Assessment of the impact of vaccine literacy on influenza vaccination practices among diabetic patients in Iraq Authors: Walid Al-Qerem, Alaa Hussein Alsajri, Anan Jarab, Judith Eberhardt, Lujain Al-sa’di, Lama Sawaftah, Safa Mohammed Aboud, Manar Ahmed Jameel, Ataa Ajeel Jassim Published: 2025 ABSTRACT ABSTRACTDiabetes mellitus (DM) increases susceptibility to infections and severe outcomes, making vaccination a critical preventive strategy. Despite this, vaccine uptake among patients with DM remains suboptimal. Vaccine literacy (VL), the ability to find, understand, evaluate, and use vaccine information, is a key determinant of acceptance, yet evidence from Iraq is scarce. This study assessed VL among diabetic adults in Iraq and examined its association with influenza vaccination. A cross-sectional study was conducted between January and August 2025 among adults with DM. Participants completed the Arabic Vaccine Literacy Assessment tool (HLVa-Ar), a sociodemographic survey, and vaccination history. Logistic regression identified factors associated with influenza vaccination, and quantile regression examined predictors of VL scores. Among 473 participants (median age 50 y; 53.1% female), the median VL score was 36 (IQR 30–41). Influenza vaccination coverage was 52.2%. Higher VL scores were positively associated with uptake (OR = 1.062, 95% CI [1.031–1.093], p < .001). Female gender and lower income predicted reduced uptake. Lower education and lower predicted lower VL. In conclusion, VL and socioeconomic factors influence vaccination behaviors among Iraqi adults with DM. Strengthening VL through targeted interventions may improve vaccine coverage and reduce complications in this high-risk population.KEYWORDS: Vaccine literacy, influenza vaccination, diabetes mellitus, Iraq, health disparities Introduction Diabetes mellitus (DM) represents a significant global health challenge, with its prevalence steadily increasing worldwide.1 The World Health Organization (WHO) defines DM as a chronic metabolic disorder characterized by elevated blood glucose (hyperglycemia) resulting from defects in insulin secretion, action, or both.2 In Iraq, the prevalence of DM has increased markedly, with approximately 17% of adults affected, including 8.1% who remain undiagnosed and therefore at higher risk of complications.3 Individuals with DM are particularly vulnerable to infections due to compromised immune systems, which can lead to severe complications and higher mortality rates.4 Vaccination remains a cornerstone of preventive medicine, particularly for individuals with chronic conditions like DM, who are at increased risk of infectious diseases, such as influenza and COVID-19.4 Despite being an important element of public health, vaccination faces a growing challenge due to increasing rates of hesitancy and refusal, which negatively affect public health by raising infection risk.5 Previous studies have demonstrated that patients with DM frequently exhibit low vaccination rates and face multiple barriers, such as inadequate knowledge and misconceptions about vaccines, limited healthcare provider recommendations, and concerns regarding adverse effects.6,7 Although many studies have identified knowledge gaps and attitudes toward vaccination in various populations, relatively few have focused on actual adherence to influenza vaccination among diabetic patients. Evidence from Saudi Arabia, Jordan, and Italy shows that influenza vaccination coverage in this high-risk group remains below recommended levels.6,8,9 Health literacy is broadly understood as the capacity to obtain and apply information needed to make sound health decisions, and it is known to influence many health outcomes, including vaccination uptake.10,11 When health literacy is limited, people may struggle to evaluate information or trust healthcare advice, which can reduce vaccine acceptance.12 The concept of vaccine literacy (VL) has therefore been developed to capture skills that are specific to immunization. VL involves being able to access, judge, and use information about vaccines to guide personal and family health choices.13 It also requires the confidence and motivation to evaluate sources critically, rather than simply knowing factual details.12,14 Because vaccination is a cornerstone of prevention, the level of VL within populations and among healthcare providers has important implications for public health programs. In Iraq, studies conducted during the COVID-19 era indicated a substantial level of vaccine hesitancy related to concerns about safety and effectiveness, unequal access, and variable trust in information sources.15–17 However, little is known about VL among patients with DM, a group particularly vulnerable to severe outcomes from vaccine-preventable infections. Evidence from other countries shows that VL influences vaccination uptake, but country-specific data are needed to design effective, context-appropriate