Legal and Policy Interventions to Improve Patient Safety
- government regulation
- liability, legal
- patient safety
- quality improvement
- reimbursement, incentive
- self regulation, professional
Twenty years ago, few systematic efforts to make health care safer existed. Today, patient safety is a priority for patients, providers, payers, and policy makers. The Institute of Medicine’s 1999 report on medical error prompted a flurry of activity, including widespread adoption of error detection and reporting programs, movement toward systems approaches for addressing error, development of new clinical interventions to reduce error, and efforts to foster stronger safety cultures within healthcare organizations.1 Although there are some indications that these activities have yielded benefits, it is also evident that much of health care, including cardiovascular medicine, remains too unsafe.2,3
Mirroring the heterogeneity and complexity of errors in health care, the myriad initiatives underway to improve safety are multifaceted and wide ranging, which makes it difficult to overview them. However, many of the initiatives that involve legal and policy-based interventions may be characterized according to their alignment with 1 of the following 4 general approaches: greater transparency; financial incentives; new forms of regulation emanating from both inside and outside the provider community; and reform of the liability system.
As patient safety improvement initiatives grow in scope and impact, understanding their precise role, effects, and limitations will become increasingly important for physicians and hospitals. Cardiologists, cardiac surgeons, and health services and policy researchers working on cardiovascular disease are especially likely to encounter these initiatives. Cardiac conditions are highly visible, frequently associated with substantial morbidity and mortality, and expensive, which helps to explain why cardiovascular care is often targeted for improvement efforts. In this article, we review the leading approaches to patient safety improvement, addressing both their rationale and available evidence of their effectiveness. We conclude with discussion of some implications for clinical practice.
Health care is an opaque enterprise. Its machinery is understood best and controlled largely by those who deliver it. Today, however, expectations are growing for providers to be more transparent with patients and the public in terms of how treatment decisions are made, how much care costs, whether conflicts of interest exist, and when and why errors and unexpected outcomes of care occur.4,5 These expectations spring partly from the trustful nature of the therapeutic relationship, but they are also fueled by hopes that that such transparency will help boost the quality and efficiency of care. Movement toward openness and candor is particularly evident in 2 areas: public reporting of data and disclosure of medical errors (Table 1).
Public Reporting of Performance Data
Public reporting of data on the quality and safety performance of institutions has expanded through a combination of external rules and voluntary initiatives from within healthcare organizations.6,7 The Center for Medicare and Medicaid Services (CMS) has, in many respects, taken the lead by leveraging its purchasing power to impose public reporting mandates. A number of influential nongovernmental organizations such as US News and World Report and the Leapfrog Group have also stepped up efforts to display, rank, or otherwise recognize healthcare organizations on the basis of performance on various metrics.8,9 The scope and sophistication of published metrics have expanded dramatically over the last decade. Quality indicators related to cardiovascular conditions at both the institution and physician levels were among the first metrics ever published.7,10,11 Today, indicators have grown to cover many conditions, including heart failure, myocardial infarction, and coronary artery bypass grafting, and often form part of publicly accessible metrics and ranking systems.9,12
A core rationale for public reporting is that it allows patients to “vote with their feet,” which in turn may press providers to compete on quality. It is well recognized, however, that various factors disrupt this aspiration. Health insurance plans often limit patients’ choices about where to obtain care. Clinical circumstances also act as a constraint. Patients in need of emergent or urgent care for conditions like chest pain are unlikely to compare performance metrics before seeking medical attention. Even for elective care, available evidence suggests that publicly reported quality data have minimal effect on the choices of patients or referring providers.13–15
However, shaping consumer choices about where to obtain care is not the only way through which publicly reported performance data may drive quality improvement. The data may motivate hospitals to pursue patient safety improvement so that they may reap the reputational benefits associated with being perceived as a market leader. This is evident in voluntary moves by healthcare organizations to publish their performance scores. Many large hospitals now report clinical outcomes for a wide range of procedures and treatments, as well as safety culture scores, patient experience scores, and patient comments.16–18 Cardiovascular conditions often feature in these reports. From a public-relations standpoint, this strategy signals a commitment to quality, continuous improvement, and transparency.
The pivotal question, of course, is not how or why public reporting may improve quality and safety but whether it actually does. Hard evidence of its effects on quality and safety remains elusive.19,20 Some studies have identified improvements in performance—as measured, for example, by National Hospital Quality Measures,21,22 cardiac surgery mortality,23 and readmissions24—after public posting of related measures. One of the most widely cited examples is the substantial improvement in door-to-balloon times seen nationally after CMS began public reporting them.25 Considered as a whole, however, the evidence demonstrating a causal relationship between public reporting and performance improvement remains thin.
