Economic Burden of COVID-19 Hospitalization in Almaty, Kazakhstan

Introduction

Over the past two decades, the world has experienced several major viral outbreaks, including coronavirus disease 2019 (COVID-19), pandemic influenza A (H1N1), Ebola virus disease (2014–2016, with recurrent outbreaks until 2021), Middle East respiratory syndrome (MERS), and Zika virus infection.¹ These emerging infectious diseases have spread regionally and globally, posing substantial challenges to public health systems and threatening the resilience and preparedness of healthcare infrastructures.

COVID-19, first identified in Wuhan, China, in late 2019 and declared a pandemic by the World Health Organization (WHO) on March 11, 2020, rapidly evolved into one of the most significant global health crises of the 21st century.² The disease is highly contagious and presents with a wide spectrum of clinical manifestations, ranging from mild respiratory symptoms to severe complications such as acute respiratory distress syndrome, multiorgan failure, and death. The rapid global spread of COVID-19 placed unprecedented pressure on healthcare systems and public health infrastructures across many regions.² Although vaccination campaigns and public health interventions have reduced the severity of the pandemic, COVID-19 continues to have long-term health consequences. Emerging evidence suggests that a substantial proportion of individuals infected with SARS-CoV-2 experience persistent symptoms after the acute phase of infection, commonly referred to as long COVID. Frequently reported symptoms include fatigue, respiratory difficulties, and reduced physical functioning.³

Beyond its clinical impact, COVID-19 has imposed a substantial financial burden on healthcare systems worldwide. Several studies have reported significant direct medical costs associated with hospitalization, particularly among patients requiring intensive care. For example, the average hospitalization cost per COVID-19 patient in the United States exceeded $13,000, while intensive care unit (ICU) treatment costs reached approximately $36,484 per patient, especially for individuals requiring advanced interventions such as extracorporeal membrane oxygenation (ECMO).⁴ Similar economic pressures have been observed in other countries. Studies conducted in Spain, Denmark, and Iran reported significant increases in hospitalization costs associated with COVID-19 treatment and intensive care utilization.^5–7 A global comparative review also demonstrated substantial variation in ICU costs across countries, ranging from approximately $5,437 per patient in Romania to $100,789 in Germany, reflecting differences in healthcare systems, treatment protocols, and resource utilization.⁸ These findings highlight the importance of evaluating the economic burden of COVID-19 across diverse healthcare settings.

In health economics, the cost-of-illness (COI) framework is widely used to estimate the economic burden of diseases. This approach measures both direct medical costs, such as hospitalization, diagnostics, and treatment, and indirect costs, including productivity losses resulting from illness, recovery, or premature mortality. COI studies provide important evidence for policymakers by identifying key cost drivers and informing resource allocation during public health emergencies.^9.

Kazakhstan operates under a universal health coverage system, in which hospital services are primarily financed through the Mandatory Social Health Insurance (MSHI) system and government budget allocations. While this model aims to ensure equitable access to healthcare services for the entire population, the COVID-19 pandemic exposed structural challenges within the national health system. Previous studies have reported that Kazakhstan experienced higher rates of avoidable mortality compared with some developed countries, suggesting gaps in emergency preparedness and response capacity.¹⁰ Between 2020 and 2022, the epidemiological situation has evolved considerably. Following the initial wave in 2020, a significantly larger surge occurred in 2021, likely associated with the emergence of more transmissible SARS-CoV-2 variants and limited vaccination coverage during the early stages of the national immunization campaign.¹¹ Differences in diagnostic coding and reporting practices during the first year of the pandemic may also have contributed to lower recorded hospitalization and mortality figures in 2020 compared with 2021. Similar patterns of increased mortality registration in 2021 have been reported in national analyzes of avoidable mortality.¹⁰ These epidemiological and systemic factors may help explain the sharp increase in hospitalizations and healthcare expenditures observed during the peak of the pandemic. Despite the growing body of research on COVID-19 epidemiology and clinical outcomes, studies examining the economic burden of COVID-19 hospitalization in Central Asian countries remain limited.¹² Previous research has investigated the burden of COVID-19 in primary care settings in Almaty, focusing mainly on outpatient service utilization;¹³ however, the economic burden associated with hospitalizations has not been comprehensively examined. Studies conducted in other countries have estimated the economic impact of COVID-19 using cost-of-illness approaches that include both direct healthcare expenditures and indirect productivity losses. Nevertheless, evidence from Central Asia remains limited. Understanding the financial impact of hospitalization is therefore essential for identifying major cost drivers and strengthening health system preparedness for future public health emergencies.

