The aim of this study was to assess the influence of perception, attitude and level of trust on the uptake of COVID-19 vaccination among pregnant women attending ANC clinic in Mbeya Urban.
Study Design and setting
The study used a cross-sectional analytical design, employing a quantitative approach in data collection so to assess the influence of perception, attitude and level of trust on the uptake of COVID-19 vaccination among pregnant women attending ANC at primary level (Kiwanja Mpaka health Centre), secondary level (Mbeya Regional Referral Hospital (MRRH)) and tertiary level (Mbeya Zonal Referral Hospital (MZRH)) in Mbeya urban in Tanzania
Study Population
The study was conducted in Mbeya urban in the health centre, one regional referral hospital, and one zonal referral hospital. All these facilities had antenatal care clinics that were being conducted in week-days. The population of this study were all pregnant women of the reproductive age (15-49 years) attending ANC clinics in Mbeya Urban.
Sample size calculation and sampling procedure
The researcher sample size was obtained basing on the formula of single proportion population (Cochran. 1977).
Where:
n= minimum sample size
P= proportion from the previous study on the prevalence of pregnant women who had received COVID 19 vaccine which was 27% 8,9
E= Margin of Error. That is, 5% in the context of this study
Z= critical value of a normal distribution of 95% Confidence Interval equal to 1.96
Non response rate 10%.
Thus:
n=333 respondents
Therefore, the minimum sample size was 333 pregnant women who met the inclusion criteria
Sampling procedure
A random sampling method was used in this study and the researcher developed the sampling frame from ANC register for each facility and the total number of the study population from the selected facilities was about 2400 pregnant women who were used to calculate sampling interval needed to enable systematic sampling during data collection for the respondents who met the inclusion criteria. The total of study population from all three facilities identified from selected facilities was divided by 333 sample sizes to get sampling interval.
i.e. (i= N/n =2400/333 =7.2 approximately 7).
Therefore, the sampling interval was every 7th pregnant woman
Where: -
N= 2400 (total population for pregnant women attending ANC in Mbeya urban),
n= 333 (sample size for the study which is 333 of pregnant women) and
i= 7 (interval during data collection)
Data collection procedure
An approval letter was sought from the hospital director for data collection in Mbeya zonal referral hospital and from the Regional Administrative Secretary for data collection at Mbeya regional referral hospital and Kiwanja Mpaka health centre. Permission was sought from the unit in charge at ANC clinic and the pregnant women were approached and the researcher explained to them the purpose of the study in order to get their informed consent. For those who consented, they were asked to fill in the questionnaires. A systematic sampling technique was employed to prevent any bias during data collection. To achieve this, each facility had a minimum of 40 pregnant women who attended ANC clinic a day. The researcher took 1 month in each facility collecting data and sampling size in each facility was divided as follows; 100 respondents in Mbeya Zonal Referral Hospital, 100 respondents in Mbeya Regional Referral Hospital and 133 respondents from the Kiwanja Mpaka Health Centre. The allocation based on the fact that at primary level, there is high ANC attendance compared to secondary and tertiary level which has more referred cases that require more specialised service 10. The reason for using only one month in each facility was mainly aimed to prevent repetition of data collection from the same respondents because pregnant women attend ANC clinic once in every month. Therefore, in each week, the researcher could get a minimum of 25 respondents in each facility to participate in the study. The selection was made to every 7th pregnant woman who arrived from home for ANC clinic that day and met the study’s inclusion criteria the same was asked for her consent to participate in the study with the assistance from the ANC nurse. This was done just after taking the vital information from the pregnant women in queue waiting to enter the doctor’s consultation room.
Recruitment and training of research assistant
After developing data collection tools, two research assistants with background in research from the National Institute of Medical Research (NIMR) were identified. Both of them had Medical Doctor (MD) qualification background. After their identification, they were trained and oriented to the respective tools for successful data collection process in the field.