Efforts to implement robust public reporting programs face a number of practical challenges, beginning with the identification of suitable measures (Table 2). To be effective, the measures must both matter and be easily comprehensible to the general public. The measures must also be technically robust, providing valid and reliable indicators of what they purport to indicate. Faulty measures risk generating the wrong incentives. Consider, for example, a recently retired CMS measure that called for preantibiotic blood cultures for patients with pneumonia. Although the measure was well intentioned, critics questioned whether it was an appropriate indicator of high-quality care26,27 and raised concerns about the unintended consequences of incentivizing unnecessary blood cultures in patients with respiratory illnesses.28,29
Another substantial challenge for public reporting is the proliferation of measures. An overabundance of measures creates confusion and dilutes the usefulness of the most informative ones. Confronted with a battery of metrics and no guidance about which are most important, providers and patients may be none the wiser about which institutions provide the best care.30 To address these concerns, measures are sometimes retired once performance “tops off” (ie, when most providers are scoring near or at 100%), on the basis that the value in ongoing measurement will be limited.31,32 For example, CMS has started retiring a number of heart failure and heart attack measures such as the prescribing of β-blockers at discharge for patients with a myocardial infarction.33 Ongoing scrutiny and willingness to retire measures help guard against the problems created by accretion, but finding the “sweet spot” for the number and type of measures and ensuring an optimal set is in place at any given point in time remain ongoing challenges.
A potential drawback to the implementation of public reporting is that it may generate disincentives for physicians and hospitals to care for the sickest patients. This issue was particularly highlighted when New York started posting coronary artery bypass graft surgery outcomes. Although the evidence has been mixed, critics have been concerned that the move may have resulted in lower use of coronary artery bypass graft surgery in patients presenting emergently or with high-risk characteristics.34,35 Similar worries have been noted with percutaneous coronary intervention in some states.36,37
Transparency Around Errors
Disclosure of medical errors to patients is another form of transparency that has captured significant interest. There is now wide agreement that providers have an ethical obligation to communicate the occurrence of harmful errors and unanticipated outcomes in care to patients who experience them and to convey what is known about why the outcome happened, including any mistakes made.38,39 Survey research suggests that patients and their families strongly desire this information.40,41 Safety experts regard disclosure programs as an essential part of institutional efforts to improve patient safety.42
The tide has clearly turned against the traditional risk-management strategy of saying as little as possible when an adverse outcome occurs. Over the last 15 years, many organizations have implemented programs designed to promote greater openness after errors and adverse events. Nonetheless, the plain fact is that many errors remain undisclosed today.38
Two elements of disclosure practice have proven particularly controversial. First, should a typical disclosure include an apology, and if so, how should that apology be phrased? Second, in circumstances in which patient harm occurred and was clearly attributable to error, should an offer of compensation routinely follow the disclosure of the error?
Several pioneering programs—initially called disclosure-and-offer programs and today generally referred to as communication-and-resolution programs (CRPs)—have embraced all of these elements and more.43 These programs promise open discussion with injured patients about what occurred in which an appropriate apology and an explanation of what is known about the cause are communicated, regardless of whether the injury is determined to have been caused by error. When the hospital (in collaboration with its insurer) determines the standard of care was not met, the programs proactively offer fair compensation without waiting for the patient to file a malpractice claim. When the care was reasonable, the institution affirms to the patient and family its commitment to stand behind and to defend the involved providers should a malpractice claim be filed.
Early data on CRPs suggest that they can help improve liability outcomes.44,45 For example, data from the University of Michigan Health System demonstrated a 36% drop in claims for compensation and a 59% drop in legal costs.29 Other institutions such as Stanford University Medical Center have also reported liability cost savings.46
It is not yet clear whether the positive liability experiences of these early adopters will be replicable elsewhere, nor is it clear the extent to which the benefits of CRPs extend beyond litigation efficiencies to include real improvements in patient safety. Theory and anecdotal evidence suggest that they do, but a causal connection with safety outcomes at the institutional level is difficult to establish.
Nevertheless, by facilitating open discussion of adverse events, CRPs may boost institutional awareness of errors and foster a culture that is more attuned and open to safety and learning. In addition, best practice demands an explanation to patients of what will be done to prevent similar events in the future. In organizations that conduct rigorous investigations, propose meaningful reforms, and keep their word, substantial safety improvements may follow. Identifying whether CRPs actually realize such patient safety benefits will not be easy, but as the programs grow, so does the imperative to investigate and answer this question.
New Reimbursement Strategies
Fee-for-service payment models incentivize physicians and hospitals to provide more care. These financial incentives join other currents, including liability pressure and physicians’ natural desire to avoid clinical “misses,” to promote overuse. Overuse not only involves costly waste but also places patients at risk of complications from unnecessary tests and procedures. Perversely, fee-for-service reimbursement may even reward some iatrogenic complications by paying providers for care and readmissions needed to treat them.
There appears to be little appetite for wholesale abandonment of fee-for-service reimbursement or for efforts to reduce costs that involve explicit rationing. Growing concerns about quality and unnecessary expenditures have instead led payers and policy makers to search for incentive-based solutions that plausibly could improve quality while simultaneously achieving cost control. Two of the methods promoted by the Affordable Care Act, direct performance-based payment incentives (either via pay for performance or financial penalties) and accountable care organizations (ACOs), are the main financing strategies for the foreseeable future.
