The case against annual profiles
Duration-specific Dutch disability weights (95% confidence interval) for three selected chronic conditions and annual profiles of conditions of short duration
Condition – duration
Severe hearing loss – chronic Severe vision loss – chronic Paraplegia – chronic
Pneumonia – 2 weeks Influenza – 2 weeks Constitutional eczema – 2 x 6 weeks Acute cystitis – 1 week Acute urethritis – 1 week Acute pyelitis – 2 weeks Peptic ulcer disease – 4 weeks Acute bronchitis – 2 weeks Acute tonsillitis – 2 weeks Acute sinusitis – 2 weeks Acute nasypharyngitis – 1 week Symptomatic acute gonorrhoea or chlamydia – 1 week Digestive tract infection (complicated) – 3 weeks Digestive tract infection (mild) – 2 weeks
weights appear to be considerably overvalued. Their use in burden of disease assessments would give undue weight to short duration conditions.
Essink-Bot and Bonsel (chapter 9.1) state that the annual profile ap- proach worked reasonably well. This may be the case for the valuation procedure. It is indeed a bit uncomfortable for panel members to imagine living a full year with pneumonia. However my experience as a member of a valuation panel in the past is that it is feasible to imagine living for a year in a state as severe as during a bout of pneumonia and to compare it with a year with a chronic condition.
A reason for the high values given to the annual profiles may be that valuation procedures have problems in determining accurate weights for low-severity conditions. Specifically, in the PTO1 and PTO2 methods the Dutch study replicated from the GBD protocol, the numbers of people to trade off against 1 000 healthy persons for a low-severity condition is close to 1 000 for the PTO1 method and a very high number in the PTO2 method. For example, the equivalent for a condition with a disability weight of 0.01 in the PTO1 trade-off is 1 010 and in the PTO2 trade-off is 100 000 (Table 1). What probably happens is that most people find it hard to make a distinction between one large and another even larger number in the PTO2 trade-off and similarly consider the difference trivial between trading off 1 010 or 1 005 people with a minor disease against 1 000 healthy people in the PTO1 version. Unfortunately, at the lower end of the severity spectrum these seemingly trivial differences in PTO values trans- late into a doubling of the corresponding disability weight. As a conse-