The study took place in the four natural regions of Guinea (Lower Guinea, Middle Guinea, Upper Guinea and Forest Guinea). Latitude North 7°30’ and 12°30’ Longitude West: 8° and 15°, with an area of 245,857 km² and 12,907,395 inhabitants Guinea-Bissau, Senegal, Mali, Ivory Coast, Sierra Leone, and Liberia .
Study type and period
We conducted a mixed cross-sectional study. The quantitative survey laid on knowledge and attitudes about COVID-19, and the qualitative part was on the perception of COVID-19. The study occurred from 23 to 25 August 2021.
For the diversity of opinion, our target population will consist of the general population on the one hand and health care staff on the other hand.
- Free and informed consent;
- At least 18 years old at the time of inclusion;
- Available and able to express themselves.
- Refusal to participate in the survey;
a) Selection method
The selection was made at two different levels.
- Health care staff: First, we randomly selected health facilities in the natural regions of Guinea based on the list of operational health facilities provided by the Ministry of Health and then chose the health personnel who met the criteria.
- The general population. We randomly selected the workplaces and then randomly recruited the participants.
Health care workers. We hypothesised that 70% of health care workers favour vaccination. With the desired precision of 5%, the sample size was calculated using the formula.
The minimum expected size is 322; considering the 10% of non-response rates, this size was increased to 370 per health district in the region.
General population. We hypothesised that 50% of the population favour vaccination. The formula calculates the sample size with the desired precision of 5%.
The expected minimum size is 384; considering the 10% of non-response rates, this size will be increased to 420 per prefecture in the natural region.
Participants from the health care staff, considering the socio-professional categories, and participants from the general population were interviewed in each area of the natural region selected. We interviewed 20 participants among the health care personnel for the four natural sites and 50 participants among the general population, i.e. 70 participants per region.
1. Training of the data collectors
The interviewers were trained in the survey’s methodology, the collection tools and the collection technique. A pre-test of the tools was organised to assess the feasibility of the field survey. This pre-test focused on areas unselected for the study.
2. Data collection
Investigators used Android phones to administer the questionnaires to patients at the workplace or by appointment at the nearest or most convenient location. The data was recorded through an Android application (ODK) downloaded and connected to the ONA server (https://ona.io/home/). For the qualitative part, we used a semi-structured interview, and the participants’ permission was obtained for the recording of the interview. The participants for the qualitative survey were from the respondents to the quantitative part; if necessary, other participants were selected from the same sites.
Theoretical framework and variables. The theoretical framework is based on the Fishbein integration model . The elicitation or preliminary analysis consisted of a literature review. Our model was adapted from existing work [18–20]. The model presented in Figure 1 incorporates socio-demographic characteristics, disease history, vaccination information, disease perception and apprehension, and barriers. These different elements influence the individual’s attitude. Finally, norms and the ability to be vaccinated potentiate the previous factors to predict vaccination against COVID-19.
- Study variables
A- Dependent variable: this is the participant’s vaccination status. The participants were asked about their vaccination status.
B- Independent variables
- Socio-demographic variables
- Age: in completed years; sex represented by male and female, marital status (single, married); residence; the level of education; occupation (job held by the participant), the number of persons in the household, number of persons with age ≥ to 18 years, the duration of stay in the place of residence (less than six months, more than six months), monthly income and current pregnancy.
- Variables on medical history. The participant’s existing medical history: asthma, hypertension, diabetes, obesity or overweight, allergic diseases (sinusitis, rhinitis, allergies, vaccinations), and other chronic diseases.
Variables for COVID-19 and vaccination. This section consists of three parts:
- Existing knowledge about vaccination: this item asked whether the participant had any prior knowledge about vaccination in general; the items were based on the definition of vaccination, types of vaccine, post-injection adverse effects, and individual and herd immunity. It was simply a question of whether they knew the general principle of vaccination for the general population.
- Sources of information: these are the primary sources of news used by the participants to discover the disease. Have you sought news about COVID-19 recently?