Performance-Based Payment Incentives
Pay-for-performance programs have assumed increasing prominence in health care over the past 15 years. Initially, CMS imposed rather modest conditions on reimbursement by requiring hospitals to report their performance on quality measures that would then be publicly displayed (ie, pay for reporting). CMS then moved to pay for performance in which payments are based on a hospital’s actual performance on a broad set of process, infrastructure, and outcomes measures, including measures of clinical quality, cost, and patient experience. The structure of incentives varies broadly and includes policies such as “no-pay” rules for preventable complications acquired during a hospital stay (eg, a patient who develops a central line–associated bloodstream infection), payments based on performance on evidence-based process measures (eg, the percentage of patients experiencing a heart attack who receive aspirin on arrival or reperfusion with angioplasty within 90 minutes), payments linked to infrastructure measures (eg, implementation of an electronic health record), and payments based on clinical outcomes (eg, mortality rates of patients admitted for heart failure).
In October 2008, CMS implemented its Hospital Acquired Conditions Program and started adjusting hospital reimbursements on a case-by-case basis by removing certain injuries and infections it deemed to be highly preventable from inclusion in the diagnosis-related group algorithm that stratifies discharges by severity and is used to determine payment.47 This meant, for example, that a patient admitted for heart failure who developed a catheter-associated line infection during that hospital stay would now have his or her episode of care coded in a manner that excludes the line infection from the payment algorithm.
Although this program generated a fair amount of attention, the financial impact is believed to have been relatively modest: <1% of a hospital’s Medicare revenue for the average hospital.48 Researchers have surmised that hospital billing departments may have adopted compensating strategies to reduce the fiscal impact of the program, for example, by improving identification of infections that are present on admission rather than hospital acquired or by improving identification of comorbidities that would increase case severity and reimbursement.48,49
Findings on the clinical effects of the program have been somewhat mixed, but most have indicated no significant reductions in the incidence of the no-pay conditions.48,50,51 One recent study identified larger effects for infections than patient falls or pressure ulcers.51 If the amount of revenue at stake is as small as reports have suggested, these results, although disappointing, are not surprising.
Following key provisions of the Affordable Care Act, CMS has now implemented 3 additional value-based payment programs that form a larger performance-based payment system: Value-Based Purchasing, the Hospital-Acquired Condition Reduction Program, and the Hospital Readmissions Reduction Program.52–54 Through Value-Based Purchasing, CMS now scores and ranks all hospitals (without regard to whether they are academic or community hospitals) on a range of process measures, outcome measures (including safety measures), cost (or efficiency) measures, and patient experience metrics. On the basis of a hospital’s score, in fiscal year 2015, a hospital would either earn or lose payments up to 1.5% of its total Medicare revenue, with the incentive projected to climb to 2% in fiscal year 2017. The program scores hospitals on their level of performance in each domain and gives points for improvement. This approach is intended to reward low performers for incremental improvement, but concerns remain that the Medicare programs disadvantage underresourced hospitals and those that serve patients with low socioeconomic status.55,56
For hospital-acquired conditions, CMS ranks hospitals on the basis of rates of specified complications and will penalize the lowest quartile of hospitals 1% of their Medicare payments in fiscal year 2015. Through the Readmission Reduction Program, CMS will assess penalties of up to 3% of all Medicare payments for hospitals that have higher-than-predicted risk-adjusted readmission rates for patients hospitalized for pneumonia, heart failure, myocardial infarction, chronic obstructive pulmonary disease, and total knee or hip arthroplasty.
Many of the measures used in these programs have come under fire. Some critics reject outright the premise of using safety metrics as the basis for financial incentives because the metrics were not developed with that use in mind. For example, PSI 90 is a composite patient safety indicator (PSI) measure that was developed by the Agency for Healthcare Research and Quality to help identify adverse events from administrative data.57 PSI 90 has been criticized for targeting the wrong clinical areas, having flawed component measures, identifying adverse events inaccurately, and risk adjusting poorly; the National Quality Forum has not endorsed it.57 However, a hospital’s PSI 90 performance significantly affects its Value-Based Purchasing and Hospital-Acquired Conditions scores, comprising 35% of the total score in the latter program, for example.57
Another metric used in the Hospital Acquired Condition Reduction program, PSI 12, evaluates facilities on the basis of the occurrence of postoperative venous thromboembolism. This metric has been shown to return higher rates of venous thromboembolism events in institutions that have higher rates of testing for venous thromboembolism, raising concerns that this metric is not a good proxy for safety because it suffers from surveillance bias.58 Mortality metrics are criticized for inadequate risk adjustment and failing to measure only preventable deaths.59 Critics also have taken aim at readmission measures. Objections include holding hospitals responsible for care administered beyond their walls and control, counting all readmissions as unwanted events even though some may reflect better care, and the recurrent concern about the adequacy of risk adjustment.60–62
How well these CMS programs work to improve patient safety in the long run remains unclear. Recent data on pay-for-performance programs in general have shown mixed results,22,63–65 and 1 analysis found no early benefits from the Medicare Value-Based Purchasing program.66 However, the federal government professes confidence in the approach and is quick to highlight even modest evidence of such positive effects.24,67,68 As with public reporting, policymakers are not waiting around for rigorous causal studies of the effects of pay-for-performance programs. Belief in the idea is propelling action, and CMS has clearly indicated that more value-based programs are on the way.69
Performance measurement and payment programs for physician care lag considerably behind hospital value-based purchasing programs. The recent elimination of the Medicare sustainable growth rate, however, gave way to a new merit-based payment system for physicians. This may present opportunities for much-needed alignment between hospitals and the physicians who work in them and who care for patients after discharge.
