Cover Image

PAPERBACK
$136.25



View/Hide Left Panel
Click for next page ( 75


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 74
~, investor Ownership and He 7 Cow of Medical Care What is the effect of investor ownership on health care costs? Some observers be- lieve that the rise in investor-owned health care organizations exacerbates the cost problem that investor-owned organiza- tions will exploit inadequacies in whatever system is used to pay for care; that cost in- creases will result from the need to attain consistent earnings growth to maintain the price of their stock; and that it is costly to pay dividends, taxes, and the salaries that business executives expect from the reve- nues they generate. Others, however, expect investor own- ership to help alleviate the cost problem, particularly if payment systems reward re- straints on expenses. Several factors are cited: the motivation and clarity of purpose that come with the for-profit form; economies of scale that result from multi-institutional ar- rangements; flexibility ofthe corporate form that permits rapid response to changing con- ditions; advantages in developing a reposi- tory of managerial skills because greater financial incentives and more extensive ca- reer paths can be offered; and cost advan- tages that can flow from cared! decisions regarding markets to be targeted Cocation? services offered, types of patients sought). (See the Appendix to Chapter 1 for ~ sum- mary of economic theories about the be- havior of for-profit and not-for-profit organizations. ~ The contradictory speculations indicate a 74 need to define different types of cost, to identify the nature of various parties' con- cerns about cost, and to examine the avail- able evidence. This chapter defines different types of cost and reviews evidence on ownership-related differences in hospital costs. Ibe impact on the cost of acquisition of hospitals by for- profit chains is discussed next, followed by a review of what is known about cost dif- ferences Mat relate to the ownership of other types of health care facilities, such as nurs- ing homes and freestanding ambulatory care centers. The chapter closes with the com- mittee's conclusions. DEFINITION AND MEASUREMENT OF COSTS- AND WHO IS CONCERNED Costs can be categorized according to who incurs them. First, providers of health care incur operating costs including capital costs in producing goods or services. They purchase resources, including plant and equipment. They also incur costs for the supplies and personnel required to provide patient care. For purposes of clarity we will call these costs "expenses." Second, the purchasers of health care pay a price for the care they buy. For some pay- ers this price is based on We hospital's charges (or a discount therefrom); for some, the price is based on reimbursement for allowable ex- penses incurred. In either case the price

OCR for page 74
INVESTOR OWNERSHIP AND THE COSTS OF CARE paid becomes cost to purchasers and reve- nue to providers. We will call costs to the purchasers of health care "price." Accordingly, two types of questions arise regarding cost and type of ownership. The first is concerned with ownership-related differences in expenses per unit of output. The second is concerned with ownership- related differences in the price that pur- chasers pay for similar services. The two questions are of interest to dif- ferent groups. Economists, who are con- cerned with efficiency, are interested in the relationship of ownership type to expense per unit of output. Some theorize that the not-for-profit organization's prohibition on distribution of profits leads to an emphasis on other goals, which are not conducive to efficient operations. For example, it is ar- gued that administrators of not-for-profit hospitals seek'added prestige for their hos- pitals by increasing the range of services, equipment, and personnel, which results in an expansion of services without adequate regard for their need or likely use (Lee, 1971~. It is commonly believed that the economic discipline that is assumed' to accompany the more singular purpose of for-profit organi- zations helps them to avoid such pitfalls and, hence, to operate more efficiently. Purchasers of care are more interested in price than in productive efficiency, espe- cially since the' move away from cost-based reimbursement by Medicare and most Blue Cross plans. Questions of relative efficiency are of little interest unless they translate into lower prices. Until recently, charge-paying purchasers have been generally inattentive to price differences among hospitals, in part because higher hospital prices could be passed along in higher premiums and be- cause patients and the physicians who or- dered services were substantially insulated from price differences because of thi'rd-par~ payments. Thus, keeping strict control of expenses was often not a necessity for hos- pitals. This picture is of course changing rap- i~y. Medicare's prospective payment system 75 and the emergence of much more price competition are putting pressure on both expenses and price. Providers who can over services for lower prices now stand to gain market share. Questions of quality wit likely become more salient. Measurement and comparisons of hospi- tal expenses end ' revenues are complicated by hospitals' being "multiproduct firms." A firm with one standard product devotes all expenses to the production of that product, making it' relatively simple to calculate the cost per item. But hospitals have numerous "products," including many types of inpa- tient services, outpatient services, ancillary services, and, at some'hospitals, educational and nonpatient care services. Even if the discussion is limited to inpatient services, these can be measured as admissions, days of care, and canny for patients with specific diagnoses. Comparisons remain Occult even after a unit of comparison has been selected. For instance,' a day of care in one hospital may differ from' a day of care in another hospital in terms of services provided and their frequency and quality. Severity of ill- ness of patients with the same diagnosis can vary widely. Another problem is that cal- culations of expense per unit of ''output are influenced by the method used to allocate indirect costs (such as administration, laun- dry, capital costs, etc.) to revenue depart- ments.~ Although payers who paid on the basis of incurred costs all specified indirect cost-allocation methods, there nevertheless was considerable leeway for hospital man- agers to manipulate allocation to maximize revenues. In addition, if a hospital was part of a multi-institutional system, some indi- rect expenses could be allocated either to a hospital or to over entities (such as the home office), depending on the incentives in reim- bursement rules. Thus, several variations in cost-allocation processes can make expense comparisons among institutions imprecise, even for a welI-defined service such as a Drain scan. Price comparisons are further compli \

OCR for page 74
76 cased by differences in the extent to which hospitals provide uncompensated care. A1- though cost-based purchasers pay on the ba- sis of expenses incurred in the care of their covered patients. others Dav what the hns FOR-PROFIT ENTERPRISE IN HEALTH CAM published studies. Only such data would al- low differences due to ownership type to be distinguished clearly from differences as- sociated with membership in a multi-insti , ~tutional system. pital charges. Such charges must be set at ~.7 11 . 1. a level that covers the institutions' expenses in caring for patients who either pay less than their share of expenses or pay nothing. Thus the price for charge payers and certain cost payers includes a subsidy to patients who cannot or do not pay for the full costs of their care.2 Differences among institu tions in the amounts of uncompensated care that they provide are thus likely to affect their charges, unless the institution has ac cess to nonpatient care revenues with which to subsidize uncompensated care. For those concerned with the price of care for partic ular groups for instance, a privately in sured employee group- such price and cost shifting and resultant price differences are important. Finally, it is important to recognize that all available studies are of hospitals oper- ating under economic incentives and con- straints that existed prior to the introduction of Medicare prospective payment.3 Al- though these studies describe expenses, prices, and margins under one set of incen- tives, comparative data are not yet available to indicate how different types of hospitals have responded to the new incentives. However, early data show that growth in hospital expenses has slowed significantly. rew avallaole studies compare me two twes of systems. However the authors of one national study that did make this com- parison note, '~e clear pattern that emerges from the study is that in 1980 there were much greater similarities among hospitals of the same ownership type (for-profit or not- for-profit) than among hospitals of the same organizational type (multi-hospital system tar ~'o~ct~nr~incr) ~^ ^~ ,/ .... Clearly in 1980 the strategies and performance of for-profit and not-for-crofit multi-hospital system hospi- tals had not converged as some observers believe they have today" (Watt et al., 1986a). Expenses According to popular belief and economic theory (see the Appendix to Chapter 1), for- profit organizations produce services less ex- pensively. Indeed, the assumed expense disadvantage of not-for-profit hospitals has prompted serious doubt about public poli- cies that are believed to encourage the not- for-profit mode (Clark, 1980~. Studies ex- amined by the committee show the postu- lated for-profit advantage in expenses to be a myth. Table 4.1 summarizes eight studies that analyzed hospital expenses as they relate to ownership. one methods, data sources and dates, and geographic scope of the studies varied considerably. As might be expected, therefore, results were not entirely consis- tent; however, the weight of the evidence and the overall direction is clear: not-for- profit hospitals controlled their expenses more effectively than did for-profit hospitals of the same general size. (All studies ex- cluded large teaching hospitals from their analyses.) Five of the studies compared the ex- penses per day in for-profit and not-for-profit STUDIES OF HOSPITAL EXPENSE AND PRICING Many studies examine the relationship between type of hospital ownership and ex- penses, prices, and profitability. The com- mittee focused on recent studies that examine investor-owned chain hospitals, rather than for-profit hospitals in general. The ideal comparisons of independent and chain hos- pitals, both for-profit and not-for-profit, are not always possible from available data and

