Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
6 Qmlityof Care The growth of for-profit health care has prompted concern and speculation about the quality of cared Some observers are doubt- ful that the increasingly visible investor- owned companies would risk their capital investments, reputations, and the possibil- ity of tighter regulation by deciding to cut corners on quality. Nevertheless, problems could develop if the upper management of art organization emphasizes economic per- formance or efficiency to such an extent that subordinates make decisions that reduce quality, even if that is not intended. Some concern about the quality of care rests on the assumption that for-profit organizations are more likely than not-for-profit organi- zations to judge the performance of man- agers on narrow economic grounds, thereby inducing them to take steps that could neg- atively affect quality. On the other hand, clearly no general assumption is warranted that goals of high quality are inconsistent with goals of profitability in an organization. The question of the relationship between for-profit status and quality of care is em- pirical, not definitional. This chapter ex- amines the limited data now available. QUALITY OF HOSPITAL CARE Approaches to the appraisal of health care quality generally fall into three broad cat- egones evaluation of structure, evaluation 727 of process, and evaluation of outcome or end results (Donabedian, 1969:2~. Structural in- dicators inclucle not only physical aspects of facilities and equipment, but also charac- teristics of the organization and qualifica- tions of its health professionals. Process measures pertain to the activities of health professionals in the care of patients. Out- come measures may be stated in terms of health or in terms of patient or family sat- isfaction. Quality indicators vary in their comprehensiveness, their importance, their face validity, and their availability in exist- ing data sources. Evidence regarding the relationship be- tween hospital ownership and quality of care is fragmentary and limited. Ideally, the committee would have liked to examine pa- tient outcome data from carefully matched for-profit and not-for-profit settings, as well as from chain and independent institutions, in order to draw inferences about quality. Unfortunately, the committee had access only to studies using statistical controls and struc- ture and process quality measures, such as hospital accreditation, board certification rates of staff physicians, and perceptions of physicians and hospital board chain, as well as some aggregate but nondefinitive pooled outcome data.
128 Accreditation Probably the most elemental single in- dicator of the overall quality of care is vol- untary accreditation by the loins Commission on Accreditation of Hospitals dCAH). Ac- creditation is based on the evaluation of re- ports by site visit teams who assess institutional compliance with more than 2,200 separate standards. Although experts agree that accreditation addresses quality, the process also has its limitations. It is based heavily on structural indicators, although some process indicators are also included. Furthermore, accreditation represents a kind of threshold or minimum baseline, and be- cause of its global nature it does not capture many important differences among institu- tions. Not all institutions seek accreditation; attitudes about the importance of accredi- tation vary in different parts of the country. The American Hospital Association, in its annual survey of hospitals, collects data on whether hospitals are accredited. A second source of data on ICAlI accreditation is from a paper prepared for the committee by Gary Gaumer (1986), using data that had been compiled for a study of state hospital rate- setting programs. The sample of hospitals included all short-term hospitals in IS states that have prospective payment programs plus a 25 percent random sample of aD other U. S. short-term hospitals. Information about fills, two-year accreditation and provisional, one- year accreditation was obtained by Gaumer directly from JCAH. Finally, in response to a request Tom the committee, ICAH itself prepared an analysis of the outcomes of one year's accreditation visits to independent and chain for-proEt and not-for-profit hospitals (Longo et al., 1986~. This analysis focused not only on accreditation itself but also on "contingencies" attached by ICAH in its ap- proval of an institution's compliance with particular ICAH standards. Table 6.1 shows that the highest per- centage of accreditation is found among hos- pitals in investor-ownecl systems, while the FOR-PROFIT ENTERPRISE IN HEALTH CARE lowest rates are found among the indepen- dent investor-owned hospitals and among hospitals owned by state or local govern- ments. There is virtually no difference be- tween investor-owned and not-for-profit chain hospitals. Since much of the past growth of investor-ownec! chains has been through ac- quisition of independent for-profit hospitals, the growth of investor-owned systems likely has resulted in the accreditation of a number of hospitals that were not previously ac- credited. Table 6.2 shows the results of the JCAH staffs own analysis, which is consistent with the AHA and Gaumer data shown in Table 6.1. Using data from accreditation visits con- ducted during the fiscal year ending May 31, 1983, the hospitals that were part of investor-owned chains were most likely to be accredited (and to be accredited without contingencies) and least likely to have failed to pass accreditation. Differences between these investor-owned and not-for profit hos- pitals were very small and again were in striking contrast to independent for-profit hospitals. Qualifications of Medical Staff There is some evidence that investor- owned hospitals are less selective in ap- proving physicians for staffprivileges.2 In a paper prepared for the committee, Morrisey et al. (1986) report data on the percentage of physician-applicants approved in 1981 for privileges in various hospital groupings. They found that hospitals in investor-owned sys- tems approved 89.8 percent of applicants, compared with 81.4 percent in freestanding hospitals (for-profit and not-for-profit com- bined), 86.8 percent in religiously affiliated system hospitals, 81.4 percent in other not- for-profit systems, and 76.4 percent in con- tract-managed hospitals. (However, the dif- ferences among hospitals from investor- owned, religious, and not-for-profit systems were not statistically significant.) Regarding board certification of sta~phy