Accountable Care Organizations
ACO models in Medicare are intended to improve quality and to decrease costs by incentivizing groups of providers to deliver coordinated care and holding them jointly accountable for clinical outcomes and total costs of care.70 State and private payers are implementing similar models that go by various names. Payment incentives are typically structured to allow providers to share in cost savings realized by the payer, so long as prespecified quality and safety thresholds are met.
ACO models are in their infancy. The 2 critical questions for their future are whether the structural reorganization, contract signing, and relationship building among providers will translate into lower costs and higher-quality patient care. One early study suggests that ACOs produce modest improvements in spending (with greater reductions coming from organizations with higher baseline levels of spending) and either unchanged or improved performance on quality measures.71 Another recent analysis revealed that ACO implementation did not affect spending on discretionary or nondiscretionary imaging and procedures for cardiovascular conditions.72 Outcomes such as performance on heart failure admissions and blood pressure control will continue to feature prominently in such evaluations; almost half of CMS’s quality measures for ACOs relate to cardiovascular care.73 More data and time are needed to properly determine whether this model will be successful in improving safety, including whether the shared savings incentives could have unintended consequences such as underuse that threaten patient safety.
The medical profession has a long history of autonomy and self-regulation, traditionally justified by the specialized nature of medical practice and a strong code of professional ethics.74–77 Self-regulation continues to play a major role in oversight of the profession. However, it coexists with external regulatory controls such as licensure rules, reporting requirements, and accreditation. Use of both forms of regulation to drive improvements in patient safety is growing.78–80
Medical professionalism requires physicians to maintain their clinical knowledge and skills and to help ensure that colleagues do the same.74–76 The rapid pace of change in clinical and scientific knowledge has led medical boards to shift from basic one-time certification requirements to maintenance of certification (MOC) requirements that seek to ensure provider competence throughout a physician’s practice career.78,81 MOC requirements often include periodic examinations and other practice assessments and participation in quality-improvement projects.
The introduction of MOC requirements has been controversial throughout medicine, and cardiovascular medicine is no exception. Many commentators question the value of these requirements.82–85 The adoption of rules by subspecialties has been especially controversial because many physicians are saddled with multiple, sometimes overlapping recertification requirements.82–85 For example, a recent web-based petition opposing the MOC requirements circulated by leaders in interventional cardiology attracted >19 000 physician signatures; about one third of the signatories pledged noncompliance.83 The American Board of Internal Medicine recently revised portions of its MOC requirements, conceding that the organization “clearly got it wrong” in producing requirements that were not sufficiently meaningful for physicians.84 The revisions include simplifications of the certification process for subspecialists such as interventional cardiologists.
Fierce debate over the merits of MOC programs looks set to continue. Many physicians remain skeptical of the value of these requirements and troubled by the burden they impose.86 Others see these programs as an important element of the professional duty to maintain competence.78 Still others fear that failure to self-regulate in this way invites more burdensome external regulation.78 Which view will prevail is unclear at this point.87 What is apparent, however, is that sensible debate will be difficult as long as a dearth of evidence regarding the effect of MOC programs on physician performance remains.
Professionalism also dictates that physicians review harmful errors to understand why they occurred, and then use this knowledge to guide prevention efforts. Peer review of adverse events is often a first step in this process. To encourage clinicians to participate in such investigations and to allay apprehension about the liability implications of candor about what happened, the federal Health Care Quality Improvement Act of 1986 provides liability (but not evidentiary) protection for statements made as part of a peer-review process.88 This means that physicians cannot ordinarily be held liable for statements made during peer review but that the statements may nonetheless be used as evidence in court proceedings. States have enacted their own layer of peer-review protection, and many of these laws go further, protecting statements and documents generated during peer review from any use in litigation.89 Court decisions in a few states have subsequently weakened that additional layer of state peer-review protections.89
Troubled by this trend and eager to encourage wider generation and use of safety-related information within and among healthcare organizations, the federal government passed the Patient Safety and Quality Improvement Act of 2005.90 This law allows providers to create patient safety organizations (PSOs), entities that facilitate open, confidential discussion of adverse events. Adoption of PSOs has been somewhat slow, probably in part because of the slew of establishment requirements.89 It is too soon to tell what advantages PSO status confers; the effects of PSO status in stimulating patient safety activities and reducing error have not been closely studied. Nevertheless, the legislation represents a welcome step by the federal government to help physicians fulfill their professional obligations to investigate and prevent error.
Perhaps the most visible external regulator physicians face is the state medical board (although, as many commentators have pointed out, these agencies also exhibit many of the hallmarks of self-regulation).91 Boards control licensure and police physician misconduct, with the stated goal of protecting the public. They have heavy sanctions at their disposal, up to and including suspension or revocation of a physician’s license to practice medicine.
