Study Reveals Complexities of Disclosing and Compensating for Medical Mistakes

[Editor’s Note, I am guest posting this on behalf of my wonderful colleague Michelle Mello, at the Harvard School of Public Health]

Gridlock in many state legislatures over proposals to reform medical liability by capping noneconomic damages—and growing recognition that caps have only modest success in addressing the problems with the malpractice system—have led  health care providers and other stakeholders across the country to think hard about alternative approaches.  Alternatives that don’t require the passage of legislation are especially appealing.  Attention has focused in the last couple of years on a promising approach pioneered by a handful of hospital systems, including the University of Michigan Health System: “disclosure-and-resolution” programs, or DRPs.  In DRPs, healthcare facilities and their malpractice insurers disclose unanticipated care outcomes to patients and their families; investigate and explain what caused them; apologize; and, where appropriate, offer compensation without waiting for the patient to sue.

Early adopters of this approach report remarkable success in reducing liability costs and believe they have markedly improved patients’ experience following a medical injury.  But they can’t tell how much of the benefit is attributable to disclosing errors and apologizing, versus offering compensation.  Is it the honesty and empathy, or the money, that matters?  And if it’s the money, how much is enough to get the outcomes healthcare providers want: reduced frequency of malpractice claims, lower defense and indemnity costs, quicker disposition, improvements in staff reporting of unanticipated care outcomes, and a clinical culture that supports open communication with patients?

A new study that I published with my colleagues, Lindsey Murtagh, Penny Andrew, and Tom Gallagher, in Health Affairs this week begins to answer these questions.  We used an experimental survey design to investigate the relative effects of disclosure, explanation, and apology on the one hand, and different kinds of compensation offers on the other, on people’s responses to learning that they were the victim of a medical error.  We fielded an online survey in which 2,112 American adults randomly received one of 16 vignettes in which they were informed of a medical error.  In all vignettes, a physician and administrator explained how the error occurred, took full responsibility, and apologized.  Some vignettes also included an offer of compensation—either waiver of medical bills, limited reimbursement of out-of-pocket expenses, or full compensation—while others included no compensation offer.  Respondents answered several questions about how they would react to the disclosure.  The survey sample was drawn from KnowledgePanel, a standing, probability-based panel of U.S. adults maintained by GfK (formerly Knowledge Networks).  The survey response rate was 65%.

What did we find?

  • Two thirds of U.S. adults feel that they should be offered compensation following a medical error.
  • Patients are likely to accept even relatively modest offers of compensation, like offers to waive medical bills (70% likely to accept) or reimburse a limited amount of out-of-pocket expenses related (79% likely to accept).
  • Patients are about half as likely to accept an offer of full compensation that requires them to waive their right to sue than to accept more modest offers that don’t require them to give up any rights.
  • Increasing the amount of the compensation offer doesn’t decrease the likelihood that people will want to consult an attorney about whether to take it.  Injury severity is the biggest predictor of wanting to seek legal advice.
  • People who believe the provider is offering compensation primarily to avoid a lawsuit are half as likely to accept the offer as people who believe the provider has a less self-interested motive.  The more money is offered, the more likely people are to perceive a self-interested motive for both offering compensation and admitting the error.

Clearly, people’s responses to offers of compensation are more complex than one might guess: patients think money ought to be offered in the wake of a harmful medical error, but more isn’t necessarily better in producing the outcomes providers seek.  It’s particularly striking that disclosures and apologies that were worded identically were viewed quite differently when they were accompanied by an offer of compensation—especially a generous offer.  To prevent compensation offers from eroding trust in clinical providers, healthcare facilities and their insurers should separate disclosure-and-apology conversations from discussions about compensation.  Hospitals and their insurers, not physicians and nurses, should handle talks about money.

It’s also clear from the survey that disclosure, apology, and explanation alone aren’t sufficient to meet the needs of many patients.  It has been fashionable in some quarters to assert that if you’re simply honest with patients, and sincerely sorry, they won’t sue you.  But two thirds of people want—and in many cases, really need—money too.  Disclosures and apologies for errors should be accompanied by fair offers of compensation.

What about the finding that patients are no less likely to seek the advice of an attorney when you offer them money than when you don’t?  At first blush, this finding may be depressing to health care providers and insurers.  But consulting an attorney isn’t the same as suing, and as we discuss in the article, a good attorney can serve a helpful role in facilitating settlement.  Recognizing that many patients will feel the need to have compensation offers evaluated by an attorney—and that attorneys can help patients understand when an offer is indeed fair—insurers should develop resources to help patients connect with capable, experienced legal counsel.

Our study has some notable limitations.  How people respond to a hypothetical vignette of medical error may not correspond to how they would actually react to such a traumatic event.  We tried to design medical error scenarios that involved health conditions (abdominal pain and an infection in the knee) that could happen to anyone and did not require people to imagine themselves with a complex illness.  Still, people’s powers of prediction may be low.  Moreover, their reactions may depend on aspects of how the physician and hospital communicate with them that we couldn’t capture in our vignettes, like “emotional intelligence” and affect.  Reactions may also vary depending on the strength of the patient’s prior relationship with the physician.

To test these dynamics, one needs to study DRPs as they operate in the real world.  We’re doing that now, in two federally funded demonstration projects involving 11 health care facilities in New York and Washington state.  But an experimental study like the one in Health Affairs enables us to isolate the different effects of disclosure/apology and compensation in a way we can’t in the demonstration projects.  We’ll never be able to control all the ways that disclosures, explanations, apologies, and compensation offers differ across cases in the real world.

Figuring out how to optimally design and operate DRPs is complicated, as this study shows.  But many forward-thinking provider organizations and insurance companies are committed to the cause.  An initiative here in Massachusetts is trying the approach in 2 major hospital systems, and additional demonstration projects are underway in Texas and elsewhere.  It’ll be fascinating to see if hospitals and insurers are able to achieve what has long eluded legislators: transformative change in how patients and providers experience the aftermath of a medical error.

The study appears in the December issue of Health Affairs (  A copy of the article can be obtained by emailing

The study was conducted by Lindsey Murtagh, JD, MPH, a former HSPH Research Associate (and graduate of Harvard’s JD/MPH joint degree program) who is now an Associate at the law firm Hogan Lovells in Washington, DC; Thomas Gallagher, MD, a professor in the Departments of Medicine and Bioethics and Humanities at the University of Washington; Penny Andrew, MBChB, LLB, a former Commonwealth Fund Harkness Scholar who is currently the clinical lead for quality at Waitemata District Health Board in New Zealand; and Michelle Mello, JD, PhD, Professor of Law and Public Health in the Department of Health Policy and Management at the Harvard School of Public Health.  We were supported by grants from the Robert Wood Johnson Foundation and the Commonwealth Fund, and in-kind support from the National Science Foundation’s TESS program.