QUALITY OF CARE TABLE 6.1 Rates of JCAH Accreditation, by Type of Hospital, 1983 129 Investor- Not-for- Not-for- State and Owned For-profit profit profit Local Chain Independent Chain Independent Government Accreditation Status (%) (%) (%) (%) (%) Accredited, AHA data, 1983a 91 52 87 83 59 Full (2-year) accreditation, 1981b 79 50 74 55 Average All accreditation, 1974-1981b 69 42 69 54 Average provisional accreditation, 197~1981b 10 10 14 12 aPeter Kralovec, Hospital Data Center, American Hospital Association, 1985. bJCAH data. Gaumer (1986:Table 10~. sicians,3 Morrisey et al. (1986) found no sta- tistically significant differences between investor-owned system hospitals and other types of hospitals in their requirements that at least some specialists be board certified, with approximately 30 percent of the hos- pitals having such requirements. "Proprie- b~ry" hospitals (a term that was used to include both chain and independent for-profit hos- pitals) have more board-cerdfiec! medical staff per 100 beds than do "nonproprietary" hos- pitals (28.8 versus 24.6) (American Medical Association, 1984~. However, this is because on average they have more physicians on their medical staffs (49. 8 versus 37.34. These data, from the 1982 AHA annual survey, show higher rates of board certification among physicians in nonproprietary hospitals than in proprietary hospitals (66 percent versus 58 percent); this is also true of the specialty groupings compared genera]Jfamily prac- lice (41 percent versus 37 percent), medical specialties (68 percent versus 60 percent), surgical specialties (73 percent versus 64 percent), and hospital-based specialties (78 percent versus 69 percent) (American Mecl- ical Association, 1984~. These same data, however, were analyzed in a paper prepared for the committee, and it was found that when chain hospitals are separated Dom independent inshtubons, He rates of board certification of staff physicians TABLE 6.2 Results of ICAH Accreditation Visits for Fiscal Year Ending May 31, 1983 Accreditation Outcome Accredited Accredited Without with Contingencies Contingencies (%) (%) Not Accredited (%) Systemlinvestor-owned 48.1 51.3 0.6 System/not-for-profit 43.2 55.9 1.0 Independent/for-profit 46.4 51.8 1.8 Independent/not-for-profit 36.5 62.4 1.1 SOURCE: Longo et al. (1986~.
130 vary little across types of hospitals (Morrisey et al., 1986~. The lowest rates of board cer- tification of staff physicians (58 percent) were found in publicly owned system hospitals and in contract-managed hospitals; the high- est rates (65 percent) were reported for hos- pitals in religious systems and in other not- for-profit systems; the rates for hospitals in investor-owned systems and freestanding hospitals of all types were in an intermediate position (61 percent). In a multivariate anal- ysis, only public hospitals were statistically different Cower) from other types of hospi- tals. In another paper prepared for the com- mittee (Musacchio et al., 1986), data from a 1984 American Medical Association survey of physicians show that the lowest rates of board certification were among physicians whose primary hospital was owned by an investor-owned system (61.4 percent) and the highest rates were among physicians practicing primarily in not-for-profit hospi- tals (almost 69 percent), with physicians as- sociated with public hospitals in an intermediate position; however, these dif- ferences were not statistically significant. At the committee's request, a table was pre- pared to display accreditation rates among several specialties. Table 6.3 shows a puz- zling and inconsistent pattern: lye for-profit chain hospitals show comparatively high rates of board certification among the medical specialties and comparatively low rates among the surgical specialties, including OB/GYN. However, the difference between for-profit and not-for-profit system hospitals was not statistically significant. Nursing Personnel If data on the relationship between phy- sicians and quality are sparse, evidence re- lating quality of care to nursing services is even more meagre. The only available proxy measures are numbers of nursing personnel per patient. Table 6.4 displays data from the 1983 AHA annual survey of hospitals and FOR-PROFIT ENTERPRISE IN HEALTH CARE shows no important differences between investor-owned chain hospitals and not-for- profit hospitals, although both inclependent for-profit hospitals and state and local gov- ernment hospitals had lower levels of reg- istered nurses. Physicians' Evaluations At the committee s request, the American Medical Association asked the physicians re- sponding to its 1984 core survey to compare their primary hospital with other hospitals that they might be familiar with on four di- mensions: the adequacy of nursing support, the responsiveness of the hospital's admin- istration, the level of patient satisfaction, and the adequacy of technical resources and equipment (Musacchio et al., 1986~. These data are perceptual in nature, and since they pertain to the hospital where physicians ad- mit most of their patients, the possibility of favorable biases cannot be dismissed. Still, these evaluations from a national probability sample of approximately 3,200 responding physicians are subject to a different set of biases that result from studies based on data provided voluntarily by institutions. No comparable source of data has previously existed. As shown in Table 6.5, when responses to the four questions are averaged, differ- ences among physicians in not-for-profit and for-profit system hospitals and independent hospitals are small; they are very favorable in comparison to the responses of physicians whose primary hospital is owned by state/ local governments. (Because independent for-profit hospitals are typically owned by physicians, that category possibly includes some evaluations from owners and therefore is not included in this chapter's tables; sta- tistics for these tables appear in Musacchio et al., 1986.) The averages displayed in Table 6.5 con- ceal some interesting differences in physi- cian evaluations on the four dimensions studied. Table 6.6 shows physician evalua
