- Email: [email protected]
The effect of third-party payers on the clinical decision making of physical therapists
Pergamon
0277-9536(94)00221-5
Soc. Sci. Med. Vol. 40, No. 7, pp. 873-879, 1995 Copyright © 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0277-9536/95 $9.50 + 0.00
THE EFFECT OF THIRD-PARTY PAYERS ON THE CLINICAL DECISION M A K I N G OF PHYSICAL THERAPISTS ROBIN MOREMENUILIj and ROSEMARYWOOD 2 IDepartment of Sociology, Northern Illinois University, DeKalb, IL 60115, U.S.A. and 2Department of Physical Therapy, Quinnipiac College, Mt. Carmel Avenue, Hamden, CT 06518, U.S.A. A~tract--According to Freidson, third-party payers have eroded the political and economic authority of medicine. To what extent is this also true for alternative practitioners such as physical therapists? The effect of Medicare's restrictive guidelines on physical therapy practice in skilled nursing facilities (SNFs) is examined. SNF physical therapists in Connecticut were surveyed (response rate 99%) using a mixture of open-ended and fixed-alternative responses. Results indicate that SNF physical therapists recognize Medicare criteria and view them as important. Twenty-fiveto 33% of SNF. therapists recommend care based on the guidelines.Younger therapists, therapists with fewer years in the field, and contract therapists are more greatly influenced by the guidelines than older, more experienced, staff therapists (P < 0.08). Those who recommend care based on the guidelines may do so because of possible nonpayment for services already rendered, or because they fear loss of their positions. The majority of Connecticut SNF patients who qualify do receive therapy and Medicare coverage. Therapists may use their knowledge of the guidelines to secure services for their patients, or SNFs may be selecting patients that have the best chances for recovery. Like physicians, SNF physical therapists are under pressure from third-party payers to economize and rationalize, but most continue to secure services for their patients.
Key words--medical dominance, physical therapy, third-party payers, Medicare
INTRODUCTION
Freidson [1-3] concedes that formalized, bureaucratic controls are chipping away at the dominance of the medical profession. Unlike others [4--9], he insists that medicine maintains significant cultural authority, but he recognizes that a "small, heavily concentrated group of public and private 'third-party' payers" has reduced the political and economic power of the profession [3, p. 7]. He suggests that third-party payers mediate the doctor-patient relationship by laying down such "restrictive rules and conditions" that doctors and patients often are prevented from getting the services they want and need [3, p. 12]. Payers impose restrictions on the resources that are available to doctors and patients by routinely reviewing and challenging claims for reimbursement [3, p. 10; 91. If third-party payers are challenging the clinical freedom of physicians, what effect are they having on practitioners outside of the mainstream? Many unconventional therapies are not even recognized or reimbursed by payers such as the government and insurance companies [10]. But, of those that are, how have the economizing and rationalizing efforts of third-party payers affected the decisions these practitioners make about the care they recommend? We use physical therapy to address these concerns. Physical therapists are reimbursed by third-party payers in all 50 states. Traditionally, the payment
of physical therapy services is subject to physician authorization, however, recent legislative efforts have challenged these institutionalized practices in some states. As of May 1994, 30 states passed direct access legislation that permits physical therapists to practice independent of physician referral [conversation with official of the American Physical Therapy Association (APTA), 5/24/94]. As physical therapists seek greater autonomy, what effect do the rules of thirdparty payers have on their clinical freedom? Physical therapists practice in a wide variety of settings. For the purposes of this study, we choose to look at physical therapy in skilled nursing facilities (SNFs). The primary third-party payer of physical therapy services in SNFs is Medicare. Using a sample of SNF physical therapists in Connecticut, we examine the effect of Medicare's Part A rules on their clinical decision making.
Medicare rules Part A Medicare was designed to pay for room, board and therapy services in a SNF when physical therapy is needed on a daily basis (daily, as defined by the government, is five days per week). A complex set of government rules has evolved to determine 'need' [11-14]. Theoretically, the Medicare statute is the source of these rules [42 CFR Chap. IV]; however, a more restrictive set of guidelines has been imposed on practitioners (Health Insurance Manual [131).
