Home  /  Issues  /

Pharmaceutical Service Utilization and Costs Among Veterans Exposed to Chemical Warfare

Article details

Abstract

Introduction

Pharmaceutical service utilisation and medication costs are essential aspects of veterans’ healthcare, especially for those exposed to chemical warfare agents. Understanding the determinants of these expenditures can support the development of equitable and efficient healthcare policies.

Methods

This cross-sectional study utilised a comprehensive national dataset of Iranian chemical warfare veterans (n=64 193). A Poisson regression model was used to examine the frequency of pharmacy visits, with results reported as incidence rate ratios (IRRs). A multiple linear regression model with log-transformed costs was applied to analyse pharmaceutical expenditures, restricted to the 26 000 veterans with at least one pharmacy visit. Independent variables included age, gender, disability percentage, household size and province of residence.

Findings

The average number of annual pharmacy visits was 5.6, and the mean annual pharmaceutical cost was approximately US$123. Older age and higher disability percentage were significantly associated with greater utilisation and costs. Veterans aged ≥70 years had 1.44 times more pharmacy visits (IRR = 1.44) and 2.15 times higher geometric mean costs (β = 2.15) compared to those aged <50 years. Veterans with 50–70% disability had 1.47 times more visits (IRR = 1.47) and 2.70 times higher costs (β = 2.70) than those with <25% disability. Male veterans had 12% lower visit rates (IRR = 0.88) and 19% lower costs (β = 0.81) than female veterans. In an exploratory secondary analysis, female veterans exhibited higher use of psychiatric medications. The study also observed marked regional variations, with certain provinces showing significantly higher rates of pharmacy visits and medication expenses.

Conclusion

The findings underscore the associations between age, disability severity, gender, household characteristics and geographic factors with veterans’ pharmaceutical utilisation and costs. Policy implications include strengthening family- and community-based support systems and revising benefit structures to minimise moral hazard while ensuring adequate protection for veterans with severe disabilities.

Keywords: veterans; pharmaceutical services; healthcare costs; chemical warfare

Introduction

Although the use of chemical weapons in warfare is strictly prohibited due to their severe health consequences,1 Iraq deployed such weapons multiple times against Iran during the Iran–Iraq War (1980–1988), targeting both military personnel and civilians, including the city of Sardasht in 1986–1987.2,3 More than three decades after the war, the number of chemical warfare survivors remains substantial due to late-onset and chronic health complications. As of 2021–2022, over 69 000 cases have been reported, accounting for approximately 10.7% of Iran’s disabled veteran population.4 Exposure to chemical agents results in a wide range of long-term health conditions, including respiratory, dermatological and neurological disorders, as well as psychological complications such as post-traumatic stress disorder (PTSD).5,6 These complex and persistent health issues necessitate specialised medical and pharmaceutical interventions to enhance survivors’ quality of life. Pharmaceutical services play a pivotal role in managing these conditions by ensuring appropriate medication use, providing counselling, supporting adherence and monitoring therapeutic outcomes.7,8 However, patterns of pharmaceutical service utilisation and related expenditures among veterans exposed to chemical warfare remain poorly understood, despite their importance for informed health policy and resource allocation. Globally, pharmaceutical expenditures continue to rise, and veterans typically require long-term, multidrug treatments, leading to substantial costs.9 Many countries have implemented targeted programs to address these challenges; for instance, the US Department of Veterans Affairs has expanded its pharmacy management initiatives to meet the increasing medication needs of veterans.7

In Iran, veterans’ healthcare benefits include comprehensive insurance coverage, notably through Day Complementary Health Insurance, which provides free services in public healthcare centres and reduces out-of-pocket expenses in the private sector. Day Insurance operates a network of more than 10 600 contracted healthcare providers, allowing veterans to access free or subsidised medical services by presenting their identification cards.10 Despite these supportive measures, ongoing policy reforms are needed to meet the evolving health needs of this population adequately. This study aims to examine pharmaceutical service utilisation and costs among Iranian veterans exposed to chemical warfare agents, offering insights to inform equitable and sustainable healthcare strategies.