OCR for page 74
INVESTOR OWNERSHIP AND TlIE COSTS OF CARE hospitals, making more or less rigorous at- tempts to control for institutional size and geographic location. On this measure, for- profit chain hospitals had higher expenses than not-for-profit hospitals in four of the five studies. The range was from 3 percent Tower in a Florida study to 10 percent higher in a national study. In another national study the difference was not statistically signifi- cant. Seven of the studies compared expenses per case (sometimes measured as "per ad- mission" or "per discharge." For-profit hos- pitals' expenses were again found to be higher than not-for-profit hospitals in six of the seven studies. The range was from 4 percent lower (again in the Florida study) to 8 percent higher in one of the two national studies. On this measure, the differences were not statisti- cally significant in two of Me seven studies; however the two studies with the largest number of observations found for-profits to have statistically significant higher costs. Studies that examined both perfidy and per- case expenses generally found the not-for- profit advantage to be larger on the former measure, largely because of shorter average lengths of stay in for-profit hospitals.4 The pattern of differences in expenses notwithstanding, two studies one national (1974-1980) and one in California (1977- 1981)-showed no major difference in the average annual rate of increase in expenses per discharge between for-profit chain hos- pitals and not-for-profit hospitals (Coelen, 1986; Pattison, 19861. During this period of rising hospital expenses, for-profit chain hospitals constrained expenses no better and no worse than not-for-profit hospitals.5 '~ ' One reason for the for-profits' expense disadvantage can be found in rates of oc- cupancy. Unit expenses (expenses per case or per day) are affected by the hospital's occupancy rates, owing to the number of cases to which fixed expenses must be al- located. In 1983, when overall hospital oc- cupancy was 73 percent, for-profit chain and for-profit independent hospitals had occu 77 pancy rates of 62 and 64 percent, respec- tively, lower than the 74 and 75 percent occupancy rates of not-for-profit chain and independent hospitals (Peter Kralovec, Hospital Data Center, American Hospital Association, 1985, unpublished data).6 One study estimated that each additional percentage point of occupancy reduces op- erating expenses by $2.00 per case, total patient care expenses by about $2.50 per case, and general and administrative ex- penses by about $3.00 per case (Watt et al., 1986b). Thus, according to this study the 1983 12-percent difference in average oc- cupancy rate between for-profit and not-for- profit chain hospitals accounts for $2S.56 in operating expenses per case. This repre- sents only 1.5 percent of the $1,712.88 av- erage operating expenses per case in for- profit chain hospitals. However, it accounts for almost one-third of the $78. 72 difference in average operating expenses per case be- tween for-profit chain and not-for-profit chain hospitals. Because of the large size offor-profit chain hospital organizations, economies of scale might be expected to occur in some areas. Investor-owned hospitals might be expected to have Tower administrative expenses be- cause of their profit orientation and their economies of scale. But both national and California data show that investor-owned hospitals operate with higher administrative expenses than their not-for-profit counter- parts (Pattison and Katz, 1983; Pattison, 1986; Watt et al., 1986a). Earlier caveats con- cerning different ways of allocating over- head costs in hospitals are particularly applicable to chain organizations. Such or- ganizations have some flexibility in spread- ing corporate overhead expenses across hospitals. Under cost-based reimbursement it was advantageous to allocate as many ex- penses as were allowable to the central office so that they could be reallocated to the hos- pitals with the greatest Medicare loads. These costs became the prices to government pay ers.

OCR for page 74
78 U. Cal x en ._ U' o ._ C) o Cal ._ ._ U) .,, Cow :- C~ - Em c . ~5 . CD :~ Cal : Cal o Cal a: Cal a: on o ~5 _ U) , - ~ ~s -I 5- -I ~ O G) O To I, ~ ~ ~ C) ~ 0, - = ~ ~= 0 Cal 0 = ~ = ~ = - Ct o sol o ~ ~ ~ ~ .~= ~ .~= ~ ~ ~ ~ t4 = =c C =o ~ =o ' e, _ c c 0 ~ ~0 0 ~0 ~ ~0 ~ ~0 ~ ~0 ~ 0 Z ~ ~s ~=5 ~=, ~C5 0 CC ~C ~eC ~C~ = 0 C =,, ~(,0 ~,ca _ ' ~ ~ ~ ' _ - V) _ ~, _ ~ et ~= ~ ~V _ 0 ~ ~ ~ ~ ~ ~ ~s:` ~O ~ ~ ~C) ~ = ~ C~ = ~; ~ 0 ~ ~ . ~ ~ O ~ i ~ e t5 c~ ~ > ~= ~ ~ ~ 0 e ~ Y E ~ y E,, ~i ,,, E ~ 8- ~ 0 = E ~ .= :> _ ~Z = x ~ c ~ oo a ~ C~ ~ a ~ V)

OCR for page 74
79 c .^ , . ~ _ ~ I= ~ ' ~ 3 ' ~ a j=~ "'is I'm ~;~= .o ~ C C) ~.~ ~ ,~ .~ :, _ x ' 3 ~ t~ O ~0 ~ - ~84 C) ~ es O ~ en O C) ~ - ~ O ~ 'a R ~ ~ ~ ~ ~ _ ~ ~ ~ ~ ~ , ~ ~ ~ o ~ x cat ~ E ~5 `==,9 j e .~ j ~ ~,: ~ ~it; ~ CD _ __ _ O ~C) Its ~ I= (C - oooo o0 ec ~ ~ ~o0 c~ oo 0 - - c~ 3 c) C~ o C~ 3 C~ ._ o - ._ - e~ ._ o c ,o~ o o ._ e~ - ._ e~ - C~ o 50 D. C~ ._ U) 3 - o o ._ C~ "s

OCR for page 74
80 Economies of scale resulting from bulk purchasing might also be expected to lower expenses of investor-owned chains for drugs and supplies. However, Wad et al. (1986a) fount] higher aggregate expenses for drugs and supplies sold to patients in for-profit chain hospitals than in not-for-profit hospi- tals on a per-day basis. If savings in the pur- chase price had been realized there, they were absorbed by the use of more charge- able items per day in for-profit than not-for- profit hospitals (Pattison and Katz, 19831. Personnel constitute a major hospital ex- pense. Data from studies of 1978 and 1980 showed that for-profit chain hospitals em- ployed fewer full-time equivalent staff per adjusted average daily census than did not- for-profit hospit~s.7 However, the for-profit hospitals paid higher salaries and benefits per employee, which largely eroded any dif- ference that might have shown up in lower total personnel expenses (Watt et al., 1986a).8 Capital costs are incurred in the pur- chase, construction, and improvement of plant and the purchase of major equipment. For-profit chain hospitals operated with sig- nificantly higher capital costs relative to op- erating costs than did not-for-profit hospitals (Anderson and Ginsberg, 1983; Watt et al., 1986a,b). This is due in part to the younger accounting age of for-profit chain facilities than not-for-profit facilities, but it is not known how much accounting age reflects the newer facilities or the timing of acqui- sitions. Nor is it known how much capital costs reflect the cost of acquisitions rather than expenditures to improve plant or equipment. Also not known is the extent to which acquisitions represent the rescue of hospitals in danger of closure or needing renovation and the extent to which con- struction represents the addition of unnec- essary capacity. Prices How do the prices paid by purchasers of care compare at for-profit and not-for-prof~t FOR-PROFIT ENTERPRISE IN HEALTH CARE hospitals? Studies have examined two as- pects of price-overall price to payers and strategies used by hospitals, such as markup and patient selection. Table 4.2 arrays six studies of price dif- ferentials between for-profit chain hospitals and not-for-profit hospitals. The studies con- sistently found that prices of for-profit chain hospitals were substantially higher than the prices of not-for-profit hospitals chain or independent. The price-per-day differential for charge payers ranged from 23 to 29 per- cent higher in for-profit chain than not-for- profit hospitals; for cost payers, from 11 to 13 percent. Although smaller, the per-case differential was still substantial, ranging Tom 17 to 24 percent for charge payers and from 8 to 15 percent for cost payers such as Med- icare. Across both types of payers the dif- ferences in the average price realized by the hospitals (net patient revenues) ranged from 12 to 14 percent; when measured per ad- mission, to about 17 percent on a per-day basis. Because investor-owned companies own less than 10 percent of the nation's hos- pital beds, these differences in price are the equivalent of less than half of one percent of the nation's hospital cost.9 It can be argued that the price paid by a community for hospital care should include nonoperating revenues provided to not-for- profit (anc! public) hospitals from philan- thropy and grants, and that adjustments should be made for the net difference be- tween tax subsidies to not-for-profit hospi- tals and taxes paid by the for-profits. Philanthropy and grants can be viewed as a price paid by the community for care and, thus, as an economic cost to the community. It can also be argued that tax payments are returned to public coffers and therefore rep- resent a reduction in net price to the com- munity. Several studies attempted to make these complex adjustments (Table 4.3~. All of the studies began with net patient service revenue the average price mea- sure described above and made additional adjustments. Lewin et al. (1981), using 1978