873
874
ROBIN MOREMENUILI and ROSEMARYWOOD
While the guidelines are intended to clarify the statute, they are decidedly more limited. Where the statute says: The restoration [i.e., recovery] potential of a patient is not the deciding factor in determining whether skilled services are needed. Even if full recovery or medical improvement is not possible, a patient may need skilled services to prevent further deterioration or preserve current capabilities [42 CFR sec. 409.32(c), 1987], the guidelines state that skilled physical therapy is needed only when there is the "expectation that the condition will improve significantly in a reasonable (and generally predictable) period of time [Health Insurance Manual 13, Section 3101.8B(c)]". Progress must be measurable and swift (usually within two weeks); if not, the services are considered 'unreasonable and unnecessary' and payment is denied. The guidelines have become the government's normative standard for Part A Medicare coverage in SNFs because, in times of tightening resources, they limit reimbursement and reduce federal expenditures. Moreover, the government does not review claims itself; it delegates this task to fiscal intermediaries (FIs) who review and pay (or deny) claims on its behalf. Fiscal intermediaries make decisions about payment retrospectively. This permits FIs to determine compliance with the guidelines after care has been given. In the 1980s, the policies of the Reagan administration made SNF Medicare coverage for physical therapy even more difficult to obtain. The administration's politically-defined economic crisis resulted in cost containment and federal funding cuts [15] that put increasing pressure on the FIs to deny even more Part A Medicare claims. Feder and Scanlon [12] discuss how Medicare restricted coverage through exclusions for custodial care, lack of significant improvement and maintenance therapy. Loeser, Dickstein and Schiavone [13] contend that the FIs went beyond their own guidelines to further limit Part A coverage, implementing an arbitrary and unpredictable set of rules. Smits, Feder and Scanlon [14] look at the claims review process of eighteen FIs and find tremendous variation in their decision making. They conclude that coverage is based on an arbitrary set of criteria that "often were implicit, unwritten, and not available for perusal or comment" [14, p. 860]. Adler and Brown [11] examine the cases of 61 people who were denied Part A Medicare coverage for physical therapy and conclude that the fiscal intermediaries did so without objective evidence (i.e. there was no medically valid reason for denying payment). The alleged purpose of these 'arbitrary and restrictive' standards is cost containment; fiscal criteria are being imposed on medical decision making in order to curtail federal expenditures. The overall effect is to deny elderly sick people access to Medicare-covered services. Individuals of modest means are hardest hit because they have neither wealth nor the welfare
system to protect them. Instead, they pay out-ofpocket for services that often amount to thousands of dollars a month [13]. Some choose to defer physical therapy treatment because it represents an additional expense they can ill afford [1 I]. The negative consequences of deferring treatment are profound; without rehabilitation, few recover sufficiently to be able to return home [16, 17]. Insurance rules place SNF physical therapists in an awkward position. How are they to get services for their patients and still get paid by Medicare? To what extent are therapists even aware of Medicare's normative standards? Do these standards limit the provision of physical therapy services? How do these standards affect physical therapists' recommendations for care? The purpose of this study is to clarify the effect of Medicare's Part A rules on the clinical decision making of SNF physical therapists.
