COVID-19 Vaccine Hesitancy Among Pregnant Women in Pakistan: A Cross-Sectional Survey

Pregnant women are at increased risk of severe outcomes from COVID-19, including higher rates of ICU admission, mechanical ventilation, mortality, and adverse pregnancy outcomes such as preterm birth. The pandemic has also exacerbated anxiety and depression, which can negatively affect pregnancy outcomes. Although vaccines are a key public health tool, early COVID-19 vaccine trials excluded pregnant women, contributing to uncertainty and hesitancy in this group. Professional bodies (e.g., ACOG, CDC, EMA) subsequently recommended COVID-19 vaccination during pregnancy, noting that authorized vaccines do not contain live virus or harmful adjuvants. In Pakistan, widespread conspiracy theories and social media misinformation have historically impeded vaccination campaigns and have been particularly influential during the COVID-19 era, potentially elevating vaccine hesitancy among pregnant women. This study aims to quantify the extent of vaccine hesitancy among pregnant women in Pakistan, identify its determinants, and compare attitudes with non-pregnant women of reproductive age.

Vaccine hesitancy is defined by WHO SAGE as delayed acceptance or refusal despite availability and is a barrier to controlling vaccine-preventable diseases. Global evidence shows variability in COVID-19 vaccine acceptance, with lower acceptance in some emerging economies and among certain groups (e.g., women, lower education/income). Studies in Europe and North America report concerns about safety, efficacy, and necessity as common drivers of hesitancy, with social media as a key influence channel. Among pregnant women, emerging data indicate that COVID-19 vaccination does not worsen perinatal outcomes and that acceptance is associated with trust in healthcare systems, perceived safety/efficacy, adherence to public health measures, and reliable information sources. In Pakistan, conspiracy narratives and distrust have historically undermined vaccination efforts, underscoring the need to understand local determinants of hesitancy during pregnancy.

Design: Cross-sectional survey conducted November 2021–February 2022 at the outpatient Obstetrics and Gynecology department of Benazir Bhutto Hospital, affiliated with Rawalpindi Medical University, Pakistan. Participants: Pregnant women (n=372 analyzable) and non-pregnant women of reproductive age (n=353 analyzable) who had not received a COVID-19 vaccine during the study period. Sampling: Convenience sampling; target sample size ≥369 per group (5% margin of error, 95% CI, estimated 60% vaccination rate). Instrument: Urdu-translated, validated Vaccination Attitudes Examination (VAX) scale (12 items; four subscales: mistrust of vaccine benefit [items 1–3], worries about unforeseen effects [4–6], concerns about commercial profiteering [7–9], preference for natural immunity [10–12]). Items are scored 1–6 (1=strongly disagree to 6=strongly agree); higher scores indicate stronger anti-vaccination attitudes. Additional binary (Yes/No) items assessed beliefs and experiences related to COVID-19 vaccination: disbelief in vaccines, belief in social media rumors, prior unpleasant vaccine side effects, insufficient information, not being afraid of COVID-19, disbelief in SARS-CoV-2. Data collection: Paper-equivalent online questionnaires were distributed via web links to minimize in-person contact; each participant could submit only once (automated filtering). Variables: Demographics (age, origin, marital status, income, education, employment), VAX scores, vaccination confidence/hesitancy (confident/unsure/hesitant), and additional belief items. Ethics: Approval by Benazir Bhutto Hospital Scientific Ethics Committee (Ref: BBH-2022/006052); informed consent obtained; adherence to Helsinki Declaration. Analysis: SPSS v26. Descriptive statistics (means±SD; frequencies/percentages). Group comparisons via t-test/ANOVA; VAX scores as median (IQR) with non-parametric median test. Correlations via Pearson/Spearman. Univariate risk analysis using odds ratios (OR) with 95% CIs. Multivariate backward stepwise logistic regression identified independent factors associated with hesitancy among pregnant women. Chi-square/Fisher’s exact tests compared proportions. Significance threshold α=0.05.

