Viewpoint: scaling up testing services for non-communicable diseases in Africa: priorities for implementation research
The burden of non-communicable diseases in Africa is rising rapidly and implementation of evidence-based control strategies is needed urgently. Testing people for hypertension and diabetes will be an important component in the fight against these diseases, as voluntary counselling and testing was for HIV-infection. We discuss the below the areas where we believe evidence is needed to inform policy.
Rates of non-communicable diseases (cardiovascular diseases, chronic respiratory diseases, diabetes and cancers) are rising rapidly in Africa and elsewhere (WHO 2009). Cardiovascular diseases carry the highest burden and are the most preventable and treatable among the non-communicable diseases. Hypertension is the most important risk factor for cardiovascular diseases and the leading risk factor for death worldwide (Lim et al. 2012). Diabetes causes significant morbidity and mortality in its own right, but is also a risk factor for cardiovascular diseases (Mbanya et al. 2010). Reliable data on the burden of non-communicable diseases are lacking from Africa, but some settings have reported rates of hypertension and diabetes among adults exceeding 25% and 5%, respectively, with little difference between rural and urban settings (Msyamboza et al. 2011). By 2030, non-communicable diseases are estimated to be the leading causes of deaths in Africa (Mathers & Loncar 2006).
Non-communicable diseases will be a massive burden to Africa's health system, which is weak and fragmented in many settings. Health systems in Africa are geared towards dealing with acute episodes of illness, such as those acquired from infection, rather than providing long-term chronic care. The exception is the recent experience gained in HIV control. HIV infection requires lifelong care and support, similar to most non-communicable diseases. The lessons learnt in delivering services for HIV infection are important in defining the research questions on non-communicable diseases control in Africa. Here, we focus on the issue of testing people for hypertension and diabetes, which will form an important component of control efforts for non-communicable diseases. Research is needed to identify testing strategies for non-communicable diseases that are accessible to people, minimise the risk of increasing stigma and can be scaled up and sustained in resource-constrained settings. We discuss below the areas where evidence is needed.
Who should be tested for non-communicable diseases?
Routine screening for non-communicable diseases within communities or indeed at clinical facilities is unlikely to be feasible in Africa. This is not normal practice in most developed country settings. However, knowledge of the high-risk groups will be important. We need data from Africa to quantify the association of non-communicable diseases with age, sex, body mass index and any other important traditional risk factors so that we can define who should be targeted for routine testing. We cannot simply extrapolate the data from developed countries because risk factors and the nature of non-communicable diseases might be very different (Ebrahim et al. 2013). For example, it has been reported that unlike in developed countries, non-communicable diseases are affecting a younger population in Africa (Yusuf et al. 2001; Miranda et al. 2008), but this association needs to be quantified accurately from different settings. We also need a better understanding of the relationships between non-communicable diseases and the infections that are highly prevalent in Africa, particularly HIV infection and tuberculosis, as these appear to affect the risks of non-communicable diseases (Currier et al. 2008; Kapur & Harries 2013). Integrating non-communicable diseases testing among HIV-infected people receiving antiretroviral therapy (Mateen et al. 2013) and vice versa might be cost-effective as these individuals will already be receiving chronic care, but this needs to be evaluated.
Where should non-communicable diseases testing be carried out?
We know from HIV infection that because of the severe shortages in healthcare staff and the high costs of transport, few people are likely to come forward for testing at primary care centres, let alone at district hospitals, which is where most diabetes and hypertension services are currently offered. Community-based strategies, such as testing in mobile vans at central locations (Corbett et al. 2010), home-based testing (Fonner et al. 2012; Kennedy et al. 2013) and self-testing (Choko et al. 2011), which have been evaluated for HIV infection and TB, need to be investigated for testing for hypertension and diabetes. Given the infrastructure now available for HIV testing, the feasibility of integrating hypertension and diabetes testing with HIV testing, for example at voluntary counselling and testing centres or antenatal clinics, needs investigation. Although an encouraging example of this integration has been described in Uganda (Chamie et al. 2012), a comprehensive assessment of the cost-effectiveness of adding hypertension and diabetes screening to a community-based HIV testing programme is needed. Integration of hypertension and diabetes testing will also contribute to reducing the stigma still associated with HIV infection. Much of our knowledge today in hypertension and diabetes management comes from clinic-based studies in developed countries. Recent data suggest significant variation in the distribution of hypertension levels when monitoring is carried out in a clinic or in the community or is ambulatory (Hodgkinson et al. 2011). Research is needed to determine variation in these measurements in the African setting.