METHODS
In April 1989 we surveyed SNF physical therapists in Connecticut (n = 184) as part of a federally-funded evaluation of admissions procedures in all Medicarecertified SNFs in Connecticut. The surveys were designed to understand how SNF physical therapists made treatment decisions, using a mixture of openended and fixed-alternative questions. The response rate was 99%. (The evaluation was ordered by the federal court in Connecticut; the court's authority enabled us to achieve such a high response rate; only two surveys were not returned.) Measures Direct measures. We included three direct measures of Medicare's normative standards. They were 'restoration potential', 'measurable progress' and 'maintenance therapy'. These measures came from the guidelines. The guidelines indicated that the FIs were to make payment decisions based on the restoration potential of the patient and the patient's ability to make measurable progress in therapy. Patients with good potential for improvement and recovery were entitled to daily physical therapy and Medicare benefits; maintenance therapy was a type of therapy reserved for patients whose improvement had plateaued or for whom full recovery was not possible (they usually received it less than daily, i.e. three times per week). We included these measures to establish the therapists' familiarity with Medicare's normative standards of care. Each measure was assessed on a four-point scale ranging from one (often important) to four (never important). A mean ranking was derived for each measure. Levels of care. The SNF physical therapists also were asked to list three criteria they used when deciding about different levels of care (daily physical therapy, less than daily physical therapy and no
875
Clinical decision making of physical therapists physical therapy treatment). We were interested in the extent to which normative fiscal standards figured into their open-ended responses. All criteria were entered into a database and sorted alphanumerically by field, line and word. The resulting lists were independently sorted by the two investigators and items associated with each level of care were grouped according to similarity of content. For example, 'potential to reach goals', 'potential for home discharge', 'good potential for rehab', 'realistic potential for positive results', 'potential for recovery' and 'restorative potential' were grouped together to comprise the following indicator for daily physical therapy: 'Rehab/Restorative Potential: Good/Excellent'. The top five indicators for each level of care were retained. Physical therapy evaluations. A third measure consisted of SNF physical therapists' responses to real patient situations. We deliberately chose cases where prospects for improvement and recovery varied. We wanted to see whether or not therapists made recommendations based on the patient's rehabilitation potential and ability to make measurable progress in therapy. Two of the four cases had complex medical problems, but presented with complications that severely affected their prospects for improvement and recovery (John Doe No. 1 had an amputation, but also had a neurological condition since birth that prevented him from using a prosthesis; John Doe No. 2 had a hip fracture, but also had poor eyesight and an old stroke on the same side as the hip fracture; neither had plans to return home). The third case had a medical problem similar to the first two, but presented with a good prognosis for improvement and recovery (Jane Doe No. 1 had a hip fracture with no complications, was beginning to walk and had definite plans to return home), and the final case was chosen as a distractor (Jane Doe No. 2 was independent and did not require the skills of a physical therapist). Please refer to the Appendix for detailed descriptions of the four cases. The case studies were taken from actual assessments of SNF residents in the larger study. A standardized admission evaluation form is used by SNF physical therapists in Connecticut. We included standardized forms for each of the four cases in the survey. There was no identifying information on the forms. Each form contained subjective and objective findings, a list of patient problems, short- and longterm goals and a recommended treatment plan. At the end of each form, the SNF physical therapists were asked to indicate whether or not daily, less than daily or no physical therapy treatment was needed, based on the information contained in the form. We asked three physical therapists with extensive long-term care experience and no involvement in Connecticut SNFs to serve as experts and to define standards of care for each patient. They were instructed to base their recommendations on the need for therapy services and not on fiscal or payment
criteria of any kind. They concluded, with 100% agreement, that Jmae Doe No. i, John Doe No. ! and John Doe No. 2 all needed daily physical therapy to achieve their short- and long-term goals and to carry out their treatment plans. They also concluded that no physical therapy treatment was needed in the case of Jane Doe No. 2. The responses of the SNF physical therapists in the survey were compared with the standards set by the experts and the percentage agreement was calculated in each of the four cases. Where there was agreement, the respondent was assigned a score of one; where there was disagreement, a score of zero was assigned. The scores for each case were averaged and the average scores were converted to percentages. The extent of agreement between the four cases was tested for significant differences. A paired-comparisons t-test was used. A new variable was created that contained the difference between each pair of scores, and a Student's t and probability value were calculated to test whether the mean difference was significantly different from zero [18]. The sample was then divided into two groups: those that agreed with the experts and those that did not. The two groups were compared for significant differences by sex, ethnicity, age, years in the field, staff/contract relationship and type of physical therapy training. RESULTS
Demographic data Seventy-nine percent of the SNFs in Connecticut are proprietary. The average facility size is 109 beds. The primary payer of SNF services is Medicaid (59%), followed by Private Payment (30%) and Medicare (8%) (approx. 3% of the beds are empty). Table 1 provides demographic information on the sample of SNF physical therapists that responded to our survey. Based on the information contained in the table, the prototypical respondent was a white woman in her thirties who had basic baccalaureate training in physical therapy and who had been out in the field for quite some time. She was more likely to be on contract to the facility than a staff member and the facility where she practiced did not have a particularly strong rehabilitation focus.