Data management and data quality assurance methods
The researcher tested the reliability through pre-testing of the questions to pregnant women attending ANC clinic in the Muhimbili National Hospital to make sure the questions were clear and understood by the respondents and for that, the questions could respond and measure what the study wanted to achieve. Nevertheless, the researcher tested the validity of the study using face validity, where he used 4 people (1 professor, 1 medical doctors with an MPH and who was working as an assistant lecturer and 2 assistant lecturers with postgraduate in Health Policy and Management). All of these were from MUHAS School of Public health and Social Sciences. They knew clearly about the topic of the study. They read the questionnaire and ensured the researcher that it had captured the necessary questions to be asked as well as the necessary data required to be collected effectively.
Many of the questions focused on the study and things that changed were the questions to clearly define the two dependent variables; perception and trust. This is because some of the questions on perception measured belief towards the COVID 19 vaccine and some questions in trust measured belief towards COVID 19 vaccine too. It should be noted that vaccination status did not use normal Likert scale of five levels of measurements because it had only two (vaccinated or not vaccinated). Finally, the gender question was removed since we were studying about pregnant women. So, changes were made to increase focus during data collection and the questionnaire was improved on the basis of reviewing the recommendations of the expert.
The questionnaire focused on collecting information related to perception, attitude and level of trust of pregnant women towards COVID-19 vaccine. The questionnaire was developed by the principal researcher through collecting questions on how to measure the variables collected from different studies written in English but which were then translated to Kiswahili then back to English.
Data analysis
The participants’ responses were entered into data collection software (Stata/SE 14.1) whereby all completed questionnaire were checked if they were error free i.e., their eligibility, range of values inconsistency and missing identification numbers. Later, all the submitted questionnaires were double entered into excel database then exported to stata for data checks and cleaning.
Data analysis corresponding to each specific objective:
To determine the proportion of the pregnant women attending ANC clinic who have been vaccinated, Socio-demographic factors associated with uptake of COVID-19 vaccine among pregnant women were measured by demographic questions in relation to individual factors associated with uptake/non uptake of COVID-19 vaccine among pregnant women. To effectively carry this out, frequency measure (n) and proportion (%) were calculated using Stata/SE 14.1 where software, continued variables were conducted to describe the study population in terms of mean, median, range and standard deviation.
To find out the perceptions influencing COVID-19 vaccine uptake among pregnant women attending ANC clinic; the perceptions associated with uptake/non uptake of the COVID-19 were measured by Likert scale from 1 to 5 responses where by the respondents mean score of <2.5 were labelled as having positive perception towards COVID 19 and the mean score of >2.5 was labelled as having negative perception towards COVID 19 11
In identifying the attitudes influencing COVID-19 vaccine uptake among pregnant women attending ANC clinic, the perception associated with uptake of COVID 19 was measured by Likert scale from 1 to 5 responses where by the respondents’ mean score of <2.5 were labelled as having positive attitude towards COVID 19 and the mean score of >2.5 was labelled as having negative attitude towards COVID 19 11
To examine the extent to which the level of trust influences the uptake of COVID-19 vaccine among pregnant women attending ANC clinic; the perception associated with uptake of COVID 19 was measured by Likert scale from 1 to 5 responses where the respondents’ mean score of <2.5 were labelled as having trust towards COVID 19 vaccine and the mean score of >2.5 was labelled as having mistrust towards COVID 19 11
To examine the individual factors influencing COVID 19 uptake among pregnant women attending ANC clinic; categorical variables and proportional frequencies were calculated. Pearson Chi square statistics test was used to compare group differences for categorical variables and all variables which were statistically significant using chi-square were included in the bi-variety and multivariate logistic regression.
To find the association between the dependent variable and all four objectives/variables; socio-demographic, perception, attitude and trust objectives, were added to software and multivariate binary logistic regression was done. Crude Odds Ratios (OR) and Adjusted Odds Ratio (AOR) with 95% confidence intervals (CI) after adjusting for confounders were reported and Fisher’s test was used to compare the mean difference between variables. Pie chart and bar charts were used for pictorial presentation of the results. Relationship, association and difference between variables were considered statistically significant at P<0.05.