OCR for page 74
INVESTOR OWNERSHIP AND THE COSTS OF CARE data, removed an estimated income tax fig- ure from the for-profit hospitals' average. Sloan and Vraciu (1983) performed the same adjustment using different data that in- cluded a more explicit measure of taxes (but not a higher income tax rate). Finally, Watt et al. (1986a) widened the adjustment to standardize for differences in public subsi- dies and contributions received by not-for- profit hospitals in 1980. Sloan and Vraciu's results from Florida, which show little re- maining price difference after removing taxes, stand in contrast to the other two studies, which cover more states and show a residual difference of about 10 percent. While the latter difference is large, the authors were unable to estimate its statistical significance. Watt et al. (1986a) point out that taxes paid by the for-profit group accounted for less than half the difference in price per day be- tween the for-profit and not-for-profit hos- pitals. These calculations of"community" cost have been criticized on the grounds that taxes accrued to the nation, not to the local community, and that only a small portion of tax money is spent on health. Furthermore, the studies measured accrued, not paid, taxes. For growing organizations, taxes paid are lower than taxes accrued (see Chapter 3; see also Lewin et al., 1983; Conger, 1983~. Fi- nally, the studies make no adjustments for differences between types of hospitals in the costs of capital, especially the subsidy rep- resented by greater access to tax-exempt bonds by not-for-profit hospitals. The studies reviewed to this point have used average hospital revenues per case as a measure of price. A small study by Blue Cross/Blue Shield of North Carolina (1983) took a different approach. This study com- pared payment claims of six for-profit chain hospitals with similar-sized not-for-~rofit hospitals for three frequently performed procedures (hysterectomy, cholecystec- tomy, and normal deliveries) in 1981 and 1982. Charges at for-profit chain hospitals were higher (ranging from 6 to 58 percent) 81 than those at not-for-profit hospitals, with the exception of one for-profit chain hospi- taT's charges for normal deliveries which were 10 percent lower than the comparison hospital's. Comparison of the for-profit chain hospitals with other hospitals in the same communities showed a near-perfect pattern of higher charges by the former. MARKUP Given the findings of equal or higher ex- penses and considerably higher prices in for- profit chain hospitals, there is likely to be a difference in the markup by hospitals of dif- ferent ownership types. The most visible component of hospital price is routine daily service-the basic room rate. Because this component is likely to be the most price sensitive, many hospitals try to keep routine service prices close to expenses, or even accept a Toss, recouping profits on ancillary services for which demand is generated once the patient is in the hospital (e.g., di- agnostic tests). This strategy appears to have been pursued more vigorously by for-profit than not-for-profit hospitals, although it is followed by both. Four studies using data from different sources and years all show that routine daily room services were priced at or below ex- penses by all types of hospitals (Lewin et al., 1981; Watt et al., 1986a; Pattison and Katz, 1983; State of Florida Hospital Cost Containment Board, 19841.~ However, an- cillary services appeared to be marked up to be highly profitable. In Florida in 1982, the markup for ancillary services at for-profit hospitals was 121 percent; at state hospitals, 88 percent; and at not-for-profit hospitals, 74 percent (State of Florida Hospital Cost Containment Board, 1984~. California data indicate that all ownership types earned in- come on clinical laboratories, central service and supplies, pharmacy, and inhalation therapy, but in each case for-profit chain hospitals made a greater profit than not-for- profit or public hospitals (Pattison and Katz,

OCR for page 74
82 . =5 sit Cal ._ Cal _ .= ~ C U) c) a U) ._ CD o et O Con Cl _ CD .0 ~ ~ ,0 O ~ O~ O ~ O~ O ~ O ~ O ~ O - 5 ~_ . - ~. - ~. ~._ ~. ~. ~.- ~ _ . ~ _ ~ Cal ~ Ct~ CC ~ Ct~ tC ~ ~ Ce 00 ~ ~ Go O _ O _O _ O _O _ O ~O ~O _ _ t ~ ~ ~ ~t ~ _,.,, ~ A. t, _ ~{l,, ~ ~ ~ C ~c: ~ c: ~ ~ {~5 ~ ~ ce :- {,5 f- C) C5 ,_ C.) ~{- ~ ~ ~ ttS ~ ~ 'C;: ~ ~ ct ~ ~ cC ~ ~ ~ ~0 C24 ~ b0 ~ -~0 ~ C) ~ O C} O C, O C) O C) O C~ - O c, ~ ~c) ~S C ~ ~ C~ ~ ~ C ~ ~C ~ ~ C~ S ~ C~ ~ ~S: C; ~C S u "0 ~ C::: J C) C - 0 O C OJ O 4= "0 O C C ~ C ~ O C 0 - C) C~ ~ {~) =0 ~ {~) - 0 ~ C' J ~ 1I' - 0 ~ {V 30 ~ (~) 0 - ="~) ~0 =, CD OJ O S:` ~O ~ S~ ~ ~ =4 0 ~ 0 o4 ~ 0 ~ 0 ~ ~O ~4 ~4 ~4 ~4 _ ~ ~ ~ -~ ~ ~ O O ~- ~- O ~ ~ ~ ~ O ~ ~o = 00 ~ ~ ~ 3 , c~ ~ ~._ ~.~ ~ ~ ., ~ _ ~ ~ ~ ~ =, 'L ~ - =4 ~o <~5 ~ ~ ~C ~ ~ =, ~C~ ~o ;~ ~ `_ ~ ;~ 3 ~ _ C ~4 P" P" ~E ~C: Z _ ~ s~ Ct o X - C~ ~ o ._ ~ o - Ct C} ._ ._ o C~ ~ ._ o~ - ._ .^ _ Ct ._ . _ _ -::1 - '- O .= ~ ~,_ ._ ~ ~ ~& C~ J ~ O U, ,- (v ~ 5 t ~o ~ ~ c~ _ c5 ~ ^ 3 ~ ~ - {ts cc - ~ ~ ~ c~ ~ ~ - ~ c) ~ - c) bo =, ~ x ~ ~ - ~ c J - C~ Ce 00 S: ~ O ~ v, ~ ._ ~ c~ cC ~ 0 ~ _ ~ _ ,0 0 u, _ O C: ~ O ~ C; 0 ~ 0 ._ _ - ._ C~ ~ 0 _ CS' ~ c~. _ w ~ _ ]_ ._ O ~ ~ U2 c,, O _ _ O D O C~ C~ _ O C) ~ ~: O ~ 00 _