Historically, state medical boards focused on forms of misconduct that are both egregious and relatively easy to monitor and prove such as substance abuse and sexual indiscretions, and they did relatively little to proactively monitor and assess physician competence.77,79,92 With growing awareness of the outsized impact—on both patient welfare and trust in the profession—of a small minority of practitioners who are incompetent, there have been increasing calls for boards to become more sophisticated in discharging their regulatory obligations to protect the safety of patients.81,93,94
Exactly what the medical board of the future will look like and do differently is not yet clear. Options under consideration include enhanced licensure requirements and more thoroughgoing and targeted enforcement. Needless to say, many physicians have already signaled their discomfort with the prospect of additional regulatory burdens and oversight.81,95
In addition to their institutional and professional licensing functions, states have leveraged their authority to require institutional reporting of adverse events. Reporting requirements vary greatly from state to state, as do the follow-up actions that a report triggers.96 Through pattern analysis, some state agencies have generated safety data and disseminated alerts and recommendations, but there is little systematic evidence that these activities improve safety.
Finally, institutional accreditation is seen as providing an additional layer of accountability that can improve the safety of care. Technically, participation in accreditation programs is voluntary and a product of market forces, but US hospitals may see it as effectively mandatory. The Joint Commission, the best-known private accreditation organization, is particularly influential. Many states have incorporated The Joint Commission’s standards into their hospital licensing requirements; indeed, some even accept Joint Commission accreditation in lieu of separate state requirements. At the federal level, participation in the Medicare program requires either periodic review by CMS or by an organization that CMS has deemed to have the authority to review, and The Joint Commission has this devolved authority.97
The Joint Commission has been seen as a national leader in the development of patient safety standards.98 To obtain accreditation, organizations must meet The Joint Commission’s National Patient Safety Goals, a program established in 2002 to help accredited hospitals address specific areas of threat to patient safety (eg, wrong site procedures, alarm fatigue, safe use of anticoagulants).99 Intuitively, accreditation is a necessary and important step toward safety improvement. However, there is surprisingly little evidence demonstrating its effects on patient outcomes.100,101
Regulation from inside and outside the profession is almost certain to form a key plank in future efforts to improve safety. Many physicians will chafe at greater oversight and continue to view many forms of regulation as costs that lack proven benefit.79,95 However, more regulation is inevitable. The public will demand it. The largest expansions may be in self-regulation as organized medicine seeks to respond to the concerns of lawmakers and the general public in ways that maintain professional control and preempt more stringent rules from government.
The medical liability system is designed not only to compensate injured patients but also to deter providers from practicing unreasonably unsafe care.102 The economic sanctions imposed by the tort system when practitioners and healthcare institutions are held legally responsible for negligent injuries have long been seen as a necessary backstop for ensuring accountability.
However, the performance of the medical liability system pleases no one, whether the metric is efficient, accurate delivery of compensation or effective deterrence.102 Among the chorus of criticisms is that the system is imprecise: it frequently fails to direct compensation to patients with meritorious claims while at the same time permitting many nonmeritorious lawsuits to be filed and sometimes result in a payment.102,103 Cardiac surgeons are among the groups of physicians most frequently sued, and cardiovascular physicians have been vocal critics of the liability system.104–107
A deeper concern relates to the lack of evidence that the system improves the safety of care. Key contributors to this problem are the relative rarity of claims compared with the incidence of negligent injury, the imprecision quandary, and the prevalence of liability insurance coverage with little or no experience rating (meaning that an individual practitioner’s insurance premiums do not go up very much after a claim).103,108,109 Adding to safety concerns is a well-known, unintended consequence of malpractice liability: defensive medicine. Although difficult to quantify, defensive practices are believed to be widespread, and they place patients at risk for complications from the unnecessary testing and treatment.110 Cardiology has been in the bull’s-eye of studies documenting defensive medicine.111
Over the last 40 years, the majority of the enacted tort reforms—such as caps on damages and pretrial screening panels—have been designed primarily to address the fears of healthcare providers and their insurers about frivolous lawsuits and high, unpredictable damages awards.112,113 Evaluations of the effects of these reforms have not demonstrated that they have improved the safety of care delivered.112,113 Indeed, they were not designed to do so. Economic theory would suggest, to the contrary, that reduced liability might reduce incentives for safety. However, advocates hoped that lower liability pressure might at least ease defensive medicine. Even on that front, however, effects have been modest or nonexistent, depending on the reform.
Such findings make it more difficult to make the political case for traditional tort reform. Because they do not appear to bolster safety, traditional reforms offer little in the way of a quid pro quo for patients. Consequently, in recent years, the policy focus has been on identifying liability reforms that plausibly could also improve patient safety.114
A great deal of attention has focused on the potential for CRPs and the policy strategies that lawmakers could pursue to encourage their growth.115 Those strategies include passage of apology laws that protect expressions of empathy or admissions of fault from being introduced into evidence at trial. Such protections may encourage greater provider dialog with patients right after an adverse event occurs.116 Some states have also enacted presuit notification laws that require a patient to give a provider advance notice of the intent to sue.116 After notice is served, a waiting period for the patient to file suit follows, giving the provider a chance to resolve the matter and to avert the lawsuit.
