Word of Mouth and Physician Referrals Still Drive Health Care Provider Choice

HSC Research Brief No. 9

December 2008

Ha T. Tu and Johanna R. Lauer

Overview

Organizations that sponsor initiatives to promote health care price and quality transparency frequently assume that all consumers represent their potential audience. In practice, however, the actual population of consumers positioned to act on such information is considerably smaller than these broad assumptions suggest. Data from the nationally representative 2007 Health Tracking Household Survey conducted by the Center for Studying Health System Change (HSC) revealed that only 11 percent of American adults sought a new primary care physician during that year. Additionally, 28 percent required a new specialist, and 16 percent underwent a medical procedure at a facility they had not previously used. Among those consumers who did obtain new providers, relatively few reported engaging in active comparison shopping or consulting price and quality data, particularly when selecting specialists or facilities for medical procedures.

How Many Consumers Are in the Market for New Health Care Providers?

The premise underlying consumer-directed health care holds that individuals who bear significant cost-sharing obligations will be sufficiently motivated to compare providers on both price and quality dimensions. Beyond the population of insured consumers facing growing cost-sharing requirements, large numbers of uninsured Americans confront even more substantial out-of-pocket financial exposure when seeking medical care. Despite the growing economic incentives that should theoretically encourage at least some consumers to shop for providers, and notwithstanding the proliferation of public and private transparency tools designed to facilitate informed decision-making, the majority of consumers continued to depend on physician referrals and informal word-of-mouth recommendations from family members and acquaintances when selecting health care providers. These patterns emerged from HSC's nationally representative 2007 Health Tracking Household Survey findings.

Sponsors of transparency programs, whether operated by state governments or private entities, commonly identify their target audience as the entirety of the consumer population within their jurisdiction. The realistic addressable market for these programs is far more circumscribed. To accurately gauge the universe of consumers who might actually engage in provider shopping at any given time, one must first establish how many adults are seeking new primary care physicians, new specialists, or new facilities for procedures. These figures serve as reasonable approximations of the target market size for transparency initiatives. The estimates presented in this analysis are likely somewhat conservative, as children were excluded from the calculations.

Consumer Shopping for Primary Care Physicians

Roughly 25 million adults -- more than one out of every ten -- indicated that they had searched for a new primary care physician at some point during the preceding 12 months. Of these, 17 million succeeded in establishing a relationship with a new primary care doctor, while approximately 8 million were still searching or had abandoned the effort without success. A considerably larger segment of the adult population reported needing new specialist physicians during the same period. Close to 63 million adults -- nearly three in ten -- stated they had required a new specialist in the prior year, with 46 million ultimately seeing one.

Approximately 35 million adults -- nearly one in six -- reported having undergone a medical procedure at a new facility during the previous year. This figure encompasses both inpatient and outpatient settings and includes both surgical and nonsurgical procedures; respondents were not asked to specify the exact nature of the procedure they received. About 19 million of these individuals, representing roughly 9 percent of all adults, had their procedures performed in hospital settings rather than in physician offices or ambulatory surgery centers.

Taken together, adults reported needing new specialists or new facilities for procedures at substantially higher rates than they reported needing new primary care physicians. Providing consumers with comprehensive information covering the full spectrum of specialists and procedure facilities they might require would present an enormous logistical challenge. Consequently, most quality transparency programs have concentrated their reporting on the most commonly utilized specialties and on procedures performed in hospital settings. The remainder of this analysis focuses specifically on consumers who actually found new primary care physicians and specialists and who underwent procedures at new facilities -- collectively referred to as "shoppers" regardless of whether they engaged in deliberate comparison shopping.

Word of Mouth and Physician Referrals Dominate Provider Selection

Among the 17 million adults who secured a new primary care physician during the study period, fully half drew upon recommendations from friends and relatives, with more than one-quarter relying on such personal endorsements as their sole source of guidance. This heavy dependence on informal social networks is consistent with prior research demonstrating that consumers place greater trust in the experiences related by friends and family members than in more formal information channels such as provider report cards. While word of mouth constituted the most frequently cited information source for primary care physician selection, substantial minorities of shoppers also consulted recommendations from other health care providers (38 percent) and information distributed by their health plans (35 percent).