OCR for page 74
83 ~ _ ~ , ~ _ oo ~ ~ '- C,] ~Hi ~m) C45 - ~ ~ ~ ~ ~ ~ - ~ - - Ct: 'Cl, ~ ~ci. is_ _ .e e - _ _ A a E _ ~ 0 c E G _ G e - G C G e ~ G _ - G - G e ~G e Lc C) C O. ~ Cam ~ ~ of Sol ~ C~ ~ ~ Con 50- ~ C; ~0 ~ Cry O ~ Cry ~ t) ~0 ~ C) ~ ~ 40 0~ ~ 3) ~ ~ (~D O ~ t) O ~ ,Jo O ~ Co O O ~ ~ O ~ ~ O ~ ~ O ~ ~ O &4 ~ O ~ ~O ~ ~0 ;; ~ ~_ a' 1 w ~E 3 E ~ ~ ~ ~ E s_ ~ c,) D~ _ ~ c) et ~=t b4 ~ ~ tt ~ .~0 ~ ~ ci .N c~ ~c:~ _ c:,, 3 0~ C) =5 ~ ce C; C~ C.) ~ ~C~ cO; .C~ Cs 00 ~C~ C) o4 Ct 3) e~ ~ ~ C: ~ ~Z ~Z ~ . _ ~E ~ _ _ Ct _ u, _ ~ 0 ,5 c) _ ~ ~ 0 .> 0 ~ c) ~ ,, ~ 0 ~u' ~ 0 o . -_ ~ - .~ ~ t4 G) X C ~ t0 O ,x~ ~ ~ ~ ,~ - ~- ~ ~- - u 2 EE ~ E o' ~ , ,, a ~ ~; ~ x _e15 ~ _, ~_ _ CDCD ~ ct ~0

OCR for page 74
Study Controls Data Sources Measure Findings 84 FOR-PROFIT ENTERPRISE IN HEALTH CARE TABLE 4.3 Summary of Study Findings Concerning Net Community Cost Lewin et al. 53 matched pairs of 1978 Medicare Cost Net patient ser- For-profit chain hos (1981) hospitals Reports, California, vice revenue pitals 18 percent Florida, Texas less taxes per higher than not-for adjusted day profit hospitals be fore tax adjustment; 12 percent higher after tax adjust menta Sloan and Nonteaching hospitals Data reported to Net operating For-profit chain hos Vraciu (1983) under 400 beds Florida State Cost fiends per ad- pitals 2 percent Containment Board lusted patient higher than not-for dayb profit hospitals Net operating Difference not statisti fi~nds per ad- cally significant justed admis sionb Watt et al. 80 matched pairs of Medicare Cost Be- Net patient ser- For-profit hospitals 17 (1986a) hospitals, adjusted ports, 1980; AHA vice revenue percent higher than for case-mix differ- Annual Survey of standardized not-for-profit hospi ences Hospitals, 1980; Of- for differences tats before adjust fice for Civil Rights in accrued meet; 10 percent Survey of lIospi- taxes, public higher afteradjust tals, 1980 subsidies, and menta contributions income, per adjusted day Methodological difficulties in calculating Me measures precluded estimation of statistical significance. The 18 percent difference in net patient revenues in the 1981 study and the 17 percent difference in the 1986 study were statistically significant (p < .01~. bNet operating funds are defined as net operating revenues (net of contractual allowances and unpaid accounts) minus income taxes. 19831. Furthermore, ancillary services that were profitable were used with greater fre- quency per patient day and per admission in for-profit chain hospitals than in not-for- profit hospitals. However, ancillary services that broke even or lost money were used at similar rates in for-profit chain and not-for- profit hospitals. In California, ancillary ser- vices contributed a higher proportion of to- tal revenues in for-profit chain than in not- for-profit hospitals in 1982 (69 percent com- pared with 63 percent), and this differential occurred after a period of years in which hospitals of all types of ownership shifted more inpatient charges to ancillary services (Pattison, 19861. In another study that used national data, differences in pricing or use of ancillary services, or both, resulted in significant differences in ancillary service charges per case: in for-profit chains, charges were almost 34 percent higher and in not- for-profit chains, a htHe over 5 percent higher than independent not-for-profit hospitals (Coelen, 1986~. The studies reviewed so far measure markup on individual components of patient care. Watt et al. (1986b) took a broader mea- sure gross patient care markup, defined as total patient charges divided by total oper- ating costs ant! found that for-profit chain hospitals averaged 14.2 percentage points higher than not-for-profit chain hospitals.

OCR for page 74
86 factors, the authors found no statistically sig- nificant difference in after-tax margins on total revenues between Florida's for-profit chain and not-for-profit hospitals. The raw data showed not-for-profit hospitals to be slightly more profitable than for-profit hos- pitals. As studies show, nonpatient care reve- nues and taxes paid have a substantial im- pact on the bottom line of both not-for-profit and for-profit hospitals. Comparing the prof- itability of hospitals by measures other than margin on total net (aDcer-tax) revenues omits an important source of revenue for not-for- profit hospitals, and an important deduction from revenues of for-profit hospitals. Al- though for some purposes it is important to compare levels of profitability achieved through patient services, such a measure does not indicate the financial health or fu- ture viability of hospitals. Indeed national figures indicate that the average u A. hos- pital had a negative margin on patient care service until the 1980s, although this is to some degree an artifact of the tendency of many not-for-profit and public hospitals to expend any surpluses as fast as they were achieved. Positive total net margins were achieved only through the contribution of Inpatient care revenues (see Table 5.1, Chapter 51. The studies reviewed have all used con- trols to try to ensure that the hospitals being compared are similar in characteristics such as so location, and teaching status that are expected to affect profitability. Thus, the studies' finding that for-profit hospitals are more profitable than not-for-profit hospitals (even when margin on after-tax total net in- come is the measure) is based on similar not- for-profit and for-profit hospitals. But for- profit and not-for-profit hospitals are not similar in all ways. For-profit hospitals, for example, are more often located in sun-belt states, are smaller than not-for-profit hos- pitals, and are virtually never teaching hos- pitals. National, uncontrolled data that compare for-profit (chain and independent) FOR-PROFI T ENTERPRI SE IN HEALTH CARE TABLE 4.4 Total Net Margin of U. S. Community arid For-profit Acute Care General Hospitals, 1980-1984 Year United States For-profit 1980 1981 1982 1983 1984 4.6 4.7 5.1 5.1 6.2 3.9 4.4 5.1 4.2 4.8 SOURCES: Office of Public Policy Analysis, American Hospital Association, Chicago, Ill.; Federation of American Hospitals, Statistical Pro- file of the Investor-Oumed Hospital Industry, 1979 and 1980, 1981, 1983, and personal communica- tion, SamuelMitchell, Director ofResearch, Fed- eration of American Hospitals, 1985. For-profit figures are after taxes. with all U. S. community hospitals (includ- ing for-profit and not-for-profit hospitals and nonfederal public hospitals), show that the for-profit group as a whole has not been more profitable than the average U. S. hospital in terms of total net margin. Indeed, in four of the past five years, the for-profit hospitals have been less profitable; in one year, prof- itability was equal (see Table 4.4~. Further- more, during the period covered by the studies reviewed, for-profit hospitals and the companies that own them did not on average generate high profit margins when com- pared to other industries. Average profit margins for two-thirds of American indus- tries exceeded margins for the hospital sec- tor (The Value IAne Investment Survey, 1981 and 1982, cited in Sloan and Vraciu, 1983~. In sum, the studies reviewed show that for-profit chain hospitals have been more profitable than similar not-for-profit hospi- tals, but not-for-profit hospitals as a group have been less profitable than the average hospital. Thus, a for-orofit chain hospital of the same size, in the same location, and similar in other aspects to a not-for-profit hospital is likely to achieve higher margins. The reverse is true for the average for-profit hospital, compared with the average com- munity hospital. . . ~