QUALITY OF CARE TABLE 6.3 Percent of Physicians flat Are Board Certified, by Hospital Ownership Status and Selected Specialty Breakdown, 1984 . Multihospital Systems 737 Independent State and Private State and Private Physicians' Local Not-for- For- Local Not-for- For Specialties Government profit profit Government profit profit All physicians 68.7 68.6 61.4 63.8 68.7 63.0 Generallfamily practitioners 33.3 43.3 40.8 53.9 43.1 18.8 Medical specialty 66.7 67.3 75.0 65.0 70.3 54.6 Surgical specialty other than OB/GYN 73.9 82.1 69.6 73.3 78.8 82.4 OB/GYN 81.1 69.8 59.1 75.0 72.3 Othera 75.9 70.1 67.2 63.2 70.0 73.9 a Psychiatrists, radiologists, anesthesiologists, and pathologists. NOTE: This breakdown showing the percentage of board-certified physicians by specialty and hospital type is statistically significant (chi-square = 111.316 with 45 degrees of freedom). Chi-square tests of significance show no diEerer~ce between not-for-profit and for-profit multihospital systems when all five sDecialtv Shrouds were considered or when only medical, surgical, and OB/GYN were considered. SOURCE: Musacchio et al. (19861. tions of how nursing, hospital administra- tion, patient satisfaction, and technical equipment in their primary hospital com- pares with other hospitals. The perception of greater administrative responsiveness at investor-owned hospitals is particularly striking; also noteworthy are the less-favor- able evaluations given to for-profit chain hospitals than to not-for-profit hospitals on the other three dimensions: nursing sup- port, patient satisfaction, and technical re- sources and equipment. Data Dom another national AMA survey conclucte`d in 1984 show that 32 percent of physicians believed that not-for-profit hos .~ < ~r~ A pitals provide better quality of care than do for-profit hospitals, while only 5 percent be- lieved the opposite to be true (Musacchio et al., 1986:Table 111. (The remainder of physicians believed there is no difference or they had no opinion.) Interestingly, of the physicians who had some staff privileges at a for-profit hospital (and data are not avail- able on whether proprietary or investor- owned hospitals are involved), 24 percent believed that not-for-profits provide better quality of care, and 8 percent believed that for-profits provided better care. Musacchio et al. (1986) attribute this difference, at least in part, to the fact Mat most physicians who TABLE 6.4 Nursing Personnel per 100 Adjusted Census, Short-Term General and Over Special Hospitals, 1983 Investor- For- Not-for- Not-for- State and Owned profit profit profit Local Chain Independent Chain Independent Government RNs/100 patients 82 70 84 83 71 LPNs/100 patients 32 28 26 25 30 SOURCE: Peter Kralovec, Hospital Data Center, American Hospital Association, 1985.
132 FOR-PROFIT ENTERPRISE IN HEALTH CARE TABLE 6.5 Physicans Average Evaluation of Their Primary Hospital on Four Dimensions, 1984a Type of Primary Hospital Investor-Not-for- Not-for- Nonfederal Nonfederal Ownedprofit profit Government Government Primal ChainChain Independent Chain Independent Hospital Is (%)(%) (%) (%) (%) Better 53.155.4 55.4 31.0 47.6 Same 41.640.3 39.1 44.1 34.8 Worse 5.34.4 5.6 25.0 17.0 a Physicians were asked to compare their primary hospital to other hospitals with which they were familiar on four dimensions: (1) adequacy of nursing support, (2) responsiveness of the hospital administration, (3) level of patient satisfaction, and (4) technical resources and equipment. Numbers in the table are averages across these four dimensions. SOURCE: American Medical Association, 1984 Socioeconomic Monitoring System Core Survey. Calculated from Table 11 in Musacchio et al. (1986~. perceive a difference in the amount of clin- ical discretion allowed physicians believe that physicians, have less discretion at for-profit hospitals. Implications for quality are quite unclear. Board Chairman Evaluations Another source of perceptions comes from a survey of hospital governing board chair- men that was conducted by Arthur Young for the American Hospital Association's Trustee Magazine in 1983 (Arthur Young, 19831. Table 6.7 shows responses on which board responsibilities were "very impor- tant" and which responsibilities were being performed "excellently." Although these are hardly direct or objective measures of in- stitutional quality, the responses of chair- men show some are interesting contrasts in the different types of institutions. In com- parison with board chairmen Dom not-for- profit hospitals, board chairmen from inves- tor-owned hospitals reported comparatively Tow levels of board involvement in financial and management issues (such decisions are TABLE 6.6 Percent of Physicians Evaluating Their Primary Hospital as Better or Worse Than Other Hospitals with Which They Were Familiar, 1984 Type of Primary Hospital Investor-Not-for- Not-for- Nonfederal Nonfederal Oaredprofit profit Government Government ChainChain Independent Chain Independent Dimension (%)(%) (%) (%) (%) Nursing support Better 48.259.7 56.3 29.0 50.0 Worse 5.12.0 3.5 11.8 7.9 Responsiveness of the Better 60.247.3 46.8 29.? ~.5 hospital Worse 7.010.2 11.3 25.3 41.8 administration Level of patient Better 51.556.6 57.6 31.9 45.7 satisfaction Worse 4.61.7 2.8 14.3 6.7 Technical resources Better 52.557.8 60.9 33.3 51.4 and equipment Worse 4.53.8 4.6 48.4 11.8 SOURCE: American Medical Association, 1984 Socioeconomic Monitoring System Core Survey; Musacchio et ad. (1986).
