Epidemic Waves Are Followed by Measurable Health Workforce Declines
Across three decades of country-level data, epidemic shocks were followed by a measurable erosion in human resources for health. The analysis covered 194 countries and territories from 1990 to 2019 and linked recorded epidemic disasters with subsequent changes in health worker density, using a longitudinal ecological design rather than individual patient data.
The pattern was blunt. An additional epidemic event in the preceding year was associated with an average decrease of 57.5 × 10−3 in health workforce density in the following year, after adjustment for geography and socioeconomic status. Small on paper, that decrement becomes consequential when it is applied repeatedly across national systems already running with thin staffing margins.
The study then translated the association into workforce losses. Assuming the observed relationship was causal, epidemics were associated with an average annual global loss of 17,549 health workers, or 2.57 per 10,000 health workers. For health systems with chronic shortages, the loss is not abstract accounting; it can mean thinner staffing on wards, slower outbreak response, fewer community contacts, and less slack when the next emergency arrives.
The Burden Fell Heaviest in Low-Resource Settings
The largest attributable fractions clustered in regions and income groups that already face the steepest structural deficits. South Asia had an average attributable fraction of 20.52 per 10,000, and sub-Saharan Africa followed at 18.03 per 10,000. Low-income countries showed an average attributable fraction of 15.89 per 10,000, while lower-middle-income countries recorded 12.76 per 10,000. By contrast, upper-middle-income countries and high-income countries showed far smaller values, at 0.96 and 0.12 per 10,000, respectively.
In practical terms, the weight of epidemic-related workforce loss did not distribute evenly across the world. It landed where staffing levels are already most fragile and where replacement capacity is least forgiving. In countries with limited training pipelines, tight fiscal space, and fewer reserve personnel, any post-epidemic decline in workforce density can deepen pre-existing access gaps for routine care as well as emergency response.
The top 10 countries with the highest average attributable fractions were mostly in sub-Saharan Africa, including Niger at 57.06 per 10,000, Somalia at 44.19 per 10,000, and Ethiopia at 37.12 per 10,000. Those figures point to a pattern of concentrated vulnerability rather than a diffuse global phenomenon.
Why Epidemics Can Drain the Workforce
The biology and the system dynamics both matter here. Epidemics do not only kill health workers directly. They can also strain institutions through mortality, illness, fear of exposure, migration, burnout, reduced service demand, and economic contraction that weakens retention. When a health system is already undercapitalized, even a temporary shock can trigger durable attrition.
Healthcare workers sit at the front line of epidemic response, so they absorb disproportionate occupational risk when infectious threats spread. The article also cites prior global estimates suggesting that 115,500 healthcare workers died during the first year of the COVID-19 pandemic, a reminder that direct mortality is only one part of the burden. In parallel, systemic disillusionment and reduced institutional capacity can push workers out of the workforce even when they survive the outbreak itself.
The study did not isolate one pathway from another. It instead captured the aggregate effect of epidemic exposure on workforce density across time and place. That broader view is useful because national health systems experience outbreaks as a compound shock: clinical danger, administrative disruption, and labor-market pressure all hit at once.
What the Data Can and Cannot Prove
The analysis used country-year data drawn from the Emergency Events Database and global health workforce estimates, with epidemic counts combined with cadre-specific density data across 16 types of health worker. The models adjusted for key contextual variables, including geography, income level, socio-demographic development, urbanization, and population density. Sensitivity analyses also examined reverse causation, epidemic intensity, baseline disease burden, socioeconomic status, and long-term trends.
Even with that breadth, the design leaves room for uncertainty. Country-level ecological data cannot fully exclude reverse causation, migration effects, or the influence of endemic disease patterns. Weak disaster-reporting systems may also lead to undercounting, especially in countries with less robust surveillance capacity. If epidemic events are missed, their workforce toll is likely missed as well.
The direction of the bias could run both ways. Countries with weaker reporting infrastructure may appear less affected than they truly are, while countries with better surveillance may register more events and sharper associations. Either way, the estimate is best understood as a conservative global signal rather than a precise census of every lost clinician, nurse, midwife, or community health worker.
Workforce Protection Belongs in Epidemic Preparedness
The study’s central message is direct: epidemic control and health workforce protection cannot be separated. Preparedness plans that focus only on vaccines, surveillance, and hospital surge capacity overlook the labor force needed to deliver all three.
Because losses were disproportionately larger in low- and middle-income countries, the findings argue for explicit staffing protections in outbreak planning. That means retention strategies, occupational safeguards, and support for cadres such as nurses, midwives, physicians, and community health workers, especially where the health system has little buffer. In settings with already sparse staffing, the loss of even a modest share of workers can push services past a functional threshold.
A Quantified Global Inequity in Health Labor
The study reframes epidemics as a driver of health workforce inequality, not just a test of it. Over 30 years, 1,185 epidemic events were recorded across 132 countries and territories, and the workforce effects were concentrated in places least equipped to absorb them. The result is a reinforcing cycle: epidemic shocks weaken the workforce, and a weaker workforce makes the next epidemic harder to contain.
Those findings carry a clear policy implication. Health emergency planning in low-resource settings needs to account for labor preservation as a core objective, not an afterthought. Without that, the next outbreak will keep draining the very workforce required to fight it.