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.
For-Profit Enterpnse in Health Care. 1~6. Nabonal Academy Press, Washington, D.C. Physician Participation in the Adniinistra~don and Governance of System and Freestanding Hospices: A Comparison by Type of Ownership Jeffrey A. Alexander, Michael A. Mornsey, and Stephen M. Shortell INTRODUCTION This paper examines the type and extent of physician participation in the administration and governance of hospitals in multihospital systems. Particular emphasis is given to phy- sician participation in investor-owned chain hospitals relative to freestanding hospitals and other system-affiliated institutions. The advent of prospective pricing, coupled with the rise of multi-institutional arrange- ments in the hospital industry is thought to have major implications for the role ofthe phy- sician in the hospital. The nature of many of these changes, however, is still open to em- pirical question. Brown (1979) and Reynolds and Studen (1978) suggest that hospital con- solidation into systems may promote more for- malizec] methods for securing physician input into hospital decision-making processes.22 Others have argued, however, that physician power and influence in hospital policymaking and administration will be reduced in the con- text of multi-institutional arrangements and the "corporatization" of medicine.34 Physician involvement in hospital decision making represents a potentially important is- sue for several reasons. First, several national groups (i.e., American Hospital Association [AHA], loins Commission on Accreditation of Hospitals MICAH], American Medical Associ- ation [AMA]) perceive and endorse physician Dr. Morrisey is in the Department of Health Care Organization and Policy and Dr. Alexander is in the Department of Health Services Administration, both at the University of Alabama at Birmingham. When this paper was written they were with the Hospital Research and Educational Trust, Chicago, Illinois. Dr. Shortell is with the J. L. Kellogg Graduate School of Management and Center for Health Services and Pol- icy Research at Northwestern University, Evanston, Illinois. 402 participation in hospital decision making as an important mechanism for conflict resolution among the administration, board members, and medical staff.5 Second, such participation is seen to foster an institution-wide perspective among medical staff members, of particular importance under diagnosis-related group (DRG) reimbursement. Finally, studies by Neuhauser (1971), Shortell et al. (1976), Shor- tel1 and LoGerfo (1981), and Morlock et al. (1979) suggest that greater medical staff par- ticipation in governance is associated with higher quality hospital care.6~9 In light of the above, this study examines a series of structural variables related to physi- cian involvement in hospital decision making. Two categories of involvement are considered: (~) physician participation in hospital gover- nance, and (2) physician participation in hos- pital management. The primary group under investigation are those hospitals owned or leased by investor- owned multihospital systems (IO hospitals). On a series of physician decision-making vari- ables, these hospitals are compared to five other hospital groups: (1) freestanding hospitals or those not affiliated with a multihospital sys- tem; (2) hospitals owned or leased, by secular nonprofit multihospital systems; (3) hospitals owned, leased, or sponsored by religious mul- tihospital systems; (4) hospitals owned or leased by public multihospital systems; and (5) hos- pitals contract-managed by all of the afore- mentioned system types. SAMPLE The sample for this study was composed of 3,027 community hospitals located in the 48 contiguous states and the District of Colum- bia. The sample hospitals were chosen by ran- dom selection and represent 25 percent of all
PHYSICIAN PARTICIPATION IN GOVERNANCE community hospitals, augmented with addi- tional, randomly selected hospitals in 22 states. These states were primarily those with man- datory or voluntary rate review programs. Sample hospitals were sent a six-page sur- vey questionnaire related to medical staff or- ganization. Of the 3,027 sample hospitals, 2,065 responded to the survey, a response rate of 69.9 percent. These hospitals represented the final sample for this investigation. It is important to note that the hospitals in this study are not representative of all com- munity hospitals insofar as sample represen- tation was greater for New England, Middle Atlantic, and West North Central regions than for the South East, East South Central, and West South Central regions. Sample hospitals were also more likely to be involved in teach- ing activities and less likely to be investor- owned than the population of community hos- pitals.~° DATA Data for this investigation were obtained from these sources: (1) the 1982 AHA Survey of Hospital Medical Staffs; (2) the 1982 AHA An- nual Survey of Hospitals; and (3) the 1982 AHA Validation Survey of Multihospital Systems. The medical staff survey provided information on the type and extent of physician/medical participation in hospital decision making. The annual survey was used to obtain data on se- lected hospital characteristics (e.g., size, teaching involvement, and regional location). The validation survey indicated whether or not a hospital was part of a multihospital system, the ownership status of that system and the type of hospital affiliation with the system (i.e., owned, leased, sponsored, or contract-man- aged). MEASUREMENT Physician participation in hospital gover- nance was measured by four variables: (1) total number of physician members on the hospital governing board; (2) whether or not physician board members have voting privileges; (3) whether or not physicians serve on the board executive committee and; (4) whether or not the hospital chief executive officer (CEO) is a voting member of the hospital governing board. 