OCR for page 74
INVESTOR OWNERSHIP AND THE COSTS OF CARE COSTS AND ACQUISITION OF HOSPITALS The grown of for-profit chains through acquisition (and, to a lesser extent, through construction) has led to two concerns about impact on cost. The first, particularly among third-party payers who pay charges, per- tains to the for-profit sector's willingness to price "aggressively" (i. e., to charge more). The second concern stems from the capital costs associated with an acquisition, which can be passed on to charge payers, as well as to cost payers, who have included capital costs (interest, depreciation, and, for for- profits, return on equity) as a reimbursable expense under cost-based reimbursement (and currently as a "pass through" under prospective payment). Until recently, ac- quisitions also put in place a new and higher depreciation schedule (based on the price paid for a facility), which was reimbursed as an expense and thereby contributed impor- tant cash flow. A General Accounting Office (GAO) re- port on the costs associated with Hospital Corporation of America's 1981 acquisition of Hospital Affiliates International found a first- year net capital cost increase of$55.2 million from the acquisition of the 54 Hospital Af- filiates hospitals $62.5 million from inter- est, $8.4 minion from depreciation, minus $15.7 million in savings from a reduction in home office costs. GAO calculated Medicare cost increases at only two of the acquired hospitals, and estimated them to be nearly $600,000 in the year following acquisition (U.S. General Accounting Office, 1983~.~2 The GAO report did not conclusively an- swer the question of the magnitude of the increase in hospital prices that results from hospital mergers and acquisitions. How- ever, the fact that very substantial increases in reimbursable expenses could result from an acquisition, even with no change in the services being provided, led Congress to change the law that permitted an acquiring company to revalue the assets it acquired 87 for purposes of being reimbursed for de- preciation expenses under Medicare. Three studies shed light on the changes made by new owners after an acquisition. One study grouped hospitals by length of ownership by a chain, noting that average expenses were Tower in hospitals that had longer affiliations with a for-profit multihos- pital system than in those with a shorter affiliation. They found support for a hypoth- esis that companies acquire poorly managed hospitals and improve their efficiency (Becker and Sloan, 1985~. Two studies conducted for the committee described some character- istics of acquired hospitals and analyzed changes in the years immediately following acquisition. Pattison (1986) studied nine California hospitals acquired by for-profit systems between fiscal years 1977-1978 and 1981-1982. The study showed evidence that the hospitals were initially in financial trou- ble and that the change in ownership gen- erated some dramatic operating changes. At the beginning of the period the hospitals showed an after-tax Toss of $1.38 per patient day, which was changed to a $23.70 after- tax profit per patient day about two-and-a- half years after acquisition a profit level below that of other for-profit chain hospitals, but substantially higher than that of com- parison for-profit hospitals that remained in- dependent. Apparently, some of the increase in prof- itability was achieved by initiating operat- ing-cost efficiencies. In 1978, expenses per day were $12 higher in for-profit hospitals that were later acquired than in those hos- pitals that remained independent. By 1982, expenses per day were $15 lower in the ac- quired than in the still independent for-profit hospitals. However, occupancy showed only a slight improvement, rising from 48 to 50 percent, while inpatient charges per day rose 89 percent" only slightly more than the 86 percent increase in the unacquired for-profit hospitals. Thus, profitability improvements in the acquired hospitals evolved partly through tight control of expenses relative to

OCR for page 74
88 charges. Capital costs per day increased from $11 to $55, partly because of additional debt, which increased from $7,800 to $57,700 per bed. This striking increase of over 600 per- cent should be put in context. For-profit hospitals that remained independent more than doubled their debt per bed to $24,000, while hospitals that were already part of investor-owned chains increased their debt per bed only 60 percent, to $39,000 (Pat- tison, 1986~. A detailed description of 15 hospitals bought by for-profit chains in Florida be- tween 1979 and 1981 (Brown and Kloster- man, 1986) confirms the California finding of low profit levels before acquisition: their pre-acquisition after-tax margin was below that of unacquired hospitals, although their pretax operating margin was higher than that of unacquirec} hospitals. The data also sug- gest that potential for improvement in the control of expense existed, because pre-ac- quisition operating expenses per adjusted admission were roughly 10 percent higher than in unacquired hospitals. Profitability after acquisition varied. Three years after acquisition by for-profit chains, 2 of the 15 Florida hospitals were operating in the red. One-third of the acquired hos- pitals achieved margins comparable to other for-profit hospitals in the state, and almost half had margins higher than unacquired hospitals. During the three years after ac- quisition, the previously independent hos- pitals acquired by for-profit chains improved their margins to become more profitable than the average unacquired hospital. This was not true of acquired hospitals formerly in for-profit chains. Despite high operating ex- penses before takeover, such expenses per adjusted admission increased at a faster rate after acquisition by for-profit chains than did operating expenses in unacquired hospitals. Gross and net revenues per adjusted ad- mission and ancillary revenues per adjusted admission generally increased at a higher rate in hospitals acquired by for-profit chains than in unacquired hospitals again despite FOR-PROFIT ENTERPRISE IN HEALTH CARE high pre-acquisition levels. These data sug- gest that any improvements in profitability were achieved not through tight control of expenses but rather through higher prices. Another strategy to enhance profitability af- ter acquisition reductions in uncompen- sated care is discussed in Chapter 5. The data from Florida and California sug- gest that acquisitions by for-profit chains had the effect of increasing the cost of care. In- creases in charges in acquired California hospitals were only slightly above those of unacquired for-profit hospitals. In contrast, in Florida some hospitals acquired by for- profit chains showed substantially larger in- creases in revenue per adjusted admission than hospitals whose ownership did not change. For hospitals that belonged to for- profit chains before acquisition, the rate of increase in net revenues per adjusted ad- mission was significantly higher than the rate of increase in other hospitals. Since finan- cially troubled hospitals seem to be common acquisition candidates, it is possible that in the absence of a change in ownership, the former owners might have had to either raise prices to improve the financial standing of the hospitals or close the hospitals. NURSING HOME COSTS Nursing homes provide a striking contrast to hospitals and their cost-based/charge-based system of reimbursement. Nursing homes operate with a payment system that has not changed fimdamentally in recent years. Also, the nursing home business unlike the hos- pit~ business is marked by constraints on growth, imposed by regulation and by lim- ited competition caused by high occupancy rates and lack of alternatives for long-term care. Nursing home care is generally charged on a per diem basis. Medicaid accounts for roughly half of all nursing home revenues. States vary in the way the Medicaid pay- ment is determined, but typically, the per diem rate is cost-based within the limits of

OCR for page 74
INVESTOR OWNERSHIP AND THE COSTS OF CARE the state-imposed cap. The cap can apply to operating or capital expenses, or both. There are no limits on charges to private-pay pa- tients, except in Minnesota, where nursing homes cannot charge private-pay patients more than the Medicaid rate.22 Estimates of the price differential between Medicaid and private pay hover around the 18-20 percent mark (Lawrence Lane, unpublished data prepared for the Institute of Medicine Com- mittee on Nursing Home Regulation, 19841. An extensive review of the literature on nursing homes was conducted for the com- mittee (Hawes and Phillips, 19861. Most available studies of nursing home expenses anc! prices fait to distinguish investor-owned chain from independent proprietary nursing homes. The usual comparison is between not-for-profit and for-profit ownership. Very few studies use national data, or control for severity of illness or quality of care. Empirical studies generally show signifi- cantly higher reported per-patient-day ex- penses in not-for-profit than in for-profit nursing homes (Hawes and Phillips, 19861. Studies that control for factors that affect expenses, such as location, patient case mix, intensity, and some proxies for quality, show that an expense differential exists, but the magnitude is smaller than found in studies lacking controls. A study of Illinois nursing homes in 1976, notable for its controls for patient mix, facility size, occupancy rate, and other variables, found that for-profit fa- cilities were more efficient, providing a given level of care at lower cost to the facility, but there was no difference in cost to payers (Koetting, 1980~. There are tentative indi- cations that for-profit nursing homes achieve expense savings through Tower patient care expenditures, but spend more on property- associated costs (CasweD and Cleverly, 1978, cited in Hawes and Phillips, 1986~. How- ever, lower for-profit expenses do not trans- late into Tower charges. The few studies that focus on charges show no or small differ- ences between the two types of ownership (Hawes and Phillips, 1986~. 89 Findings of lower expenses in for-profit nursing homes was confirmed by a study that used a four-way classification to inves- tigate the effects of ownership and affiliation of selected efficiency measures (Hiller and Sugarman, 1984~. A sample of approxi- mately 1,100 nursing homes was drawn from the 1977 National Nursing Home Survey. Chain for-profit homes had the lowest av- erage cost, operating cost, and nursing cost per resident day, followed by for-profit in- dependent nursing homes (see Table 4.4~. However, regression analysis suggested that lower for-profit average and nursing costs were due to such factors as occupancy rates and location, not ownership or chain affili- ation. Finally, the data in Table 4.5 show chain for-profit nursing homes to have the highest before-tax profit per resident day after inclusion of nonpatient care revenues. Independent not-for-profit nursing homes had a negative bottom line. For both for- profit and not-for-profit homes, chain-affil- iated facilities were substantially more prof- itable than independent facilities. Regression analysis suggested that ownership accounts for only a small (3 percent) part of the dif- ferences among ownership groups in this measure of profit. In an effort to differentiate more finely among ownership groups than did earlier studies, Hawes and Phillips (1986) analyzed data from private not-for-profit nursing homes that participated in the Ohio Medicaid pro- gram in 1977 and three types of for-profit nursing homes: those owned by individuals operating 3 or fewer homes; intrastate chains, usually of 4 to 10 homes; and interstate, publicly held chains. Not-for-profit homes were found to operate with expenses higher by $6 to $8 per patient day than the average for-profit homes, and interstate chains had expenses of $1.50 higher per patient day than the other two for-profit forms. A similar pattern (Table 4.6) was found in expendi- tures for direct patient care (nursing, social services, supplies, etch. Administrative and general service expenses are items for which