The Internet was consulted by approximately one in nine primary care physician shoppers. Around 37 percent of those seeking a new primary care doctor utilized multiple information sources during their search. Younger adults and those with higher levels of educational attainment were more inclined to consult the Internet and to draw upon multiple sources simultaneously when choosing primary care physicians.

The information-gathering patterns observed among specialist and procedure shoppers diverged substantially from those seen in primary care physician shopping. When seeking a specialist, the predominant source of guidance was a referral from the consumer's primary care physician: nearly seven in ten specialist shoppers utilized this channel, and almost six in ten relied on it to the exclusion of all other sources. One in five specialist shoppers drew upon recommendations from friends and relatives, and only 15 percent consulted multiple information sources. Individuals with chronic health conditions and those reporting fair or poor health status showed a stronger tendency to rely exclusively on their primary care physician's referral, while younger and more highly educated consumers were more apt to seek out alternative sources including the Internet and health plan directories.

When selecting facilities for medical procedures, consumers exhibited an even greater reliance on physician guidance. Nearly three-quarters of procedure shoppers followed the recommendation or referral of the physician who would be performing the procedure, and almost all of these individuals used no other source of information whatsoever. Alternative information sources attracted relatively few procedure shoppers, and only about one in twelve consulted multiple sources when choosing a facility. Older consumers and those managing chronic conditions displayed a pronounced tendency to rely solely on their physicians' referrals or recommendations when selecting a procedure facility.

The greater reliance on word of mouth for primary care physician selection compared with specialist or procedure facility selection is logical. Because most adults maintain an ongoing relationship with a primary care doctor, they can readily identify acquaintances who can offer a personal recommendation. Locating someone in one's social circle who has visited a particular specialist or undergone a specific procedure proves considerably more difficult, which helps account for the outsized role of physician referrals in those contexts. Seeing a specialist or undergoing a procedure also tends to be associated with more serious medical circumstances, inclining consumers to defer to professional clinical judgment. Furthermore, some health maintenance organizations mandate a primary care physician referral before a member can access most specialist categories, and HMO enrollees may also be required to have procedures performed at designated facilities, thereby narrowing consumer choice and reducing the impetus to consult independent information sources.

Limited Consumer Use of Price and Quality Information

Relatively few consumers who needed new providers reported incorporating either price or quality information into their selection process. Utilization of price information was exceedingly rare across every category of shopper, a pattern that likely reflects both the scarcity of publicly available price data that consumers consider relevant and actionable, and the reality that most insured consumers face little or no financial incentive to compare provider prices. For the majority of insured individuals, out-of-pocket expenses remain constant across in-network providers, effectively eliminating any economic motivation to seek out lower-cost options.

Self-reported utilization of quality information was modest among specialist and procedure shoppers but somewhat higher (23 percent) among those shopping for primary care physicians. Because the survey did not ask respondents to specify what types of price or quality information they had consulted, it is plausible that some consumers interpreted favorable recommendations from health care providers or positive accounts from friends and relatives as constituting "quality information." Earlier research has found that most consumers do not perceive clinical quality as varying meaningfully across physicians and therefore exhibit limited demand for formal clinical quality report cards. At the same time, consumers care deeply about finding physicians who are attentive listeners and who demonstrate approachability and compassion -- attributes most readily assessed through the personal accounts available from friends and family. To the extent that primary care physician shoppers treated anecdotal patient-experience reports as quality information, the 23 percent quality-information usage figure for primary care shopping may be an overestimate of formal quality data consultation.