Although several decades have passed since the end of the war, few comprehensive studies have focused on Iranian veterans exposed to mustard gas. Amini et al.6 were the first to report the number of affected veterans by injury severity, age and gender. Mehdizadeh et al.11 investigated healthcare utilisation and costs among mustard gas–exposed veterans at the national level. At the same time, Hosseini-Shokouh et al.12 described healthcare utilisation among sulfur mustard survivors in Iran from 2018 to 2021, identified high-risk regions and assessed inequality across socioeconomic groups. However, additional research—particularly at the subnational level—is still needed.

While several studies have examined the overall burden of injuries among chemical warfare survivors, far less attention has been paid to their patterns of pharmaceutical service utilisation and associated costs, especially in low- and middle-income settings such as Iran. Understanding these patterns is crucial, as medication use is a major component of long-term healthcare expenditures and indicates both disease burden and the effectiveness of the health system’s response.

Methods

Study design and setting

This applied, analytical, cross-sectional study was conducted from 21 March 2019 to 19 March 2020 (01/01/1398 to 29/12/1398 in the Iranian calendar). It aims to investigate the utilisation of pharmaceutical services and the associated costs among veterans exposed to chemical warfare. The study population included all veterans in Iran with complete medical records and billing data in the Iran Health Insurance Organization’s (IHIO) administrative database during the study period. Because the study used a census approach that included all veterans with complete administrative records, the dataset had minimal missing information. Cases with incomplete key variables were excluded from analysis (<2% of total observations).

Study population: The total number of chemical veterans whose medical records and billing data were fully recorded in the administrative database of the IHIO during the study period. The number of these chemical veterans is 64 193. For the regression analyses, the study population was restricted to chemical warfare veterans who received at least one outpatient pharmaceutical service during the study period (n = 26 000). This restriction was applied because the primary objective of the study was to examine patterns of pharmaceutical service utilisation and associated costs among service users, rather than to investigate determinants of service use versus non-use.

Research sample: All chemical veterans who used outpatient pharmaceutical services in the study year were selected as the research sample (26 000 people). The study inclusion criterion was use of outpatient pharmaceutical services in the study year. The number of visits is the average number of times a chemical veteran visits a pharmacy and receives medication. The cost of medication is the average cost of medication per visit to the pharmacy.

Data sources and collection

The study used secondary data from the IHIO administrative database (Day Insurance Company), which contains detailed records of insured individuals, including veterans receiving pharmaceutical services. Data were collected through a structured data extraction form, designed to capture key variables related to pharmaceutical utilisation and associated costs. The extracted data included the number of pharmacy visits per veteran and the total cost of pharmaceutical services for each veteran, as well as the demographic and socioeconomic characteristics of chemical veterans, including age, gender, household size (number of family members), province of residence, and their disability percentage (which was an ordinal variable categorised into groups: less than 25%, 25–49%, 50–70% and above 70%). In Iran, the disability percentage is officially determined by the Martyrs and Veterans Affairs Foundation based on the severity of war-related injuries or illnesses. Veterans with lower disability percentages (e.g., less than 25%) typically have mild impairments. In contrast, those in higher categories (e.g., above 70%) suffer from severe or total disabilities, often requiring extensive medical care, financial support and specialised rehabilitation services.13 This classification system is crucial in determining the level of healthcare benefits and social assistance provided to veterans. Veterans exposed to chemical agents during the Iran–Iraq War (1980–1988) are assessed separately due to the progressive nature of chemical-related diseases (e.g., pulmonary fibrosis, chronic bronchitis, immune system disorders). Many of these veterans experience worsening health conditions over time, leading to periodic reassessment of their disability percentage.

In this study, a fixed exchange rate of US$1 = RIs 130,000 (Iranian Rials (has been used for all currency conversions. This exchange rate came from the Central Bank of Iran in 2020. The model also included province of residence as a categorical variable, with Tehran Province as the reference category. All other provinces were entered in the regression models as indicator categories and compared against the reference category (Tehran Province).

