Global economic burden per episode for multiple diseases caused by group A Streptococcus

Group A Streptococcus (GAS) causes a spectrum of diseases from mild pharyngitis and skin infections to severe invasive and immune-mediated conditions such as ARF and its sequela, RHD. The health and economic burden is concentrated in low- and middle-income countries and certain Indigenous populations. RHD accounts for substantial mortality (estimated 319,400 deaths in 2015), and with over 616 million pharyngitis and 111 million pyoderma cases annually, the overall disease and economic burden is large. Progression from untreated pharyngitis to ARF and then to RHD highlights the escalating clinical and economic consequences. Despite this, economic burden estimates, especially outside high-income settings, are scarce. This study aims to estimate average economic burden per episode for key GAS-related diseases across World Bank income groups to inform prevention strategies and vaccine policy via the Strep A Vaccine Global Consortium (SAVAC).

Background literature indicates high prevalence and mortality of GAS-related diseases, especially RHD in LMICs. Prior economic burden estimates for GAS diseases are limited and uneven, with more attention historically to ARF/RHD than to skin or invasive infections. Existing work has shown wide variability in costs by country and disease, and previous burden syntheses for other diseases have used model-based approaches. The authors note a paucity of studies from low-income countries and limited age-stratified cost reporting in the literature, necessitating the use of adjustment factors and proxy data sources.

Study scope included seven GAS-related disease categories with sufficient data: pharyngitis, impetigo, cellulitis, invasive and toxin-mediated infections (ITMI), ARF, RHD, and severe RHD (defined as RHD with heart failure or surgical intervention). Costs were estimated per episode by World Bank income group and disaggregated into direct medical costs (DMC), direct non-medical costs (DNMC), and indirect costs (IC). Literature identified by Telethon Kids Institute provided observed cost data for each component when available. All costs were inflated to 2018 using GDP deflators and converted to 2018 USD using official exchange rates. International dollars (I$) were also reported using PPP.

• DMC: Components included consultations, medications, diagnostics, inpatient care, surgery, and other medical tests. WHO-CHOICE unit costs for outpatient visits and hospital bed-days at relevant facility levels (primary, secondary, tertiary; mapped by disease severity) were combined with average utilization (outpatient visits, inpatient bed-days) to construct crude DMC. Average utilization per episode was derived from South Korea’s Health Insurance Review & Assessment Service (HIRA) Healthcare Bigdata Hub for 2010–2019 across disease subgroups (e.g., streptococcal meningitis for invasive disease), using mean outpatient visits and bed-days per episode. DMC adjustment factors were calculated as the ratio of observed DMC (from literature) to crude DMC for each disease and income group, then applied globally to generate adjusted DMCs.
• DNMC: Included transportation, food, lodging. Observed DNMC values from literature were compared to GDP per capita to compute DNMC adjustment factors (observed DNMC/GDP per capita) by disease and income group. These factors were applied to GDP per capita for countries lacking data. If data were missing for an income group/disease, factors were borrowed from similar income groups or adjusted by duration of illness relative to a similar disease.
• IC: Productivity losses during illness were estimated by multiplying country minimum wage by duration of illness. Due to limited disease-specific duration data (except pharyngitis), duration was proxied by aggregated outpatient visits plus inpatient bed-days from HIRA data; for pharyngitis, literature-derived sick days were used.
• Productivity loss due to premature death: Considered for invasive infections, RHD, and severe RHD (deaths from pharyngitis/impetigo are rare). Weighted mean age at death was estimated using IHME data for RHD and, for invasive infections, pneumococcus as a proxy. Productivity years lost equaled life expectancy (World Bank) minus mean age at death, valued at minimum wage and discounted at 3%.
• Sensitivity analysis: Probabilistic multivariate sensitivity analysis used beta-PERT distributions for DMC/DNMC adjustment factors and duration of illness (min, mode, max from inputs), and uncertainty in age at death for mortality-related IC. Monte Carlo simulation with 5000 iterations (Ersatz) produced 95% CIs. Facility mappings by disease: primary outpatient for pharyngitis; primary inpatient/outpatient for impetigo and cellulitis; secondary or tertiary inpatient/outpatient for ARF, RHD, severe RHD; tertiary for ITMI.