QUALITY OF CARE TABLE 6.7 Perceptions of Hospital Governing Board Chairmen of Their Own Board's Activities, by Type of Hospital (percent of chairmen giving each response) 133 Investor- Owned (%) Religious (%) Other Not-for- profit (%) Government Total (%) (%) Which board responsibilities are viewed as "very important"? Ensure integrity of operations Perfonn strategic planning Monitor CEO Monitor financial reporting Ensure competitive position Ensure quality patient care Ensure standards are met Monitor legal liability Approve medical stag Which board responsibilities are being performed "excellently"? Ensure integrity of operations Perform strategic planning Monitor CEO Monitor financial reporting Ensure competitive position Ensure quality patient care Ensure standards are met Monitor legal liability Approve medical stab 88 31 28 22 33 84 71 38 47 52 14 21 17 24 64 57 36 40 78 48 45 39 31 83 59 29 44 50 22 24 41 15 52 39 23 34 81 43 42 42 28 84 56 28 32 54 22 21 42 16 49 35 23 25 15 33 39 46 29 83 60 29 37 51 18 24 38 20 57 39 ~2 28 80 42 41 41 29 84 58 29 36 53 21 22 40 17 52 38 23 28 SOURCE: The Hospital Governing Board Chairman: Profile and Opinions: A National Study by Trustee Mag- azine and Arthur Young (1983~. Often made at the regional or corporate level in multi-institutional systems) and as much or more concern with issues that bear on quality. This may reflect another finding from the same survey 43 percent of the board chairmen in investor-owned hospitals were physicians, compared with 2-3 percent of chairmen at government, religious, and other not-for-profit hospitals. Because several studies have provided ev- idence that greater medical staB participa- tion in governing is associated with higher- quality hospital care (Neuhauser, 1971; Shortell et al., 1976; Shortell and LoGerfo, 1981; Flood et al., 1982; Scott et al., 1979), it is worth noting that investor-owned hos- pitals have particularly high levels of phy- sician participation in hospital governing (Alexander et al., 1986; Arthur Young, 19831. These data are discussed at greater length in Chapter 9. Outcome Measures Of all the possible measures of quality of care, outcomes are the most important. Ideally, comparisons of quality in for-profit and not-for-profit settings would include standardized case-fatality ratios; rates of such events as intra- and postoperative compli- cations, wound infections, and drug reac- tions; and unnecessary care. Unfortunately, few such comparisons exist. The only major attempt at examining the relationship between patient outcomes and investor ownership of hospitals was done for the committee by Gaumer (19861. He ex- amined postoperative mortality (during the
134 hospital stay and within 180 days of dis- charge) and 90-day post-discharge readmis- sion rates among Medicare patients who had undergone one of eight types of elective sur- gery between 1974 and 1981.4 Gaumer con- cluded that hospital ownership was not a "strong or consistent influence on post-op- erative mortality rates," in that significant differences that were found were "not prop- agated across all procedures categories." For example, proprietary status was associated with lower in-hospital mortality in several cases, but chain affiliation was sometimes associated with higher mortality, and the data on mortality within 180 days after dis- charge tended to be the reverse of findings for in-hospital mortality. Chain affiliates tended to have higher 90-day readmission rates. Conceding that his study had not been a conclusive test of the quality-of-care ques- tion, Gaumer noted the methodological problems that are involved in studying in- stitutional differences in relatively rare events (such as mortality from elective surgery). Even though several years' data from a 25 percent sample of U. S. hospitals were used, only differences larger than 10-12 percent would have been detected by this analysis. However, no pattern of large and persistent ownership differences could be detected for the serious, postsurgical outcomes that Gau- mer studied. The possibility that the profit-seeking ori- entation milt lead to more unnecessary surgery at for-pro~Rt hospitals could not be examined directly, but some data are avail- able on Caesarean sections. Although high rates of elective Caesarean sections are con- sidered by some as an indication of excessive surgery, the great professional debate about indicators for elective Caesarean sections makes it difficult to place too much reliance on such rates as a measure of excessive sur- gery. However, data from the National Hos- pital Discharge Survey show Caesarean section rates to be higher in "proprietary" hospitals than in governmental hospitals in all four areas of the country examined and FOR-PROFIT ENTERPRISE IN HEALTH CARE higher than voluntary not-for-profit hospi- tats in three of four areas of the country, as is shown in Table 6.8 (Placek et al., 1983~. These data do not distinguish chain from independent hospitals, unfortunately, nor do they control for possible variations in pa- tient characteristics. The analysis of primary Caesarean section rates in California by Wil- liams and Chen (1983) fails to show signifi- cant differences by ownership category. These data are also shown in Table 6.8. QUALITY OF CARE IN NURSING HOMES Other than hospitals, nursing homes are the only type of health care providers about which there is literature linking quality-of- care concerns and types of ownership. A1- though the committee did not focus its at- tention heavily on nursing homes, because of time and budget constraints and because the problem of nursing home regulation is the subject of another Institute of Medicine study (1986), it did commission an extensive review of the nursing home literature (Hawes and Phillips, 19861. There are important similarities and dif- ferences between hospitals and nursing homes in the United States. Both have early histories as charitable institutions and both have been significantly affected by govern- mental policies. Since the passage of the Medicare and Medicaid programs in 1965,5 both hospitals and nursing homes have wit- nessed a substantial decline in locally owned proprietary facilities and the development and rapid growth of investor-owned com- panies that own and manage large numbers of facilities. However, in contrast to hospi- tals, where the not-for-profit sector has long predominated (and is still more than 80 per- cent of the total), more than 75 percent of America's nursing homes are proprietary. Another important difference between the hospital and long-term-care sectors is that while the investor-owned hospital compa- nies developed during an era of generous
QUALITY OF CARE TABLE 6.8 Caesarean Section Rates for Nonfederal Short-Stay Hospitals, by Hospital Ownership 135 For-pro~t Not-for-pro~t Government United States --1981 (% of births) 22.0 18.5 15.4 Northeast 25.5 20.4 14.9 Norm Central 30.4 15.9 14.5 Soup 18.9 15.8 20.4 West 22.5 17.4 15.5 Primary Ceasarean sections, California 1978-1980 Observed rate (% of births) 12.4 13.3 12.5a Expected rateb 11.4 12.4 11.9 Standardized ratio 108.8 107.9 105.2 a"District" hospitals only. bExpected rate based on risk factors such as maternal age; panty; infant's sex, race, and birthweight; and type of presentation (e.g., breech). SOURCE: U.S. data: Placek et al. (1983:862~; California data: Williams and Chen (1983:865~. funding (Medicare's cost-based reimburse- ment, with most other third-party payers paying charges), prospectively set rates have characterized much of the financing of nurs- ing home care. Ibus, hospitals have had fewer financial incentives which might have led them to consider cost reductions that had quality-of-care implications; reductions in nursing home expenses have long translated much more directly into profits, because revenues were substantially based on fixed rates, rather than reimbursement for costs