Given that medical expenses for chemical warfare veterans in Iran are fully covered, the total pharmaceutical cost per veteran included both the insurer’s share (Day Insurance) and the cost of inpatient and outpatient pharmaceutical services. Also, the costs are nominal. All collected data were anonymised to protect patient confidentiality, and all personal identifiers were removed before analysis. After data collection, raw records were cleaned and organised in Microsoft Excel and then imported them into STATA 14 for statistical analysis. Both descriptive and inferential statistical methods were employed to examine associations between veterans’ demographic and socioeconomic characteristics and their pharmaceutical utilisation patterns. To identify factors associated with pharmaceutical service utilisation and costs, two regression models were applied

1. Poisson regression model:

    he Poisson regression model was used to analyse the number of pharmacy visits per veteran, a non-negative count variable. This approach was selected because the distribution of pharmacy visits followed a Poisson pattern, making it suitable for modelling count data. The model estimated the relative likelihood of service utilisation across demographic and socioeconomic groups. Results were presented as incidence rate ratios (IRRs), indicating the expected change in pharmacy visits for each unit change in an explanatory variable. A standard Poisson regression was initially fitted. Diagnostic plots (e.g., observed vs expected counts) did not reveal pronounced overdispersion, and the potential for overdispersion was not detected. Model fit assessment also indicated no substantial overdispersion in the count data (Pearson χ²/df = 1.17), supporting the use of the standard Poisson regression model.

    2. Multiple linear regression (ML):

      For the pharmaceutical expenditures analysis, the dependent variable was the total annual cost per veteran. Because cost data are expected to be right-skewed, a logarithmic transformation was applied to approximate a normal distribution. Consequently, a multiple linear regression model was fitted using the log-transformed cost as the outcome variable. Diagnostic assessment indicated no major violations of homoscedasticity or residual normality assumptions. The model was applied only to the subset of veterans who had received at least one pharmaceutical service during the study period. The model estimated the effect of independent variables—including province of residence, gender, age group, disability percentage and household size—on the geometric mean of pharmaceutical costs. The estimated coefficients (β) were exponentiated for interpretation, representing the multiplicative factor by which the geometric mean cost for a given category differs from the reference category, while holding all other variables constant.

      Compliance with research ethics guidelines

      All patient identifiers (e.g., name, national ID, hospital ID) were removed prior to data processing to ensure participant anonymity. Access to the dataset was restricted to authorised researchers, and the published results could not identify any individual. Because this study was based entirely on secondary data analysis, it posed no physical, psychological or financial risk to participants. The study received official ethical approval, confirming adherence to national and international research ethics standards.

      Results

      Table 1 summarises the effects of province of residence, gender, age group, disability percentage and household size on pharmaceutical service utilisation.

      Table 1. Descriptive statistics of pharmaceutical service utilisation and costs among Iranian chemical warfare veterans