Who should do the non-communicable diseases testing and diagnosing?
Given the severe shortages of clinically qualified healthcare staff, trained lay workers and lower cadre clinic staff, working with clear referral pathways and under supervision, will need to play a major role in the testing for and diagnosis of hypertension and diabetes and also in the monitoring and support of patients following a diagnosis. Evidence of the use of trained lay workers in HIV infection is accumulating (Jaffar et al. 2009), but their use outside research studies remains limited. Integration of non-communicable diseases and HIV testing may make their use more cost-effective. However, the ability of lay workers to handle multiple infectious and non-infectious conditions needs evaluation.
How frequently should people be tested for non-communicable diseases?
For those found to have normal blood pressure and blood glucose, we need to know when they should be retested. Moreover, how often should those with pre-diabetes and borderline hypertension, who are placed onto lifestyle modification interventions, be retested? We cannot extrapolate data from developed countries as the introduction of non-communicable diseases in Africa appears to be occurring at a much faster pace than was the case in developed countries (Yusuf et al. 2001). This could mean that the period of pre-diabetes or borderline hypertension might be short in Africa. Finally, we will need data from Africa on how often people under treatment for hypertension or diabetes should be monitored, where the monitoring should occur and what is the lowest cadre of healthcare worker who could do this monitoring effectively.
Which factors are associated with people coming forward for non-communicable diseases testing?
We need research to understand better the type of person who comes forward for testing or chooses not to do so and the determinants of this decision-making. What proportion of people come forward for hypertension and diabetes testing even in settings where treatment services are severely resource-constrained and are located far away and drug supply is unreliable? What role does stigma play in hypertension and diabetes? What proportion never test for hypertension and diabetes and are first identified through a major clinical event, such as a stroke?
In HIV infection, testing rates have remained low. There has been greater progress recently, but the number tested annually is still less than 10%, and just more than half of HIV-infected persons do not know their HIV status (WHO/UNAIDS/UNICEF 2010). This is despite the widespread availability of antiretroviral therapy and the drives to get HIV-infected people into treatment programmes as early as possible. In all likelihood, given the experience with HIV infection, we can expect low levels of coverage of non-communicable diseases testing whether similar models of testing and counselling are provided as for HIV infection.
Should counselling be offered pre- and post-non-communicable diseases testing?
In an ideal world, we would like to offer people both pre- and post-test counselling in order to reduce their risk. This is the norm within HIV infection, which involves confidential one-to-one counselling, including evaluation of personal risk of HIV transmission to facilitate preventive behaviour (UNAIDS 2000). There is a large body of evidence showing that voluntary counselling and testing reduces HIV-related sexual risk behaviours (Denison et al. 2008; Fonner et al. 2012). However, it is unlikely to be practical to scale up non-communicable diseases testing combined with counselling because it is time-consuming and expensive to implement; even if this were possible, its coverage and its cost-effectiveness are likely to be low as we have seen for HIV infection (WHO/UNAIDS/UNICEF 2010). It is also possible that the formation of groups of patients will be critical to bring about lifestyle changes to tackle non-communicable diseases, and one-to-one counselling is irrelevant.
We need to investigate simplified models of non-communicable diseases testing, which can be scaled up rapidly. In particular, we need to know how effective is the hypertension or diabetes screening without the one-to-one counselling?
It is also critical that stigma is kept minimal around non-communicable diseases, as it will impede control efforts. Simplified testing protocols that are in line with approaches used for testing other conditions, rather than specialised protocols as is the case for HIV infection, might ensure that stigma of non-communicable diseases is limited, but this needs evaluation.
Non-communicable diseases pose a major imminent threat to sub-Saharan Africa, and their control presents major challenges for the continent's health systems. An important control strategy will be the scale-up testing for non-communicable diseases with linkage into care, particularly for hypertension and diabetes, which already appear to have high prevalences on the continent. It is essential that the control strategies be accompanied by research, which is informed by our experience in HIV infection.