Three other liability reform proposals that have been prominent on the policy scene are enterprise liability, safe harbors for evidence-based practice guidelines, and health courts or administrative compensation systems.116,117 These reforms hold theoretical promise for improving patient safety, but they have not been tested in situ in the United States except for a few, narrow policy experiments.
Enterprise liability involves changing the rules of liability so that when a patient wishes to seek compensation, rather than filing a claim against a clinician, the patient would pursue a claim against the clinician’s institution. The idea is to hold the institution, or enterprise, that is benefiting from the clinician’s actions liable for the clinician’s errors, as well as any broader systems failures. This includes both human errors (eg, misplacement of a left ventricular pacemaker lead) and system factors that are more under the control of the enterprise than the clinician (eg, understaffing). Focusing liability on enterprises targets the accountability signal to parties with the resources and organization to build safer systems of care. Because the enterprise will receive a greater number of claims over time than any one of its practitioners, the signal will also be stronger than would be the case in an individual liability system.
Enterprise liability holds potential to improve safety by effecting cultural shifts. In reducing blaming of individual clinicians, it can help build a “Just Culture” in which individuals do not feel singled out for inadvertent errors that could happen to other reasonable providers and could have been prevented by better systems of care.4 Reducing blame and stigmatization may stimulate providers to be more open about errors and curb defensive practices. Enterprise liability also permits better apportioning of liability costs through experience rating.109 It is difficult for insurers to effectively calibrate premiums to claims risk for individual practitioners’ experience because incurring a claim is, for the vast majority of physicians, a rare event. That problem is less likely to occur at the institutional level.
Benefits notwithstanding, a shift to enterprise liability raises countervailing safety concerns because it removes the physician accountability that arises from the threat of being personally sued. Proponents of the reform argue that this potential decrement in individual deterrence will be outweighed by the improved incentives for safety at the institutional level. How well institutions would actually act on that signal (including holding providers accountable when necessary) remains a matter of debate.
Safe harbor legislation, a second avenue of liability reform that has patient safety in its sights, aims to assuage physicians’ fears about being sued when they have done everything right but nonetheless experience an adverse outcome.116,118 States can enact legislation providing a strong defense to a malpractice claim in cases where the provider can show that a clinically relevant, state-sanctioned practice guideline was appropriately followed. By offering physicians an incentive to follow evidence-based guidelines, safe harbors theoretically should improve quality and safety. For example, safe harbors could provide liability protection to a physician who, in accordance with guidelines, does not order stress testing for a low-risk patient presenting to the emergency department with chest pain.
Empirical data on how well safe harbors perform on safety improvement are limited because there has been minimal experimentation with the approach.118 One recent simulation-based analysis using a large sample of closed claims from Oregon found that safe harbors would not have provided much legal benefit for physicians because physicians already tend to prevail in claims concerning incidents in which they followed practice guidelines, even absent a formal safe harbor.118 However, the study found that safe harbors have the potential to induce physicians to provide safer care by adhering to guidelines. In at least 5% of the closed claims, reviewers judged that the injury would have been avoided had the physician followed an applicable guideline.
Implementation of safe harbor rules is not without a number of practical challenges. One obstacle is finding a sufficient number of acceptable guidelines, given that most care delivered is without an evidence base. Disputes over who chooses the approved guidelines and whether the guidelines apply in a case also appear inevitable. Finally, the administrative costs of guideline selection and updating may prove prohibitive.
A third model for reform is to replace judicial resolution of malpractice claims with an administrative compensation system, also known as a health court (a somewhat confusing moniker because the model does not involve a court).114,117 In an administrative compensation system, claims are filed with and resolved by an administrative body, with judicial courts involved only if one of the parties chooses to appeal the decision of the administrative body. The administrative body (for example, a government agency) uses neutral experts, evidence-based decision guidelines, and precedent from similar past cases to make determinations about the validity of claims. In most proposals, the health court applies a compensation standard other than negligence such as whether the injury was preventable in an optimal system of care. This can expand access to compensation while removing the stigma and adversarialism associated with negligence judgments. To control costs, administrative compensation models limit noneconomic damages according a schedule based on injury severity. They also save on overhead costs when compared with tort litigation by sparing litigants the need to conduct costly battles with hired experts.
Proponents of administrative compensation systems point to potential design features that would promote patient safety improvement.114 Electronically capturing detailed information about errors and contributing factors in a centralized database could prove extremely valuable in powering analyses of where and why errors occur across our healthcare system. Currently, such information resides in hundreds of small, proprietary databases maintained by insurance companies and self-insured health systems. These data are rarely pooled or publicly shared. Administrative systems could also promote safety by contributing to a Just Culture. Removing the stigma associated with negligence investigations and findings could make providers more comfortable discussing and admitting errors, in turn supporting learning and prevention.
Administrative compensation systems have been in operation in New Zealand and Scandinavia for decades.117 They enjoy widespread support among physicians and, if the low rate of appeal is any indication, among claimants. Their overhead cost rates are less than a third of the costs in the US system. Their effects on patient safety have not been systematically measured, but evidence suggests that system administrators make good use of their databases to identify safety problems and that they disseminate lessons learned nationally.114 Politically, however, the prospects for adoption of administrative compensation in the United States (beyond the context of state-based no-fault systems for severe neurological injuries to newborns) appear poor.