Therefore, this study aims to analyze the direct and indirect costs associated with COVID-19 hospitalizations in the Almaty metropolitan area, a city with a population exceeding 2.19 million residents, during the period 2020–2022. By estimating the economic burden of COVID-19 hospitalizations, this study seeks to provide evidence that can support more effective health system planning, resource allocation, and pandemic preparedness strategies.

Material and Methods

An applied research study was conducted that included an analysis of all patients hospitalized with COVID-19 in the city of Almaty between 2020 and 2022. The data were obtained from the administrative hospital reporting database maintained by the National Scientific Center for Health Development Ministry of Health (NSCHD), which compiles standardized inpatient records from medical institutions across Kazakhstan. COVID-19 cases were identified based on hospital discharge diagnoses coded according to the International Classification of Diseases, 10th Revision (ICD-10), specifically codes U07. The dataset includes all hospitalized patients with a recorded COVID-19 diagnosis; however, the available administrative data did not allow differentiation between admissions primarily due to COVID-19 and incidental SARS-CoV-2-positive hospitalizations.

Direct medical costs per patient were estimated using hospital administrative billing data and national cost information obtained from the NSCHD. A top-down costing approach was applied using standardized reimbursement tariffs provided by NSCHD for inpatient treatment under Kazakhstan’s Mandatory Social Health Insurance system. These tariffs represent the average cost per treated case and include expenditures related to hospitalization, medications, diagnostic procedures, ICU services, and other medical resources. Hospital administrative data were used to determine the number of hospitalized patients, ICU admissions, and length of stay. The standardized per-case cost estimates were then applied to calculate the total direct medical costs associated with COVID-19 hospitalizations in Almaty. In addition, detailed hospital-level data on specific cost components, including medication costs, ICU care, diagnostic procedures, inpatient stay, and other medical expenses, were collected. The aggregated hospital-level costs were consistent, on average, with the standardized per-case estimates provided by NSCHD, supporting the consistency of the cost calculations used in this study.

The following variables were included in the analysis: number of COVID-19-related inpatient cases in Almaty hospitals, including those treated in ICUs, broken down by sex and age, average length of hospital stay (ALoS), and discharge status (recovery or death). In addition, regional wage data for Almaty stratified by sex were incorporated into the analysis.

Indirect costs were estimated from a societal perspective using the human capital approach, where lost productivity is valued using wage data as a proxy for the economic contribution of individuals unable to work due to illness or premature death. In this framework, productivity losses represent economic losses to society rather than direct expenditures of the healthcare system. This approach is widely used in cost-of-illness studies and recommended in health economic evaluations. Because employment status was not available in the administrative dataset, productivity losses were estimated using working-age population averages (18–65 years). This approach is commonly used in cost-of-illness studies when individual employment data are unavailable.

• Productivity loss due to hospitalization was calculated by multiplying the average length of hospital stay (ALoS) by the minimum daily wage, based on official data from the National Bureau of Statistics of Kazakhstan.¹⁴ The daily wage was used because productivity losses related to hospitalization and recovery were calculated based on the number of days patients were unable to work. The minimum wage was used in the base-case analysis to provide a conservative estimate of productivity losses and avoid potential overestimation of the economic burden. Sensitivity analyses using alternative wage assumptions were conducted to assess the robustness of the results.
• Productivity loss due to the recovery period was estimated by multiplying a 14-day recovery period based on Rajabi et al assumptions by the minimum daily wage applied to recovered patients aged 18–65.⁷
• Productivity loss due to premature mortality was calculated using the Forgone Labor Output (FLO) equation (1):

where, w is the minimum wage per person per year, (the annual wage was used in the premature mortality calculation because the Forgone Labor Output model estimates productivity losses over future years of working life).

G is the annual growth rate, r is the discount rate,

i is the number of years of life lost, and

P_i is the current value of the predicted future income per workforce.¹⁴

Following previous economic burden studies and World Bank recommendations for upper-middle-income countries, an annual growth rate of 3% and a discount rate of 6% were applied in the base-case analysis.¹⁵ These parameter values are commonly used in economic burden and cost-of-illness studies in middle-income countries and are consistent with recommendations used in previous global health economic evaluations. To calculate the average years of life lost (YLL), the average age of the patients who died from COVID-19 was subtracted from the maximum productivity age (65 years).