403 The latter item, while not a direct indicator of physician involvement on the board, is indic- ative of participation and influence by the ad- ministrative component ofthe hospital, possibly complementing physician influence. Physician involvement in hospital manage- ment activity was assessed by six variables: (1) whether or not any staff physicians hold salar- ied positions as part of the hospital's manage- ment team; (2) whether or not the chief of the hospital medical staff is compensated by the hospital; (3) whether or not the director of medical education is compensated by the hos- pital; (4) whether or not the hospital has a medical staff committee for long-range plan- ning; (5) whether or not the hospital has a medical staff committee for cost containment/ cost awareness; and (6) the total number of medical staff committees in the hospital. Because many of the outcome measures de- scribed above may be influenced by factors other than multihospital system participation, several control variables were incorporated in the analysis. These included hospital size as measured by the total number of beds set up and staffed for use; regional location of hos- pital, indicated by whether the hospital is sit- uated in the Northeast, South, North Central, or Western areas of the country; size of stan- dard metropolitan statistical area (SMSA), measured by seven population categories ranging from 0 to over 2.5 million; teaching involvement, assessed by the number of house staff in the hospital; and system size as indi- cated by the number of owned, leased, spon- sored, or managed hospitals in the system. Hospitals not affiliated with a system were as- signed a value of 0 on this variable. Descriptive statistics for the dependent, in- dependent, and control variables are pre- sented in Table 1. ANALYTIC APPROACH Zero-order comparisons among investor- owned and other hospital categories were per- formed using chi-square tests of significance in the case of dichotomous variables, and d~f- ference-of-means l-tests in the case of contin- uous, dependent variables. IO hospitals were compared separately to each of the other five hospital categories. Following zero-order comparisons, all de
404 FOR-PROFIT ENTER[FUSE IN HEALTH CARE CABIN 1 Descriptive Statistics for Hospital Sample (N = 2,067) Vanables Standard Number Mean Deviation in Samplea . .. . Independent Variables Freestanding hospital.71.45 1,469 System characteristics System control Religious nonprofit.12.32 247 Secular nonprofit.03.18 70 Investor-owned.06.24 126 Public.01.07 11 Contract-managed.07.25 144 System size21.0072.28 Hospital characteristics Number of beds190.76179.53 Regional location Northeast.23.42 468 South.26.44 542 Norm Central.32.47 660 West.19.39 397 SMSA sizes Non-SMSA.45.50 929 Under 100,000.02.12 32 100,000-250,000.09.28 178 250,000-500,000.09.29 187 500,000-1 million.08.27 160 1 million-2.5 million.14.35 297 Over 2.5 million.14.34 284 House staff11.6050.14 Dependent Variables Whether M.D.s receive salaries as administrators.16 Whether director of medical education paid.19 Whether chief of staff paid Whether chief executive officer is voting member of governing board Whether M.D.s on governing board of executive committee Whether M. D. governing board members have voting rights Whether hospital has committee for cost containment/awareness Whether hospital has committee for medical staff long-range planning .36 .39 .07 .26 34 . .35 .47 .48 323 392 145 697 720 .70 .46 1,437 .16 .37 .28 .45 329 584 a For dichotomously scored variables only. bSize of standard metropolitan statistical area (SMSA).
PHYSICIAN PARTICIPATION IN GOVERNANCE TABLE 2 Comparison of Hospitals in For-profit Systems with Other Hospital TvDes Number of Physicians on Governing Boards . , ~ Hospitals in Hospitals in Hospitals in Hospitals in For-profit Freestanding Religious Nonprofit Public Contract Systems Hospitals Systems Systems Systems Managed _ Mean Mean Mean Mean Mean Mean N (S.D.) ~(S.D.) N (S.D.) N (S.D.) N (S.D.) N (S.D.) 125 3.83 1,442 (2.32) .86*** 24 2. 12) 1.76*** 68 (1.66) 2. 13*** 7 (1.95) . . _ . _ _ _ **AT significant at p c .01 when compared with hospitals in for-profit systems. **T significant at p c .05 when compared win hospitals in for-profit systems. pendent variables were subjected to multi- variate analysis to assess the impact of hospital and environmental controls on the bivariate relationships. Two regressions were per- formed for each dependent variable. The first compared all system categories to freestanding hospitals. The second compared IO system hospitals to all other system categories and freestanding hospitals. In the case of dichoto- mous depenclent variables, multiple logistic regression was performed. Ordinary least- squares regression was used for continuous de- pendent variables. RESULTS OF COMPARISONS Governance Results of the zero-order comparisons be- tween IO hospitals and other hospital CLOUDS are presented in Tables 2-11. Results of the multivariate analysis are contained in the ap- pendix. In general, our findings suggest that TABLE 3 Comparison of Hospitals in For-profit Systems with Other Hospital Types: Do Hospital Physician Governing Board Members Have Voting Privileges? Yes No Total 405 7.00** 142 ~ 42~*' (14.~) (1.77) physician participation in hospital governance is greater in IO hospitals than in the five com- parison groups. Table 2 indicates that with the exception of public system hospitals, IO hos- pitals have on average a greater number of physicians (X = 3.83) serving on their boards than the other system hospital groups or free- standing hospitals. These results are main- tained even after the introduction of controls for system size, hospital size, region, SMSA size, and teaching involvement. It is important to note that these board po- sitions can y some influence insofar as 91 per- cent of the sample IO hospitals indicated that physician board members had voting privi- leges on the board (Table 3~. Zero-order com- parisons between IO hospitals and over groups also suggest that physician board members are more likely to have voting privileges in IO hospitals than in the other five hospital groups. With the introduction of the control variables, however, only differences between the IO and He secular nonprofit and public system hos- pitals are sustained. Hospitals in Hospitals in Hospitals in Hospitals in For-profit Freestanding Religious Nonprofit Public Contract Systems Hospitals Systems Systems Systems Managed Response NMeanNMean NMean ~ Mean N Mean NMean . . . _ 115 91.27 991 67.46 193 78.14 5071.43 545.45 83 57.64 11 8.73 478 32.54*** 54 21.86*** 2028.57*** 654.55*** 61 42.36*** 126 100.040 1,469 100.00 . 247 100.00 70100.00 11100.00 144 100.00 _ . . $**Chi-square significant at p c .01 when compared with hospitals in for-profit systems.