OCR for page 74
go FOR-PROFIT ENTERPRISE IN HEALTH CARE TABLE 4.5 Selected Nursing Home Cost and Profit Measures by Ownership and Affiliation, National Nursing Home Survey, 1977 For-profit Not-for-profit Measure Chain Independent Chain Independent Net profit (before tax) per resident day $ 1.00 $ 0.19 $ 0.32$-0.66 Average cost per resident daya 22.38 31.43 40.5037.38 Operating cost per resident dayb 4.67 7.52 9.878.18 Nursing cost per resident day 7.25 10.14 13.7112.06 aIncludes all costs (including operating, labor, and fixed). b Operating cost includes the cost of food, drugs, and services such as laundry or nursing services purchased from outside sources. Excluded are labor costs (payroll) and fixed costs of capital expenses and rent. SOURCE: Hiller and Sugannan (1984~. chain operators might be expected to show economies of scale, and Table 4. 6 shows that interstate chains in Ohio had the lowest ex- penses per patient day for these two cate- gories. However, the costs of ownership (interest, depreciation, and rent payments) are lower for not-for-profit than for all types of for-profit nursing homes. Controlling for such factors as location, certification, and quality (by proxy measures), the differences among ownership types are reduced, but remain significant. Ibe higher for-profit ownership expenses could be due in part to acquisitions and sell- ing of nursing homes to take advantage of tax and reimbursement benefits. Capital payments for depreciation, interest, and a return on equity are generally reimbursed by public programs, particularly Medicaid. The 1981 tax provisions on accelerated cost recovery also spur acquisition activity. How- ever, some states have moved to slow ac- quisition activity by limiting reimbursement for capital and interest costs. As in the tos- pi~l industry, chain grown in nursing homes has been achieved mainly through acquisi TA;BLE 4.6 Comparison of Average Nursing Home Expendi- tures per Patient Day in Major Cost Centers, State of Ohio, 1977 Type of Ownership ~- For-profit For-profit For-profit Individual Intrastate Interstate Not-for Ownership Chain Chain profit Cost Center (N = 382) (N = 78) (N = 32) (N = 87) General services costs 2.47 2.44 2.04 3.27 Ownership or rent 3.12 3.12 3.10 2.69 Patient care costs 9.05 8.90 9.52 13.16 Administrative costs 2.73 2.90 2.53 3.17 SOURCE: Hawes and Phillips (1986).

OCR for page 74
INVESTOR OWNERSHIP AND THE COSTS OF CARE tion partly because construction has been limited by states, constraining the expansion of the supply of nursing homes through cer- tificate-of-need restrictions, and partly be- cause inflation of building costs has made acquisition cheaper than construction. Al- though payment by Medicaid is frequently inadequate, the nursing home market con- tinues to attract for-profit enterprises that generate sufficient total revenue to provide an adequate return on investment. In sum, for-profit nursing homes, oper- ating with reimbursement that has similar- ities to the Medicare prospective payment system being implemented for hospitals but lacking the competition that hospitals are now confronting, operate with lower ex- penses but with similar charges than not- for-profit nursing homes. Lower for-profit expenses are in some cases the result of lower patient care expenditures, and there is some evidence that chain operations are more ef- ficient in the provision of general and ad- ministrative services. Limited evidence from Ohio suggests that interstate chain homes operate with expenses that are substantially lower than not-for-profit homes, but some- what higher than other for-profit homes. Despite Tow Medicaid reimbursement lev- els, profitability is sufficiently high to attract for-profit providers to the nursing home business in many states. OTHER FOR-PROFIT PROVERS Although there is no literature available on the comparative cost of freestanding for- profit and not-for-profit urgent care centers, surgery centers, imaging centers, and the like indeed, there are little data available on the extent of not-for-profit ownership of such centers the growth of freestanding centers represents a dynamic entrepreneu- rial trend in American health care. It un- doubtedly has important cost implications, although these may shed little light on the comparative behavior of for-profit and not 91 for-profit institutions. Both for-profit and not- for-profit hospitals and investor-owned com- panies have established new ambulatory care centers, but it is not known whether the practice is more common among not-for-profit than among for-profit organizations. Many of the centers established by not-for-profit organizations are set up as for-profit subsidi- aries. Some of the new types of ambulatory care centers are based on innovations. The in- novations can be in marketing, such as pri- mary care centers' convenient hours and location or lack of an appointment require- ment. Other innovations, particularly in am- bulatory surgery centers, involve new technologies and new ways of providing ser- vices. Trough such innovations and through various means of reducing overhead, in- cluding specialization, selection of particu- lar types of patients, and limiting hours of service, many new types of ambulatory care centers are able to charge less for services than do the hospitals with which they com- pete, a finding documented in several stud- ies reviewed by Ermann and Babel (1986~. There are also many anecdotes of hospitals Towering their charges to compete with free- standing centers. Many innovations in various types of am- bulatory care centers have undoubtedly led to significant unit-cost savings to many pay- ers and greater convenience for many pa- tients. The competitive responses they have stimulated on the part of hospitals may also help control certain health care costs. Whether the thousands offreestanding cen- ters that now exist represent a net reduction in health care costs is much less certain, however, because they constitute additional capital investment and because existing in- stitutions that lose patients to the centers must spread their overhead over fewer cases. Also, if existing institutions lose revenues from services on which they can make money, they may be less able to provide unprofit- able services and uncompensated care. These

OCR for page 74
92 are matters that deserve scrutiny as the cost implications of freestanding centers are as- sessed. MONITORING COST TRENDS Almost all the analyses of expenses and prices reported here were based on data collected in an environment that differs sig- nificantly from that in which hospitals and physicians find themselves today. Much of the data in this report provides an important baseline, but continued and expanded data collection and monitoring of expenses and prices are essential. The move away from cost-based reimbursement; the rapid growth of preferred provider organizations, health maintenance organizations, and many new forms of ambulatory care; and the expansion of home care all indicate that major changes are taking place in the way medical care is organized and paid for in the United States. These changes, which alter many of the eco- nomic incentives to providers and create a competitive environment, are likely to affect provider behavior with respect to prices, control of expenses, and in many other ways. Dramatic changes are occurring and seem likely to continue. Data systems to monitor operating and capital expenses and prices wit} need to be sensitive to subtle changes in the hospital product as well as to changes in national health care expenditures, in providers' uses of resources, and in expenses and prices for different payers and different population groups. Moreover, to determine savings and/ or cost shifting in the health care system as a whole, the various new types of nonhos- pital providers should be monitored. In light of the findings from the studies reviewed in this chapter that suggest a greater respon- siveness of for-profit providers to economic incentives, it is important to Pack differ- ences between not-for-profit and for-profit providers as the new incentives take hold. Multi-institutionaI systems, also separated into for-profit and not-for-profit, shouted be FOR-PROFIT ENTERPRISE lN HEALTH CARE monitored to determine the relationships between ownership and system member- ship. The major sources of data today include Medicare costs reports, which are the only existing source of national hospital expense information; some very valuable state re- porting systems; and regular surveys by or- ganizations such as the American Hospital Association. No sources were identified for data on national nursing home expenses. The committee believes that Medicare cost reports provide valuable information and should be maintained even as cost-based reimbursement is phased out. The commit- tee would also like to see other states adopt hospital reporting systems uniform with those already in place in some states. States should develop similar kinds of uniform reporting systems to monitor operating and capital ex- penses and prices in nursing homes and other out-of-hospital facilities. Until quite recently, most analyses of hos- pital expenses were conducted without ref- erence to case mix or case severity, which affect the costs of care. Medicare's case-mix index, a measure of costliness of cases treated by a hospital relative to the average of all Medicare patients, is for hospitals only and is based on Medicare program data. It is the only national source of case-mix informa- tion. The Prospective Payment Assessment Commission (ProPAC) is working to refine and improve the index. The commission has not yet decided on how it will respond to the needs of non-Medicare private and pub- lic payers who are beginning to use the sys- tem for their own purposes (Prospective Payment Assessment Commission, 19851. The committee emphasizes the need for continued improvement of case-mix and case- severity measures, and the use of such mea- sures in studies of expenses and prices. The committee is aware that much useful data is being collected by insurers, employ- ers, providers, and other organizations con- cerned with the price of health care. However, for reasons of confidentiality and

