If all healthcare worker positions defined in the Government of Malawi’s establishment were filled, it could lead to a 13.6% overall reduction in DALYs over 2023–2040, equivalent to 43.1 million DALYs averted under the assumption of current resource allocation across diseases. Among the top 10 causes of DALYs, the largest difference would be for AIDS (41% lower with all positions filled), TB (26% lower), Malaria (24% lower), and Measles (31% lower): this underscores that scaling up workforce capacity can alleviate health burdens for high-priority conditions, including infectious diseases that rely heavily on continuous service delivery Expanded workforce availability substantially increases the system’s overall capacity to deliver care over time (Fig. 4), providing a clearer indication of the scale of services that can be supported for planning, budgeting, and forecasting of commodities and workforce requirements. Although some services exhibit modest changes, either because demand is already close to being met or because staffing is not the primary constraint, the broader pattern confirms that workforce expansion has system-wide benefits. These additional services represent meaningful opportunities to avert disease burden; for example, earlier detection and treatment through increased HIV-related HSIs could help reduce progression to AIDS³¹.

Increased officer availability in the Target scenario does facilitate the ability to deliver more treatments for conditions directly; however, it still may not translate linearly to a proportional reduction in DALYs, as increased treatment delivery does not necessarily guarantee a reduction in morbidity or mortality. Not all interventions have the same effect on reducing ill health, depending on treatment efficiency, accuracy of diagnostics, and stage of intervention, all of which are captured by the TLO model. In line with the 88.6% shortfall to the Target in Mental officers shown in Fig. 1, addressing this gap would substantially expand officer availability and enable delivery of additional services. For instance, while antidepressants and talking therapies show efficacy in improving depression outcomes (the relative risk of resolution for depression is increased by 1.5 with the use of antidepressants and 1.1 with talking therapies), their ultimate effect on DALYs will rely on the proportion of cases achieving remission and the degree to which reduced symptoms lessen the associated disability burden³². With Malaria, many services are delivered by the DCSA cadre³³, particularly rapid diagnostic tests (RDTs) performed at community or Level 0 facilities (Fig. 7 in ¹⁷ supports this interpretation). In the Target scenario, DCSA staffing is already at the establishment level (0% shortfall, Fig. 1), reflecting the relative maturity of this cadre under existing programmes. Therefore, expanding the workforce does not translate uniformly across diseases or service areas, and these complex, non-linear relationships emphasise the value of a model such as TLO.

The main assumption taken to undertake this analysis is a rigid healthcare system approach (also explained in ^17,26) that applies “hard constraints” to HCW utilisation, whereby no level of overtime work and/or task-shifting is allowed, and HSIs must take the time that is specified (based on HCWs’ elicited expectation¹⁷). It is likely that, in reality, some level of elasticity regarding assumptions takes place, such as informal task-shifting, overtime, or varying patient throughput²⁷ may influence actual service delivery, although these processes are difficult to quantify. Similarly, both scenarios do not incorporate unexplained staff absences, meaning daily capabilities may be overestimated to a similar extent. The opposing effects of working overtime and some staff absence create some hesitancy around the DALYs incurred and should be noted. However, because both scenarios rest on the same assumptions regarding overtime, absence, and other unobserved processes, the relative difference between scenarios is less likely to be affected. Given that both scenarios considered make identical assumptions about these constraints, our results should therefore be interpreted as reflecting the health-cost of maintaining 2019 staffing levels versus filling all established positions, rather than an absolute forecast of future DALYs.

Another limitation is that the analysis, although dynamically capturing population growth, assumes no growth in the healthcare workforce numbers over the entire simulation period of (2023-2040), and that consumable availability levels remain constant. These simplifying assumptions were taken for tractability. As the same assumptions are applied in both scenarios, we do not expect them to strongly bias our results.

Another factor not captured by the model is the possibility that understaffing perpetuates a self-reinforcing cycle: too few HCWs increase the health burden, which in turn overloads the system, driving burnout among staff, higher rates of absence, and lower care quality³⁴. This reduces service delivery further, increasing the health burden again, and can ultimately drive HCWs to leave, worsening shortages. One 2018 cross-sectional study in Malawi of 520 HCWs involved in HIV care (a cause with potential for substantial DALYs reduction in our analysis (Fig. 3)) 62% met the criteria for burnout, defined as scoring in the mid-to-high range on either the Emotional Exhaustion or Depersonalization subscales of the Maslach Burnout Inventory and burnout was also significantly associated with self-reported suboptimal patient care³⁵. Additionally, absenteeism among HCWs in countries across Sub-Saharan Africa has shown rates as high as 50% due to a range of reasons^36,37, including emotional demand and their health, with far-reaching implications such as causing patients to forego treatment-seeking due to delays ^38,39. As these factors are not explicitly included in our modelling framework, the relative health gains that could be achieved from fully filling establishment positions could be underestimated.

Finally, our analysis does not address whether filling established posts is the right strategy for health system strengthening. Previous analyses^17,20,40 suggest that established positions may not correspond well to population health needs. Therefore, the conclusion of this study is not necessarily that government and donors should immediately fill all unfilled posts, but rather that there is a measurable health cost to continued shortfalls under current establishment assumptions. Malawi faces many competing demands on its limited health-sector resources, and we do not compare the health gains from staffing increases with alternative uses of funds. Future analyses should consider HRH expansion paths that are explicitly aligned with the health needs of the population, rather than relying on workforce establishment sizes that may have been based on limited evidence.