Key Factors Influencing Consumer Provider Choices

Survey respondents were asked about several factors -- including cost, insurance network inclusion, perceived quality and provider reputation, a doctor's recommendation, and convenience -- that might have played a significant role in their choice of a new physician or facility. Regardless of the type of provider being sought, consumers assigned roughly comparable weight to perceived quality, convenience considerations such as location and appointment availability, and whether the provider belonged to their health plan's network. Cost considerations trailed these other factors by a wide margin across all shopper categories, almost certainly because most insured consumers face no differential in out-of-pocket costs when choosing among in-network providers. Uninsured shoppers were more than twice as likely as their insured counterparts to identify cost as a major consideration. Younger adults and individuals with lower incomes were also considerably more likely to weigh cost heavily. Consumers with greater educational attainment demonstrated a stronger tendency to cite quality-related considerations, including provider reputation, as major factors in their decisions.

A notable finding was that every individual factor was cited much less frequently by specialist and procedure shoppers than by those shopping for primary care physicians. Although physician referrals represented the dominant information source for specialist and procedure shoppers, these consumers were far less inclined than primary care shoppers to identify a doctor's recommendation as a major factor in their ultimate selection decision. These patterns may indicate a tendency among many consumers to view primary care physician selection as a more active and deliberate process -- one demanding conscious evaluation of provider reputation and convenience -- while the selection of specialists and procedure facilities is often perceived less as a genuine choice and more as a matter of following clinical instructions.

Consumers who relied on a single information source, as well as those who depended exclusively on their physician's referral, were substantially less likely to report that any particular factor -- whether cost, quality, insurance coverage, or convenience -- played a major role in their provider selection. This finding underscores the relative passivity of shoppers who did not seek out alternative or supplementary information. For these individuals, choosing a provider may have been functionally equivalent to filling a prescription: they were simply carrying out their doctor's directions rather than exercising independent judgment. Passive shoppers, even if presented with accurate and usable price and quality data, would be unlikely to consult such information absent strong motivating circumstances.

Policy Implications

The consumer-directed health care model, which envisions patients as active and informed purchasers who systematically compare providers, is far removed from the observed reality of how the majority of consumers actually navigate health care provider selection. Few consumers make meaningful use of Internet resources, price data, or formal quality metrics. Instead, most continue to depend on the longstanding channels of physician referrals and word-of-mouth endorsements from friends and family members.

One significant structural barrier to more active provider shopping is the limited supply of physicians in certain geographic markets and within certain health plan networks. Consumers seeking a new primary care physician may discover that very few doctors within a convenient distance are currently accepting new patients. Similarly, a given health plan network might include only a single specialist or subspecialist of a particular type in a market area. Consumers facing such supply constraints have limited opportunities for comparison regardless of the information available to them.

Even where provider choice is not supply-constrained, other impediments to active shopping persist. This analysis demonstrates that passive shoppers -- those who relied on a single information channel or did not independently seek out information -- considered fewer factors before selecting a provider and may not have perceived the selection process as involving a genuine choice. Such individuals would be unlikely to engage with price and quality information even if it were readily accessible, unless given compelling reasons to do so.

For price information to drive consumer behavior, insured individuals would need to encounter meaningful out-of-pocket cost differentials based on which provider they chose. At present, such incentives are weak or absent for most insured consumers who remain within their health plan's provider network. On the quality side, consumers would likely find quality data more compelling if they believed that clinical quality varied substantially across providers and that these variations could have serious -- even life-altering -- consequences. For policymakers seeking to promote more engaged consumer participation in provider shopping and quality improvement, a central challenge lies in educating the public about the existence and significance of quality differences among providers.

Data Source

This Research Brief draws on findings from the HSC 2007 Health Tracking Household Survey, which was supported by the Robert Wood Johnson Foundation and administered between April 2007 and January 2008. The nationally representative survey encompassed approximately 18,000 individuals in 9,400 families, with a study-specific sample of roughly 13,500 adults. The survey achieved a response rate of 43 percent. Samples were constructed using random-digit dialing methods, and data were collected via telephone using computer-assisted telephone interviewing (CATI) protocols. The Household Survey module addressing consumer shopping behavior for health care providers is available from the authors upon request. This research was funded by the California HealthCare Foundation.

Related Resources

For additional HSC research and publications, visit the HSC archives at hschange.com