Economic Burden of COVID-19: A Systematic Review

Introduction

Since late 2019, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) has infected more than 180 million people globally and caused more than 3.9 million deaths worldwide as of June 30th 2021.¹ The virus and the measures taken to control its spread have profoundly impacted people’s lives. Since the start of the Coronavirus disease 2019 (COVID-19) pandemic, significant efforts have been made by the industry and academia to discover promising treatments and vaccines that can improve the clinical disease course and patient outcomes or prevent infection, and therefore reduce the burden to public health systems. Quantifying the economic burden of COVID-19 is an essential consideration for evaluating the value of therapeutic and preventive interventions against the COVID-19 disease.

The macroeconomic impact of the COVID-19 pandemic has been overwhelming, with data from Europe showing an average of 7.4% reduction of GDP in 2020, with significant variations across countries.^2,3 The GDP losses were shown to be highly correlated with drops in employment rates, especially in regions that rely on tourism industry. Negative economic growth was just one of the unintended consequences associated with the adoption of non-pharmaceutical interventions (NPIs), such as lockdowns. Governments worldwide have all struggled in balancing the trade-offs between controlling the spread of the virus to limit the burden of the disease against the unintended economic, socioeconomic and other health consequences of the imposed NPI measures.⁴

It was estimated that without adequate policy measures, the pandemic would have the most significant economic impact on those of lower income, therefore increasing poverty rates and overall inequality.² Similar predictions have been made by the Council of Economic Advisers to the US government, who predicted that without economic support policies, people in the lower-income bracket would have suffered an average reduction of disposable income of more than 10% during Q1 of 2020.⁵

As with any other new disease, initially, the costing data related to the burden of COVID-19 was scarce. Due to the profound economic burden of the disease, several studies related to the medical costs of COVID-19 have recently been published in different regions. Due to this increase in the number of studies, and its profound impact on healthcare budgets as well as global economy, it is essential to understand and synthesize this data to inform policy makers on the overall burden of COVID-19.

The objective of this study is to systematically review and synthesize the evidence on the economic assessment of burden of COVID-19 since the start of the pandemic.

Methods

A systematic literature review based on a protocol was conducted utilizing EMBASE, MEDLINE, MEDLINE-IN-PROCESS, and The Cochrane Library using the search terms presented in Appendix 1. The search was conducted on April 1, 2021. In addition to the searches of electronic database, manual searches were also conducted to capture data from recent studies not yet published. Manual searches included searches of conference proceedings (published from January 2020 onwards) and online information repositories, the MedRxiv preprint server, and health technology agencies from the UK, France, Germany, Canada, US, South Korea, Taiwan, and Australia.

Eligibility Criteria and Study Selection

The study question of this Systematic Literature Review (SLR) was specified using the PICOS (Population, Intervention, Comparison, Outcomes and Study design) framework. The population of interest included patients with COVID-19. People indirectly affected, including populations who suffered from the consequences of COVID-19 prevention measures, caregivers, or carers, were also included. Studies reporting on any population subgroup (for example, patients with a different socioeconomic status, different demographic characteristics) that met the inclusion criteria of being affected by COVID-19 directly or indirectly were included and data from these studies were extracted. All interventions for COVID-19 treatment and strategies for prevention or control of COVID-19 were included.

The study types of interest were observational costing studies and assessments of the economic impact of COVID-19. Systematic reviews were included and extracted separately from the individual studies identified in this review. Individual studies identified from those SLRs were not extracted but are listed in Appendix 2. Randomized controlled trials (RCT), reviews, case reports, and case series were excluded. No date restrictions were placed on the search strategies for the electronic databases. For the manual searches, the date was restricted to the last year.

The titles and abstracts of records identified from the search strategy, where available, were reviewed according to the pre-specified inclusion/exclusion criteria reported in Appendix 3. A second reviewer independently screened the titles and abstracts of the identified studies. Any discrepancy was resolved by discussion. Articles identified as potentially relevant based on screening of titles and abstracts were then reviewed in full and selected according to the list of pre-specified inclusion/exclusion criteria. After the full-text review was completed, a list of the excluded studies was created with the reason for exclusion. The complete list of excluded studies during the full-text review stage is presented in Appendix 4. Meta-analysis synthesis was not conducted due to the broad study heterogeneity.

The outcomes of interest included resource use related to the disease and the associated intervention and measures of equity. These are included but are not limited to hospital admission, length of stay, physician visits, emergency department visits, and pharmacy costs. Indirect costs included, but not limited to, were decreased productivity, lost wages, or caregiver costs. Total costs per health state and patient were also of interest.

Results

A total of 18,621 citations were identified through the electronic database search. Duplicates were identified and compared based on an exact match for author, year, title, and abstract. After removal of duplicates, 351 unique citations were obtained and screened.

After application of the pre-specified selection criteria and title, abstract, and full-text stages, a total of 35 individual publications and 2 systemic literature reviews were included in this SLR (Figure 1: PRISMA diagram). Most of the publications (n=247) were excluded because they did not have the outcomes of interest. Furthermore, 25 publications were excluded based on study type, 33 did not include the population of interest, and four were previously missed duplicates. A full list of included studies is presented in Appendix 2.

Figure 1 PRISMA diagram.

In exploring the literature, we found an abundance of literature assessing direct and indirect medical costs. The studies included here evaluated some of the following objectives: identifying risk factors for clinical burden and resource use, resource use of hospitalized patients, determining the association of COVID-19 severity with costs as well as assessing the direct impact of COVID-19 on GDP.