interventions. Therefore, this study was designed to evaluate VL among diabetic adults in Iraq and explore its association with influenza vaccination practices. The findings will help clarify how VL shapes vaccination behavior in this high-risk population and provide an evidence base for interventions to improve vaccine uptake and reduce infection-related complications. Materials and methods This cross-sectional investigation enrolled adults with clinically diagnosed diabetes mellitus attending clinics in Iraq between January and August 2025. Participants were recruited from seven outpatient clinics, including five diabetes/endocrinology clinics and two general internal medicine clinics located in the Baghdad and Al-Anbar governorates. These clinics serve diverse urban and peri-urban populations and include both public and private healthcare facilities. Eligibility required age ≥18 y and willingness to participate. Potential participants were identified through clinical record review and invited during routine visits. A trained research assistant explained the study aims, assured confidentiality, and emphasized the voluntary nature of participation, including the right to withdraw at any time without penalty. The self-administered questionnaire took approximately ten minutes to complete. It included items on influenza vaccination history, covering whether the participant had ever received the influenza vaccine. This information was used to classify participants into ‘vaccinated’ and ‘never vaccinated’ groups for the logistic regression analysis. All procedures adhered to the Declaration of Helsinki, and ethics approval was obtained from the Department of Pharmacy, Al-Rasheed University College (Ref. RUCPD10012507). Data collection and study instrument To assess vaccine-related literacy, we employed the Arabic adaptation of the Vaccine Literacy Assessment tool (HLVa-Ar),18 which has been validated for use in Arabic-speaking settings based on previously published studies.19,20 The scale has two dimensions: one focusing on functional aspects (five items capturing ease of reading and comprehension, each rated on a four-point Likert scale) and another reflecting interactive and critical skills (nine items that address how individuals evaluate and apply vaccine information). In addition, participants provided demographic information (e.g. age, sex, education, marital status, household income, health insurance coverage, and medications). Finally, the questionnaire included items on vaccination practices, including influenza and other adult immunizations. To ensure data quality, several quality-control procedures were implemented throughout the survey process. First, all research assistants received standardized training on how to introduce the study, obtain consent, and assist participants without influencing their responses. Second, they were instructed to check each questionnaire for completeness on-site, allowing immediate correction of missing responses. The HLVa-Ar instrument used in this study has undergone rigorous psychometric validation, including both factor analysis and Rasch analyses, supporting its structural validity and reliability in Arabic-speaking populations. Internal consistency of the scale in our sample was also assessed using Cronbach’s α. These procedures collectively ensured the accuracy, validity, and consistency of the collected data. Sample size calculation The required sample size was determined according to the events-per-variable principle for logistic regression,21 which advises a minimum of ten outcome events per predictor. Because our analysis included seven predictors, a minimum of 70 cases in the smaller outcome group was needed. The final sample consisted of 473 participants; of those, 226 were in the smaller group (never vaccinated), greatly surpassing this threshold and providing sufficient statistical power for the planned models. Statistical analysis Data analysis involved descriptive and inferential statistics. Continuous data were summarized as the median with interquartile range, and categorical data as counts with percentages. Participants were grouped by their primary source of vaccine information: those reporting only healthcare professionals, official campaigns, internet searches or printed medical materials were classified as “scientific,” whereas those citing family, were placed in the “non-scientific” group. Reliability of the HLVa-Ar subscales was checked using Cronbach’s α. To explore factors linked to influenza vaccination (vaccinated vs never vaccinated), binary logistic regression was performed, including predictors such as age, sex, education, income, marital status, and VL score. A separate quantile regression model was fitted to examine socio-demographic and informational predictors of VL score distribution. The underlying assumptions for both the binary logistic and the quantile regression models were examined. For the logistic model, the