incurred. As Beverly Enterprises Chairman Robert Van Tuyle noted in 1982, because Beverly would serve 60,000,000 meals that year, "every penny saved per meal trans- lates into $600,000" (Los Angeles Times, May 25, 1982). The presence and influence of medical personnel, particularly physicians, who play an important agency role in hospitals but not in nursing homes is another difference with important implications. As physician- oriented institutions, hospitals have a Tong history of genuine self-regulatory activity; nursing home quality has long been seen primarily as a problem of governmental reg- ulation, not of self-regulation. The relatively minor role of physicians in nursing homes also contributes to the comparatively lim- ited role of market forces in the Tong-term- care sector. Whereas most American hos- pitals have long had surplus capacity and have vigorously competed for the loyalty of physicians who controlled the flow of pa- tients, nursing home beds in most parts of the country are in very short supply, and physicians tend not to be involved in the referral process or in the day-to-day care of nursing home residents. Because of their physical and mental disabilities and social isolation, and owing to the scarce financial resources of Medicaid patients, many nurs- ing home residents are less able than hos- pitalized patients to look out for their own interests or to have advocates to do so. In other words, economic incentives in support of quality of care are less apparent in nursing homes than in hospitals. Except for the relatively few nursing homes whose facilities and prices are aimed at a wealthy clientele, the primary incentives that op- erate in favor of quality are based either on ethics and community or religious values or on avoidance of regulatory problems. Arid, nursing home regulation has generally con- centrated enforcement on building and fire safety standards rather than on health and
136 patient care standards. All of these factors help to explain the pattern of scandal in the history of American nursing homes, as well as the frequently expressed concern about providers whose motivation lies in econom- ics rather than "religious responsibility or fraternal solidarity" (VIadeck, 1980:1261. Regarcling quality of care and type of nursing home ownership, two major limi- tations of much of the available literature should be noted (Hawes and Phillips, 19869. First, as with hospitals, most studies use resource measures rather than outcome measures and, as such, are subject to meth- odological dispute and difference of inter- pretation. Second, and more unfortunate for a study that is oriented toward understand- ing implications for the future, most studies lump together as "proprietaries" the inde- pendent "mom-and-pop" nursing homes that have been on the decline since the very different investor-owned chain facilities have appeared and have been rapidly growing.6 To broadly summarize available evi- dence, most studies on quality (or surrogate measures) of nursing home care tend to fa- vor the not-for-profit mode of organization. This finding holds up across a wide range of measures amount of patient care staff, ex- penditures on food, complaints to state reg- ulatory agencies, nonconformity with regulatory requirements, and, in one study, on the following outcome measures: care planning, quarterly review of patients, room conditions, and quality of living environ- ment (Lee, 19841. These studies are sum- marized in some detail by Hawes and Phillips (1986), who conclude that "the preponder- ance of evidence suggests the superiority of not-for-profits-particularly of the church- related not-for-profits." Similar conclusions, and an important caveat about variations, were also reached by VIadeck in his book, Unloving Care. . . . on the average, voluntary facilities are some- what better than proprietary ones. The worst nursing homes are almost exclusively proprie- tary. But in the middle ranges, there is substan FOFt-PROFIT ENTERPRISE IN HEALTH CARE tial overlap. The best way to visualize the difference might be to conceive of the range of quality in each of the two types of nursing homes as a quasi-normal distribution.... The two dis- tributions overlap markedly, with the mean for the voluntaries slightly higher thar1 the mean for proprietaries, and with the voluntaries having a shorter low-quality tail (Vladeck, 1980:123~. However, Hawes and Phillips also note that most studies have been flawed in one way or another, and "that the lumping to- gether of various types of not-for-profits (government, church-related, and private) and of independent and chain for-profits may obscure some significant differences in per- formance." A particular problem in making quality comparisons across the for-profit/not- for-profit line is that not-for-profit nursing homes may have, on average, higher reve- nues than for-profits. For example, Hawes anc! Phillips cite Texas data showing that not-for-profits have 19 percent higher rev- enues (as well as 31 percent higher expen- ditures) than for-profits. - Not-for-profits have a nigher percentage of non-Medicaid pa- tients, who in most states pay higher charges than patients covered by the state Medicaid program. Many not-for-profit facilities also benefit from subsidies from their sponsoring organizations (e.g., church or synagogue) or from other sources. Hospitals, Nursing Homes, and Quality Although the limitations of existing stud- ies must be acknowledged, it is clear that there are some differences associated with type of ownership of nursing homes that do not appear among hospitals. The reasons for this are not fully clear, but their implications are important. If the relatively favorable comparisons of quality between for-profit and not-for-profit hospitals are due to factors that are associ- ated with the growth of investor-owned chains, then the continued growth of the for- profit nursing home chains may lead to im- provement in nursing home quality. The
QUALITY OF CARE analysis by Hawes and Phillips (1986) does suggest that, as was the case with hospitals, the replacement of small proprietary nurs- ing homes by better-financed and better- managed investor-owned chain nursing homes has frequently improved quality. However, the difference between hospi- tals and nursing homes could be due to other factors, such as the greater presence and influence of physicians in hospitals (as well as other health professionals), who fee] a responsibility to act as agent or advocate for their patient. It is important to consider why physicians do not play a similar role in nurs- ing homes. Is it a lack of patient need, a lack of money to pay for physician services, or a lack of physician interest in the kinds of problems that are presented by residents of long-term-care facilities in contrast to the more acute and dramatic cases seen in the hospital? The most likely answer is the last. A second question to be considered is how the agency aspect of the physicians' role in the hospital setting may be affected by the entrepreneurial trends that are taking place in health care. This issue is discussed in Chapters 8 and 9 of this report. Another possible explanation of the dif- ference between hospitals and nursing homes is the lack of competitive conditions among nursing homes because ofthe tight supply of beds. If so, differences between hospitals and nursing homes will persist so long as the hospital world is relatively competitive (including competition on the basis of qual- ity as well as price) in comparison with the nursing home market. Still another possible explanation of the difference in the relative performance offor- profits and not-for-profits in nursing homes and hospitals is the stringent financial con- straints under which nursing homes oper- ate, which are a reflection of societal values and public policy decisions. If constrained resources do play a