Richard Smith: NCD among the bottom billion

Richard SmithMy main job these days means thinking about non-communicable disease (NCD) in low and middle income countries (LMIC), but a paper in the Lancet suggests that I may be thinking in the wrong way. It’s always hard to shift your mental model dramatically, but perhaps I need to do so.

I and the 11 centres in LMIC that I work with are largely following the WHO model, which says that there are four main NCD—cardiovascular disease, diabetes, chronic obstructive pulmonary disease, and some cancers—caused by four risk factors—tobacco, poor diet, physical inactivity, and the harmful use of alcohol. The main approach to these conditions is preventive, through steps like raising the price of tobacco, but also through simple treatments for risk factors like hypertension. The United Nations has ratified this approach, and WHO has set a target of reducing deaths under 70 from NCD by 25% by 2025, the 25 x 25 target.

Before the United Nations meeting in 2011 various groups argued that mental health should be included, but it was not—on the grounds that the four conditions share risk factors that are not quite the same as those for mental health problems, that we are less sure about prevention of mental health problems, and that if everything is a priority nothing is a priority. Others argued that the emphasis should be on universal healthcare, meaning that all conditions could be treated, but such an approach failed to understand that most of the response to NCD needs to be outside the health system.

The current challenge to the WHO model comes from a meeting in Rwanda and starts by arguing that the NCD of the very poorest in the world—those, for example, in Rwanda—do not suffer primarily from the four WHO conditions but rather from a “long tail” of other NCD—rheumatic and congenital heart disease, postinfectious renal failure, malignancies (such as Burkitt’s lymphoma), sickle cell anaemia, type I diabetes, asthma, appendicitis, suicide, epilepsy, and road traffic or workplace injuries. Together, say the authors quoting data from the Global Burden of Disease, these conditions account for 36% of deaths, 33% of life years lost, and 44% of disability adjusted life years (DALYs) in those under 40 in developing countries.

Agnes Binagwaho, the charismatic minister of health in Rwanda, has conceived a target of 80x40x20 target—to reduce deaths under 40 by 80% by 2020. The commentary in the Lancet has been developed with the NCD Synergies Group, which includes input from 10 African ministers of health. One of the leaders of the group is Gene Bukhman, a cardiologist from Harvard and a friend of mine with whom I’ve debated these issues.

The group believes that the target is achievable largely through universal healthcare plus mutisectoral action on indoor air pollution, safety on the roads and at work and at home, and extreme poverty. Indeed, they argue that Rwanda is on track to achieve this as it has between 2000 and 2010 seen a 49% reduction in deaths from NCD and injuries in those under 40. During this time, they point out, annual health expenditure was less than $27 for each person. So should I change my thinking?

To some extent I have already. I teach on NCD in LMIC in Amsterdam, and I certainly include some of Gene’s data and proposals in the discussion.  Gene has spoken to the centres I work with, and I circulated the paper to all of the researchers in our network. And I don’t deny the importance of the conditions that the Lancet paper describes.
But I have my worries.

Firstly, I’m not convinced by the data—either on the scale of the response or the reduction in deaths in Rwanda. Data on NCD from LMIC are weak and were even weaker in 2000. Most of our centres are collecting more reliable data and do find high rates of hypertension, diabetes, obesity, tobacco use, COPD, and cardiovascular disease. This is not to argue that the other conditions do not exist, but it’s hard to be sure that the Global Burden of Disease has got the scale right.

Secondly, the conditions listed in the Lancet paper are a ragbag, whereas the four WHO conditions share common risk factors and preventive strategies. It will be hard to devise a coherent strategy for attacking all the conditions identified in the Lancet.

Thirdly, we know little or nothing about how to prevent some of the conditions—for example, congenital heart disease, type I diabetes, and sickle cell disease. The response to some of these conditions—for example, congenital heart disease—is likely to be highly expensive, and it’s hard to see how concentrating resources on those patients, important as they are, will promote equity.

Fourthly, this would be a response primarily through health systems, and it seems unlikely that high quality health systems will be in place by 2020 in many LMIC. And concentrating on health systems, particularly specialist hospitals, may divert resources and attention from political and public health interventions that are likely to be better buys for reducing the burden of NCD.

Fifthly, there is a problem generalising from Rwanda to other LMIC when Rwanda has impressively concentrated on health and probably had more support, from groups like Partners in Health International, than other countries.

Finally, and perhaps most powerfully, too many targets can be counterproductive. My impression is that despite the United Nations meeting many countries are making slow progress with countering NCD, partly because they are understandably concentrating on the MDGs, which do not include targets for NCD. Adding another target may take things backwards.

So my thinking is changing, but not quite as far as Minister Binagwaho and colleagues would hope. But this is an important debate that needs to continue, and it’s right to concentrate on the bottom billion.

Competing interest: As the blog states, RS works with 11 centres in LMIC, all of whom are essentially following the WHO model. In addition, he is on the International Advisory Board of Children’s Heartlink, which promotes access to high quality cardiac surgery for children in LMIC.

Richard Smith was the editor of the BMJ until 2004 and is director of the United Health Group’s chronic disease initiative.