406 FOR-PROFIT ENTERPRISE IN HEALTH CARE TABLE 4 Comparison of Hospitals in For-profit Systems ninth Other Hospital Types: Are There Physicians on the Hospital Governing Board Executive Committee? Hospitals in For-profit Systems Freestanding Hospitals Hospitals in Religious Systems Response N % N % N % Yes No Total Hospitals in Nonprofit Systems N % Hospitals in Public Systems N % Contract- Managed N % 57 45.24 524 35.67 65 26.32 22 31.43 4 36.36 4833.33 69 54.76 945 64.33** 182 73.68*** 48 68.57* 7 68.64 9666.67* 126 100.00 1,469 100.00 247 100.00 70 100.00 11 100.00 144100.00 ***Chi-square significant at p ' .01 when compared with hospitals in for-profit systems. **Chi-square significant at p c too when compared with hospitals in for-profit systems. *Chi-square significant at p c .10 when compared with hospitals in for-profit systems. Table 4 suggests that physicians are more likely to be members of the important board executive committee in IO hospitals relative to all other hospital groups except public sys- tem hospitals. However, only the difference between IO hospitals and religious system hospitals holds after the introduction of the condos variables. Finally, Table 5 indicates strong involvement of IO hospital CEOs in hospital governance. Eighty percent of the sample IO hospital CEOs participated as vot- ing members of their governing boards. This finding is particularly striking when compared to the low percentage of CEO board mem- bership in freestanding hospitals (26 percent), secular nonprofit system hospitals (31 per- cent), and public system hospitals (9 percent). As the percentage differences would indicate, these IO hospital CEOs are significantly more likely to be members of their hospital govern- ing board than the CEOs of the other five hospital groups. The introduction of hospital and environmental controls does not alter these differences. Management Comparisons related to physician involve- ment in hospital management suggest a pat- tern of results opposite that of physician participation in hospital governance. In gen- eral, physicians in IO hospitals appear to be less integrated into hospital managerial activ- ities than physicians in other hospital catego- nes. Table 6, for example, indicates that physi- cians in IO hospitals are significantly less likely to hold salaried positions as part of the hos- pital's administrative team. Only 2 percent of the IO hospitals in the sample indicated that their physicians held such positions, by far the TABLE 5 Comparison of Hospitals in For-profit Systems win Other Hospital Types: Is the Hospice CEO a Voting Member of the Governing Board? . Hospitals in For-profit Freestanding Systems Hospitals Hospitals in Religious Systems Hospitals in Hospitals in Nonprofit Public Contract Systems Systems Managed Response N % N % N % N % N % N % Yes No Total 101 80.1638526.21 25 19.841,08473.79*** 126 100.001,469100.00 162 65.59 22 31.43 1 9.09 2618.06 85 34.41*** 48 68.57*** 10 90.91*** 11881.94*** 247 100.00 70 100.00 11 100.00 144100.00 ***Chi-square significant at p ' .01 when compared with hospitals in for-profit systems.
PHYSICIAN PARTICIPATION IN GOVERNANCE 407 TABLE 6 Comparison of Hospitals in For-profit Systems with Other Hospital Types: Do Physicians lIold Salaried Positions as Part of Hospital's Administrative Team? . . Hospitals in Hospitals In Hospitals in Hospitals in For-profit Freestanding Religious Nonprofit Public Contract Systems Hospitals Systems Systems Systems Managed Response N % N % N % N % N % N % Yes 3 2.38 230 15.66 52 21.05 20 28.57 5 45.45 13 9.03 No 123 97.62 1,239 84.34*** 195 78.95*** 50 71.43*** 6 54.55*** 131 90.97** Total 126 100.00 1,469 100.00 247 100.00 70 100.00 11 100.00 144 100.00 ***Chi-square significant at p c .01 when compared with hospitals in for-profit systems. **Chi-square significant at p 5 .05 when compared with hospitals in for-profit systems. lowest proportional representation among the six hospital groups. These finclings remain un- altered, holding constant the five hospital and environmental control variables. Consistent with these findings, key medical staff officers in IO hospitals are also less likely to be com- pensated by the hospital (see Tables 7 and 81. Specifically, chiefs of staff and directors of medical education in IO hospitals are signifi- cantly less likely than those in all comparison groups, except contract-managed hospitals, to be paid by the hospital. It should be noteil, however, that when hos- pital and environmental variables are held constant, chiefs of staffin IO hospitals are less likely than those in secular nonprofit system hospitals to be compensated. Medical educa- tion directors in IO hospitals are less likely than their counterparts in freestanding hos- pitals and religious system hospitals to be com- pensated after introducing-~e control variables. A second arena of administrative activity for physicians is reflected in the medical staffcom- mittee structure of the hospital. Table 9 sug- gests that with an average of 11, IO hospitals have fewer medical staff committees than all comparison groups except contract-managed hospitals. These differences, with the excep- tion of hospitals in secular nonprofit systems, hold even after controlling for other hospital and environmental characteristics. We also examined the presence or absence of two medical staficommittees that have par- ticular relevance for hospital administration: cost containment/awareness anil long-range planning. IO hospitals appear to be signifi- cantly less likely to have medical staff com- mittees on long-range planning (Table 10) relative to all other groups except public sys- tem and contract-managed hospitals. Few sig- nificant differences were obtained between IO hospitals and other hospital groups on the presence of a cost containmentlawareness committee (Table 111. These differences are TABLE 7 Comparison of Hospitals in For-profit Systems wig Other Hospital Types: Is Hospital Chief of Staff Compensated? . . . Hospitals in Hospitals in Hospitals in Hospitals in For-profit Freestanding Religious Nonprofit Public Contract Systems Hospitals Systems Systems Systems Managed . Response N % N % N % N % N % N % Yes No Total 126 100.00 1,469 100.00 2 1.59103 124 98.411,366 7.01 208.10 13 92.99** ~791.90** 57 ~ 247100.00 70 100.00 18.57 R1 *** 218.185 3.47 981.82**139 96.53 11100.00144 100.00 ***Chi-square significant at p ' .01 when compared with hospitals in for-profit systems. **Chi-square significant at p < .05 when compared with hospitals in for-profit systems.