role, quality problems could increase in investor-owned hospitals (or, perhaps, in all hospitals in which bot- tom-line considerations are important) un 137 der Medicare's prospective payment and increasing financial constraints. In any event, there is a need for improved monitoring of quality-related patient care outcomes. The possibility that financial goals may supplant quality of care at least under some circum- stances must be recognized, although some reassurance comes from studies of quality of care in HMOs (Luit, 1981). Although quality care clearly requires a certain level of funding, research has yet to establish a clear relationship between the level at which long-term care is funded and the quality of care that institutions provide. Unless a change takes place in the cultural values that afford high prestige and funding for hospitals and low prestige and, fre- quently, meager funding for long-term care, the unfortunate contrast between hospitals and nursing homes may remain. QUALITY OF CARE AND THE NEED FOR MONITORING Assurance of health care quality has long relied heavily on professional and institu- tional self-regulatory mechanisms and on monitoring efforts focusing primarily on structural and procedural measures- staff- ing patterns, requirements for obtaining staff privileges, existence of certain facilities and procedures, and the operation of institu- tional quality assurance systems. However, as our health care system becomes more competitive and as effects of economic in- centives gain recognition, it is becoming more apparent that there is a need to supplement self-regulatory mechanisms and structural indicators of quality with more ongoing monitoring mechanisms that focus on utili- zation patterns and outcome measures. For Medicare, professional review organizations (PROs) have been assigned this responsi- bility. Because of concerns about cost con- trol, as well as about quality, insurance companies and large employers are showing increased interest in monitoring patterns of care provided to beneficiaries.
138 The task is not confined to hospital care. For various reasons, continued expansion of the outpatient and Tong-term-care sectors seems inevitable. Quality-of-care concerns wiB increase as these sectors expand and for- profit participation accelerates. These qual- ity-of-care concerns wiD culminate in a de- mand for quality monitoring in all settings and under all financing mechanisms. Data systems already exist or are being developed that wiD allow monitoring of such things as · Caesarean section rates · wound infection rates after surgery · nosocomial infections · readmission rates · complication rates · fatality rates (with adjustments for di agnoses and severity) · functional status of nursing home pa- tients · health status measures in geographic areas. Because the results of one institution's quality problems may turn up at another institution's door, focusing on events that take place within the wads of particular in- stitutions is not adequate. Thus, it is highly desirable that the monitoring of quality of care be done by organizations that have ac- cess to information about the total health care utilization experience of a population. Data systems are needed that allow linkage of patients' experiences in one institution with subsequent utilization of other insti- tutions; for example, to establish the rates of hospitalization among patients who had undergone a particular surgical procedure in venous institutions within the previous six months. For the foreseeable future, the data systems that have the most potential value in monitoring the health care expe- rience of defined populations are connected with payers, including large employers. Several statewide data systems look prom- ising, but they face formidable problems in moving beyond a focus on particular insti FOR-PROFIT ENTERPRISE IN HEALTH CARE tutions and types of institutions to a focus on populations. CONCLUSION Evidence now available does not support the fear that for-profit health care is incom- patible with quality of care, nor the belief Mat public ownership might provide some assurance of quality. Among hospitals, most available (although rudimentary) measures show that investor-owned hospitals are sirn- ilar in quality to not-for-profit hospitals, and on some measures they are better. Investor- owned hospitals also appear superior to the type of hospitals they have substantially re- placed independent proprietary hospitals. However, as with many other topics in this report, the past may not predict the fixture. Some sources of support for quality care are changing as hospitals find ways to market around physicians (e.g., by selling health care plans directly to employers) and as payment methods are changed to reward institutions for finding ways to cut costs. Al- though there are many differences from hos- pitals, evidence from nursing homes is not reassuring regarding how investor-owned institutions will behave if profits require that quality be traded off against cost. Entrepre- neural trends that are stimulating the growth of diagnosis and treatment in various types of freestanding centers and through home care organizations, 7 also raise concerns about quality, particularly since quality assurance mechanisms and standards are less well de- veloped outside of hospitals. For these reasons, the committee con- cludes that the monitoring of health care quality, and the research required to in- crease the sophistication of such monitor- ing, should be major items on the nation's health policy agenda.8 (The need for more resources devoted to long-term care by the public and private sectors is also apparent, but is a subject for another study.) A worth- while goal is the development of quality as- surance systems Mat would make their results
QUALITY OF CARE available to the public, although safeguards are needed to avoid unwarranted harm to individual and institutional reputations (e.g., mortality rates for excellent surgeons or in stitutions would be misleading when their referrals include disproportionate numbers of complicated cases). The publicly funded PRO system is a place where this can begin, although the PRO program (and its PSRO ancestor) has relied to some degree on vol untary participation by physicians, who re flect the long-standing resistance among providers against public disclosure of the results of review activities (Institute of Med icine, 1981). The possibility of public disclosure of the results of privately funded quality assurance activities is limited by several factors. The loins Commission for Accreditation of Hos pitals is a vol,,~ry organization, Ended and controlled by providers who have strong his torical attachments to self-regulatory mech ani.cms in which confidentiality is a hallmark. Also, payers and large employers have little incentive to make public the results of their utilization review and quality assurance ac tivities. However, if providers fail to respond to information developed through these still embryonic monitoring activities, the threat of public disclosure will undoubtedly come into play. The possibility must also be rec ognized that institutions that have high stan dards of care and that are proud of the outcomes of the care they provide may find public relations advantages in mailing known NOTES such facts as how their surgical mortality or readmission rates compare with other insti tutions' rates. In sum, if market forces are given in creasingly free reign in health care and if cost containment pressures increase, both the interests of patients and the continuing debate over public policy in health care will require an increasing need to monitor He quality of care. Quality should be monitored in all settings, regardless of type of owner ship. Although lack of uniform standards for 139 reporting outcomes other than death make the monitoring of outcomes more