Disease severity and public preferences

Does including the name of a specific severe disease in preference surveys affect public attitudes toward resource allocation? A recent proof of concept study suggests that ‘naming’ does not increase preferences for allocating more resources to severe diseases when improvement is likely to be modest.

Allocating limited health care resources is a continuing challenge.  Maximizing the overall health benefit realized from health care spending entails giving priority to patients whose health is most likely to improve as the result of treatment.  Research suggests, however, that the public often is willing to sacrifice as least some health gain to treat those who are severely ill, even if their gain from treatment is limited.

Surveys aimed at understanding allocation preferences rarely name the ‘severe disease’.  In part, this has been based on concerns that naming the disease — using the word ‘cancer’, for example — may elicit an emotional response.  OHE’s 2010 summer intern, He Guo, designed a survey to test the effect of naming the ‘severe disease’ on such public preferences.  The online survey included 134 respondents from the student bodies of the London School of Economics and Political Science and Peking University.  Three sets of questions were used: one did not name the severe disease, one named ‘cancer’ and one named ‘congestive heart failure’.  No additional information was provided about either disease.  The survey allowed four choices: (1) allocate most to a moderate illness, likely to improve considerably with treatment, (2) allocate most to the severe illness, likely to improve only a little, (3) allocate equally between the two and (4) ‘I am not able to make a decision and would prefer that the choice be made by others’.

The study results showed, first, that naming the disease did not increase the preference for treating the severe disease at the expense of the moderate disease and, second, that cancer was not given priority over heart failure.  Overall, respondents preferred to allocate more resources to the moderately ill than to the severely ill, whether or not the severe disease was named.

Because the study has some important limitations, it is a ‘proof of concept’ study rather than one with immediate implications for policy making.  It suggests, however, that understanding public preferences for treatment requires further research on the impact both of naming diseases and providing more contextual information about them.  Such research is important for both individual-level health valuation studies and societal-level resource allocation preference studies.