- Perception/fear of COVID-19: for this section, we have three items on a scale of 1 to 5 (strongly disagree, disagree, neutral, agree, strongly agree).
- Perceived Susceptibility of the disease. I am likely to get COVID-19, I am at risk of COVID-19, I may get COVID-19.
- Perceived seriousness of the disease. I think that COVID-19 is a severe health problem. I believe that COVID-19 has negative consequences. I think that COVID-19 is highly harmful.
- Fear of COVID. We used the French version of the COVID-19 fear scale (I am terrified of COVID-19. Thinking about COVID-19 makes me feel uncomfortable; my palms get sweaty when I think about COVID-19. I am afraid of dying of COVID-19. Watching the news about COVID-19 on social networks makes me nervous or anxious. I cannot sleep because I am fearful of catching COVID-19. My heart rate increases, or I have palpitations when I think about COVID-19).
Attitudes and beliefs are composed of three elements, similar to perception. The rating is on a scale of 1 to 5 (strongly disagree, disagree, neutral, agree, strongly agree).
Perceived benefits of COVID-19 vaccination: taking COVID-19 vaccination will help prevent coronavirus, and taking COVID-19 vaccine will help strengthen the body’s immunity in fighting viruses.
Perceived barriers to COVID-19 vaccination. The vaccine is reserved for a group of people, and the vaccine is too expensive for me; the procedure to get the vaccine is laborious.
Perceived threats to COVID-19 vaccination. I am afraid of the side effects of the vaccine. I have learnt that the vaccine induces the disease; the vaccine is made to prevent reproduction. I don’t trust vaccines.
- Ability. Rating on five points. I will be able to get the vaccine to avoid getting COVID-19. It will be easy for me to get a vaccine to protect me from COVID-19.
- Intention to receive COVID-19 vaccine. I am seeking the COVID-19 vaccine. I will vaccinate myself if my family or friends ask me to do so.
Quality control and assurance: A system was put in place that included strict monitoring of the fieldwork progress, including the geolocation of interviewers. Data were regularly checked and corrected as they were collected. A data manager checked the internal consistency and validity of the data. Any inconsistencies were reported and dealt. Finally, integrity constraints will limit the occurrence of missing data.
Statistical analysis. We recoded some variables before data analysis. Age into two classes (under 40 considered young, over 40 adults). For the following variables: perception/fear, attitudes/beliefs, subjective norms, ability and intention to receive the vaccine, we classified them according to the average of the scale scores. Thus, for perception, participants with a score above or equal to the mean were considered to have a positive perception. Otherwise, the perception was negative. For attitude and belief, we divided into two parts: items related to negative attitude (when the score is lower than the mean, the attitude is less negative; otherwise, the attitude is more negative) and those related to positive attitude (when the score is lower than average, the attitude is less positive; otherwise, the attitude is more positive). For norms, when the score of the scales was below average, the norms are considered favourable; otherwise, the norms are unfavourable). When the score was below average for ability, the participants are deemed unable; otherwise, they are able.) Finally, when the score was below the mean of income for the intention, the participants had less intention to be vaccinated; otherwise, they had more intention to be vaccinated. Quantitative variables were analysed using the median and interquartile range, and qualitative variables using the percentage. We considered households with high income when the mean income is ≥2000000GNF, and the number of people in the household is ≤10. For low-income households, when the mean income is <2000000GNF and the number of people in the household is >10, all other cases were considered middle income. The Chi-square or Fisher test and the Student or Wilcoxon test were used for the descriptive analysis. We used multivariate logistic regression between the participants’ vaccination status and the independent variables to identify facilitators and barriers. Then, we put in the classification and regression tree (CART) the significant variables of the previous regression models for the health care staff and the general population while keeping the dependent variable. The last analysis was backed by qualitative research with the thematic content method. The statistical tests were considered significant at the risk α = 0.005. The software R version 4.1.2 and Stata 15 were used.