To assess the overall economic burden of COVID-19 in Almaty, the following formula was applied: Equation (2)

(2)

Where: iₓ = the number of affected patients, pₓ = the unit cost per patient, and C is the total cost.¹⁶

All costs were calculated in Kazakhstani Tenge and converted to U.S. dollars using the average annual exchange rate for each respective year (2020–2022).

Statistical Analysis

Descriptive statistics were used to summarize the data, including sex, age, ALoS, discharge status (recovery or death), mean direct and indirect costs, and total expenditures for the key cost groups. This approach is commonly used in cost-of-illness studies aimed at estimating the overall economic burden rather than identifying causal determinants of cost variation. A sensitivity analysis was conducted by applying a range of discount rates (from −20% to +20%) and different wage levels (minimum and maximum salaries) to estimate the best- and worst-case scenarios. Additionally, given the variability in direct medical costs, primarily driven by differing treatment approaches, a ±20% range was applied to minimize potential errors. Consequently, both optimistic and pessimistic estimates of the economic burden of COVID-19 on Almaty were calculated.

Results

In 2020 and 2021, older adults (especially those aged > 65 years) represented the largest share of cases, ICU admissions, and deaths, highlighting their vulnerability. ICU admissions and deaths were particularly high in male aged > 56 years old. Interestingly, by 2022, in addition to older adults, there was a marked increase in younger patients (especially 0–17 years), albeit with low mortality rates (Table 1).

Table 1 Demographic Characteristics of COVID-19 Patients Studied Between 2020 to 2022

The number of hospitalized patients increased sharply in 2021 and declined in 2022. Direct medical costs per patient were also highest in 2021, mainly due to higher drug and hospital service expenditures. ICU patients and those who died were consistently older than the general hospitalized population. The length of hospital and ICU stays peaked in 2021 and decreased slightly in 2022, reflecting the highest healthcare burden during that year (Table 2).

Table 2 COVID-19 Hospitalization and Its Direct Cost (US $)

Indirect Cost

Indirect costs per patient due to hospitalization ranged from $429.2 to $578.0, peaking in 2021. Total productivity losses related to hospitalization increased substantially from $4.71 million in 2020 to $25.42 million in 2021, before declining to $1.95 million in 2022, reflecting changes in the number of working-age patients affected across the three years (Table 3).

Table 3 Indirect Cost Due to COVID-19 Between 2020–2022 (US $)

Productivity loss during the recovery period at home increased from $379.64 to $510.68 per patient during the study period. Accordingly, total productivity losses during recovery reached $7.91 million in 2020, $35.48 million in 2021, and $3.54 million in 2022, corresponding to 10,971, 42,972, and 4,038 affected working-age patients, respectively (Table 3).

The economic burden of premature mortality also contributed substantially to indirect costs. Total productivity losses due to premature death increased from $2.70 million in 2020 (177 deaths) to $14.96 million in 2021 (1,013 deaths), before decreasing to $1.37 million in 2022 (71 deaths). These findings highlight the significant but fluctuating economic impact of lost workforce productivity during the pandemic (Table 3).

The total direct and indirect costs of COVID-19 in Almaty amounted to $26.7 million in 2020, surged to $258.0 million in 2021, and declined to $15.1 million in 2022, reflecting the peak financial burden during the height of the pandemic (Table 4).

Table 4 Sensitivity Analyzes for Indirect Costs of COVID-19 (US $)

Sensitivity Analysis

Sensitivity analysis showed that the estimated economic burden of COVID-19 between 2020 and 2022 remained relatively stable under different discount rates (4.8%, 6%, and 7.2%) with a fixed growth rate of 3%. Across all scenarios, the economic burden consistently peaked in 2021, confirming it as the year with the greatest financial impact. Estimated costs for 2020 and 2022 were substantially lower but showed only minor variation across different discount rates and wage assumptions, indicating that the overall estimates of the economic burden are robust to changes in key parameters (Table 4).