408 FOR-PROFIT ENTERPRISE IN HEALTH CARE TABLE 8 Comparison of Hospitals in For-profit Systems win Other Hospital Types: Is Me Director of Medical Education Compensated? Hospitals in Hospitals in Hospitals in Hospitals in For-profit Freestanding Religious Nonprofit Public Contract Systems Hospitals Systems Systems Systems Managed Response N % N To N % N % N % N % Yes No Total 126 100.00 1,469 100.00 9 7.14286 117 92.861,183 19.47 67 27.13 16 80.53*** 180 72.87*** 54 . ~= 247 100.00 70 100.00 22.86 77 1 A*** ***Chi-square significant at p c .01 when compared with hospitals in for-profit systems. eliminated with the introduction of the control variable set. Summary In summary, IO hospitals, relative to free- standing hospitals, appear to have greater phy- sician representation on hospital governing boards although these physicians are no more or less likely to have voting privileges or to serve on the board executive committee than physicians on freestanding hospital boards. Relative to hospitals in other system cate- gories, IO hospitals also have greater physician representation on hospital governing boards. Only hospitals in public systems have a higher mean number of physicians on the board. In addition, physician board members of IO hos- pitals are more likely to have voting privileges than those in secular nonprofit and public sys- tems and are more likely to serve on the ex- ecutive committee of the board relative to physicians in religious system hospitals. 545.459 6.25 654.55***135 93.75 11100.00144 100.00 It is important to note that CEOs of IO hos- pitals are more likely than those in freestand- ing hospitals to be voting members of the hospital board. This may suggest a more in- fluential position of the IO hospital CEO on the board relative to physicians since no dif- ferences were obtained between IO hospitals and freestanding hospitals on physician voting . . pI.lVl. Ages. This same pattern holds when comparing IO hospitals win hospitals in over system groups. IO hospitals tend to be significantly more likely to have Heir CEOs serve as voting members of the board than hospitals in religious, sec- ular, nonprofit, or public systems. In terms of physician involvement in hos- pital a~ninistra~on, IO hospitals are less likely to have physicians as salaried administrators and to have fewer medical staff committees than freestanding hospitals. When controlling for other hospital and environmental attri- butes, however, no significant differences were obtained between IO and Freestanding hos TABLE 9 Comparison of Hospitals in For-profit Systems with Other Hospital Types: Number of Medical StaE Committees Hospitals in Hospitals in Hospitals in Hospitals in For-profit Freestanding Religious Nonprofit Public Contract Systems Hospitals Systems Systems Systems Managed Mean Mean Mean Mean Mean Mean N (S.D.) N (S.D.) N (S.D.) ~ (S.D.) N (S.D.) N(S.D.) 125 11. 14 1,442 13.55*** 241 15.97*** 68 14.06** 7 16.29* 1429.68* (6.26) (9.36) (8.70) (10.89) (6.58) (6.96) ***T significant at p s .01 when compared with hospitals in for-profit systems. **T significant at p s .05 when compared with hospitals in for-profit systems. *T significant at p ' .10 when compared with hospitals in for-profit systems.
