This article is more than 5 years old. Information may no longer be current.
Practice patterns, not patients, may drive overutilization in oncology
The United States spends more on health care than any other country in the developed world — an estimated $2.9 trillion in 2013.
Increasingly, there are concerns that a substantial proportion of health care spending is spent on unnecessary or unhelpful care. Using data from 2009, the IOM estimated that $750 billion is wasted every year in the U.S. health care system; of that figure, $210 billion is spent on unnecessary care, according to IOM estimates. A frequently cited driver of overutilization is patient demands for services.
Interestingly, although patient demands are frequently cited as drivers of overutilization, until the recent publication by Gogineni and colleagues, there was little formal documentation of actual practice in this realm.
Lisa Hicks
Gogineni and colleagues report on the frequency and appropriateness — as determined by the treating clinician — of patient requests for services in a large study of more than 5,000 outpatient, oncology encounters. In their study, patient requests for additional tests and/or treatments were uncommon, occurring in only 8.7% of encounters. Moreover, clinicians judged most requests to be appropriate (72%) or equivocal (17%), and they felt a minority of requests were inappropriate (11%).
Importantly, when clinicians did view a request for treatment/testing as inappropriate, they rarely felt compelled to act on the request. Of the 50 “inappropriate” requests in the sample, only seven were acted upon.
Gogineni and colleagues provide important information about which types of tests and treatments are most frequently requested by oncology patients. The most frequently requested service in their sample was imaging (49% of all requests). Of these, only a small number were considered inappropriate by clinicians (8.7% of all imaging requests). Requests for supportive care — including pain and insomnia treatments — also were relatively common (15.5% of all requests), and were overwhelmingly considered appropriate by clinicians.
Requests for genetic or chemosensitivity tests were uncommon (only 5.2% of all requests), but it is notable that more than one-third (39%) of these requests were considered inappropriate by clinicians.
The findings in this study contradict those of past publications, which have tended to support the notion that patients’ demands are an important factor in health care overutilization. Why the difference? First, past publications have tended to ask physicians about their general perceptions of patients’ influence on health care expenditures, or to ask physicians to recall distant patient interactions. Both methodologies invite recall bias and may lead to inappropriate weighting of rare but memorable interactions.
A single past publication by Kravitz and colleagues reported on actual physician–patient interactions (545 encounters) and reported a higher rate of patient demands (23%) than the Gogineni study. Importantly, however, the study by Kravitz and colleagues included only patients seeking care for a new or worsening problem or for an undiagnosed disease, and it did not include any estimation of the appropriateness of requests.
It is perhaps not surprising that patients seeking help for an unresolved medical problem would make more requests of their physicians than those receiving unfiltered oncologic care. What is unknown is how many of the patient requests in the Kravitz study were appropriate.
Ultimately, patient requests for care are integral to the practice of medicine: Some requests are direct, rare requests are demanding, and many are implied simply by walking through the clinic door.
The study by Gogineni and colleagues makes it clear that patient requests, or even demands, are not inherently negative.
In their study, direct requests for tests and treatments were uncommon, and when they occurred, they were mostly appropriate. Moreover, when inappropriate requests occurred, clinicians rarely felt obliged to act upon them.
Taken together, these results suggest that in oncology — and perhaps more broadly — overutilization is not driven by patients. Now it is time to turn our attention to other factors that may drive overutilization, including our own practice patterns.
References:
CMS. National Health Expenditures 2013 Highlights. Available at: www.cms.gov/research-statistics-data-and-systems/statistics-trends-and-reports/nationalhealthexpenddata/downloads/highlights.pdf. Accessed on Feb. 20, 2015.
Kravitz RL, et al. JAMA. 2005;293:1995-2002.
Kravitz RL, et al. Arch Intern Med. 2003;163:1673-1681.
Smith MD, et al. Best care at lower cost: The path to continuously learning health care in America. IOM. Washington, D.C. National Academies Press; 2012.
Tilburt JC, et al. JAMA. 2013;doi:10.1001/jama.2013.8278.
For more information:
Lisa Hicks, MD, MSc, is staff physician in the department of medicine at St. Michael’s Hospital and assistant professor in the department of medicine at University of Toronto. She also is chair of ASH’s Choosing Wisely Task Force. She can be reached at hicksl@smh.ca.
Disclosure: Hicks reports no relevant financial disclosures.