difficult, research, standard development, and mon- itoring of other outcomes infection rates, readmission rates, complication rates, changes in functional status of nursing home patients are clearly needed. To assess broader changes in the health care system not only the growth of the for-profit sector, but also of multi-institutional arrangements, new freestanding centers, and vertically in- tegrated organizations, the monitoring of national and regional health status measures is essential. Important health status indices include infant and maternal mortality, mor- tality of children, and death from potentially controllable diseases (e.g., diabetes and such infectious diseases as tuberculosis). Formidable methodological, bureau- cratic, and political problems are raised by the idea of more monitoring of, and public accountability about, the performance of health care organizations. Yet, the multiple changes taking place in the organization and ethos of health care, as well as in the reg- ulatory and economic environment, should place the monitoring of the performance of health institutions and of the health status of the population high on the public policy agenda ant] on the agenda of organizations (such as third-party payers) that are in a po- sition to monitor key aspects of institutions' performances. 11he theoretical relationship between quality and for-profitinot-for-profit status is complex. Some of the concern about fior-profits and quality may result from the sense that the orientation offor-profit organizations is toward markets and marketing, although a marketing orientation increasingly pervades both for-profit and not-for-profit health care. Excessive attention to mar- ket factors Could affect aspects of quality either posi- tively or negatively. Weisbrod' and Schlesinger (1983) observe that in fields characterized by "asymmetric information" (e.g., the seller knows much more than the buyer), a potential for misrepresentation exists in the form of promises to deliver one level of quality while actually delivering lower, less costly, quality.
140 When it is difficult for the purchaser to write and en- force contract contingencies against such misrepresen- tation, consumers seek protective mechanisms such as governmental regulation or devices to limit market en- try of providers who are thought likely to exploit their informational superiority. Weisbrod and Schlesinger hypothesize that proprietary firms are more likely than not-for-profit or public firms to take advantage of an informational superiority. Medical care is a classic ex- ample of a service characterized by asymmetric infor- mation (Arrow, 1963~. However, Weisbrod and Schlesinger (1983) also recognize that patients (or, pre- sumably, their physician-agents) are aware of some as- pects of institutional performance, such as the responsiveness of floor nurses, while other aspects are more difficult or costly to monitor. Thus, they hypoth- esize that proprietary firms will perform as well as or better than not-for-profit firms on "outputs" that can be monitored by consumers, but that they will perform worse than not-for-profit firms on "outputs" that the consumer cannot readily monitor. The premise that for-profit organizations have greater sensitivity toward markets also suggests that in com- petitive markets proprietary firms will perform as well as or better than not-for-profit firms, but in noncom- petitive situations, such as when all facilities are near capacity or when a community has only one provider, they will perform worse than would a not-for-profit institution in similar circumstances. 2There are also differences between for-profit and not-for-profit hospitals regarding the specialty com- position of their medical staff (Musacchio et al., 1986), with for-profit hospitals having fewer internists and more generaVfa~nily practitioners. Interpretion of this spe- cialty composition in terms of quality must be tem- pered by several considerations. First, the relationship between specialty composition of staff and institutional quality of care has not been demonstrated. Second, no data are available regarding restrictions on hospital privileges of general/family practitioners (or other spe- cialties) in different types of hospitals. Third, the not- for-profit percentages undoubtedly reflect, in part, that many large teaching hospitals are in this category. In this regard, it is important to note that the data come from a survey of physicians, not of institutions. Because the probability that a particular hospital's staff physi- cians would appear in the sample is related to the size of its medical staff, the not-for-profit category reflects Me influence of very large institutions, of which vir- tually none are for-profit. Finally, the specialty com- position of for-profit hospitals is very similar to public hospitals. 3 Board certification is generally accepted by the medical profession as an indicator of competency at a given time in a professional career. As with all indi- cators, it must be kept in perspective when judging the overall quality-of care questions raised in this study. FOR-PROFIT ENTERPRISE IN HEALTH CARE Board certification is conveyed by the 23 specialty boards upon candidates who have completed an accredited training program and then passed a cer~ing exami- nation. 4Wil]ia~lS (1979) reports another outcome study, which found that "for-profit" hospital status had a very small, but significant, relationship to perinatal mortality in California. Unfortunately, the date were for 1960-1973, a period that preceded most of the substantial growth of the modern investor-owned hospital companies. 5However, whereas Me federal Medicare program is the most important source of dollars for hospitals, it is the state/federal Medicaid program that has become the single most important financier of nursing home care. 6A notable exception in a literature that generally does not distinguish between chain and independent proprietary nursing homes are two studies that focused or the single largest nursing home company: "Beverly Enterpnses Patient Care Record" and "Beverly En- terprises in Michigan: A Case Study of Corporate Take- over of Health Care Resources." These studies, which were done in 1983 by the Food and Beverage Trades Department of the AF~CIO and two other AFL-CIO affiliates, both suggest a disturbing pattern of quality- of-care problems in Beverly Enterprises nursing homes with respect to complaints of patient abuse and viola- tions of state regulations. However, because these re- ports were prepared as part of an innovative "corporate campaign" strategy (English, 1985) to exert pressure on Beverly Enterprises to come to the bargaining table with the unions, the fairness and accuracy of these studies' findings are open to question. Because the committee did not have the time and resources to make its own assessment of the catalog of charges raised by the union against Beverly Enterprises, it chose not to use the information in these reports. 7Quality of care in ambulatory care settings has not been examined in this chapter because of the absence of studies in which for-profit/not-for-profit comparisons are made. Ike few existing studies of ambulatory sur- gery centers are confined to the experience of partic- ular centers. Some studies with quality implications regarding for-profit and not-for-profit hemodialysis centers do exist; however, because of the specialized aspects of that technology and of Medicaid's End Stage Renal Disease Program, the committee decided not to review these studies. Researchers at the Rand Corporation have pub- lished a very valuable and detailed research agenda for monitoring the effects of Medicare's prospective pay- ment system on the quality of care (Lohr et al., 1985~. Virtually all of the ideas in the Rand report could be usefully applied to studies of ownership-related differ- ences in the quality of care.