• Sample: 372 pregnant and 353 non-pregnant women completed the survey. Groups were similar in age, origin, marital status, and education; pregnant women were more often unemployed (23.66% vs 12.75%, p=0.013) and had lower income (below average: 58.33% vs 50.01%, p=0.011).
• Trust in vaccines: About one-third trusted the COVID-19 vaccine (pregnant 33.33%; non-pregnant 30.31%); trust in other vaccines was higher (pregnant 82.53%; non-pregnant 89.52%).
• VAX scores: Median total VAX score was higher in pregnant women (32 [IQR 9]) vs non-pregnant (28 [IQR 10]), p<0.001. Pregnant women scored higher on all subscales: mistrust of benefits (8 vs 6, p<0.001), worries about unforeseen effects (12 vs 10, p=0.012), commercial profit concerns (7 vs 5, p<0.001), and preference for natural immunity (9 vs 8, p=0.011).
• Vaccination stance: Pregnant vs non-pregnant—Confident: 40.86% vs 29.46%; Unsure: 7.80% vs 32.29%; Hesitant: 51.34% vs 38.24% (p<0.001). Thus, hesitancy was significantly higher and uncertainty markedly lower among pregnant participants.
• Other reasons for hesitancy (pregnant vs non-pregnant): Trusting social media rumors 39.52% vs 31.16% (p=0.029); previous unpleasant side effects 23.39% vs 21.25% (ns); insufficient information 15.05% vs 19.83% (ns); not afraid of COVID-19 12.37% vs 14.45% (ns); not believing in SARS-CoV-2 5.38% vs 5.38% (p=0.022 reported); not believing in vaccines 4.30% vs 7.93% (p=0.054).
• Among pregnant women, distribution by decision factors showed higher hesitancy among those trusting social media (63.35% of hesitant), with significant associations for social media trust and previous side effects.
• Univariate risk factors (OR, pregnant): Rural origin 1.89 (1.21–2.07), p=0.010; below-average income 2.68 (1.88–3.25), p=0.001; trusting social media rumors 3.29 (2.56–4.08), p<0.001; not afraid of COVID-19 2.78 (2.01–3.33), p<0.001; not believing in SARS-CoV-2 3.64 (2.64–5.21), p<0.001; not believing in vaccines 5.18 (3.31–7.29), p<0.001. Trusting the COVID-19 vaccine was protective: OR 0.55 (0.31–1.46), p=0.033.
• Multivariate adjusted factors (pregnant): Trusting social media rumors adj OR 2.58 (2.09–3.22), p<0.001; not afraid of COVID-19 adj OR 2.01 (1.76–2.56), p=0.010; not believing in SARS-CoV-2 adj OR 2.53 (2.01–3.12), p<0.001; not believing in vaccines adj OR 4.25 (2.68–6.43), p<0.001. Rural origin was not a significant independent factor (adj OR 1.47 [0.87–2.15]).

This study demonstrates that pregnant women in Pakistan exhibit higher COVID-19 vaccine hesitancy and stronger anti-vaccination attitudes than non-pregnant women. Despite similar general trust in routine vaccines, pregnant participants expressed more mistrust of COVID-19 vaccine benefits and greater concerns about unforeseen effects and commercial motives. Social media misinformation emerged as a major driver of hesitancy, alongside low perceived risk of COVID-19 and disbelief in the virus and vaccines. These findings align with international literature indicating the influence of information sources and trust in healthcare systems on vaccine acceptance. Given the heightened risk of severe COVID-19 in pregnancy and evidence supporting vaccine safety during pregnancy, targeted interventions addressing misinformation, enhancing risk perception, and building trust are crucial. Tailored communication via trusted healthcare providers and community channels may improve uptake.

COVID-19 vaccination is urgently needed among the general population and particularly among pregnant women in Pakistan. Vaccine hesitancy in this group is prevalent and strongly associated with social media rumors, low perceived risk, and disbelief in COVID-19 and vaccines. Targeted, evidence-based strategies and policy measures—focused on countering misinformation, improving health literacy, and leveraging trusted healthcare professionals—are recommended to increase vaccine acceptance in pregnant women.

• Convenience sampling and single-hospital, population-based survey may limit generalizability to other regions or populations.
• Reliance on an online questionnaire format may introduce selection bias.
• Sample size estimation was not stratified for all subgroups, potentially limiting power for subgroup analyses.
• The study did not include men, precluding assessment of gender differences in hesitancy beyond pregnancy status.
• Context-specific factors in Pakistan (e.g., media environment, socioeconomic disparities) may limit applicability to other settings.