the context and not previously validated and the reliability was checked using Cronbach's alpha.

Strength of the study
Through this community based survey, it was possible to conduct a face-to-face interview and Observation with maximum precaution than a simple telephone survey as others during the pandemic to evaluate the real response and adherence of the community towards mitigation measures. This study conducted in a highly spreading time of the pandemic being an input for the government and others actors to intervene. prevention has to be revitalized as well as possibility with serious precaution to be followed and implemented.            Percentage Tables   Table 1: Proportion of Sample size allocated to zones and towns of the study area, Oromia Region, September 2020 to March 2021