OCR for page 74
INVESTOR OWNERSHIP AND THE COSTS OF CAM competitive position, such organizations have been reluctant to make their data publicly available. The committee believes that with the proper assurance of confidentiality and the removal of individual identifiers, valu- able inflation could be released. The committee urges those in control of relevant data to work with agencies, such as the Na- tional Center for Health Statistics, to ar- range for increased release of information. CONCLUSIONS The evidence reviewed in this chapter re- veals several patterns in the cost-related be- havior of for-profit and not-for-profit institutions. First, studies of hospital costs that control for size (and in some cases for case mix and other factors) show for-profit hospitals to have slightly higher expenses than not-for-profit institutions ranging from a statistically insignificant level to 8-10 per- cent higher when payment is based on costs incurred. Second, studies (again, controlling for size and sometimes other variables) show that for-profit institutions charge more per stay than not-for-profit institutions ranging from 8 percent for cost payers to 24 percent for charge payers. Using the difference in for- profit chain and not-for-profit chain prices found in a national study, for-profit chains' high prices added approximately $470 mil- lion, or less than 1 percent to the nation's bill for hospital care. If for-profit chain hos- pitals increase their share of the hospital market and maintain the price differential, the price impact will increase. Bird, controlled studies comparing the profitability of for-profit chains and not-for- profit hospitals show that for-profit chains have achieved higher levels of profitability before and after taxes. They are also less likely than not-for-profit hospitals to have losses on patient revenues and more likely to have high margins on patient revenues. Nonoperating revenues such as charitable contributions and investment income are 93 important sources of revenues for not-for- profit hospitals, contributing substantially to their overall margins. But national, uncon- trolled data show that for-profit hospitals have lower (after-tax) margins on total net reve- nues than the average U. S. hospital. Fours, data from nursing homes suggest that when payment levels are fixed on a per diem basis, for-profit institutions restrict ex- penses more than not-for-profit institutions, which have lower margins, obtain revenues from philanthropy, and have more charge- paying patients. The cost savings are ob- tained on patient care expenses, not on cap- it~ or administrative expenses. These findings support the broad conclu- sion that for-profit institutions responc! more precisely to economic incentives than do not- for-profit institutions. (The for-profit hos- pitals' shorter lengths of stay and their lower occupancy rates are inconsistent with this broad interpretation.) Under circumstances in which it is economically rewarding to charge more (e.g., when payers pay charges and when there is little price competition), for-profit institutions have charged more. Under circumstances that reward invest- meet, they have invested more. Under cir- cumstances in which it has been economically rewarding to have higher expenses, even if allowable expenses are sharply defined, ex- penses of the for-profits have been slightly higher. Under circumstances where prices are fixed, such as nursing homes, Weir ex- penses have been lower. Economic rewards that derive from revaluation of assets and increased reimbursement may also explain why changes of ownership have occurred rather frequently among for-profit hospitals and nursing homes. In addition to comparisons of costs asso- ciated with different types of ownership, the question ofthe effect offor-profit health ser- vices on the nation's health care costs also deserves comment. Two areas of heavy for- profit influence nursing homes and di- alysis treatment are characterized by fixed- payment methods that provide strong in

OCR for page 74
94 centives to limit expenses per unit of service (per day or per dialysis treatment). In other areas of for-profit growth, including both ambulatory care centers of various sorts and health maintenance organizations, new en- trepreneurial organizations have been com- peting on the basis of price, services, and convenience. As noted earlier, the emer- gence of ambulatory care centers may have resulted in many reductions in unit price and cost to payers, although the increased capacity these centers represent may well have increased national health care costs. A comparison of expenses and charges on a per-day or per-case basis, however, does not adequately answer the question of im- pact on total community costs of hospital services. It can be argued that the true costs of care to communities must take account of government grants and charitable contri- butions received by not-for-profit facilities and taxes paid for for-profit facilities. The few studies that make these complex and ~cult-to~uanti* adjustments suggest that for-profit hospital care is more expensive, but less so than indicated by simpler mea- sures of prices and cost. Comparisons of hospital charges and ex- penses also overlook the fact that the for- profit sector has made substantial capital in- vestments in the modernization and con- struction of health care institutions and the acquisition of financially weak institutions that in some cases might otherwise have perished. In this regard, an important his- torical question about the impact of the for- profit sector on the nation's health care costs is not whether their investments in health care have increased costs but whether the capacity that has resulted from their in- vestment is a net gain. This is partly an issue of values about the operation of markets and the objectivity of community health care needs. However, occupancy rates in hos- pitals in early 1985, at 66 percent nationally, suggest overbedding, and the investor-owned chains, with occupancy rates of 10-12 per- cent less than the national average, have FOR-PROFIT ENTERPRISE IN HEALTH CARE more unused capacity than the average hos- pital. Only the future will reveal whether the costs of unused capacity will be absorbed by communities or by the hospitals them- selves. Although some predict that for-profit health care organizations will lead the way toward lower-cost care now that economic incen- tives reward such behavior, this should not be regarded as a forgone conclusion. Their expenses and prices have been higher in the past; they have high ongoing capital ex- penses because of past acquisition activities; their average lengths of stay are already shorter; they have Tower occupancy rates across which to spread fixed costs; they have less access to nonpatient care revenues; and they must not only earn a profit to stay vi- able but they must also pay taxes on these profits (as well as on their property.) The strategies they (and not-for-profit hospitals) adopt will undoubtedly affect Me shape and configuration of our health care system, the provision of at least some services to those patients who are unable to pay, and perhaps the quality of care. NOTES Cost allocation is an accounting technique that is used to spread the common overhead expenses of an institution (e.g., the administrator's salary, insurance) among its revenue-producing departments. This re- quires identifying outputs, defining expenses to be al- located, and establishing rules for allocating expenses to outputs (Yoder, 19801. 2 Some cost payers allowed some costs of bad debts and charity care in their calculation of reimbursable costs. Medicare so treated the bad debts of only its beneficiaries (incurred when deductibles and co-pay- ments were not paid), not of other patients. Some cost- based Blue Cross plans treat charity care as an allow- able expense, and in some areas in which Blue Cross pays charges less a discount, payment includes a factor for bad debt and charity care incurred by all patients in the hospital. 3Under retroactive cost-based reimbursement, hos- pitals were paid on the basis of the cost of providing a service; thus, the greater the expenditure, the greater the reimbursement. Cost-based reimbursement con- tained few restraints on the provision of costly services