      GroupCategoryMean number of pharmacy visits per veteranMean pharmaceutical costs per veteranSD of pharmaceutical costsMedian pharmaceutical costsInterquartile range
      RIsUS$RIRIRI
      Total–5.6615,998,955 12311567313138630788894012
      Age group<50 years4.8810,799,692 834926645126649405816415
      50–59 years5.4314,786,266 11412367560835654288199825
      60–69 years6.3720,096,761 15595376870535615211636483
      ≥70 years7.7126,284,168 202114394605645243014311500
      Disability percentage<25%5.1712,795,191 9910802364232408647195983
      25–49%6.2718,210,242 140124167798470807010685165
      50–70%7.1140,888,123 315135198809882335222738460
      GenderFemale6.7315,328,212 11812230092636109456675600
      Male5.6416,018,295 12311547835938742008991374
      Household size<4 members6.6718,309,357 14115242328141714009633474
      ≥4 members7.1814,027,158 1089997554136036108043990
      Province of residenceMarkazi 6.2112,321,694 954934153827627206190190
      Gilan 6.0419,047,733 1476704457439174928753463
      Mazandaran 4.5815,003,114 1165170533427762857817371
      East Azerbaijan 5.5121,948,572 16931086638142096409936547
      West Azerbaijan 6.379,906,469 764899793029489995958979
      Kermanshah 4.559,816,471 763597153229581215291742
      Khuzestan 4.412,590,847 973875251440333808902950
      Fars 4.8712,962,741 10020141881633008256936722
      Kerman 5.0812,255,955 943890016634076769445320
      Razavi Khorasan 7.039,675,188 752464009539022087868227
      Isfahan 5.0716,396,632 12669355777400656910229214
      Sistan & Baluchestan3.1421,219,473 16316470880340331207943716
      Kurdistan 4.0318,958,839 14614854575427325506497775
      Hamadan 7.5913,681,361 1053659996341369159207451
      Chaharmahal & Bakhtiari9.1119,343,419 14988656089575184212189108
      Lorestan 7.317,378,319 13451979379507868910166590
      Ilam 7.3912,348,537 956230322233391005537307
      Kohgiluyeh & Boyer-Ahmad8.3310,437,873 802776949645837107955048
      Bushehr 5.229,366,887 723308332636245206898072
      Zanjan 7.3222,439,912 1738878704842093508647652
      Semnan 5.2411,660,570 902371538337780359515108
      Yazd 8.1131,389,208 242421113782459910012011655
      Hormozgan 4.8216,006,402 12346277506328323611703873
      Tehran 3.8829,517,519 227100686017495380014916260
      Ardabil 7.1320,244,394 15621528691045253408681840
      Qom 9.0422,827,423 17693431078602096015146541
      Qazvin 5.7220,107,183 1557598572730774049469618
      Golestan 7.6116,007,859 12355394439478631010006116
      North Khorasan 6.939,655,244 743627851434287606256548
      South Khorasan 5.699,841,324 762429774332147858684512
      Alborz 3.0533,715,952 260156854073400660012783670

      Table 2. Effects of socioeconomic variables on pharmaceutical service utilisation and pharmaceutical costs among chemical warfare veterans

      GroupCategoryService utilisation (Poisson Regression) (IRR [p-value])
      Pharmaceutical costs (Multiple linear) (β [p-value])
      Age group<50 years1.000 (Ref.)1.000 (Ref.)
      50–59 years1.147 (<0.001)1.234 (<0.001)
      60–69 years 1.409 (<0.001)1.732 (<0.001)
      ≥70 years1.435 (<0.001)2.153 (<0.001)
      Disability percentage<25%1.000 (Ref.)1.000 (Ref.)
      25–49%1.186 (<0.001)1.377 (<0.001)
      50–70%1.474 (<0.001)2.699 (<0.001)
      GenderFemale1.000 (Ref.)1.000 (Ref.)
      Male0.884 (<0.001)0.807 (0.001)
      Household size<4 members1.000 (Ref.)1.000 (Ref.)
      ≥4 members0.971 (<0.001)0.905 (<0.001)
      Province of residenceMarkazi 1.271 (<0.001)0.777 (0.013)
      Gilan 1.223 (<0.001)0.886 (0.076)
      Mazandaran 0.899 (<0.001)0.657 (<0.001)
      East Azerbaijan 1.083 (<0.001)0.84 (0.008)
      West Azerbaijan 0.937 (<0.001)0.573 (<0.001)
      Kermanshah 0.882 (<0.001)0.629 (<0.001)
      Khuzestan 1.106 (<0.001)0.855 (0.008)
      Fars 1.041 (0.001)0.698 (<0.001)
      Kerman 1.088 (<0.001)0.716 (<0.001)
      Razavi Khorasan 1.107 (<0.001)0.784 (<0.001)
      Isfahan 1.489 (<0.001)0.815 (<0.001)
      Sistan & Baluchestan0.818 (<0.001)0.756 (0.021)
      Kurdistan 0.744 (<0.001)0.544 (<0.001)
      Hamadan 1.177 (<0.001)0.924 (0.184)
      Chaharmahal & Bakhtiari1.369 (<0.001)1.197 (0.011)
      Lorestan 1.298 (<0.001)1.147 (0.058)
      Ilam 1.181 (<0.001)0.68 (<0.001)
      Kohgiluyeh & Boyer-Ahmad1.429 (<0.001)1.016 (0.781)
      Bushehr 1.081 (<0.001)0.726 (<0.001)
      Zanjan 1.228 (<0.001)0.901 (0.305)
      Semnan 1.079 (<0.001)0.847 (0.025)
      Yazd 1.37 (<0.001)1.033 (0.672)
      Hormozgan 1.117 (0.001)0.715 (0.033)
      Tehran 1.000 (Ref.)1.000 (Ref.)
      Ardabil 1.186 (<0.001)0.93 (0.355)
      Qom 1.361 (<0.001)1.155 (0.151)
      Qazvin 1.009 (0.691)0.711 (0.001)
      Golestan 1.361 (<0.001)1.139 (0.048)
      North Khorasan 1.055 (0.071)0.67 (0.003)
      South Khorasan 0.998 (0.97)0.757 (0.255)
      Alborz 0.983 (0.423)0.942 (0.543)