linearity-in-the-logit requirement for continuous predictors was evaluated using the Box – Tidwell procedure and visual inspection of partial residual plots, with no indication of systematic departures from linearity. For the quantile regression models, conditional distributional assumptions were examined using residual density plots and quantile – quantile diagnostics across the estimated quantiles, which did not reveal meaningful deviations from the assumed structure. Variance inflation factors were examined to rule out multicollinearity in both models. In the regression analyses, the highest socioeconomic categories (e.g., university education, highest income bracket) were used as the reference groups. Although lower categories are often selected as the reference in logistic regression, this approach was chosen a priori to quantify the extent to which less advantaged groups lag behind the most privileged stratum. This choice affects the direction and interpretation of the regression coefficients but not model fit. Covariate selection was tailored to the specific aims of each model. In the logistic regression, the VL score was entered as the principal explanatory variable to evaluate its association with influenza vaccination uptake. In contrast, in the quantile regression, VL was treated as the dependent variable, and ‘sources of information’ were included as key predictors to identify determinants of literacy levels, while controlling for other potential socioeconomic confounders. Statistical analyses were performed using SPSS version 26, with p < .05 taken as the threshold for significance. Results Participants’ sociodemographic characteristics are presented in Table 1. Most participants were female (53.1%), with a median age of 50 y (31–60). The majority were married (69.8%), and 43.3% reported a monthly income below 500,000 Iraqi Dinar (IQD). Approximately 53.9% of participants were taking metformin. Table 1.Demographic characteristics of the study participants.VariableCount (%) or median (IQR)Age50 (31–60)GenderFemale251 (53.1%)Male222 (46.9%)Education levelHigh school or less222 (46.9%)University/college degree251 (53.1%)Marital statusNot married143 (30.2%)Married330 (69.8%)Monthly income (Iraqi Dinar)<500,000205 (43.3%)500,000–1,000,000175 (37.0%)>1,000,00093 (19.7%)Private health insuranceNo425 (89.9%)Yes48 (10.1%)MedicationsMetformin255 (53.9%)Insulin195 (41.2%)Sulfonylureas37 (7.8%)SGLT2 Inhibitors40 (8.5%)GLP-1 and dual GLP-1/GIP receptor agonists37 (7.8%)Thiazolidinediones (TZDs)12 (2.5%)DPP-4 inhibitors40 (8.5%) Demographic characteristics of the study participants. Figure 1 shows the sources of vaccination information. The most frequently reported source was “general practitioner or other healthcare professionals” (46.7%), followed by “social media” (35.9%), while the least frequently reported source was “search engines” (12%). About 44.6% of participants relied exclusively on scientific sources for information on vaccination. Sources of information on vaccination. Participants’ responses to the vaccine literacy questionnaire are presented in Table 2. The item with the highest “often” response was “Did you need much time to understand them?” (30.7%), while the item with the highest “never” response was “Have you considered the credibility of the sources?” (25.4%). The median VL score was 36 (IQR 30–41) out of a maximum possible score of 56. Internal consistency for the total scale was strong (Cronbach’s α = 0.8), indicating high reliability in our sample. The VL score showed a median score of 36 (30–41). Table 2.Participants’ responses to the vaccine literacy questionnaire.ItemOftenSometimesRarelyNeverFunctional itemsDid you find that the material as a whole (texts and/or images) was difficult to read?130 (27.5%)124 (26.2%)206 (43.6%)13 (2.7%)Did you find words you didn’t know?126 (26.6%)170 (35.9%)159 (33.6%)18 (3.8%)Did you find that the material as a whole (texts and/or images) was difficult to read?123 (26.0%)104 (22.0%)229 (48.4%)17 (3.6%)Did you need much time to understand them?145 (30.7%)126 (26.6%)183 (38.7%)19 (4.0%)Did you or would you have needed someone to help you understand them?106 (22.4%)139 (29.4%)212 (44.8%)16 (3.4%)Informative VLHave you consulted more than one source of information?82 (17.3%)178 (37.6%)128 (27.1%)85 (18.0%)Did you find the information you were looking for?123 (26.0%)192 (40.6%)92 (19.5%)66 (14.0%)Did you understand the information found?147 (31.1%)193 (40.8%)80 (16.9%)53 (11.2%)Have you had the opportunity to use the information?73 (15.4%)170 (35.9%)124 (26.2%)106 (22.4%)Did you discuss what you understood about vaccinations with your doctor or other people?133 (28.1%)159 (33.6%)84 (17.8%)97 (20.5%)Did you consider whether the information collected was about your condition?126 (26.6%)183 (38.7%)90 (19.0%)74 (15.6%)Have you considered the credibility of the sources?118 (24.9%)124 (26.2%)111 (23.5%)120 (25.4%)Did you check whether the information