PHYSICIAN PARTICIPATION IN GOVERNANCE TABLE 10 Comparison of Hospitals in For-profit Systems with Other Hospital Types: Does He Hospital Have a Committee for Medical Staff Long-Range Planning? 409 Hospitals in Hospitals in Hospitals in Hospitals For-profit Freestanding Religious Nonprofit in Public Contract Systems Hospitals Systems Systems Systems Managed Response N % N % N % N % N % N % Yes 19 15.08 436 29.68 77 31.17 22 31.43 3 27.27 27 18.75 No 107 84.92 1,033 70.32*** 170 68.83*** 48 68.57*** 8 72.73 117 81.25 Total 126 100.00 1,469 100.00 247 100.00 70 100.00 11 100.00 144 100.00 ***Chi-square significant at p s .01 when compared with hospitals in for-profit systems. pitals on compensation for key medical staff members or the existence of a medical staff committee on cost containment/awareness. However, IO hospitals are less likely than free- standing hospitals to have a medical staEcom- mittee on long-range planning. In general, comparisons of IO hospitals with other system groups on variables related to administrative involvement vaned by com- parison groups. For example, IO hospitals were less likely to compensate their chief of medical stallonly relative to secular nonprofit hospitals and less likely to compensate their director of medical education when compared to religious system hospitals. However, IO hospitals, rel- ative to all other system types, were less likely to have physicians salaried as part of the hos- pital's administrative team. DISCUSSION AND TRIPLICATIONS The findings on physician involvement in hospital management and governance are of interest primarily in terms of their fixture im- plications. Increasingly, hospital trustees, managers, and medical stalimembers are going to have to make rli~cult trade-off decisions involving the volume and mix of services to offer to different kinds of potential consumers. These decisions will increasingly involve eth- ical considerations. Hospitals wilIbe caned upon to balance, in some way, health care as an economic good with the conception of health care as a social good. Physician and medical staff input into this process is critical and, therefore, the way in which hospitals solicit this input becomes important. The new en- vironment requires a more integrated man- agement structure involving closer relationships among trustees, hospital managers, and phy- sicians in order to males decisions that balance economic and social considerations to the ex- tent possible. The present findings are of interest because they suggest that, to date, IO hospitals have solicited physician input in different ways from TABLE 11 Comparison of Hospitals in For-profit Systems with Other Hospital Types: Does Hospital Have Committee for Cost Containment/Cost Awareness? Hospitals in Hospitals in Hospitals in Hospitals in For-profit Freestanding Religious Nonprofit Public Contract Systems Hospitals Systems Systems Systems Managed Response N % N % N% N % N% N % l Yes 13 10.32 238 16.20 4618.62 12 17.14 436.36 16 11.11 No 113 89.68 1,231 83.80 201 81.38* 58 82.86 7 63.64* 128 88.89 Total 126 100.00 1,469 100.00 247 100.00 70 100.00 11 100.00 144 100.00 *Chi-square significant at p s .10 when compared with hospitals in for-profit systems.
410 other kinds of hospitals, namely, greater in volvement at the individual hospital board level than at the managerial level of the organiza tion. In contrast, other hospitals have essen tially done the reverse. No studies exist to suggest that one type of involvement is su perior to another. It is likely that both types NOTES will be increasingly needed in the fixture. To the extent that this is true, one might expect to find the better performing IO hospitals moving to involve their physicians somewhat more in the managerial activities of the hos pital, and the better performing voluntary hos pitals moving toward greater involvement of their physicians in the governance and poli cymaking activities of the organization. This kind of"convergence" would be consistent with those who suggest that IO and nonprofit sys tems will become more similar in the fixture. It is clearly an issue for fixture research with the present findings serving as usefi~1 baseline data. A final issue relevant to both investor-owned and voluntary systems is the level of the system at which physician involvement occurs. The findings discussed above have pertained to physician involvement at the individual hos pital level. There are data that suggest that physicians are less involved at the divisional and corporate levels of systems.4 22 To the ex tent that systems, whether investor-owned or voluntary, become larger and more central ized, physician involvement at these higher corporate levels would appear to be essential to addressing the larger socioeconomic trade offissues that will exist. Several systems, both investor-owned and voluntary, are already moving toward greater physician involvement at higher levels, and fixture research should continue to document this trend and examine its impact on corporate decision malting. ACKNOWLEDGMENT This paper was Minded in part by contract 0551-4176 Tom the Institute of Medicine, Na FOR-PROFIT ENTERPRISE IN HEALTH CARE tional Academy of Sciences. The opinions and conclusions expressed herein are solely those of the authors. ~Brown, M. (1979) Multihospital systems: Implica- tions for physicians. The Hospital Medical Staff (Au- gust):2. Reynolds, J., and A. E. Stunden (1978~1le orga- nization of not-for-profit hospital systems. Health Care Management Review 3:23. 3Starr, P. (~1982) The Social Transformation of Amer- ican Medicine. New York: Basic Books. gMorlock, L. L., J. A. Alexander, and H. Hunter (1985) Governing board-CEO-medical staff rela- tions in multi-institutional arrangements. Medical Care 23:1193-1213. s The Hospital Med*al Staff (1978) AHA issues guidelines on physician involvement in hospital gov- ernance. 7:15. 6Nenhauser, D. (1971) The relationship between ad- ministrative activities and hospital performance. Re- search Series No. 28. Center for Health Administrative Studies, University of Chicago. 7Shortell, S. M., S. W. Becker, and D. Nenhauser (1976) The effects of management practices on hospital efficiency and qualify ofcare. In Organization Research in Hospitals, S. M. Shortell and M. Brown, eds. Chi- cago: Blue Cross Association. 8Shortell, S. M., and J. P. LoGerfo (1981) Hospital medical staE organization and quality of care: Results for myocardial infarction and appendectomy. Medical Care 19:1041. 9Morlock, L. L., C. A. Na~anson. S. D. Horn. and D. N. Schumcher (1979) Organizational factors asso- ciated with quality of care in 17 general acute care hospitals. Paper presented at the annual meeting ofthe Association of University Programs in Health Admin- istration, Toronto. 2°Noie, N., S. Shortell, and M. Mornsey (1983) A survey of hospital medical stabs-Part 1 (1983) Hos- pitalsJAHA 57~1~. Alexander, J., and D. Cobbs (1984) MDs and mul- tihospital systems. The Hoszntal Medical Staff 13~6~:8- 14.