      In 2019, veterans averaged 5.6 pharmacy visits per year, with a corresponding mean annual pharmaceutical expenditure of RI 15,998,955 (approximately US$123).

      Age significantly determined pharmaceutical utilisation. Veterans aged over 70 years recorded the highest average number of visits (7.71 visits annually) and the greatest pharmaceutical expenditure (US$202 per year). A similar pattern was observed across disability levels, with medication use increasing in proportion to disability severity. For example, veterans with 50–70% disability averaged 7.11 visits per year, indicating higher service utilisation among more severely disabled individuals.

      Marked geographic disparities were also identified. Veterans in Chaharmahal and Bakhtiari (9.11 visits), Qom (9.04 visits) and Kohgiluyeh and Boyer-Ahmad (8.33 visits) had the highest average annual pharmacy visits. Conversely, those living in Alborz (3.05 visits), Sistan and Baluchestan (3.14 visits) and Tehran (3.88 visits) reported the lowest utilisation rates.

      Pharmaceutical expenditures exhibited a different regional pattern. Veterans in Alborz (US$260), Yazd (US$242) and Tehran (US$227) incurred the highest average annual costs, while those in Bushehr (US$72), North Khorasan (US$74) and Razavi Khorasan (US$75) incurred the lowest.

      Regression analysis results

      The findings indicate that age significantly predicted both service utilisation and pharmaceutical costs. Veterans aged 70 years and older had, on average, a 1.44-fold higher rate (44% more) of pharmacy visits than those under 50 years of age. Similarly, the geometric mean pharmaceutical cost for this oldest group was 2.15 times higher (115% greater) than the youngest reference group, indicating substantially greater cost burdens among older veterans.

      An increase in disability percentage was also strongly associated with higher pharmaceutical utilisation and expenditures. Veterans with 50–70% disability had, on average, 1.47 times more pharmacy visits (47% higher) than those with less than 25% disability. Furthermore, the geometric mean pharmaceutical cost for this high-disability group was 2.70 times higher (170% increase) compared to the low-disability reference group. These results highlight the greater healthcare needs and treatment intensity among veterans with more severe disabilities. By gender, male veterans showed significantly lower utilisation and costs than female veterans. On average, men had 0.88 times (12% lower) the rate of pharmacy visits compared with women. Their geometric mean pharmaceutical cost was also 0.81 times (19% lower) than the female reference group.

