When Journalists Play Scientist
When Journalists Play Scientist
Abstract & Commentary
Synopsis: When journalists perform "research" as part of their investigative reporting for a news story, they need to adhere to the same rules and conditions as research scientists if the results are to be valid.
Source: Gottlieb M, et al. New York Times 1997;Section 1, April 6:1.
In a front-page article titled "the doctor as Investor—A special report," New York Times journalists reported on the business practices of Columbia/HCA, the nation’s largest health care company, when it moved into Dade County, FL. In addition to uncovering incentives such as lucrative consulting contracts and real estate purchases benefitting physicians, the journalists performed an analysis of computerized hospital discharge data to discover what effect investing in Columbia/HCA had on physicians’ hospital admission practice. This study showed that as a group the 62 investors increased their referrals to Columbia/HCA hospitals 13.4% after investment and reduced their referrals to other hospitals by 22.4%. One orthopedic surgeon moved his entire practice to Columbia/HCA hospitals after investment. When his data were removed, the remaining 61 showed a 6% increase in admissions to Columbia/HCA hospitals and an 11% reduction to other hospitals.
COMMENT BY GORDON D. RUBENFELD, MD
In the current chaos of competition and business arrangements loosely called health care "reform," the effect of financial incentives on physician practice is front page news. Three reporters from the New York Times set out to show that physicians who invested in Columbia/HCA hospitals moved their business to the hospitals owned by Columbia/HCA. Hard-hitting journalists uncover questionable business deals all the time, and there is nothing particularly new about the troubling conflicts of interest posed by physicians who profit by ownership in a concern whose use they control. What was unique about this article was that the journalists went out and did their own "scientific study" to strengthen their arguments. This newspaper article might have been submitted to a medical journal as health services research. It might have been submitted—but it would likely have been rejected.
A study to evaluate the effect of investment on admissions is deceptively simple: Identify the investors and plot where they admitted their patients over time compared to non-investors. The reporters failed on both counts. They couldn’t really identify the actual investors because Columbia/HCA was not willing to tell them who their local investors were. Like any researcher faced with this problem, the Times reporters used a proxy for the variable they couldn’t measure. Unfortunately, their brief "methods section" doesn’t tell us exactly how they defined "investor status." It suggests that they used information on which physicians had attending privileges at the Columbia/HCA hospitals. Obviously, you can only admit where you attend and if you define "investor" status, in part, on attending privileges, then you necessarily will find a linkage between admissions and investor status. The journalists’ proxy for investor status biases the study toward finding a link between investing and admission to one of the Columbia/HCA hospitals.
The central focus of the study should have been a comparison of investor to non-investor admission practice. In the New York Times article, only the investor behavior is presented. Without a control group, the authors can’t exclude the possibility that every physician in Dade County, FL, investors and non-investors alike, moved some patients to the Columbia/HCA hospitals. It is certainly possible that managed care contracts, or the cost of care, or for that matter better patient parking, may have shifted the whole market toward Columbia/HCA hospitals. In any event, the reader can’t draw any conclusions about one group of physicians without an explicit comparison.
This article is superb investigative journalism, and does what a scientific piece could never do by interviewing participants, gaining access to financial records, and bringing an important issue to the public in a timely fashion. It is also a very mediocre example of scientific research that used a potentially biased definition of investor status and failed to present a control group for comparison. When journalists do science, clinicians should read the newspaper with the same care and skepticism they devote to medical journals.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.