Non-communicable diseases (NCDs) are the leading cause of mortality and morbidity worldwide,  with around 80% of NCD deaths occurring in low- and middle-income countries (LMICs). An important obstacle to addressing the growing burden of NCDs in LMICs is the limited set of indicators to monitor NCD service coverage and the quality of primary health care (PHC). For example, the WHO’s Global Reference List of 100 Core Health Indicators contains only two NCD service coverage indicators (cervical cancer screening and coverage for severe mental health disorders) and one indicator relevant to NCD quality of care (facility service availability and readiness). WHO’s PHC monitoring framework has a greater focus on NCDs (e.g. existence of NCD strategies, guidelines and protocols). However, it is still limited in the number of indicators that directly monitor NCD service coverage. This is problematic because NCDs represent close to 60% of the burden of disease in LMICs,  and NCD prevention and management relies on population-wide interventions to promote healthy behaviours combined with quality PHC to diagnose and provide long-term treatment.

Care cascades are robust, well-established indicators that could be more widely applied to monitor health service delivery for NCDs as well as the quality of PHC in LMICs. A care cascade is a simple model of the sequential steps of care that a patient would ideally undertake after becoming ill. Typically, this is simplified into 3 steps: diagnosis/awareness, treatment and cure (or rather ‘controlled’ or managed in the case of chronic conditions). At each step in the care cascade, there is a chance that a patient may not proceed to the next (e.g. some people with a condition will go undiagnosed or untreated). The concept is named after the three consecutive steps shown graphically as bars or columns, which resemble a descending ‘cascade’. First proposed by Maurice Piot in the context of WHO’s TB control programme, care cascades have mainly been used to analyse progress and barriers to interventions for infectious diseases such as tuberculosis (TB), HIV/AIDS, malaria, and hepatitis C. Data is often obtained using nationally representative household surveys that include individual-level biomarkers. However, it can also be collected through health facility surveys or routine health-information systems. The UNAIDS 90-90-90 target to end the AIDS epidemic is probably the most prominent example of the care cascade concept as a monitoring tool. Although a lot is known about care cascades for diseases like TB and HIV/AIDS, it is possible that they do not reflect the general quality of care in a country because these disease programmes have been heavily supported by the international community for several decades, often operating vertically in parallel to the rest of the health system. An indicator that reveals the proportion of people successfully diagnosed, being treated and in good control of their NCD (e.g. diabetes) could potentially tell us more about the overall performance of the health system.

NCD research-policy workshop in Lausanne

This potential use of care cascades to monitor the success of health systems in meeting needs for NCD care emerged as a key message from a recent workshop organised by the universities of Lausanne and Erasmus Rotterdam together with the World Health Organization. A recent study of 44 LMICs found that on average only 39% of people with hypertension were diagnosed, 30% were receiving treatment and 10% had controlled blood pressure. In the best performing high-income countries, by comparison, awareness-treatment-control rates are approximately 80-70-60. Similar studies from LMICs for hypercholesterolemia and diabetes reveal the average care cascade to be 31-29-7 and 44-38-23 respectively. The workshop added evidence on NCD care cascades from India, Sri Lanka, Cambodia and the Philippines. It also presented evidence of missed opportunities to diagnose people with hypertension when seeking care in public and private facilities. This accumulating evidence suggests that only a very small proportion of the  population living with NCDs in LMICs are diagnosed and treated effectively. There is also a strong socioeconomic gradient, with the disadvantaged being less likely to be diagnosed and effectively treated.

In short, NCD care cascades have considerable potential as indicators, and it would be positive to see this agenda further developed by the global community of policy makers and researchers. These indicators not only tell us about effective NCD service coverage levels and the extent to which the current health system is providing quality PHC, they also highlight general health system problems that need fixing. For example, a high proportion of people who are undiagnosed with hypertension could point to issues of service coverage as well as health seeking behaviours, which are influenced by health literacy and trust in providers. Integrated, people-centered care and affordable medicines are also likely to be crucial at each step of the cascade. Initial work might concentrate on the metadata for (a) standardised indicator/s (i.e. numerator, denominator, method of measurement, preferred data sources). Dissemination and common reporting could also be promoted in standard surveys such as the NCD module of the World Health Survey Plus. Indeed, benchmarking and tracking progress in NCD care cascade indicators could be a feasible way to monitor the effectiveness of policy and health system reforms (both within and across countries). Importantly, many countries have started to monitor NCD care cascades as part of their national plans and programmes. To date, NCD morbidity and mortality rates in LMICs suggest that the world is not on track to achieve SDG target 3.4 – a one-third reduction in premature mortality from NCDs by 2030. A focus on NCD care cascades would bring attention to accessible and affordable quality PHC, which in turn could provide a needed jumpstart to accelerate NCD mortality reduction.