Delivery arrangements in low-income countries show a broad evidence base, but uneven certainty
Health systems in low-income countries are often asked to do more with fewer trained staff, thinner infrastructure, and limited room for inefficiency. This Cochrane overview assembled a large body of systematic review evidence on delivery arrangements — who provides care, when care is delivered, where care happens, and how services are coordinated — to determine which strategies improve outcomes that matter to patients, providers, and health systems.
The review is striking for both its breadth and its restraint. Investigators identified 7,272 systematic reviews and included 51 that met quality criteria and addressed delivery arrangements relevant to low-income countries. Six of those reviews carried important methodological limitations; the rest had only minor limitations. Across the evidence base, the strongest signals clustered in a few recurring system pressures: workforce shortages, fragmented referral pathways, and the challenge of reaching patients far from fixed facilities.
Shifting who delivers care changes outcomes in multiple settings
One of the most consistent themes was the effect of changing who provides care. Lay health workers improved care in several contexts, including hypertension, maternal and child health, infectious diseases, and community-based neonatal care packages. Midlevel health professionals also performed well for abortion care, while social support for pregnant women at risk and midwife-led care for childbearing women produced beneficial effects in the evaluated outcomes. Non-specialist providers delivered effective care in mental health and neurology, and physician-nurse substitution showed benefit in the included comparisons.
This pattern makes biological and operational sense. In settings where specialist clinicians are scarce, a trained non-physician workforce can extend contact time, improve continuity, and bring care closer to households that might otherwise disengage. For chronic illnesses such as hypertension, the value is not only in task shifting but also in maintaining repeated follow-up, reinforcing adherence, and catching uncontrolled disease earlier.
Lay health workers improved care across maternal, child, and infectious disease programs.
Coordination tools helped more when they closed gaps between encounters
The evidence for coordination of care was equally important. Hospital clinical pathways, case management for people living with HIV and AIDS, interactive communication between primary care doctors and specialists, discharge planning, referral from primary to secondary care, and team midwifery all showed desirable effects in at least one outcome with moderate- or high-certainty evidence and no moderate- or high-certainty evidence of harm.
These interventions share a common clinical logic: they reduce the probability that a patient disappears between decision points. In low-resource systems, a care plan can fail not because the treatment is wrong, but because no one is clearly responsible for the next step. Hospital discharge planning and case management address that fault line directly by linking inpatient decisions to community follow-up, medication access, and reassessment. For HIV care, where retention and adherence are decisive, the coordination effect is especially plausible because continuity is inseparable from viral suppression and long-term treatment success.
Location of care mattered as much as the service itself
Evidence on where care is provided also pointed to practical gains. High-volume institutions performed better in the reviewed comparisons, and home-based care — with or without a multidisciplinary team — benefited people living with HIV and AIDS. Home-based management of malaria, home care for children with acute physical conditions, community-based interventions for childhood diarrhoea and pneumonia, out-of-facility HIV and reproductive health services for youth, and decentralised HIV care all showed effective results in the evaluated outcomes.
In low-income countries, geography often functions as a clinical variable. A correct diagnosis or appropriate drug can still fail if a patient cannot return for review, afford transport, or wait in a crowded clinic. Home-based and community-based delivery models reduce those barriers by moving part of the health system toward the patient. For infectious diseases in children, where rapid treatment can determine survival, decentralised care can shorten delays between symptom onset and therapy. For HIV services, decentralisation can lower attrition and widen access for patients who are otherwise lost to follow-up.
Mobile phones and simple information tools delivered measurable gains
The evidence for information and communication technology was more focused but still clinically relevant. Mobile phone messaging improved outcomes for patients with long-term illnesses, increased attendance at healthcare appointments, and supported adherence to antiretroviral therapy. Women carrying their own case notes during pregnancy also benefited, and interventions to improve childhood vaccination were effective.
These are modest interventions on the surface, but their clinical value comes from reducing administrative friction. Messaging reminders can prevent missed appointments, while adherence prompts can support long-term treatment routines that are easily disrupted by poverty, travel, or competing household demands. Case notes carried by the patient can preserve essential information across settings where records are fragmented or unavailable. In systems with intermittent connectivity and limited electronic infrastructure, even basic communication tools can function as a bridge between contact points.
Decision support strengthened quality and safety in HIV care
Within quality and safety systems, decision support with clinical information systems improved care for people living with HIV/AIDS. This finding is especially relevant where clinicians face high caseloads and inconsistent access to specialist support. Decision support can standardize prescribing, flag missing steps, and reduce variation in care when frontline teams are stretched thin.
Complex interventions spanning more than one delivery category also showed benefit. Emergency obstetric referral interventions were effective, which is unsurprising given the time-sensitive nature of obstetric emergencies. Delays in deciding to refer, reaching a higher-level facility, and receiving definitive treatment are common contributors to maternal harm, and referral systems are designed to compress those delays.
What the evidence can support, and where it still falls short
The overview makes a strong case that delivery redesign can improve health system performance in low-income countries, but the certainty is uneven. Most of the available evidence concentrated on who provides care, where care is provided, and how care is coordinated. Other delivery questions remain less thoroughly studied, and several findings still face uncertainty about how well they generalize across different low-income settings.
Another constraint is the evidence base itself. Some reviews were methodologically limited, and many interventions were supported by low- or very low-certainty evidence or by a small number of studies. The review therefore offers direction more than finality: it identifies strategies that appear useful, but it also exposes where primary research is still too sparse to support confident policy transfer. Future evaluations need to test not just whether an arrangement works in one setting, but whether it can be adapted across health systems with different staffing patterns, referral capacity, and community access to care.