      Household size exhibited a modest inverse relationship with pharmaceutical utilisation and expenditures. Veterans from households with four or more members had, on average, a 0.97 times (3% lower) rate of pharmacy visits and a 0.91 times (9% lower) geometric mean pharmaceutical cost compared with veterans from smaller households (<4 members). Veterans residing in Isfahan (IRR = 1.489), Kohgiluyeh and Boyer-Ahmad (IRR = 1.429), and Yazd (IRR = 1.370) had the highest pharmacy visit rates relative to the reference province (Tehran). Conversely, those living in Kurdistan (IRR = 0.744), Sistan and Baluchestan (IRR = 0.818), and Kermanshah (IRR = 0.882) had the lowest visit rates. For pharmaceutical expenditures, veterans in Chaharmahal and Bakhtiari (β = 1.197), Lorestan (β = 1.147) and Qom (β = 1.155) had the highest geometric mean costs compared with Tehran. In contrast, residents of Kurdistan (β = 0.544), West Azerbaijan (β = 0.573) and Kermanshah (β = 0.629) had the lowest geometric mean pharmaceutical costs.

      Discussion

      The Iran–Iraq War, which took place between September 1980 and August 1988, remains one of the most devastating conflicts of the twentieth century,14 resulting in over 200 000 deaths and more than 400 000 injuries. Currently, more than 400 000 disabled veterans reside in Iran.15

      Due to the physical and psychological injuries sustained during the war, these veterans are at increased risk of chronic diseases and other long-term health conditions compared with the general population. They have a substantially higher demand for therapeutic and pharmaceutical services, which contributes to elevated medical expenditures.16

      This study examined patterns and determinants of pharmaceutical service utilisation and associated costs among Iranian chemical warfare veterans, with a focus on how demographic, clinical and socioeconomic factors are associated with access and spending. Notable regional differences emerged: veterans in Isfahan, Kohgiluyeh and Boyer-Ahmad, andChaharmahal and Bakhtiari had higher pharmacy visit rates and were more likely to incur higher costs than those in Tehran. These disparities likely reflect differences in healthcare infrastructure, access to specialised services and regional healthcare policies. Regions with more developed healthcare systems generally provide better access, which may improve health outcomes and reduce long-term healthcare needs.

      Veterans living in rural areas face barriers similar to those documented worldwide. For example, in the United States, approximately 24% of veterans reside in rural regions and face substantial obstacles to healthcare access, including geographic isolation, limited availability of healthcare professionals, and transportation challenges.17 These factors often lead to lower service utilisation, particularly among socioeconomically disadvantaged groups.

      Consistent with prior studies, age was significantly associated with pharmaceutical service utilisation and expenditures. Veterans aged over 70 years had 1.4 times more pharmacy visits and 2.1 times higher odds of being in higher pharmaceutical cost quintiles, likely reflecting age-related comorbidities and the long-term consequences of war-related injuries.6,11 Similar patterns have been documented in the US, where veterans aged 65 and older use significantly more medications, with a median of five drugs per individual, increasing the risk of adverse outcomes.18

      Khan et al.19 conducted a comprehensive analysis of health services research data to examine prescription drug coverage for elderly populations in the United States. Their study concluded that adequate access to prescription medications significantly improves health outcomes for older adults, highlighting the importance of maintaining comprehensive drug coverage programs for seniors.19

      The high reliance of veterans aged over 70 on therapeutic services is likely driven by the increased healthcare needs associated with ageing, including the management of chronic conditions that require continuous pharmacotherapy.19,20 Similarly, Marchand et al.21 demonstrated that veterans’ age significantly affects their health status and utilisation of therapeutic services. Other studies have also emphasised that veterans, particularly older individuals, are more likely to utilise healthcare services compared with the general population, and this tendency increases with advancing age.22

      Research further shows that veterans who participated in combat or who have higher disability levels require more frequent healthcare services, resulting in an increased number of medical visits.23 Consistent with these findings, the present study indicates that veterans with a disability rating of 50–70% had, on average, 1.4 times more pharmacy visits than those with less than 25% disability. Higher disability levels were also associated with increased service utilisation and pharmaceutical costs, particularly among those with severe conditions such as pulmonary injuries, which contribute substantially to overall medication expenditures.11 Policy mechanisms, including free prescriptions for highly disabled veterans under disability-based health benefits, may further incentivise higher medication use.24

      The regression analysis also showed that increasing age and disability percentage significantly increase both the likelihood of pharmacy visits and the probability of falling into higher cost categories. These findings suggest that as veterans age and their disability severity increases, the complexity of their therapeutic needs intensifies, leading to higher pharmaceutical utilisation and associated costs.