Discussion

To our knowledge, this study provides one of the first comprehensive assessments of the economic burden of COVID-19 hospitalizations in a major Central Asian metropolitan area, offering important regional evidence that has been largely absent from the global literature. In our study, the average age of hospitalized patients ranged from 40.2 to 56.8 years across the study period. According to previous studies, the average age of hospitalized COVID-19 patients ranged from 44.6 to 60 years.^7,17 Overall, evidence indicates that the risk of severe outcomes and mortality increases with age. Consistent with these findings, mortality in our study was higher among females over 65 years old, while among males it was most frequently observed in the 46–65 age group.^18,19 These results are consistent with international clinical evidence showing that older adults experience higher mortality and ICU utilization during COVID-19 outbreaks.

A study conducted in England reported an increase in pediatric hospitalizations due to COVID-19 between 2020 and 2022, with 62.1% of hospitalized children under the age of 5 and 4.3% requiring PICU care, particularly among children aged 5–17 years with serious comorbidities.²⁰ Martella et al suggested that the increase in pediatric hospitalizations during the 2021 wave may partly be attributed to vaccination prioritization among adults and elderly populations, which reduced infection rates in older groups and shifted the burden of disease toward the still unvaccinated pediatric population.²¹ Another study demonstrated that prior SARS-CoV-2 infection significantly reduced the risk of breakthrough COVID-19 infection after full vaccination across different vaccine types, indicating additive protection that could improve the effectiveness and cost-efficiency of vaccination strategies.²² This pattern may also be related to the emergence of the Omicron variant, which was associated with increased transmission among younger populations, as well as differences in vaccination coverage among minors.

Our study found similar patterns, with a notable number of hospitalized patients aged <17 years in 2021, many of whom required ICU admission, although mortality remained relatively low in this group (Table 1). Additionally, research conducted in Kazakhstan suggests that parents are more likely to leave their children unvaccinated against COVID-19 due to concerns about vaccine safety and effectiveness, insufficient information, and their own unvaccinated status.²³ This vaccine hesitancy may partly explain the increase in pediatric hospitalizations observed in 2021 and its contribution to higher healthcare costs during this period.

International comparisons also demonstrate substantial variation in COVID-19-related healthcare costs. In the United States, the average inpatient cost increased from $10,394 in early 2020 to $13,072 by March 2022, while ICU costs could reach $36,484, particularly for patients requiring advanced treatments such as ECMO.⁴ In Denmark, hospitalizations and deaths related to COVID-19 between 2022 and 2024 were reported to be more than double those associated with influenza, highlighting the continued demand for hospital resources.⁶ In contrast, studies from Brazil reported higher hospitalization costs during the first year of the pandemic, particularly among men and populations living in regions with greater socioeconomic vulnerability.²⁴ In Iran, the economic burden of COVID-19 hospitalizations reached approximately US$44 million in 2021.⁷ Similarly, our findings indicate that the economic burden in Almaty increased sharply in 2021, reflecting the peak pandemic wave (Table 4). By incorporating productivity losses through the Forgone Labor Output approach, the study highlights the broader macroeconomic consequences of pandemic-related mortality beyond healthcare system expenditures.

This period also coincided with the global spread of the Delta variant, which has been associated with higher hospitalization rates and greater disease severity. These factors likely contributed to the increased ICU utilization, longer hospital stays, and higher medication use observed during the 2021 peak. In addition, productivity losses related to recovery periods and premature mortality represented a substantial share of the total economic burden, emphasizing the broader socioeconomic consequences of the pandemic beyond direct healthcare expenditures. The higher ICU utilization among older male patients, particularly those aged over 56 years, highlights the importance of targeted monitoring and hospital resource planning during future respiratory disease outbreaks. This peak corresponds to the period with the highest mortality and ICU demand observed in the dataset, reflecting the severe strain placed on the healthcare system during the most critical phase of the pandemic.

Overall, these findings—particularly the sharp increase in economic burden observed during the 2021 peak—provide important evidence for health system planning, pandemic preparedness, and economic resilience in Kazakhstan and other middle-income countries. By quantifying the economic burden associated with hospitalizations, ICU care, and productivity losses, the study highlights the importance of early public health interventions and efficient resource allocation to reduce both healthcare costs and societal impacts during future pandemics. The magnitude of the economic burden observed during the 2021 peak suggests that establishing emergency healthcare funding mechanisms could help mitigate financial shocks during future pandemics.