was correct?128 (27.1%)136 (28.8%)110 (23.3%)99 (20.9%)Did you find any useful information to make a decision on whether or not to get vaccinated?127 (26.8%)154 (32.6%)92 (19.5%)100 (21.1%) Participants’ responses to the vaccine literacy questionnaire. Participants’ responses to vaccination-related practice items are presented in Table 3. The majority reported being vaccinated against influenza at least once (52.2%); however, only 7.1% reported receiving the vaccine in the past year. Most participants had not received the pneumococcal vaccine (83.5%). About 40.0% had received two doses of the COVID-19 vaccine, and more than half (52.9%) were not vaccinated against tetanus. Table 3.Participants’ responses to vaccination-related practice items.PracticesCount (%)Have you been vaccinated against tetanus?Never250 (52.9%)Yes, more than 10 y ago125 (26.4%)Yes, in the last 10 y98 (20.7%)Have you been vaccinated against influenza?Never226 (47.8%)Yes247 (52.2%)Have you been vaccinated against influenza in the past year?No439 (92.8%)Yes34 (7.2%)Have you been vaccinated against pneumococcus?Never395 (83.5%)Yes78 (16.5%)How many times have you previously received the COVID-19 vaccine?Never134 (28.3%)Once110 (23.3%)Twice189 (40.0%)Third33 (7.0%)More than three times7 (1.5%) Participants’ responses to vaccination-related practice items. A binary logistic regression model was used to identify variables significantly associated with participants’ influenza vaccination (Table 4). Females had lower odds of receiving the influenza vaccine (OR = 0.606, 95% CI [0.395–0.931], p = .022). Participants with a monthly income below 500,000 IQD or between 500,000 and 1,000,000 IQD also had significantly lower odds of vaccination (OR = 0.182, 95% CI [0.091–0.362], p < .001; OR = 0.292, 95% CI [0.151–0.567], p < .001, respectively). In contrast, participants with higher VL scores had significantly higher odds of receiving the influenza vaccine (OR = 1.062, 95% CI [1.031–1.093], p < .001). Table 4.Binary logistic regression of variables associated with influenza vaccination (yes vs. never). ORP-value95% C.I for ORLowerUpperAge 1.001.9110.9851.017GenderFemale0.606.0220.3950.931Male(REF)Education levelHigh school or less0.943.8130.5791.536University/college degree(REF)Marital statusNot married0.817.4960.4571.462Married(REF)Monthly income (Iraqi dinars)<500,0000.182<.0010.0910.362500,000– 1,000,0000.292<.0010.1510.567>1,000,000(REF)HbA1c 1.035.4060.9541.122Vaccine literacy score 1.062<.0011.0311.093 Binary logistic regression of variables associated with influenza vaccination (yes vs. never). Table 5 shows the results of a quantile regression that was employed to identify variables associated with participants’ vaccine literacy. Participants with a high school education or less had lower vaccine literacy compared with those holding a university or college degree (Coefficient = −3.545, 95% CI [−5.618, −1.473], p = .001). Participants with a monthly income between 500,000 and 1,000,000 IQD also had lower vaccine literacy compared with those earning more than 1,000,000 IQD (Coefficient = −2.848, 95% CI [−5.384, −0.313], p = .028). Table 5.Quantile regression model of variables associated with vaccine literacy.Parameter CoefficientP-value95% Confidence IntervalLower BoundUpper BoundAge −0.061.091−0.1310.010GenderFemale−0.364.695−2.1861.458MaleReferenceEducation levelHigh school or less−3.545.001−5.618−1.473University/college degreeReferenceMarital statsNot married−1.939.120−4.3860.507MarriedReferenceMonthly income (Iraqi dinars)<500,000 JOD−2.606.054−5.2610.049500,000– 1,000,000 JOD−2.848.028−5.384−0.313>1,000,000 JODReferenceSources of vaccine informationNonscientific sources0.333.719−1.4862.152Scientific sourcesReference Quantile regression model of variables associated with vaccine literacy. Discussion Our study contributes new evidence on vaccine literacy among Iraqi adults with diabetes mellitus and its relationship to influenza vaccination. The results show that VL scores were moderate overall and that individuals with higher scores were more likely to have been vaccinated. At the same time, women and participants from lower-income groups reported lower vaccination rates, suggesting that both informational and socioeconomic factors influence behavior. These findings underline the relevance of literacy and social context when considering preventive practices in diabetic populations. Our findings align with global research highlighting similar determinants of vaccine hesitancy and uptake among individuals with diabetes. For example, a study examining influenza vaccination intentions among diabetic adults reported that concerns about vaccine safety, limited trust in information sources, and insufficient knowledge were key barriers to vaccination; factors that closely parallel the lower VL levels and reliance on nonscientific information observed in our sample.22 The study also emphasized the role of healthcare provider