APPENDIX TABLE A-1 Number of Physicians on the Governing Board System Control Coefficients Relative to Freestanding Hospitals Investor-Owned Systems Standard Standard Variables Coefficient Error Coefficient Error Freestanding hospital -1.55 .23*** System characteristics System control Religious nonprofit-.26 .14*- 1.81 .26*** Secular nonprofit-.06 .25- 1.49 .33*** Investor-owned1.55 .23*** Public4.23 .80***2.68 .83*** Contract-managed- . 29 .21-1.85 . 25* ** System size (100s).22 .08***.22 .08*** Hospital characteristics Number of beds (100s).25 .04***.25 .04*** Regional location Northeast.42 .15***.42 .15*** South- .27 .14*- .27 .14* North Central-.49 .13***- .49 .13*** SMSA sizea Under 100,000.40 .37.40 .37 100,000-250,000.46 . 17*.46 . 17* 250, 000-500, 000. 89 . 17** *.89 . 17* * * 500,000-1 million.81 .19***.81 .19*** 1 million-2.5 million1.00 .15***1.00 .15*** Over 2.5 million1.47 .16***1.47 .16*** House sta8~100s)-.54 .11***- .54 .11*** Constant1.04 .12***2.59 .25*** Model Statistics R2 .23 .23 N 2,025 2,025 aSize of standard metropolitan statistical area (SMSA). ***Chi-square significant at p ' .01. **Chi-square significant at p s .05. *Chi-square significant at p ' .10. 417
412 FOR-PROFIT ENTERPRISE IN HEALTH CARE TA;BLE A-2 Do Hospital Physician Governing Board Members Have Voting Privileges? System Control Coefficients Relative to- Freestanding Investor-Owned Hospitals Systems Standard Standard Variables Coefficient Error Coefficient Error Freestanding hospital -1.12 .35*** System characteristics System control Religious nonprofit.44 .18**- .68 .38* Secular nonprofit.30 .29- .82 .44* Investor-owned1.12 .35*** Public-1.47 .75**- 2.59 .82*** Contract-managed- . 58 . 23* *-1. 70 . 38 * * * System size (100s).46 .12***.46 .12*** Hospital characteristics Number of beds (100s).30 .05***.30 .05*** Regional location Northeast1.31 .19***1.31 .19*** South.18 .16.18 .16 North Central-.14 .15- .14 .15 SMSA sizes Under 100,000.78 .51.78 .51 100, 000-250, 000. 40 . 20* *.40 . 20* * 25O, 000-500, 000.89 .22***.89 .22*** 500,000-1 million.28 .23.28 .23 1 million-2.5 million.95 .19***.95 .19*** Over 2.5 million.39 .20**.39 .20** House staE(100s)-.88 .14***-.88 .14*** Constant- . 25 .13*.88 . 36** Model statistics Pseudo R .35 ·35 N 2,067 2,067 aSize of standard metropolitan statistical area (SMSA). ***Chi-square significant at p c .01. **Chi-square significant at p c .05. *Chi-square significant at p c .10.
PHYSICIAN PARTICIPATION IN GOVERNANCE TABLE A-3 Are There Physicians on the Governing Board Executive Committee? System Control Coefficients Relative to- . Freestanding Hospitals _ Investor-Owed Systems Standard Standard VariablesCoefficient ErrorCoefficient Error Freestanding hospital - .17 .24 System characteristics System control Religious nonprofit-.57 .16***- .74 .28*** Secular nonprofit-.26 .27- .43 .35 Investor-owned.17 .24 Public- .49 .67- .66 .71 Contract-managed- . 03 .23- .20 .26 System size (100s).12 .09.12 .09 Hospital characteristics Number of beds (100s).15 .04***.15 .04*** Regional location Northeast.31 . 15**.31 .15** South- .03 .15- .03 .15 North Central-.31 .15**- .31 .15** SMSA sizea Under 100,000.03 .40.03 .40 100,000-250,000.55 . 18***.55 . 18*** 250,000 500,000.59 .17***.59 .17*** 500,000-1 million.42 .19**.42 .19** 1 million-2.5 million.59 .16***.59 .16*** Over 2.5 million.55 .16***.55 .16*** House staEflOOs)-.29 .12**- .29 .12** Constant- 1.13 .13***- .96 .26*** Model Statistics Pseduo R.19 .19 N 2,067 2,067 a Size of standard metropolitan statistical area (SMSA). ***Chi-square significant at p c .01. **Chi-square significant at p s .05. 413
414 FOR-PROFIT ENTERPRISE IN HEALTH CARE TABLE A-4 Is the Hospital Chief Executive Officer a Voting Member of the Governing Board? System Control Coefficients Relative to Freestanding Hospitals Variables Standard Coefficient Error Investor-Owned Systems Standard Coefficient Error -2.67 Freestanding hospital System characteristics System control Religious nonprofit Secular nonprofit Investor-owned Public Contract-managed System size (100s) Hospital characteristics Number of beds (100s) Regional location Northeast South North Central SSISA sizea 1.82 .32 2.67 -2.24 - .15 .05 .23 .36 -.16 .34 .16*** .29 .30*** 1. 11** .28 - .11 .04*** .17*** .17 .65** - -.86 -2.35 -4.92 -2.82 .05 .23 .36 - .16 34 . .30*** .32*** .39*** 1.14*** .33*** .11 .04*** .17*** .17 .16** Under 100,0001.06 .41***1.06 .41*** 100,000-250,000.28 .20.28 .20 250,000-500,000.28 .20.28 .20 500,000- 1 million.45 .21 * *.45 .21 * * 1 million-2.5 million.51 .17***.51 .17*** Over 2.5 million.59 .18***.59 .18*** House staff (100s)- .38 .13***- .38 .13*** Constant-1.89 .15***.79 .31** Model Statistics Pseudo R .42 .42 N 2,067 2,067 aSize of standard metropolitan statistical area (SMSA). ***Chi-square significant at p ' .01. **Chi-square significant at p ' .05.