OCR for page 74
INVESTOR OWNERSHIP AND THE COSTS OF CARE and virtually no restraints on capital spending. On the contrary, expenditures on plant and equipment could be recaptured more swiftly by higher utilization, as well as through depreciation and other capital pay- ments. However, some restraints were imposed by dis- allowance of certain costs, by limits on costs per day, by utilization controls by professional services review organizations, and by capital controls under certificate of need. By paying a flat rate per diagnosis, the DRG approach severs the direct link between expenses and reimbursement and reverses the incentive for greater intensity of services in particular cases. A hospital with expenses that exceed the payment rate will lose money, and a hospital with expenses below the payment rate will earn income. Thus, new incentives exist to provide cost-effective care. 4In 1983 the average length of stay in for-profit chain hospitals was 6.4 days, compared with 6.7 days in for- profit independent and public state and local hospitals, 7.2 days in not-for-profit chain hospitals, 7.4 days in not-for-profit independent hospitals. One reason for the relatively shorter length of stay in for-profit chain hospitals might be that they treat less sick patients, which would presumably be shown by a lower Medi- care case-mix index; regional factors may also play a role, since hospitals on the West Coast have a sub- stantially shorter length of stay than East Coast hos- pitals. Indeed, two studies have found that after adjusting for differences in case mix, location, and other factors, differences in length of stay among ownership types virtually disappear (Coelen, 1986; Freund et al., 1985~. Watt et al. (1986b), however, in their multiple regres- sion analysis of national data, found that for-profit chain hospitals had only slightly lower Medicare case-mix indices than did not-for-profit chain hospitals (0.999 compared with 1.0171. 5One study (Ashby, 1982) showed more rapid growth of expenses in for-profit than not-for-profit hospitals between 1973 and 1978, but this study does not dis- tinguish investor-owned from proprietary for-profit hospitals. 6This differential was sustained as overall occupancy rates dropped to 66 percent in early 1985. The major for-profit chains had occupancy rates ranging from 55 percent at Hospital Corporation of American to 43 per- cent at Republic Health Corporation (Tatge, 1985a). Some not-for-profit multIhospital systems were run- ning closer to the national average-three of five listed in one recent report had occupancy rates between 60 and 64 percent, although one was 45 percent (Tatge, 1985b). 'Adjusted daily census, adjusted patient day, and adjusted admissions reflect inpatient care adjusted to reflect the volume of outpatient care provided. 8Data provided to the committee by the American Hospital Association show investor-owned chain hos- pitals had lower labor expenses per adjusted patient 95 day than did not-for-profit hospitals in three out of four regions of the United States and for the United States as a whole ($183.49 versus $212.31 in 1983~. However, because of the absence of controls for hospital size or range of services, it cannot be determined if these fac- tors, rather than type of ownership, are responsible for the difference. 9Data supplied to the committee by the American Hospital Association show 2,544,000 admissions in for- prof~t hospitals in 1983. Watt et al. (1986b) found net patient care revenues adjusted per admission at inves- tor-owned hospitals to average $184.55 more than the similar figure for not-for-profit chain hospitals and $137.94 more than independent not-for-profit hospi- tals. Thus, the difference is equivalent to $351-$469 million of the nation's total hospital expenditures of more than $100 billion. 20A problem of cost allocation should be noted with respect to routine daily service. Under Medicare cost- based reimbursement, it was advantageous to allocate expenses to routine daily service up to a set limit, beyond which payment was not made. The advantage occurred because Medicare paid hospitals the average cost for all patients for the routine daily service com- ponent of care, and, until 1982 paid hospitals an ad- ditional "nursing differential" payment on the assumption that Medicare patients required more nursing care than the average patient. However, the average cost for Medicare patients, who tend to have long stays, may actually have been lower than the average cost for other patients. 22 Hospital Corporation of America challenged sev- eral of GAO's findings and noted some omissions in the analysis-particularly that GAO omitted that Med- icare's share of the $55 million cost increase was only about $8 million. Furthermore, over $135 million in one-time federal capital gains tax liabilities were in- curred, which more than offset Medicare's total future cost increase James P. Smith, Hospital Corporation of America, Washington, D.C., personal communication, 1984~. 22 For a discussion of differences in Medicaid nursing home reimbursement policies and incentives among states, see Holahan (1983~. Some changes in reim- bursement methods are being considered, and some are being implemented. Medicare, which pays only 2 percent of nursing home costs, is considering moving to resource utilization groups similar in concept to Medicare's hospital prospective payment system. Some state Medicaid programs are using or considering reim- bursement systems that recognize differences in case max. REFERENCES American Hospital Association (1985) Economic Trends l(Summer):4.

OCR for page 74
96 Anderson, Gerard F., and Paul B. Ginsberg (1983) Prospective Capital Payments to Hospitals. Health Af- fairs 2(Fall):53-63. Ashby, John L., Jr. (1982) An Analysis of Hospital Costs by Cost Center, 1971 through 1978. Health Care Financing Review 4(September):37-53. Becker, Edmund R., and Frank A. Sloan (1985) Hos- pital Ownership and Performance. Economic Inquiry 23January):21~6. Blue Cross/Blue Shield of North Carolina (1983) Pro- prietary Hospitals. Charges to BCBSNC Subscribers. Durham, N. C: Blue CrosstBlue Shield of North Car- olina. Brown, Kad~ryn J., and Richard E. Klosterman (1986) Hospital Acquisitions and Their Effects: Florida, 1979- 1982. This volume. Clark, Robert (1980) Does the Nonprofit Forrn Fit the Hospital Industry? Harvard Law Review 93(May):1417-1489. Coelen, Craig G. (1986) Hospital Ownership and Comparative Hospital Costs. This volume. Conger, J. N. (1983) Letter to the Editor. Health Affairs 2(Fall):138-141. Ermann, Dan, and Jon Gabel (1986) Investor-Owned Multihospital Systems: A Synthesis of Research Find- ings. This volume. Freund, Deborah, et al. (1985) Analysis of Lengths- of-Stay Differences Between Investor-Owned and Vol- untary Hospitals. Inquiry 22(Spring):3344. lIawes, Catherine, and Charles D. Phillips (1986) The Changing Structure ofthe Nursing Home Industry and the Impact of Ownership on Quality, Cost, and Access. Ibis volume. Hiller, Marc D., and David B. Sugarman (1984) Pri- vate Nursing Homes in the U. S.: Effects of Ownership and Affiliation. Unpublished paper. University of New Hampshire, Durham, N.H. Holahan, John (1983) State Rate Setting and the Ef- fects on Nursing Home Costs. Working Pacer 3172- 05. Washington, D.C.: Urban Institute. Koetting, M. (1980)Nursing Home Organization and Efficiency. Lexington, Mass.: Lexington Books. Lee, Maw Lin (1911) A Conspicuous Production Theory of Hospital Behavior. Southern Economics Journal 28July) 48-58. Lewin, Lawrence S., Robert A. Derzon, and Rhea Margulies (1981) Investor-owneds and Nonprofits Dif FOR-PROFIT ENTERPRISE lN HEALTH CARE fer in Economic Performance. Hospitals ssuuly 1~:52- 58. Lewin, Lawrence S., Robert A. Derzon, and J. Mi- chael Watt (1983) Letter to the Editor. Health Affairs 2(Fall): 134-137. Pattison, Robert V. (1986) Response to Financial In- centives Among Investor-Owned and Not-for-Profit Hospitals: An Analysis Based on California Data, 1978- 1982. Ibis volume. Pattison, Robert V., and Hallie M. Katz (1983) Inves- tor-Owned and Not-for-Profit Hospitals. The New En- gland Journal of Medicine 309(Aug. 11~: 347-353. Prospective Payment Assessment Commission (198 Report and Recommendations to The Secretary, U.S. Department of Health and Human Services. Washing- ton, D.C.: U.S. Government Printing Office. Sloan, Frank A., and Robert A. Vraciu (1983) Inves- tor-Owned and Not^For-Profit Hospitals: Addressing Some Issues. Health Affairs 2(Spring):25-37. State of Florida Hospital Cost Containment Board (1984) Annual Report 1983-1984. Tallahasse: State of Flonda Hospital Cost Containment Board. Tatge, Mark (1985a) For-profits' Inpatient Occu- pancy Drops in Quarter; Outpatient Visits Are Rising. Modern Healthcare 15(May 10~:23. Tatge, Mark (1985b) Occupancy Rate Declines Level Off While Outpatient Visits Continue to Climb. Mod- ern Healthcare 15(May 10~:94. U. S. General Accounting Office (1983) Hospital Mergers Increase Medicare and Medicaid Payments for Capital Costs. (GAO/HAD 84-10~. Washington, D.C.: U.S. General Accounting Office. Watt, J. Michael, Robert A. Derzon, Steven C. Renn, and Carl J. Schramm (1986a) The Comparative Eco- nomic Performance of Investor-owned Chain and Not- for-profit Hospitals. The New England Journal of Med- icine 314(January 9~:89-96. Watt, J. Michael, Steven C. Renn, James S. Hahn, Robert A. Derzon, and Carl J. Schramm (1986b) The Effects of Ownership and Multihospital System Mem- bership on Hospital Functional Strategies and Eco- nomic Performance. Ibis volume. Yoder, Sunny G. (1980) Financing Graduate Medical Education. Pp. 107-147 in Graduate Medical Edllca- tion Present awl Prospective. A Call for Action. New York: Josiah Macy, Jr., Foundation.