Health service research coming from companies with strong financial incentive for positive results A “Special Article” in the Sept. 23, 2010 New England Journal of Medicine entitled, “A Randomized Trial of a Telephone Care Management Strategy” by Wennberg, et. al., concluded that a targeted telephonic care-management program was successful in achieving a net reduction in medical costs of $6 per person per month ($72 per person per year). Among complex chronically ill Medicare beneficiaries, this kind of telephonic only health coaching and education related to chronic and “preference-sensitive” conditions as used in the Wennberg study has not been demonstrated to affect hospitalization or total medical cost.
The evidence from the Medicare Coordinated Care Demonstration (in which HQP has been a participating site) clearly indicates that for complex, chronically ill, Medicare beneficiaries, the ‘high touch’ approach with frequent in-person contacts correlates with effectiveness. About half of all of HQP’s contacts with our high-risk participants is telephonic, but the other half consist of in-person encounters (total contacts average 19.4 per year of which 10.3 are face-to-face encounters). There are many other significant ways HQP’s model differs from the intervention used in the Wennberg study - click here for more information.
Some readers might wonder about the potential for bias in the Wennberg article, given that the study was funded, designed, undertaken, analyzed, and written by a for-profit company (Health Dialog) for whom the value of a positive report in the NEJM from a marketing perspective is very significant. Less cynical readers might only have desired more explicit articulation of what elements of the study plan were defined a priori or a few more methodological details like whether the telephonic care managers were blinded to subjects’ study group assignment. At least with regard to transparency, the authors acknowledge that they “… performed several post hoc analyses to identify subgroups in which the effect of enhanced support was greatest …” and that their grouping of patients into 4 hierarchical cohorts occurred “… After outreach targeting, but before examination of the results …”
Other readers might take exception to parts of the Discussion section that read more like a marketing pitch than a careful synthesis of research findings;
“ … The group studied was a commercially insured population; however, 7000 of the subjects were 65 years of age or older … Medicare was the primary insurer for most of the subjects who were 65 years of age or older, and we estimate that over half the savings in such cases accrued to this publicly funded program.” To what savings do they refer? There is no evidence offered in the paper that the ‘Enhanced-Support Group’ lowered medical cost in this age group. In the only comparison of outcomes I could find for this age cohort (‘Age 65+’, in the online Supplementary Appendix, Table 5) there was no meaningful difference in hospitalizations; -3.7%, p=0.717.
Cynical readers might be compelled to remain open-minded if supporting data demonstrated that person-centered outcomes were improved by the intervention. Alas, the authors found “… no significant difference between the two groups with respect to effective care measures (laboratory tests or pharmacy services)…” and “…We could not analyze mortality or changes in functional status, owing to a lack of data.”
Some readers may be reassured by the footnote at the end of the article, “Supported by Health Dialog Services. Drs. Wennberg, Lang, Marr, and Bennett and Mr. O’Malley all report being employees of Health Dialog Services. Drs. Wennberg and Bennett formerly held stock options in Health Dialog Services, but all options have been exercised. No other potential conflict of interest relevant to this article was reported.”
For doubting Thomases though, the totality of the factors noted above, may still cast doubt on the overall conclusion reached by the authors. But remember, just because there may be the appearance of a conflict of interest, doesn’t mean that the results might not still be true – at least in the population studied. Moreover, you can’t blame a company for seeking to optimize their marketing. It’s an entirely appropriate and necessary part of free enterprise. Pharmaceutical and medical device companies certainly face this challenge too.
Perhaps the most fascinating question is, How does anyone get this much ad space in, of all publications, the New England Journal of Medicine? Anyone interested in this question should certainly read the full article for themselves and make their own assessment. But, you’ll need a paid subscription to the NEJM. This would not be the case, had the study been published in an open access initiative like the PublicLibrary of Science or had the NEJM and Health Dialog had the largess to make this “Special Article” freely available to the public. Then again, NEJM has its own business needs to consider, but that’s a whole other topic.
Bottom line: Policy makers and taxpayers need to discern between research undertaken primarily to discover new ways to address the fundamental challenges facing our health care system versus that undertaken primarily to enhance commercial marketing of a product or service. Neither is intrinsically bad, both can provide useful insights, and the line between the two can be blurry, but an appreciation of where one is along this ‘spectrum’ is essential.
Some readers might wonder about the potential for bias in the Wennberg article, given that the study was funded, designed, undertaken, analyzed, and written by a for-profit company (Health Dialog) for whom the value of a positive report in the NEJM from a marketing perspective is very significant. Less cynical readers might only have desired more explicit articulation of what elements of the study plan were defined a priori or a few more methodological details like whether the telephonic care managers were blinded to subjects’ study group assignment. At least with regard to transparency, the authors acknowledge that they “… performed several post hoc analyses to identify subgroups in which the effect of enhanced support was greatest …” and that their grouping of patients into 4 hierarchical cohorts occurred “… After outreach targeting, but before examination of the results …”
Other readers might take exception to parts of the Discussion section that read more like a marketing pitch than a careful synthesis of research findings;
“ … The group studied was a commercially insured population; however, 7000 of the subjects were 65 years of age or older … Medicare was the primary insurer for most of the subjects who were 65 years of age or older, and we estimate that over half the savings in such cases accrued to this publicly funded program.” To what savings do they refer? There is no evidence offered in the paper that the ‘Enhanced-Support Group’ lowered medical cost in this age group. In the only comparison of outcomes I could find for this age cohort (‘Age 65+’, in the online Supplementary Appendix, Table 5) there was no meaningful difference in hospitalizations; -3.7%, p=0.717.
Cynical readers might be compelled to remain open-minded if supporting data demonstrated that person-centered outcomes were improved by the intervention. Alas, the authors found “… no significant difference between the two groups with respect to effective care measures (laboratory tests or pharmacy services)…” and “…We could not analyze mortality or changes in functional status, owing to a lack of data.”
Some readers may be reassured by the footnote at the end of the article, “Supported by Health Dialog Services. Drs. Wennberg, Lang, Marr, and Bennett and Mr. O’Malley all report being employees of Health Dialog Services. Drs. Wennberg and Bennett formerly held stock options in Health Dialog Services, but all options have been exercised. No other potential conflict of interest relevant to this article was reported.”
For doubting Thomases though, the totality of the factors noted above, may still cast doubt on the overall conclusion reached by the authors. But remember, just because there may be the appearance of a conflict of interest, doesn’t mean that the results might not still be true – at least in the population studied. Moreover, you can’t blame a company for seeking to optimize their marketing. It’s an entirely appropriate and necessary part of free enterprise. Pharmaceutical and medical device companies certainly face this challenge too.
Perhaps the most fascinating question is, How does anyone get this much ad space in, of all publications, the New England Journal of Medicine? Anyone interested in this question should certainly read the full article for themselves and make their own assessment. But, you’ll need a paid subscription to the NEJM. This would not be the case, had the study been published in an open access initiative like the PublicLibrary of Science or had the NEJM and Health Dialog had the largess to make this “Special Article” freely available to the public. Then again, NEJM has its own business needs to consider, but that’s a whole other topic.
Bottom line: Policy makers and taxpayers need to discern between research undertaken primarily to discover new ways to address the fundamental challenges facing our health care system versus that undertaken primarily to enhance commercial marketing of a product or service. Neither is intrinsically bad, both can provide useful insights, and the line between the two can be blurry, but an appreciation of where one is along this ‘spectrum’ is essential.