Knowledge, attitudes, and practices towards seasonal influenza vaccination among pregnant women and healthcare workers: A cross‐sectional survey in Afghanistan

Abstract Background Despite recommendation by the World Health Organization (WHO), influenza vaccination coverage among high‐risk groups remains suboptimal in Afghanistan. This study aims to document the knowledge, attitudes, and practices of seasonal influenza vaccine uptake among two priority groups, pregnant women (PWs) and healthcare workers (HCWs). Methods This cross‐sectional study enrolled PWs and HCWs in Kabul, Afghanistan, from September to December 2021. Data on vaccine intention and uptake, knowledge, and attitudes towards vaccination were collected. Simple linear regression was used to predict the impact of sociodemographic characteristics on the KAP score. Results A total of 420 PWs were enrolled in Afghanistan. The majority (89%) of these women had never heard of the influenza vaccine but 76% intended to receive the vaccine. Of the 220 HCWs enrolled, 88% were unvaccinated. Accessibility and cost were factors which encouraged vaccination among HCWs. Fear of side effects and affordability were identified as key barriers. HCWs reported high level of vaccine intention (93%). PWs aged under 18 years (β: 6.5, P = 0.004), between 18 and 24 years (β: 2.9, P = 0.014), currently employed (β: 5.8, P = 0.004), and vaccinated against COVID‐19 (β: 2.8, P = 0.01) were likely to have a higher attitude score. Among HCWs, being female was a predictor for poor vaccination practice (β: −1.33, P < 0.001) whereas being vaccinated against COVID‐19 was a predictor for higher practice score (β: 2.4, P < 0.001). Conclusion To increase influenza vaccination coverage among priority groups, efforts should be made to address issues such as lack of knowledge, limited availability, and cost barriers.


| Study design
We conducted independent community-based cross-sectional surveys for PWs and HCWs in Kabul, Afghanistan, from September to December 2021. PWs and HCWs were enrolled from the French Medical Institute for Children (FMIC), the Malalai Maternity Hospital, and Rabia Balkhi Maternity Hospitals in Kabul city and randomly selected clinics in the districts of Bagrami and Deh Sabz, Kabul. A convenience sample was taken where HCWs at the above-named health facilities and PWs coming for antenatal visits were enrolled after obtaining written informed consent.

| Study tools
Quantitative data were collected using separate standardized tools for The second section asked about sociodemographic characteristics such as gender, age, employment status, and COVID-19 vaccination status. The English language questionnaire was first translated into a regional language (Dari) and a back translation was done to ensure consistency. The translated tools were pretested in a small sample. The questionnaire took 15-20 min to complete and was administered by trained community health workers (CHWs)/data collectors at all study sites on an online mobile app built on Redcap software. Data were then transferred and stored at a local server at Aga Khan University Hospital daily with restricted access.

| Sample size
We assumed an adequate knowledge level of 50% among HCWs and PWs for a maximum sample size with a margin of error of ±5% and an alpha level of 0.05 resulting in a sample size of 220 in HCWs and 420 in PW after considering 10% incomplete responses. The sample size assumptions were varied to account for feasibility given the COVID-19 and security conditions in Afghanistan.

| Statistical analysis
Mean ± SD were reported for continuous variables, median (IQR) were reported for non-normal discrete variables, and numbers with percentages were reported for categorical variables. Attitude scores were obtained by adding Likert scale points for each question, values ranged from strongly agree = 5 to strongly disagree = 1. Similarly, knowledge and practice scores were calculated. Response values were set as "yes" = 2, "no" = 1, and "don't know" = 0. Simple linear regression model was used to predict the knowledge, attitude, and practice scores at univariate level using sociodemographic characteristics; all the factors with P value < 0.10 were considered for multivariate analysis. A final multivariate model was build using stepwise backward elimination, P value < 0.05 was considered statistically significant. The level of significance was set at 0.05. STATA version 17.0 was used for all analyses.   Table 1 describes the sociodemographic characteristics of the 420 enrolled PWs. Most respondents (33.5.0%) were aged between 25 and 29 years, 67.1% received no formal education, and only 4.5% were employed. Mean gestational age at the time of interview was 25.3 ± 11.0 weeks. Median number of total pregnancies was 4 (2-5).