What Does This Mean for Clinicians?
The imperative to reduce the considerable burden of harm caused by medical errors is likely to remain high on the policy agenda. There are no signs that payers, policymakers, or the public will let up in this area. For clinicians, this means a continued push from within organized medicine for credible self-regulation of quality and safety and continued calls from outside the profession for sterner external oversight. Formal regulation will be bolstered by market forces and financial incentives that reward safety.
What is most striking in reviewing the approaches taken to date is how little we know about their effectiveness. For most legal and policy levers in the safety domain, there is little or no solid evidence of their efficacy in improving patient outcomes (and often, the limited evidence that is available is rather discouraging). This can be explained in part by the short experience of most institutions with these initiatives and in part by the methodological difficulty of linking upstream interventions, such as changes in culture and incentives, to patient-level outcomes.
As long as the imperative to “do something” remains, the strategy of “act now and evaluate later” is probably reasonable for interventions with a plausible mechanism for improving safety.119 However, this is not a viable long-term strategy.120 As patient safety interventions proliferate and accumulate, priorities must be set, and the only sensible way to do that is by reference to what works best in attaining the desired outcomes. It is also important to understand how much safety we are buying for the dollar because these interventions are not cost free. The imperative to understand which safety improvement approaches work and do not is especially pressing in cardiology and cardiac surgery. These specialties deal with prevalent, high-cost conditions, and life or death can hinge on the quality of treatment delivered.
One clear insight gained in patient safety circles over the last 15 years is that silver bullets—interventions that reduce a wide range of errors across multiple clinical areas—are elusive. The multifaceted causes of these events eliminate that possibility. Rather, patient safety initiatives of the future will involve an ensemble of strategies arrayed across different clinical areas.
It is also important to recognize that success is rarely “baked” into the intervention itself. How it is implemented and the environment in which it is thrust matter enormously and independently determine effectiveness. For example, moves to promote transparency will not take root in a hospital culture that is not attuned and committed to change. Public reporting may help spur performance improvement by leveraging providers’ desire to be the best, but its effectiveness may be limited if there are too many measures or if the medical community perceives the measures to be invalid. Newer methods of professional and external regulation have potential to improve physician knowledge, skills, and accountability, but none of this is possible if they come with an administrative burden that alienates physicians. Performance incentives could promote patient safety but will not if the measures that are rolled out lack validity. Finally, nontraditional tort reforms may reduce long-standing tensions between the liability system and safety improvement efforts, but they will probably flounder if they lack broad stakeholder support, including champions in the highest levels of leadership in healthcare organizations.
What can physicians, particularly those who are not patient safety experts or leaders of error reduction initiatives, do to help ensure that the efforts to improve safety are both palatable and successful? Continuing the pursuit of a culture of safety, especially in terms of commitment to openness and improvement after medical errors, is critical. In the absence of such a culture, sustainable improvement may not be possible. Supporting and leading the development of better safety measures can help boost the efficacy of public reporting and make financial incentives more rational.
Further introspection within the medical profession about how to make the case that it can effectively police its own safety problems is also needed. This is more than a public-relations task; it requires real improvement in accountability mechanisms and the will to apply them consistently and fairly. Openness to more robust self-regulation could help forestall forms of external regulation that will be less nuanced and more burdensome. Taking the lead may be especially important in specialties such as cardiovascular medicine and surgery where rapid scientific and technological advances are occurring, because insiders will understand these changes and their potential effects on patients much better than outsiders. However, the most impactful legal and policy reforms are likely to be motivated by concerns for patient well-being and a shared vision of accountability for safe care rather than by self-interest and a desire to maintain control.
We thank Meredith Rosenthal for her thoughtful comments and suggestions on this manuscript.
Dr Kachalia received an honorarium for a presentation on liability reform at the Medical Mutual Insurance Co of Maine. Dr Nallamothu serves on the UnitedHealthcare Cardiac Scientific Advisory Board. The other authors report no disclosures.
- © 2016 American Heart Association, Inc.
- Kohn LT,
- Corrigan JM,
- Donaldson MS
- Levinson D
- 8.↵US News and World Report. U.S. News best hospitals 2014–15. http://health.usnews.com/best-hospitals/rankings. Accessed April 4, 2015.
- 9.↵Leapfrog Group. Top hospitals. http://www.leapfroggroup.org/Hospitals/TopHospitals. Accessed April 4, 2015.
- 12.↵US News and World Report. Which hospitals are best for patients who need common care. http://health.usnews.com/health-news/articles/2015/05/20/which-hospitals-are-best-for-patients-who-need-common-care. Accessed August 24, 2015.
- 16.↵Cleveland Clinic. Treatment outcomes. http://my.clevelandclinic.org/about-cleveland-clinic/quality-patient-safety/treatment-outcomes. Accessed June 5, 2015.
- 17.↵University of Michigan Health System. Patient safety culture. http://www.uofmhealth.org/quality-safety/patient-safety-culture. Accessed June 5, 2015.
- 18.↵University of Utah Health Care. Find a doctor. http://healthcare.utah.edu/fad/. Accessed June 5, 2015.