recommendation as a facilitator of vaccine acceptance. Together, these results highlight that vaccine hesitancy in diabetic populations is shaped by a combination of informational, psychological, and socioeconomic factors that recur across diverse settings, reinforcing the need for targeted, evidence-based educational strategies. The association we observed between higher VL and greater likelihood of influenza vaccination is consistent with earlier research. Although little has been published on diabetic patients specifically, studies in other chronic disease groups point to similar trends. For instance, research with Jordanian adults living with asthma demonstrated that stronger VL was linked to increased influenza vaccine uptake.23 A systematic review of 21 studies also concluded that individuals with higher VL are more likely to engage with recommended vaccination programs, reinforcing the role of VL as a predictor of preventive health behaviors.24 This study also highlighted the way socioeconomic position interacts with literacy. Participants with less education and lower income had lower VL scores. These findings echo broader evidence that socioeconomic status shapes the development of VL.25 Importantly, although higher VL was positively associated with uptake, our regression analyses showed that women and those with lower income remained less likely to be vaccinated, even when literacy was taken into account. This suggests that barriers such as access to services, affordability, or competing priorities extend beyond literacy alone. The overlap between limited resources and lower literacy may therefore reinforce inequalities in vaccine coverage. Although little is known about gender differences in vaccine acceptance among diabetic populations, prior studies suggest that women may be more cautious about vaccine safety and potential adverse effects.26,27 Such concerns could partly explain the lower uptake observed in our female participants. Communication strategies that acknowledge and directly address these gendered concerns may therefore be important for improving vaccine confidence among women with DM. Despite recommendations for annual vaccination in people with diabetes,28 just over half of participants in this study reported receiving the influenza vaccine. This rate is comparable to that seen in other high-risk populations, where uptake has often fallen below targets.29,30 Such shortfalls carry serious clinical implications, given the role of influenza as a trigger for hospitalization and severe complications in diabetic patients. A meta-analysis confirmed that individuals with diabetes face higher risks of influenza-related morbidity and mortality than those without the condition but also demonstrated that vaccination substantially reduces these risks.31 Additional studies have reported protective effects against hospitalization for cardiovascular events following vaccination in this group.32 Together, these findings show the need for intensified efforts to raise coverage in diabetic populations. Many studies have examined factors associated with the suboptimal acceptance of vaccines among patients with DM. For example, a cross-sectional study of 418 diabetic patients in Jordan found that 70.6% had never received the influenza vaccine, with insufficient awareness of its benefits identified as the most significant reason.33 Enhancing VL, particularly its functional and informative components, may therefore be an effective strategy to reduce the burden of preventable respiratory complications in diabetic patients. A key observation is that participants frequently cited healthcare professionals as their primary source of vaccine information, followed by social media. While trust in healthcare providers is encouraging, reliance on social media exposes patients to misinformation, which has been linked to vaccine hesitancy.15 As noted earlier, our regression analysis showed that reliance on nonscientific sources was associated with lower VL. This trend is concerning in the context of widespread online misinformation, where failure to identify credible information can lead to poor health choices. Interventions should therefore focus not only on improving the clarity and accessibility of information delivered by health professionals but also on countering misinformation spread online. Another notable finding was the variability in uptake across different vaccines. While just over half of the participants were vaccinated against influenza, vaccination rates for pneumococcus and tetanus were markedly lower. Similar discrepancies have been reported in other populations, where uptake often varies depending on vaccine-specific public health campaigns, perceived risk, and physician recommendations.7,29,30,33 The COVID-19 vaccination data, with 40% having received two doses, further illustrate the complex landscape of vaccine acceptance shaped by recent global health events. Previous studies in Iraq