| Sociodemographic characteristics of the study participants
Only 17.4% of the PW had received the COVID-19 vaccine.  disease severity (91%) and provided immunity (89%). According to HCWs, the three most important groups to be vaccinated should be healthy adults (83%), older adults (65 years and above) (55%), and school-aged children (31%). Most common reasons for vaccine refusal among HCWs were fear of side effects (79%), cost of the vaccine (79%), and unavailability of vaccine (41%) whereas accessibility (69%) and cost (65%) were the most common reasons for vaccine acceptance. More than half the HCWs (78%) were willing to take the vaccine if an international organization (such as the WHO) recommended it. Eighty-eight percent of the HCWs were actually unvaccinated; however, 93% of them intended to get vaccinated and 97%

| Knowledge, attitudes, and practices of HCWs regarding influenza vaccine
intended to recommend the vaccine to their patients (Figure 4). Tables 3, 4, S1, and S2 describe the adjusted and unadjusted simple linear regression analysis of sociodemographic variables on KAP scores. PWs under the age of 18 years (β: 6.5, P = 0.004) and 18-24 years (β: 2.9, P = 0.014) had higher attitude scores as compared with older women. Similarly, women who were currently employed (β: 5.8, P = 0.004), or had received the COVID-19 vaccine (β: 2.8, P = 0.01) were a predictor for higher attitude scores. An education level of higher secondary or above (β: À4.51, P = 0.01) was significantly associated with a poor attitude score.

| Sociodemographic predictors of KAP scores among PWs and HCWs
F I G U R E 4 Attitudes and practices of the healthcare workers.

| DISCUSSION
We assessed knowledge, attitudes, and practices related to influenza vaccination uptake among two priority groups, that is, HCWs and Our study revealed that vaccine accessibility and cost were key drivers for vaccine acceptance. Similarly, a previous study from Qatar showed that free vaccination was associated with significant improvement in vaccine uptake among the HCWs. 11 Majority of the HCWs were willing to receive the influenza vaccine when recommended by international organizations (such as the WHO) or local government.
This suggests an opportunity to conduct national immunization campaigns and establish guidelines to improve vaccine adherence among this group. We report 81.8% of the HCWs to be vaccinated against COVID-19. Additionally, being vaccinated against COVID-19 was found to be positively associated with higher vaccination practice scores among HCWs, indicating that the COVID-19 pandemic might have increased their intention to get vaccinated against influenza as reported by Kong et al. 12 We also assessed the knowledge, attitudes, and practices of 420 PWs from Kabul, Afghanistan. A substantial deficit was seen in the knowledge of the participants regarding the disease and its prevention. Only, 23% of the PWs were aware of seasonal influenza, 11% had ever heard of its vaccine, and 28% strongly believed that the vaccine protected their unborn baby. Although majority (76%) of the PW wanted to receive the vaccine, only 11% had received a India also found that the women were willing to be vaccinated if it were recommended by a healthcare provider and if they were informed of vaccine safety during pregnancy. 13 In our study, being currently employed was a predictor for higher attitude scores. This may be due to greater autonomy in making personal decisions.
To the best of our knowledge, this is the first KAP study of seasonal influenza vaccination among two priority groups, HCWs and PWs in Afghanistan. We used a standardized tool for data collection and enrolled participants from both public and private healthcare facilities. One of the limitations of our study was that the responses could be subject to recall bias. The participants were conveniently selected from hospitals in Kabul and clinics in the adjoining districts and therefore might not be representative of the whole country. In addition, our study was conducted during the waxing and waning COVID-19 waves, and thus, we could not achieve a higher sample size.

| CONCLUSION
In conclusion, this study reveals substantial gaps in the knowledge, attitudes, and practices of both HCWs and PWs regarding influenza vaccine which need to be systematically addressed. This study can provide a framework upon which targeted interventions can be developed.