From Information on Quality to Quality Information

Steady increases in health care expenditures have stimulated a number of contemporary trends focused on value or return on investment in health care. Interest in clarifying the outcomes or end results of health care in terms that people experience and care about, combined with growing public interest in health information and playing a stronger role in their own health care decisions, are now considered to be an intrinsic part of the current landscape. The health care research community has played a vital role by developing valid tools for assessing outcomes, examining the link between processes and outcomes of care and conducting studies that identify organizational and financial determinants of high-quality care. Leading edge purchasers and a few states have led the way by demanding evidence of performance through public reporting of clinical performance and patients' perspectives on care. An implicit assumption underlying policy makers' efforts and insights from research is that the results would enhance decisions and ultimately health care delivery.

The initial premise that broad dissemination of such information would prompt consumers to “vote with their feet” has proven difficult to demonstrate. First, a substantial portion of insured Americans don't have the opportunity to choose among plans. Second, efforts to present the information effectively are still evolving. Third, some observers note that available information may not be relevant to either issues of concern or the specific actions available to individuals. For example, some studies have postulated that information at the level of a health plan may be less meaningful to consumers' decisions than information on specific providers. Notwithstanding these issues, continued frustration with rising expenditures and suboptimal quality has resulted in new challenges for individuals. Today consumers are likely to confront increased cost sharing for prescription medications, and may well soon face substantial differences in out-of-pocket spending associated with insurer choices of networks based on hospital quality. While the potential decision points are clear, it is far from self-evident that our expanding capacity for assessing multiple dimensions of health care quality is directly relevant to the new demands confronting consumers. In addition to refining strategies for providing useful information and communicating the results effectively, the challenge of making available this information at the point and time of decision making must be addressed for information on quality to be useful and used.

In this issue, Abraham describes the results of a telephone survey of employees of 16 firms in Minnesota that are members of the Buyers Health Care Action Group (BHCAG). The authors also conducted an employer survey to assess how information was disseminated to employees. Of note, these employees had access to information on providers which is thought to be more useful than just information on health plans. Unlike the usual situation in which employees choose among health plans, many of which offer access to the same providers, BHCAG organized primary care physicians into distinct sets and assessed their quality and cost. Similar to previous studies, overall awareness of information on quality was low. Employers' efforts such as distributing booklets to all employees or making them available on request were associated with an increased probability of awareness about quality, as was employee level of education and recent utilization of providers' services. The authors conclude, however, that there is little empirical evidence to support the assertion that providing information on quality at the level of providers is associated with incrementally greater awareness. Importantly, the authors did not attempt to address the extent to which the information was useful—or used.

The resources allocated for assessment of quality in this study appear to exceed those provided for dissemination, so the results may not be surprising. An agnostic about the importance of assessment and reporting might well ask, however, whether the effort invested is worth the net results. Improved strategies for presenting and communicating information on quality may yet lead to increased awareness—indeed, demand for information. It is also worth inquiring why interest and awareness remains anemic, when surveys of public attitudes reveal enormous concerns about quality and safety, and the public's interest in health information online continues to increase. A related line of investigation could focus on why the response to dissemination efforts appears less than enthusiastic. Perhaps information describing compliance with evidence-based care processes may still be too far removed from the dimensions of health and health care that are of great interest or within the control of individual consumers.

It would be reasonable to postulate that purchasers' demands for reporting are based on an expectation that information on care processes is a useful proxy for what matters most in health care delivery. Indeed, Marshall's study of the impact of performance measurement found that public reporting is associated with a clear impact on providers and subsequent improvements. It is far less apparent that the information itself is directly useful to either purchasers' or consumers' decisions. We lack reliable measures for many clinical domains (e.g., mental health; maternal and child health) and it is not possible to predict how performance in one clinical area is related to performance in other areas. Indeed, a common complaint of clinicians is that the overlap between what is measurable and what is important is far less than optimal. Continued work to develop and refine valid quality measures can ameliorate many of these challenges, but such work may not be useful to decision makers unless it is both more meaningful to decision makers and linked to specific actions within their control.

This issue also includes several papers that encourage optimism that collaboration within the health care research community can help us move from information about quality to quality—or meaningful information for decision making. For example, while a few studies have examined the relationship between health care outcomes and results directly relevant to employers, most often the relationship between health care quality and productivity or lost time from work is not addressed. Moreover, the expected impact of new innovations in care on an employers' overall bottom line is not clear. Goldman's study examines the relationship between highly active anti-retroviral therapy (HAART) for employed adults infected with HIV and workforce participation. Using data from the HIV Cost and Services Utilization Study (HCSUS), a nationally representative sample of adults who had made at least one visit for HIV services in 1996–97 that included queries about the use of HAART and effect on employment, the investigators examined the impact HAART on returning to work within six months of treatment, remaining employed, and hours of employment. Their results indicate that HAART increases the probability of remaining employed and the number of hours worked for those working within six months of treatment. Of note, the increased hours worked is both statistically significant and substantial; resulting income increases compare favorably with the cost of treatment. These findings are relevant to both employers and individuals.

In another paper, Vijan used data from the Health and Retirement Survey (HRS), a nationally representative longitudinal survey of adults born between 1931 and 1941, designed to follow adults as they make the transition from active work to retirement. Using cross-sectional analyses of baseline data and longitudinal assessment from eight years of follow-up, the authors model the impact of diabetes on work force participation and estimate the impact of diabetes-related losses in productivity. For this cohort, an estimated $60 billion in lost productivity was associated with diabetes. Additional average annual losses of $7.3 billion over the next eight years resulted in an estimated total lost productivity of $120 billion by 2000. Expected increases in the prevalence of diabetes suggest additional important negative effects on productivity absent substantial enhancements in quality of care for affected individuals.

Blustein also uses the HRS to examine the impact of grandparents' caregiving on their own depressive symptoms. Their findings indicate that the period that one or more grandchildren are residing with grandparents is associated with a significant increase in depressive symptoms, particularly for grandparents who are single women or those who do not have an adult child also in residence. This paper does not examine the impact of income directly, but the finding that in 1994 8.4% of adults were caring for grandchildren for some period of time has implications for employers and grandchildren, in addition to the direct effect on the grandparents themselves.

These topically distinct papers suggest that within the broad health care research community the capacity is well established to link quality of care to important outcomes beyond those typically included in “report cards.” In other words, one can begin to envision a future generation of reports on quality of care for which the relationship between compliance with evidence-based care processes and outcomes most relevant to different decision makers can be more clearly explicated. Current efforts to accelerate the adoption and use of health information technology in health care delivery will reduce the current reporting burden and support efforts to provide information in a much more timely fashion. It is not too early to consider now how that same infrastructure can be exploited by health care researchers to develop information on the value of information on quality to decision makers. Such efforts can't compensate for lackluster dissemination efforts, but could help decision makers “connect the dots” between information on quality and quality information.