Mediators of Patient Trust

Trust—a firm belief or confidence in the honesty, integrity, reliability, and justice of another person or thing¹ —is the critical foundation of an effective patient-physician relationship. Not long ago, trust was paramount; it certainly never was a problem for Marcus Welby. However, some of the tools that managed care has introduced to influence physician behavior toward efficient and high-quality medical care (although some critics doubt its commitment to the latter) have made evaluation of patient trust a legitimate scientific question. These managed care tools include rules (eg, mandatory clinical protocols, retrospective and prospective utilization review) and financial incentives (eg, various methods of payment).²

In this issue of THE JOURNAL, Kao and colleagues³ report the results of a study in which they asked a large cross-section of patients a variety of questions about trust, including "How much do you trust your doctor to put your health and well-being above keeping down the health plan's costs?" Their most reassuring finding is that 84% of patients do trust their physician in this way; their most alarming finding is that fully 16% do not. Lack of trust is more apparent among patients who report poor health status, just the group of patients one might hope had the most trust in the guardian of their health.

The findings of this study are numerous, provocative, complicated, and somewhat cryptic. The composition of each physician's patient panel was not considered, which could have mediated results because most physicians now accept patients from several insurance companies, under a variety of payment and contractual arrangements. In general, physicians do not actively distinguish among their patients according to payment status. Instead, they form an approach to their patients that is a blend of all the rules and incentives under which they practice. For example, patient A might incorrectly identify that his or her physician is paid capitation when the physician actually is paid traditional fee-for-service (FFS) for patient A's health care. Patient A may be one of only a few patients for whom that physician is paid traditional FFS. The physician is generally paid capitation (or any other mix of payment methods), which drives his or her behavior. this provides one explanation for the differences found among patients who accurately identified how their physicians were paid, compared with those who gave incorrect answers. Those giving incorrect answers may be responding to behavior mediated by that physician's predominate method of payment, not the actual method of payment for any given patient.

The study also could not account for other important factors such as ancillary forms of payment (eg, bonuses, withholdings), rules and regulations that come in various forms and magnitude with each type of payment mechanism, or the number of tiers of management separating payer and physician.^{2 ,4} Since the underlying contractual framework in which a physician operates influences the impact of payment incentives, it will be important in the future to include these factors in similar outcomes and satisfaction studies.

In addition, the deck may have been stacked from the beginning. A self-selection bias could influence the findings, despite best efforts to minimize it by including as covariates the results of questions about trust, benevolence, perceptions of managed care organizations in general, and feelings about the patient's own health care system. Patients who elect more traditional forms of health care (ie, traditional FFS) probably do so in part because they value the perceived (or real) additional choice it offers. Having that choice, traditional FFS patients likely shop until they find a physician they trust; that is, they self-select into a system and chose a "trustworthy" physician, making traditional FFS the a priori winner, as Kao and colleagues³ found. This is supported by other findings. For instance, patients who reported enough choice of physicians trusted their physicians more and patients of physicians actually paid by traditional FFS had lower levels of trust in managed care.

Several important lessons emerge from this interesting study. First, as the authors conceptualize, although actual method of payment seems to correlate somewhat with trust, it is mitigated substantially by physicians' interpersonal skills and bedside manner. In other words, physician payment mediates patient trust both directly (how patients feel about their perceptions of how their physicians are paid) and indirectly (through patients' responses to their physicians' responses to these factors). This robust finding is so consistent with common sense that it has to be correct. Clinicians must evaluate, and maybe harness, how their own feelings about managed care, specific contractual obligations, and payment arrangements influence their relationships with patients.

Second, perceptions seem paramount, perhaps rivaling reality when patients evaluate their health care practitioner, and only one third of patients correctly identified how their physicians were paid. Trust in physicians was higher among patients who did not know the method of payment, indicating that changing perceptions about managed care, rather than ensuring accurate patient education about contractual arrangements, may be the more important way to improve patient trust and satisfaction, an unhappy thought that underscores how important the commercial medical marketplace has become in shaping patient expectations and perceptions.

Third, policymakers, managers, and researchers, when making or evaluating policy, must remember that all methods of paying physicians involve some kind of influence on behavior. Many observers feel that salary is the least worrisome method of payment; as the authors agree, salary allegedly encourages neither too few nor too many services. However, salary turned out to be significantly associated with lack of patient trust in their physicians. Without the influence of bonuses or other forms of ancillary incentives, salary may cause physicians (like all other human beings) to perform just enough to get by.

Fourth, the results of the study by Kao and colleagues³ do not justify substantially changing or eliminating financial incentives to physicians, which capture physicians' attention to practice more efficiently. Rather, the findings remind us that these incentives should be counterbalanced by physicians' concern for their patients, by physicians' bedside manner, and by the kinds of institutional rules and regulations that protect patients' interests and sustain their trust.²

Despite the continuing clarification of how management initiatives and contractual arrangements affect physician (and now patient) behavior and beliefs, answers to some of the most important health services questions remain as elusive as ever. The impact of payment methods and related influences on actual (rather than perceived) quality of care remains conspicuously, and almost completely, unknown.