Health and related indicators: assessing the data foundation of a learning health system

Aim South Africa produces extensive routine, survey, surveillance and financial health data, but their value depends on translation into decisions that improve health and service delivery. This paper assesses whether available indicators support system learning across a continuum from data availability and signal identification, to interpretation, documented response, and demonstrated improvement. It describes key South African health and related indicators, assesses what they reveal about progress, inequity and system responsiveness, identifies data constraints, and examines where learning cycles are completed or break down. Method Secondary data were synthesised from routine health information systems, Statistics South Africa, disease registers, surveillance platforms, the National Treasury and international estimation series. Indicators were selected for continuity, policy relevance, currency and provincial disaggregation, with attention to source divergence, denominator effects, timeliness, completeness and disaggregation. Divergent estimates were presented side-by-side, and implausible values interrogated and flagged. Fourteen domains were assessed across three learning arcs: practice into data, data into knowledge, and knowledge into changed practice. Results Indicators show simultaneous progress and persistent strain. Life expectancy continues to recover; HIV diagnosis and viral suppression remain strong; and tuberculosis and maternal and newborn health demonstrate established learning and feedback processes. However, substantial provincial inequities persist across social determinants, service access, outcomes, workforce and financing. Immunisation remains below target, adolescent pregnancy and injury burdens are unevenly distributed, non-communicable disease surveillance is weak, and real public health spending has stagnated. Complete signal-to-response loops were traceable in only three domains: tuberculosis, HIV, and maternal and newborn health. The weakest domains, particularly non-communicable diseases and injuries, faltered primarily at data capture. Conclusion South Africa has substantial data assets but uneven system learning. Weak surveillance, inconsistent denominators, incomplete private sector integration and limited sub-national disaggregation constrain timely action. Strengthening denominator governance, non-communicable disease surveillance, private sector reporting and routine plausibility screening is essential if indicators are to drive, rather than merely describe, improvement.