The most exciting development in the plan for DCP3 is that it will now include an overall comparison of health interventions across all areas, in terms of cost-effectiveness. That comparison will form part of the 1st volume, which will be published last, since it draws the other volumes together. Although there was a comparison of this type in DCP2, the initial plan was not to include one in DCP3. The reason was that comparing numerical cost-effectiveness of interventions across areas might give a feeling of spurious precision. In many cases, the estimate of the cost-effectiveness of a particular intervention has a large error margin, which makes people hesitant to compare it to the estimate for a very different intervention. It’s important to be clear that the estimates are uncertain and vary by region. On the other hand, these estimates are our expectation values. We cannot be certain that the interventions will have the impact these estimates predict; but these are the best indication we have and so acting on them gives us the best chance of helping people as much as we can. The challenge is presenting them without giving a false sense of precision.
This week’s meeting of the advisory panel was to discuss the two volumes of DCP3 which will be published first: those on surgery and on cancer. These two areas aren’t ones that we usually think of as particularly cost-effective, but the current drafts do contain some promising intentions. One of these is the current estimate for training first responders in first aid (in the surgery volume), which is just $7 per DALY. This is in keeping with findings in DCP2 that health education is often very cost-effective. The current estimate for providing essential surgical facilities at a first level (district) hospital is also very cost-effective, at $11-223 per DALY.
Cancer treatments are sometimes given as paradigm examples of hugely expensive interventions which serve only to marginally extend the lives of the elderly. Although in the developed world we do spend large sums on cancer drugs, some cancer treatments are extremely cost-effective. This is sometimes obscured by the fact that improving cancer treatments has a high diminishing marginal return. An example given in DCP3 is that for a childhood cancer we used to prescribe a $50 treatment, which had a 50% survival rate. We now prescribe a $100,000 treatment, for which the survival rate is over 90%. We should strive for a world in which all children can receive the latter. But in the mean time, the former can be used to help many more children.
Also in the cancer volume is information about tobacco, which is still causing more than a million deaths per annum. The current draft of the volume estimates that doubling the price of cigarettes in low and middle income countries could reduce by a third the number of people who smoke, and that that would prevent over a 100 million deaths (meaning averting deaths before 70). Unfortunately, increasing tax on cigarettes by that amount is politically difficult. If it was possible, it would not only save lives, but also generate government revenue that could be used to help people in other ways.
Going forward, Toby will be primarily concerned with the 1st (summary) volume of DCP3, for which he may write a chapter. He is also on the technical advisory group, which discusses the best ways to disseminate the information in DCP3 – for example, what kinds of executive summaries should be provided. The estimated time for the 1st volume appearing is early 2016. In the mean time, look out for the volumes on cancer and surgery, which will be coming out in a few months time.
Image credit: DCP2.org