Countries analyzed in the individual studies included 12 studies from United States, 5 studies from the UK, 3 from Germany, 3 multinational studies, 2 studies from Turkey and China and one study from the following countries: Italy, France, Greece, Saudi Arabia, Japan, Pakistan, Iran and Brazil.

Objectives and key findings of the included articles are summarized below in Table 1 Economic Burden of Patients and Table 2 Economic burden of the General Population, extraction tables with complete study characteristics and outcomes are in Appendix 5.

Table 1 Economic Burden of Patients

Table 2 Economic Burden of the General Population

Discussion

A total of 37 publications were included in this SLR, 35 individual publications, and 2 SLRs. The SLRs included evaluated age-adjusted risk factors associated with mechanical ventilation and racial and ethnic disparities in hospitalizations and death.^43,44 Consistently reported across all studies were high medical costs and resource use by COVID-19 patients. The increased resource use required for the most severe patients influenced costs substantially.^{9,11,12,14–19}

Across multiple countries in the hospital, the costs for COVID-19 patients admitted in an ICU were higher than those not admitted in an ICU. Within the ICU, mechanical ventilation contributed to a further increase in costs $2082.65 ± 345.04 to $2990.76 ± 545.98.^9,15–17,19 Overall, studies from Europe, the US, and Asia showed that more severe COVID-19 patients had higher costs and resource use than milder COVID-19 cases. On a macroeconomic level, the COVID-19 pandemic was a direct cause of GDP losses, mainly due to loss in productivity and implementation of non-pharmaceutical interventions.^1–3

Economic models have assessed the prevention, screening and treatment of COVID-19, In longer-term models, screening tests were considered cost-effective in all economic models. Similarly, social distancing was more cost-effective than quarantine, herd immunity, and having no intervention. Antiviral treatments and curative treatments were considered the most cost-effective option compared with any other measures against COVID-19.^36,45–48

An economic evaluation conducted in South Africa and the UK evaluated the economic effects of social distancing.²⁶ The model results showed that if strict lockdown measures were taken in the UK at the beginning of the pandemic, the pandemic would have resolved in 1.5 months with approximately 21,000 deaths. However, the UK decided to implement 4.5 months of semi-lockdown, which resulted in 80,000 deaths. A similar trend was observed in South Africa. Thus, the model results showed that social distancing shortens the pandemic duration and decreases the number of deaths by decreasing the number of infected people.

Another economic evaluation concluded that treatment with remdesivir for non-ventilated patients and dexamethasone for ventilated patients would maximize lives saved and save $11.5 million.²⁷ The main drivers were the efficacy of the drugs and reduction of ICU-time required.

A stochastic compartment model from the US estimated how hospitalizations and ICU admissions would decrease if the duration of infectious rate was shorter and when symptomatic patients were treated. When high proportions (>50%) of symptomatic patients were treated, the resource use decreased. Similarly, shorter infectious periods were associated with reduced resource use.²⁸

One key strategy to prevent the burden associated with COVID-19 is vaccination. Currently, the CDC estimates that available vaccinations such as the yearly influenza vaccine to have prevented 4.4 million illnesses, 2.3 million medical visits, and 58,000 hospitalizations during the 2018–2019 influenza season.⁴⁹ A Markov cohort model was used to estimate COVID-19 related direct medical costs and deaths in the United States, With the most optimistic projections, a 60% efficacious COVID-19 vaccine can prevent 31% of expected COVID-19 deaths in the United States versus no vaccines.⁵⁰ As COVID-19 vaccines enter the market, more specific cost-effective evaluations will be needed. As of June 2021, there are three COVID-19 vaccines approved by Emergency Use Authorization vaccines in the US—BNT162b2 mRNA (Pfizer-BioNTech), mRNA-1273 (Moderna), and Ad26.COV2.S (Janssen).⁵¹ The European Medicines Agency has approved an additional vaccine ChAdOx1-S by AstraZeneca.⁵² Additionally, there are ten other vaccines approved in various countries globally.⁵³

Vaccines may also offer additional benefits beyond mitigation case burden alone. Estimates for herd immunity threshold for COVID-19 range from 60–90% of the population. This can be more easily achieved through vaccination supplementing natural immunity from past exposure. There is uncertainty on whether an initial infection leads to long-lasting immunity against COVID-19.^29,54 Furthermore, the emergence of new variants of concern, such as those from the Omicron (B.1.1.529) lineage, would require the demonstration of vaccine effectiveness.

Vaccinations have made an essential contribution to the decreased incidence of infectious diseases and are considered to be a cost-effective public health intervention. Regarding childhood vaccination, it was predicted that for every dollar spent, it saves $3 for the US payer.⁵⁵ The eradication of smallpox through vaccines has resulted in a direct medical cost savings of $300 million in the US. The eradication of polio is expected to save the world $1.5 billion yearly.⁵⁶ Missed opportunity for the four most common vaccine-preventable diseases in adults over 50 years contributes to an estimated cost of $26.5 billion in the US (medical and indirect).⁵⁷ Vaccines demonstrate a considerable return on investment for payers.

Conclusion

The COVID-19 pandemic has an immense impact and current efforts implementing NPIs only have partial success in controlling the humanistic and economic costs of the pandemic. Vaccination is a strategy used to mitigate the evolving landscape of COVID-19, and policymaking bodies will consider their cost-effectiveness in making recommendations for routine use in the future.