QUALITY OF CARE REFERENCES Alexander, Jeffrey A., Michael A. Morrisey, and Ste- phen M. Shortell (1986) Physician Participation in the Administration and Governance of System and Free- standing Hospitals: A Comparison by Type of Own- ership. This volume. American Medical Association (1984) SMS Report 3Uune). Arrow, Kenneth (1963) Uncertainty and the Welfare Economics of Medical Care. American Economic Re- view 53(December):941-973. Arthur Young (1983) The Hospital Governing Board Chairman. Los Angeles: Arthur Young. Donabedian, Avedis (1969) A Guide to Medical Care Adrrunistration. Vol. II: Medical Care Appraisal Quality and Utilization. Washington, D.C.: American Public Health Association. English, Carey W. (1985) When Unions Turn Tables on the Bosses. U.S.News ~ World Report (February 4~:69-70. Flood, A. B., W. R. Scott, W. Ewy, and W. H. Forrest, Ir. (1982) E£ectiveness in Professional Orga- nizations: The Impact of Surgeons and Surgical Staff Organization on the Quality of Care in Hospitals. Health Services Research 17(Winter):341-366. Gaumer, Gary (1986) Medicare Patient Outcomes and Hospital Organizational Mission. This volume. Hawes, Catherine, and Charles D. Phillips (1986) The Changing Structure of the Nursing Home Industry and the Impact of Ownership on Quality, Cost, and Access. This volume. Institute of Medicine (1981) Access to Medical Re- view Data. Washington, D.C.: National Academy Press. Institute of Medicine (1986) Improving the Quality of Care in Nursing Homes. Washington, D.C.: Na- tional Academy Press. Lee, Y. S. (1984) Nursing Homes and Quality of Health Care: The First Year Result of an Outcome- Oriented Survey. Journal of Health and Human Re- sources Administration 7(Summer):32-60. Lohr, Kathleen N., et al. (1985) Impact of Medicare Prospective Payment on the Quality of Medical Care: A Research Agenda. Santa Monica, Calif.: Rand Cor- poration. Longo, Daniel R., Gary A. Chase, Lynn A. Ahlgren, James S. Roberts, and Carol S. Weisman (1986) Com ]4] pliance of Multihospital Systems with Standards of the Joint Commission on Accreditation of Hospitals. This volume. Left, Harold (1981) Health Maintenance Organiza- tion: Dimensions of Performance. New York: Wiley. Mornsey, Michael A., Jeffrey A. Alexander, and Ste- phen M. Shortell (1986) Medical Staff Size, Hospital Privileges, and Compensation Arrangements: A Com- panson of System Hospitals. This volume. Musacchio, Robert A., Stephen Zuckerman, Lynn E. Jensen, and Larry Freshnock (1986) Hospital Own- ership and the Practice of Medicine: Evidence from the Physician's Perspective. This volume. Neuhauser, Duncan (1971) The Relationship Be- tween Administrative Activities and Hospital Perfor- mance. Research Series No. 28. Chicago, Ill.: University of Chicago Center for Health Administrative Studies. Placek, Paul L., Selma Tassel, and Mary Moien (1983) Caesarean Section Delivery Rates: United States, 1981. AmencanJournal of Publ* Health 73(August):861-862. Scott, W. Richard, Ann M. Flood, and Wayne Ewy (1979) Organizational Determinants of Services, Qual- ity, and Cost of Care in Hospitals. Milbank Memorial Fund Quarterly 57(Spring):234 264. Shortell, Stephen M., and James P. LoGerfo (1981) Hospital Medical Staff Organization and Quality of Care: Results for Myocardial Infarction and Appendectomy. Medical Care l9(October):1041. Shortell, Stephen M., Selwyn W. Becker, and Dun- can Neuhauser (1976) The Effects of Management Prac- tices on Hospital Efficiency and Quality of Care. Pp. 90-107 in Stephen M. Shortell and Montague Brown (eds.), Organizational Research in Hospitals. Chicago, Ill.: Blue Cross Associations. Vladeck, Bruce (1980) Unloving Care. New York: Basic Books. Weisbrod, Burton A., and Mark Schlesinger (1983) Public, Private, Nonprofit Ownership and the Re- sponse to Asymmetric Information: The Case of Nurs- ing Homes. Discussion Paper #209. Madison: University of Wisconsin Center for Health Economics and Law. Williams, Ronald L. (1979) Measuring the Effec- tiveness of Perinatal Medical Care. Medical Care 17(February):9~110. Williams, Ronald L., and Peter M. Chen (1983) Con- trolling the Rise in Caesarean Section Rates by the Dissemination of Information from Vital Records. AmencanJournal of Public Health 73(August):863-867.