PHYSICIAN PARTICIPATION IN GOVERNANCE TABLE A-5 Do Any Physicians Hold Salaried Positions as Part of HospitaT's Administrative Team? System Control Coefficients Relative to Freestanding Investor-Owned Hospitals Systems Standard Standard Variables Coefficient Error Coefficient Error Freestanding hospital 1.40 .64* System characteristics System control Religious nonprofit.39 .19**1.79 .65*** Secular nonprofit.88 .31***2.28 .69*** Investor-owned-1.40 . 64** Public.27 .761.67 .98* Contract-managed.11 .381.51 .67*** System size (loos)- .14 .18- .14 .18 Hospital characteristics Number of beds (lOOs). 19 . 04* * *. 19 . 04* ** Regional location Northeast.87 .21***.87 .21*** South- .32 .23- .32 .23 North Central-.06 .21- .06 .21 SMSA sizer Under 10G,0001.19 .47**1.19 .47** 100,000 250,000.65 .27**.65 .27** 250,000-500,0001. 12 .24***1. 12 .24*** 500,000 1 million1.12 .25***1.12 .25*** 1 million-2.5 million.91 .23***.91 .23*** Over 2.5 million1.11 .23***1.11 .23*** House staE (loos).19 .13.19 .13 Constant-3.07 .21***- 4.47 .66*** Model Statistics Pseudo R .38 .38 N 2,067 2,067 aSize of standard metropolitan statistical area (SMSA). ***Chi-square significant at p s .01. **Chi-square significant at p s .os. *Chi-square significant at p c .10. 475
416 FOR-PROFIT ENTERPRISE IN HEALTH CARE TABLE A-6 Is the Hospital Chief of Staff Compensated? System Control Coefficients Relative to Freestanding Investor-Owned Hospitals Systems Variables Standard Standard CoefFicient Error Coefficient Error _ Freestanding hospital .82 .79 System characteristics System control Religious nonprofit.17 .27.99 .81 Secular nonprofit1.11 .37***1.94 .84 Investor-owned-.82 .79 Public-.86 .97-.03 1.24 Contract-managed.29 .531.12 .87 System size (100s)-.51 .37- .51 .37 Hospital characteristics Number of beds (100s).16 .05***.16 .05 Regional location Northeast.50 . 28*.50 . 28 South- .10 .31- .10 .31 North Central-.20 .30- .20 .30 SMSA sizea Under 100,000- 6.15 24.96-6.15 24.96 100, 000-250, 000. 32 . 54.32 .54 250,000-500,0001.34 .40***1.34 .40 500,000-1 million1.99 .37***1.99 .37 1 million-2.5 million1.91 .34***1.91 .34 Over 2.5 million1.67 .36***1.67 .36 House staff (100s).23 .14*.23 .14 Constant-4.32 .35***- 5.14 .84 Model Statistics Pseudo R .39 .39 N 2,067 2,067 aSize of standard metropolitan statistical area (SMSA). ***Chi-square significant at p 5 .01. **Chi-square significant at p ' .05. *Chi-square significant at p ' .10.
PHYSICIAN PARTICIPATION IN GOVERNANCE TABLE A-7 Is the Director of Medical Education Compensated? System Control Coefficients Relative to Freestanding Investor-Owned Hospitals Systems Standard Standard Variables Coefficient Error Coefficient Error Freestanding hospital .71 .45 System characteristics System control Religious nonprofit .06 .19 .78 .46* Secular nonprofit -.18 .37 .54 .55 Investor-owned - .71 .45 Public - .44 .80 .28 .91 Contract-managed - .50 .45 .21 .54 System size (100s) -.17 .20 -.17 .20 Hospital characteristics Number of beds (100s) .51 .05*** .51 .05*** Regional location Northeast .03 .20 .03 .20 South - 1.01 .23*** - 1.01 .23*** North Central - .14 .20 - .14 .20 SMSA sizea Under 100,000 1.48 .53*** 1.48 .53*** 100,000-250,000 1.48 .28*** 1.48 .28*** 250,000-500,000 1.73 .27*** 1.73 .27*** 500,000-1 million 2.02 .28*** 2.02 .28*** 1 million-2.5 million 1.91 .25*** 1.91 .25*** Over 2.5 million 2.13 .26*** 2.13 .26*** House staff (100s) - .52 .14*** - .52 .14*** Constant -3.68 .24*** - 4.39 .48*** Model Statistics Pseudo R .51 .51 N 2,067 2,067 aSize of standard metropolitan statistical area (SMSA). ***Chi-square significant at p s .01. *Chi-square significant at p ' .10. 417