• Average per-episode economic burden in 2018 USD across income groups: pharyngitis $22–$392; impetigo $25–$2,903; cellulitis $47–$2,725; ITMI $662–$34,330; ARF $231–$6,332; RHD $449–$11,717; severe RHD $949–$39,560. Highest burdens were for severe RHD and ITMI; lowest for pharyngitis.
• In general, USD-denominated costs increased with country income level. However, in PPP terms (I$), burdens for ARF, RHD, and severe RHD in UMIC approximated or exceeded those in HIC (e.g., ARF mean I$: HIC $7,147 vs UMIC $12,283; RHD mean I$: HIC $13,221 vs UMIC $13,029; severe RHD mean I$: HIC $44,306 vs UMIC $42,275).
• Cost composition: For pharyngitis, IC predominated across income groups, followed by DMC and DNMC. As severity increased (ARF, RHD, severe RHD, ITMI), DMC became the dominant cost driver. In skin infections, DMC dominated in HIC/UMIC, whereas IC dominated in LMIC/LIC. DNMC was consistently the smallest component.
• Mortality-related productivity losses: Despite fewer productive years lost in HIC (older age at death), the monetary value of premature death was highest in HIC and lowest in LIC due to income differences.
• Sensitivity: Economic burden estimates were most sensitive to DMC adjustment factors (especially in HIC) and duration of illness. For pharyngitis, duration of illness had the largest impact. For severe RHD, DMC adjustment factors across income groups dominated; for impetigo and cellulitis, duration of illness was also influential.

The study addresses the evidence gap by providing per-episode economic burden estimates for major GAS-associated diseases across income settings using a standardized framework. Findings show substantial costs, particularly for severe RHD and invasive infections, highlighting the escalating economic impact along the disease progression from pharyngitis to ARF and RHD. The PPP-based comparisons reveal that, when adjusting for cost of living, burdens in UMIC can match or exceed those in HIC for ARF/RHD, underscoring significant financial strain in these settings. Cost component analyses indicate that indirect costs dominate mild disease (pharyngitis), whereas direct medical costs dominate severe and invasive conditions, reflecting intensive care needs. The results emphasize the public health importance of preventing initial GAS infections and progression to severe disease, informing vaccine development and health policy prioritization, especially through the SAVAC initiative.

This study compiles and extrapolates existing evidence to estimate per-episode economic burdens for seven GAS-related diseases by income group, showing high costs for severe RHD and invasive infections and notable burdens in UMIC when adjusted for purchasing power. The work underscores the urgency of preventive strategies, including vaccine development, and the need to strengthen primary data collection on costs, especially in low-income countries and with age stratification. Future research should refine disease-specific resource use across diverse settings, improve attribution for non-GAS-specific diagnoses (e.g., cellulitis), capture adverse event costs from antibiotics, and incorporate more granular, country-specific utilization and wage data to enhance precision.

• Utilization inputs (outpatient visits, hospital days) were derived from a single national source (South Korea HIRA) and may not generalize to other income settings.
• Sparse age-specific cost data; limited studies only broadly split pediatric vs adult. Averaging across age groups may mask heterogeneity.
• Missing DMC and DNMC data, especially in LICs, required borrowing adjustment factors from similar income groups or adjusting by duration of illness.
• Some studies lacked explicit duration of illness; costs were treated as per-episode per year, potentially misclassifying recurrent events.
• Potential costs from antibiotic adverse reactions (e.g., penicillin) were not included due to limited data.
• Duration of illness proxied by health facility use (visits and bed-days) may underestimate total illness duration and thus IC.
• Cellulitis is etiologically heterogeneous; the proportion attributable to GAS is uncertain, leading to wide estimate ranges.