- Ketelaar NA,
- Faber MJ,
- Flottorp S,
- Rygh LH,
- Deane KH,
- Eccles MP
- Chassin MR
- 24.↵Statement of Jonathan Blum, acting principal deputy administrator and director, Center for Medicare, Centers for Medicare & Medicaid Services on delivery system reform: Progress report from CMS before the U.S. Senate Finance Committee. http://www.finance.senate.gov/imo/media/doc/CMSDeliverySystemReformTestimony2.28.13%28J.Blum%29.pdf. Accessed May 2, 2015.
- Krumholz HM,
- Herrin J,
- Miller LE,
- Drye EE,
- Ling SM,
- Han LF,
- Rapp MT,
- Bradley EH,
- Nallamothu BK,
- Nsa W,
- Bratzler DW,
- Curtis JP
- Austin JM,
- Jha AK,
- Romano PS,
- Singer SJ,
- Vogus TJ,
- Wachter RM,
- Pronovost PJ
- 32.↵Centers for Medicare & Medicaid Services. CMS to improve quality of care during hospital inpatient stays. http://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2014-Fact-sheets-items/2014-08-04-2.html. Accessed May 2, 2015.
- Omoigui NA,
- Miller DP,
- Brown KJ,
- Annan K,
- Cosgrove D 3rd.,
- Lytle B,
- Loop F,
- Topol EJ
- Mello MM,
- Boothman RC,
- McDonald T,
- Driver J,
- Lembitz A,
- Bouwmeester D,
- Dunlap B,
- Gallagher T
- Driver J
- 47.↵Centers for Medicare & Medicaid Services. Hospital-acquired conditions. http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/HospitalAcqCond/Hospital-Acquired_Conditions.html. Accessed May 3, 2015.
- 52.↵Centers for Medicare & Medicaid Services. Hospital-Acquired Condition (HAC) Reduction Program. http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/HAC-Reduction-Program.html. Accessed May 3, 2015.
- 53.↵Centers for Medicare & Medicaid Services. Readmissions Reduction Program. http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/Readmissions-Reduction-Program.html. Accessed May 3, 2015.
- 54.↵Centers for Medicare & Medicaid Services. Hospital Value-Based Purchasing. http://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/hospital-value-based-purchasing/index.html. Accessed May 3, 2015.
- Rajaram R,
- Chung JW,
- Kinnier CV,
- Barnard C,
- Mohanty S,
- Pavey ES,
- McHugh MC,
- Bilimoria KY
- Gilman M,
- Adams EK,
- Hockenberry JM,
- Wilson IB,
- Milstein AS,
- Becker ER
- Epstein AM,
- Joynt KE,
- Jha AK,
- Orav EJ
- 67.↵Agency for Healthcare Research and Quality. Efforts to improve patient safety result in 1.3 million fewer patient harms. http://www.ahrq.gov/professionals/quality-patient-safety/pfp/interimhacrate2013.html. Accessed March 2, 2015.
- Brennan N,
- Oelschlaeger A,
- Cox C,
- Tavenner M
- McClellan M,
- McKethan AN,
- Lewis JL,
- Roski J,
- Fisher ES
- Colla CH,
- Goodney PP,
- Lewis VA,
- Nallamothu BK,
- Gottlieb DJ,
- Meara E
- Ojeifo O,
- Berkowitz SA
- Starr P
- Brennan TA,
- Berwick DM
- 84.↵American Board of Internal Medicine. ABIM announces immediate changes to MOC program. http://www.abim.org/news/abim-announces-immediate-changes-to-moc-program.aspx. Accessed May 2, 2015.
- Clancy CM
- Paterson R
- Bismark MM,
- Spittal MJ,
- Gurrin LC,
- Ward M,
- Studdert DM
- 97.↵The Joint Commission. CMS renews Joint Commission hospital accreditation deeming authority. http://www.jointcommission.org/cms_renews_joint_commission_hospital_accreditation_deeming_authority/default.aspx. Accessed May 6, 2015.
- Mello MM,
- Kelly CN,
- Brennan TA
- 99.↵The Joint Commission: Facts about the national patient safety goals. http://www.jointcommission.org/facts_about_the_national_patient_safety_goals/. Accessed June 5, 2015.
- Greenfield D,
- Braithwaite J
- Flodgren G,
- Pomey MP,
- Taber SA,
- Eccles MP
- Kereiakes DJ,
- Willerson JT
- Cline RE,
- Pepine CJ
- Brennan TA,
- Leape LL,
- Laird NM,
- Hebert L,
- Localio AR,
- Lawthers AG,
- Newhouse JP,
- Weiler PC,
- Hiatt HH
- Mello M,
- Brennan T
- Mello MM,
- Chandra A,
- Gawande AA,
- Studdert DM
- Kessler DP,
- McClellan MB
- Mello MM,
- Kachalia A
- Sage WM,
- Gallagher TH,
- Armstrong S,
- Cohn JS,
- McDonald T,
- Gale J,
- Woodward AC,
- Mello MM
- Kachalia A,
- Little A,
- Isavoran M,
- Crider LM,
- Smith J