418 FOR-PROFIT ENTERPRISE IN HEALTH CARE TABLE A-S Number of Medical Staff Committees System Control Coefficients Relative to Freestanding Investor-Owned Hospitals Systems Standard Standard Variables Coefficient Error Coefficient Error Freestanding hospital 2.27 System characteristics System control Religious nonprofit .55 .49 2.82 Secular nonprofit -1.02 .86 1.25 Investor-owned -2.27 .80*** Public - 8.17 2.74*** - 5.92 Contract-managed -1.06 .71 1.21 System size (100s) .11 .28 .11 Hospital characteristics Number of beds (100s) 2.82 .12*** 2.82 Regional location Northeast 1.02 .50** 1.02 South -2.61 .47*** -2.61 North Central -2. 99 .45*** - 2.99 SMSA sizea Under 100,000 .88 1.25 .88 100,000-250,000 3.79 .ss*** 3.79 2s0,000-s00, 000 1.55 .67*** 1.55 500,000-1 million 3.12 .64*** 3.12 1 million-2.5 million 4.23 .51*** 4.23 Over 2.5 million 5.10 .54*** 5.10 House staff (100s) -3.44 .39*** - 3.44 Constant 8.02 .42*** 5.75 Model statistics R2 .45 45 N 2,025 2,025 CSize of standard metropolitan statistical area (SMSA). ***Chi-square significant at p ~ .01. **Chi-square significant at p ' .05. .80*** .88*** 1.12 2.84** .86 .28 . 12*** .50** .47*** .4s***
PHYSICIAN PARTICIPATION IN GOVERNANCE TABLE A-9 Does the Hospital Have a Committee for Cost Containment or Cost Awareness? System Control Coefficients Relative to Freestanding Hospitals Investor-Owned Systems Standard Standard Variables Coefficient Error Coefficient Error Freestanding hospital .19 .36 System characteristics System control Religious nonprofit.14 .19.37 .39 Secular nonprofit.05 .34.24 .47 Investor-owned-.19 .36 Public.14 .72.33 .80 C o n t r a c t - m a n a g e d- . 0 5 . 3 2. 1 4 . 4 1 System size (100s)-.16 .14- .16 .14 Hospital characteristics Number of beds (100s).19 .04***.19 .04 Regional location Northeast.29 .19.29 .19 South- .02 .19- .02 .19 North Central-.44 .19**- .44 .19 SMSA sizea Under 100,000-.34 .51- .34 .51 100,000-250,000- . 12 .24- . 12 .24 250, 000-500, 000- .43 . 25*- .43 . 25 500,000-1 million-.16 .25- .16 .25 1 million-2.5 million-.10 .20- .10 .20 Over 2.5 million-.16 .21- .16 .21 House sta~4100s).03 .13.03 .13 Constant-1.89 .17***- 2.07 .38 Model statistics Pseudo R .14 .14 N 2,067 2,067 aSize of standard metropolitan statistical area (SMSA). ***Chi-square significant at p ' .01. **Chi-square significant at p c 05. *Chi-sqmre significant at p ' .10. 479
420 FOR-PROFIT ENTERPRISE IN HEALTH CARE TABLE A-10 Does the Hospital Have a Committee for Medical Staff Long-Range Planning? System Control Coefficients Relative to Freestanding Investor-Owned Hospitals Systems Variables . Standard Standard Coefficient Error Coefficient Error Freestanding hospital .93 .32 System characteristics System control Religious nonprofit-.04 .16.90 .34 Secular nonprofit.03 .28.96 .40 Investor-owned- .93 .32*** Public- .65 .73.29 .79 Contract-managed- .52 . 27*.42 .34 System size (100s).08 .11.08 .11 Hospital characteristics Number of beds (100s).20 .04***.20 .04 Regional location Northeast.28 .16*.28 .16 South.01 .16.01 .16 North Central-.06 .15- .06 .15 SMSA sizea Under 100,000.15 .40.15 .40 100, 000-250, 000. 14 . 19. 14 . 19 250,000-500,000- . 16 . 19- . 16 . 19 500,000-1 million.02 .21.02 .21 1 million-2.5 million.10 .17.10 .17 Over 2.5 million.22 .17.22 .17 House staff (100s)-.34 .13***- .34 .13 Constant- 1.33 .14***-2.26 .33 Model statistics Pseudo R .14 .14 N 2,067 2,067 aSize of standard metropolitan statistical area (SMSA). ***Chi-square significant at p c .01. **Chi-square significant at p s .05. *Chi-square significant at p c .10.
PHYSICIAN PARTICIPATION IN GOVERNANCE TABLE A-ll Descriptive Statistics for Hospital Sample Subset (N = 2,025) Variables Standard Number Mean Deviation in Samplea Independent Variables Freestanding hospital.71.451,442 System characteristics System control Religious nonprofit.12.32241 Secular nonprofit.03.1868 Investor-owned.06.24125 Public.003.067 Conkact-managed.07.26142 System size (100s)21.3172.97 Hospital characteristics Number of beds (100s)190.53178.50 Regional location Northeast.23.42460 South.26.44533 North Central.32.41647 West.19.39385 SMSA sizes Non-SMSA.45.50906 Under 100,000.02.1232 100,000 250,000.09.28178 250,000 500,000.09.29184 500,000-1 million.08.27155 1 million-2.5 million.14.35293 Over 2.5 million.14.34277 House staff (100s)11.4950.04 Dependent Variables Number of M.D.s on governing board 1.96 2.27 Number of medical staff committees 13.44 9.13 aFor dichotomously scored variables. bSize of standard metropolitan statistical area (SMSA). 421