Limitation and Future Steps

Our analysis relied on aggregated administrative data from hospitals in Almaty, which may lack sufficient granularity regarding patients’ comorbidities, severity of illness, and specific treatment protocols, potentially influencing cost variability. Moreover, indirect cost estimates were based on assumptions of a uniform recovery period and minimum wage levels, which may not fully reflect the actual productivity losses for individuals with varying income levels or longer recovery periods.

Additionally, the economic evaluation focused only on direct and indirect medical costs from a societal perspective, excluding broader social impacts such as mental health consequences, long-COVID-related disability, and caregiver costs. Another limitation of this study is that multivariable regression analysis was not performed to identify independent predictors of higher healthcare costs. The administrative dataset did not contain detailed clinical variables such as comorbidities or disease severity that would allow robust multivariable regression analysis of cost determinants. Future studies using more comprehensive clinical datasets may further investigate determinants of cost variation.

The use of a uniform 14-day recovery period and minimum wage in the base-case analysis may underestimate the true productivity losses associated with COVID-19. However, this approach provides a conservative estimate of indirect costs. Sensitivity analyses using alternative wage assumptions were conducted to assess the robustness of the results. Additionally, the calculation of premature mortality costs assumed a fixed retirement age (65 years) and constant annual growth and discount rates, which may not fully capture the dynamic economic conditions during the pandemic.

Gazezova et al found that in Kazakhstan the treatment of COVID-19 involved the widespread use of anticoagulants and antibiotics, often without proper indications, raising concerns regarding patient safety and antimicrobial stewardship, while glucocorticoid use improved with updated clinical guidelines and antiviral use remained limited due to availability and uncertain benefits.¹² Therefore, future research should explore the economic and health consequences of ineffective or potentially harmful treatments in order to better inform clinical guidelines and optimize resource allocation. Additionally, the study focuses on acute hospitalizations and does not account for long-term health consequences such as Long COVID, which may further increase the economic burden of the pandemic.

The findings of this study have important implications for pandemic preparedness and health system resilience in Almaty and other comparable urban centers. The disproportionately high economic burden observed in 2021 highlights the critical need for early targeted public health interventions during peak transmission periods to minimize healthcare costs and productivity losses. Integrating economic modeling into public health decision-making could help policymakers better anticipate and mitigate financial risks during health crises.

The substantial economic impact among the working-age population highlights the importance of investing in preventive strategies such as timely vaccination campaigns, flexible work policies, and effective public health communication. Particular attention should be directed toward vulnerable groups, including older adults and children, whose hospitalization rates vary across different waves of the pandemic. Previous studies suggest that increased mobility and urban density may facilitate the transmission of COVID-19.²⁵ In rapidly urbanizing middle-income settings, these structural factors may increase vulnerability during future pandemics. To prevent increased vulnerability during future pandemics, middle-income countries must align economic development with proactive and integrated public health planning. For Kazakhstan, this involves adopting context-specific and economically balanced strategies, such as investments in healthcare infrastructure, localized risk communication, and digital surveillance systems, to strengthen pandemic preparedness while minimizing socioeconomic disruption.²⁶

Although conducting cost-utility analyses of vaccination programs during pandemics can be challenging, health authorities should, whenever possible, consider such analyses in the future to balance price and efficacy, understand the broader benefits of vaccination, and ensure affordability and effective public health outcomes.²⁷ Finally, strengthening real-time health data systems to monitor hospital utilization, patient outcomes, and economic impacts is essential for agile and effective responses to future public health emergencies.

Conclusion

This study provides one of the first comprehensive estimates of the economic burden of COVID-19 hospitalizations in a major Central Asian city. The economic burden in Almaty peaked sharply in 2021, corresponding to the highest levels of hospitalization, ICU utilization, mortality, and total costs. Older adults and males over 56 years of age experienced the greatest clinical and economic burden, while the increase in hospitalizations among younger patients may partly reflect lower vaccination coverage among minors, vaccine hesitancy among parents, and the earlier prioritization of vaccination for older adults. These findings suggest that future pandemic preparedness should include targeted vaccination and early protection strategies for high-risk groups, as well as measures to reduce productivity losses among the working-age population (18–65 years), which accounted for the majority of indirect costs associated with hospitalization, recovery, and premature mortality. Strengthening real-time health data systems to capture patient comorbidities, disease severity, and treatment patterns will be essential for more precise economic analyzes and more effective health system responses to future public health emergencies.