Date Published: February 20, 2007
Publisher: Public Library of Science
Author(s): Glenn Salkeld, David Henry, Suzanne Hill, Danielle Lang, Nick Freemantle, Jefferson D’Assunção
Abstract: The authors set out to compare spending priorities for health care, across a selection of largely middle-income countries, through a survey of current and future decision makers.
Partial Text: Making the rules of health-care resource allocation transparent is a challenge for all governments. The Oregon Health Plan in the late 1980s was one such attempt to prioritise expenditure of limited Medicaid funds, based on public values . For decision makers, asking the general public and health professionals to express their preferences for health-care spending priorities can be a way of ensuring that the process and resultant spending priorities are seen as legitimate and fair . In a study comparing the preferences of health professionals and members of the public for setting health-care priorities, Wiseman found considerable uniformity in preferences between the two groups . However, some members of the public argued that it would be better to trust health professionals to make the correct decision in the first place.
A summary of the intervention rankings, pooled across countries, is shown in Box 1. Across all countries, childhood immunisation was the highest ranked intervention and cancer treatment for smokers was ranked as the least important priority for health-care spending (Box 1). There was little variation across countries in the median rank score for preventive health care and greatest variation for “lifesaving” interventions (Figure 1). The Kruskal-Wallis test for the null (that the median ranks were equal across countries) could not be rejected at the 5% significance level for the following interventions: childhood immunisation (p = 0.114), antismoking education for children (p = 0.327), screening for breast cancer (p = 0.355) and treatment for people with schizophrenia (p = 0.317). For all other interventions the null hypothesis was rejected at the 5% level, suggesting that the median ranks for these interventions are significantly different across countries. The Kruskal-Wallis test results did not change at the 5% significance level for the all country sample that excluded the South African pharmaceutical industry respondents.
This survey was intended as an educational exercise to introduce workshop participants to the notion that priority setting in health care is a value-laden exercise and one that should be informed by evidence-based medicine and economics. The interventions used in the survey, replicated from the Groves study , are formulated in very general terms. For example, GP care for everyday illness covers a wide category of services, from preventive measures to curative services. This limits our ability to make strong conclusions about one type of intervention versus another. There is a risk of confounding in the results due to the method of selection of our sample. The study participants were self-selected; they chose to attend the course. To the extent that policy makers who attend courses are systematically different from those who do not, this may have affected the extent to which subjects are representative of a population of health decision makers.
Whilst the results of this survey do not allow for a comparison between the preferences of health professionals and the general population, other studies have shown a reasonable level of uniformity of opinion, with a few exceptions. Wiseman found that the public gave equal weighting to health professionals for public health/prevention interventions but more weight (for spending) to coronary artery bypass grafting and less to hip replacement than did the health professionals . But overall, there was considerable uniformity of preferences between the two groups. Similarly, Myllykangas et al. found that the views of health professionals, local politicians, and the general public were generally similar, although the views of doctors differed substantially on some matters . On the other hand, Groves found that the public tended to put life-saving interventions such as heart transplants and intensive care for babies higher up the spending priority list than doctors or National Health Service managers (who themselves ranked life-improving treatment as twice as important as lifesaving ones) .
The strongest opinions elicited from our sample of health professionals, a general preference for prevention and for spending on the young over the old, bear little semblance to how health care dollars are actually spent in many countries. Other opinions, such as a preference to rescue an identifiable life in danger and a tendency to assign blame for disease, seem to exert more influence over current health care spending. The values expressed here transcended national and sectoral boundaries. Across the world many countries are struggling with the health and financial implications of a rapid rise in non-communicable disease. If health care professionals and policy makers believe that prevention and targeting the young is an important principle for health spending priorities, then health care funders should examine the cost effectiveness evidence for intervening early in life. Whilst the “rule of rescue” will always be a significant influence in health-care spending priorities, greater attention needs to be given to those interventions that are life improving as well as life extending. 3