      The present study also identified significant gender differences in pharmaceutical service utilisation. Female veterans demonstrated higher frequencies of pharmacy visits and greater pharmaceutical expenditures compared with male veterans, reflecting distinct healthcare needs and potential gender-related barriers to access. These findings are consistent with prior research indicating that women have higher use of psychiatric medications and health services related to chemical injuries.25,26 In contrast, male veterans exhibited lower utilisation rates. Cultural and social factors may influence gender disparities in pharmaceutical services by shaping healthcare-seeking behaviour and access.20,27

      Household size was another factor associated with pharmaceutical service utilisation. Veterans from larger households showed slightly lower service use and costs, which may reflect informal support networks or financial constraints. This pattern aligns with broader evidence in health disparities research, which highlights the role of household dynamics in shaping healthcare access and expenditures.28–30

      An apparent inconsistency is noted between the unadjusted and adjusted associations for household size. The descriptive statistics show a slightly higher mean number of pharmacy visits for veterans in larger households (≥4 members). However, the adjusted regression models indicate a small but significant inverse association: larger households are associated with slightly lower visit rates and lower pharmaceutical costs. This reversal in direction is likely due to confounding by other sociodemographic and need-based factors. For instance, veterans in larger households may, on average, be younger or live in different regions than those in smaller households. When these factors (age, disability percentage, province of residence) are not accounted for, they can create a spurious positive association in the unadjusted data. The regression models adjust for these potential confounders, revealing the independent, albeit modest, effect of household size.

      A limitation of this study is that the analyses were restricted to veterans who utilised outpatient pharmaceutical services during the study period. Consequently, the findings may not be fully generalisable to all chemical warfare veterans, particularly those who did not use pharmaceutical services. Future research may benefit from including both users and non-users of pharmaceutical services and applying two-part or zero-inflated modelling approaches to examine both the likelihood of service utilisation and the intensity of utilisation and expenditures. However, the present study intentionally restricted the Poisson regression analysis to veterans with at least one pharmacy visit, as the primary objective was to examine the intensity of pharmaceutical service utilisation among service users. In addition, the cross-sectional design and reliance on secondary administrative data limit causal inference about the associations observed between veterans’ characteristics and pharmaceutical service utilisation patterns. Despite these limitations, the findings highlight important determinants of pharmaceutical service utilisation among chemical warfare veterans and provide valuable evidence to inform targeted policies aimed at reducing regional disparities and addressing the specific healthcare needs of this population.

      Conclusion

      This study demonstrates that pharmaceutical service utilisation and costs among Iranian veterans are significantly associated with age, disability percentage and province of residence. Older veterans and those with higher disability ratings had more pharmacy visits and a greater likelihood of being in higher pharmaceutical cost quantiles. Additionally, gender and household size were associated with service utilisation, reflecting differences in clinical needs, healthcare access, regional resources and medication-use policy frameworks.

      To enhance pharmaceutical care for veterans, it is recommended to develop targeted medication management programs for older veterans and those with severe disabilities. Efforts should also focus on improving access to medications in underserved regions and revising drug coverage policies to ensure equitable support. Promoting family- and community-based support systems, particularly in larger households, may help reduce dependence on formal care while maintaining appropriate service use. Finally, periodic reviews of benefit structures and prescription policies are essential to minimise moral hazard while providing adequate protection for veterans with high healthcare needs.

      Conflicts of interest

      The authors declare that there are no conflicts of interest related to this study.

      Ethical permission

      This study was conducted in accordance with all ethical principles in research and was approved by the National Committee for Ethics in Biomedical Research of Iran under ethical approval code IR.BMSU.REC.1400.141