FROM SGIM 2017
Displaying Medicare allowable fees in the electronic health record at the time of order entry did not significantly reduce the number of inpatient lab tests at three Philadelphia hospitals.
In a study involving 98,529 patients and 142,921 admissions, Medicare payment information popped up randomly in the EHR when standard tests including complete blood cell counts, metabolic panels, and liver function tests were ordered. The costs of the labs varied depending on their extent. The message mentioned that “the dollar amount represents Medicare reimbursement for the test. Actual costs to the consumer may vary by patient insurance status.” Just over a third of the patients were actually on Medicare; most had private insurance.
The idea of the study was to see if cost information would curb unnecessary testing, and save money. “There is growing interest in using price transparency to influence medical decision-making toward higher value care,” Mina Sedrak, MD, and her colleagues said in a paper presented at the annual meeting of the Society of General Internal Medicine.
It didn’t work out that way. Four tests ordered per patient day when the messages appeared, and 2.34 when they did not. With messaging, the mean lab fee per patient day was $38.85, versus $27.59 without it. In an adjusted analyses comparing the intervention to the control group, there were no significant changes in overall test ordering (0.05 tests ordered per patient day, P = .06) or associated fees when pricing information was displayed ($0.24 per patient day; P = .47).
In a subset analysis, the investigators did find a small decrease orders for the most expensive labs and a small but significant increase in orders for the least expensive ones when physicians aware of cost (top quartile of tests based on fee value: -0.01; P = .04; bottom quartile: 0.03, P = .04).
Despite the overall negative results, there’s still a likely role for cost information in value improvement programs; what the study shows is that there’s a better way to use it, according to Dr. Sedrak, currently of the City of Hope Comprehensive Cancer Center in Duarte, Calif., and colleagues.
The investigators made several suggestions when reviewing their work.
“First, the price transparency intervention in this study was always displayed regardless of the clinical scenario. The presence of this information for appropriate tests may have diminished its impact when tests were inappropriate. Future efforts may consider more selective targeting of price transparency.” It might also be a good idea to price out different testing options for providers, and use actual charges and other more on-point forms of cost estimates, they said, instead of Medicare fees that have little to do with what many patients are actually charged. Targeting only the most expensive tests might also help (JAMA Intern Med. 2017 April 21. doi: 10.1001/jamainternmed.2017.1144).
The investigators also noticed a problem when labs are ordered to repeat automatically; clinicians did not see the price information every day, and so missed cost information “when it would be most salient.”
The mean age in the study was 54.7 years; 52% of the patients were white, 39% black, and 57% women. The mean length of stay was about 6 days, and over 80% of the patients were discharged home.