have also shown that COVID-19 vaccination attitudes were influenced by concerns about safety, trust in information sources, and broader societal factors.15–17 The challenge is therefore not only to improve overall VL but also to provide clear, targeted information on a range of recommended adult immunizations. Strengths, limitations, and future directions There are a number of strengths to note. This is, to our knowledge, the first study in Iraq to examine vaccine literacy in relation to influenza vaccination among people with diabetes. The use of a validated Arabic-language tool and a relatively large sample size increases confidence in the robustness of the findings. At the same time, several limitations should be considered. The cross-sectional design does not permit conclusions about causality. Data on both VL and vaccination status relied on self-report, which may introduce recall bias or social desirability effects. The study was carried out within a single country, which may restrict wider applicability, and we did not measure clinical outcomes such as hospitalization due to influenza. Future studies should adopt longitudinal or interventional designs to test whether strengthening VL improves uptake and extend this work to other chronic conditions and populations in Iraq. Another important limitation concerns the use of a clinic-based convenience sample, which may introduce selection bias. Individuals who attend routine clinic visits are often more engaged with healthcare services, have better access to medical resources, or exhibit higher health awareness than diabetic individuals who do not seek regular care. These characteristics may be associated with higher vaccine literacy, greater exposure to health information, and potentially different vaccination behaviors. Vaccine literacy may also fluctuate depending on the timing and intensity of recent vaccination campaigns, which was not captured in this study. Future studies using population-based or multi-setting sampling frameworks would help improve external validity and produce more generalizable results. Additionally, the use of self-reported vaccination data may have introduced recall bias, which could lead to under- or overestimation of true vaccination coverage. Future studies may benefit from validating vaccination status through healthcare records. Conclusion This study demonstrates that vaccine literacy, together with socioeconomic and demographic characteristics such as gender, income, and education, plays a central role in vaccination behavior among Iraqi adults with diabetes. Strengthening VL and reducing socioeconomic barriers are essential steps toward improving influenza vaccine coverage and mitigating infection-related risks in this population. Public health programs should prioritize accessible education and outreach to low-income groups, alongside efforts to address systemic obstacles to vaccination. The relatively modest uptake observed in this study underscores the urgency of designing tailored interventions that combine improved information with practical support to ensure equitable access to recommended vaccines. Acknowledgments Conceptualization, W.A.-Q and A.A.; methodology, W.A.-Q.; software, W.A.-Q.; validation, J.E. and A.J; formal analysis, W.A.-Q., A.J, and A.A.; investigation, A.A; data curation, W.A.-Q., A.J. S.A. and A.A.J; writing original draft preparation, W.A.-Q., A.A., M.J. and L.A.S.; writing review and editing, W.A.-Q., L.A, L.S. and J.E.; visualization, W.A.-Q.; supervision, W.A.-Q.; project administration, W.A.-Q. All authors have read and agreed to the published version of the manuscript. Biography Judith Eberhardt is Professor of Psychology and Public Health Equity in the School of Social Sciences, Humanities and Law at Teesside University, UK. She is a Chartered Psychologist and Associate Fellow of the British Psychological Society, a Fellow of the Higher Education Academy, and a member of the American Psychological Association. Her research focuses on health psychology and public health, with particular interests in vaccine acceptance, cancer screening, health literacy, health inequalities, and psychosocial aspects of long-term conditions. She has secured funding from national research councils, health agencies, and local authorities, and collaborates with academic and community partners in the UK and internationally to improve health outcomes in underserved populations. Her work seeks to address disparities in healthcare access and outcomes across diverse communities. Funding Statement The author(s) reported there is no funding associated with the work featured in this article. Disclosure statement No potential conflict of interest was reported by the author(s). Data availability statement The data that support the findings of this study are available from the corresponding author upon reasonable request. References Associated Data Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.