Strategic Flexibility in Health Practice GuidelinesUnder Litigation Concerns
Introduction:Recent healthcare reforms and advancements in technology have enabled widespread health information sharing, which may in return lead to better care coordination. However, widespread information sharing may also increase physicians' vulnerability to legal responsibility. In particular, the rise in malpractice litigation as a result of health information technology (IT) adoption has been widely discussed.
Recent studies have pointed out the effects of health IT on three keyareas of medical malpractice litigation. First, ubiquity of health information heightens health care providers' duties to search for patient information generated by other caregivers and creates a legal responsibility to act on that information. Second, more extensive documentation of clinical activity creates more discoverable evidence for plaintiffs in the lawsuit and the mining of the metadata from the patient records can make it easier to find evidence for physician's sub-optimal decision. Third, guidelines incorporated into decision support systems could become the standard of care that courts rely upon in litigation decision.
Care protocols such as practice guidelines (PGs), which have already played an important role in dealing with the malpractice litigation cases, became even more critical in resolving litigation cases in the post-widespread health information era.However, there is disagreement within the health care community regarding how rigid or flexible the PGs should be; while some favor rigid PGs that can serve as legal standards, others prefer flexible PGs with the freedom to adjust the care according to specific patient circumstances and to accommodate different views on care practice.
In general, flexibility in PGs refers to the case where the description of the PG does not single out a clinical action from many alternatives. Although the physician's autonomy in making decisions may cause a variation in the delivery of care, flexibility in PGs is often used in clinical practice for reasons such as insufficient medical evidence, lack of consensus within the health care community, cost, and concern over setting a legally binding standard. Furthermore, flexibility can make PGs less effective as legal evidence in holding the physician responsible for deviating from them. This is because flexible PGs do not allow evidence to be treated as completely exculpatory or inculpatory for determining the physician's responsibility for a false decision. That is, besides the inherent uncertainty in the medical evidence itself, flexibility in PGs also introduces uncertainty in the outcome of a medical lawsuit. Therefore, given the potential impacts of widespread health information generation and sharing on malpractice litigation, a fundamental question faced by policy makers is how rigid or flexible the PGs should be so that social welfare is maximized.
Methodology:To study this question, we develop a game-theoretical model of the following context. A physician conducts a medical test on a patient and recommends a follow-up based on the test results. A social planner develops the screening guidelinesregarding who should take the test based on risk indicators as well as follow-up guidelinesfor the physician's recommendation.Screening and follow-up guidelines can be formulated either as flexible or rigid. Unlike the rigid guidelines, flexible guidelines allow flexibility to the screening or physician’s recommendations and results in only a fraction of patients should be tested or followed up.We refer to the information for making a recommendation regarding whether to take the test as risk information and the information collected from the test itself as test information. Risk information could be available to the physician through a health IT while he makes his decision. For given guidelines, the physician makes a follow-up decision with available information sources to maximize the physician’s own payoff. When litigation is a concern, the physician may bear additional cost that is a function of guidelines if a false-negative decision is made.
Results:We show that, when litigation is a concern for the physician, while rigid follow-up guidelines are optimal, rigid screening guidelines are not always optimal. In particular, flexible screening guidelines are optimal when the benefits from and costs of testing are moderate.On the other hand, if litigation is not a concern or the patient's risk information is not shared with the physician, then it is optimal for the social planner to set both screening and the follow-up guidelines rigid. We also extend our model by introducing malpractice damage caps as an instrument to deal with defensive medicine, and find the main results still hold.
Conclusion:This study is the first attempt to analyze the new risk of health IT by using an economic model that captures the underlying proliferation of defensive medicine subsequent to wider health information sharing. Although guidelines are already admissible in many liability cases, the question regarding whether the legal standards of care should be conclusive or not is still unanswered. Our findings provide new evidence on the potential harm of rigid guidelines that provides too much safe harborthat itmay potentially do more harm than benefit to the society in the era of widely shared health information, and the conduct of a cost-benefit analysis is a prerequisite to the implementation.Moreover, while data sharing is being touted to improve health systems, there is a critical need to examine the litigation concerns raised by such data sharing.
Selected References
[1]Green, L. V. 2012. OM forum - The vital role of operations analysis in improving healthcare delivery. Manufacturing & Service Operations Management 14(4) 488-494.
[2]Mangalmurti, S. S., L. Murtagh, M. M. Mello. 2010. Medical malpractice liability in the age of electronic health records. New England Journal of Medicine, 363(21) 2060-2067.
[3]Shumsky, R. A., E. J. Pinker. 2003. Gatekeepers and referrals in services. Management Science 49(7) 839-856.
1