PharmacoEconomics & Outcomes News 330 - 15 Sep 2001 Doubts cast on success claims of DM programmes Disease management (DM) vendors are producing misleading outcomes information, raising questions about the success claims used to promote many DM programmes, according to a report in Disease Management News. There are 2 main types of questionable data collection and reporting methods, says executive director of Disease Management Purchasing Consortium, Massachusetts, US, Al Lewis. First, when outcomes data are only collected and reported for people who enrol in a particular DM programme, as opposed to the entire covered population; this creates a motivation bias as only the more motivated people tend to enrol in the programme. Second, when DM vendors select participants for a programme based on high utilisers of a health plan in the previous year; this results in a statistical bias known as ‘regression to the mean’, comments Lewis. This bias suggests that, over the course of a year, a certain percentage of high utilisers would improve regardless of whether they received an intervention. The use of study periods that are too brief to be statistically valid, or small sample sizes, also raises questions over the statistical validity of some vendors’ outcomes data, according to executive director of American Healthways Bob Stone. He suggests that the only way to truly determine if a DM programme is a success is to conduct prospective, double-blind studies, but he acknowledges that such studies are difficult and expensive to perform; the level of scientific sophistication afforded by such studies is not really something that health plans are interested in, he adds. DM analysts say that vendors need to conduct statistically valid outcomes studies based on large numbers of health plan members (preferably the entire health plan population). DM vendors need to report true outcomes success rates, otherwise their companies, and even the DM industry as a whole, will come into disrepute. DM outcomes: industry’s seamy underside critics say. Disease Management News 800876040 6: 1 & 4, 10 Aug 2001

1173-5503/10/0330-0001/$14.95 Adis © 2010 Springer International Publishing AG. All rights reserved

PharmacoEconomics & Outcomes News 15 Sep 2001 No. 330

1