Direct measures Table 2 indicates a high degree of familiarity with Medicare's standards. 'Restoration potential' and 'measurable progress' were rated often important and 'maintenance therapy' was rated occasionally important by the sample.
Criteria for levels of care The top five open-ended responses for each level of care are listed in rank order in Table 3. As Table 3 indicates, restoration potential and the ability to make progress or improve in physical therapy figure
876
ROBIN MOREMEN UILI a n d ROSEMARY WOOD Table 1. Demographic characteristics of SNF physical therapists 1989 (n = 182) (%)
Measure Sex
Female Male
82 18
Table 3. Rank ordering of criteria used by Connecticut physical therapists for determining level of care needs (n = 177)~
Criteria .for daily physical therapy 1. 2. 3. 4. 5.
Rehab/restoration potential: good/excellent Diagnosis: acute/recent/new Prior level of function Ability to progress/improve: good/excellent Need for skilled services
Criteria Jbr less than daily physical therapy
Ethnicity African american Asian Caucasian Hispanic Other
1.0 0.5 97.5 0.5 0.5
I. 2. 3. 4. 5.
Ability to progress/improve: slower/none/plateaued Self/others carry out/assist with program Rehab/restoration potential: fair/limited/poor Diagnosis: acute/chronic/old Prior level of function
Criteria for no physical therapy treatment
Age Under 30 yr 30-39 yr 40-49 yr 50-59 yr 60-69 yr
13 41 26 13 7
Number of years in the field <2 2-5 6-9 10-14 15-19 >19
5 11 I1 21 15 37
Rehab focus of facility Limited Moderate Strong
22 38 40
Relationship to facility Contract P.T. Staff P.T.
60 40
Type of P.T. training Bachelors program Certificate program Entry-level masters program Post professional masters program Doctoral program
83 7 2 7 1
heavily in each of the levels of care. Also important are the patient's diagnosis, his or her prior level of function, the need for skilled services and the availability of other personnel. When the top five indicators are used to create 'ideal-typical' patients for each level of care, the following characterizations emerge. Daily physical therapy. The SNF therapists in the survey felt that daily physical therapy was most appropriate for a person with a recent, acute medical problem (such as a hip fracture) who had 'good' or 'excellent' potential for recovery and the ability to make 'good' or 'excellent' progress toward a prior level of function, and for whom skilled physical therapy was necessary to achieve those goals.
1. 2. 3. 4.
Self/others carry out program/patient independent Rehab/restoration potential: poor/none Diagnosis: long-term/chronic/old/unstable/terminal Ability to progress/improve: plateaued/none, or goals met/ optimal benefit 5. No need for skilled services
aUsable responses.
Less than daily physical therapy. The respondents felt that less than daily physical therapy was appropriate for a person with an acute or chronic problem who had 'fair' or 'limited' potential for recovery, whose progress toward a prior level of function was slower or had plateaued, and for whom skilled services may or may not be required (i.e. other staff members or the patient her- or himself might be able to carry out or assist with the program). No physical therapy. The SNF physical therapists felt that no physical therapy treatment was appropriate for a person with a chronic long-term problem, or for a person who already was independent. If the problem was chronic or long-term, there usually was 'little' or 'no' potential for progress or recovery and no need for skilled physical therapy because other staff members (i.e. nurses, aides) could provide the needed care. If the patient was independent, the independence was present on admission or was achieved earlier in the stay. Physical therapy evaluations The majority of SNF physical therapists in the survey agreed with the standards set forth by the experts (i.e. that Jane Doe No. 1, John Doe No. 1 and John Doe No. 2 all needed daily physical therapy to achieve their goals and that Jane Doe No. 2 did not require care) (see Table 4). The extent of agreement varied, however, and significant differences were noted. There was significantly more agreement about the need for daily services for Jane Doe No. 1 than for either John Doe No. 1 or John Doe No. 2, with
Table 2. Physical therapists' mean rankings on the degree of importance of restoration potential, measurable progress and maintenance therapy Measures Restoration potential ( n = 179)a Measurable progress (n = 180) Maintenance therapy (n=177)
Mean R a n k i n g + S D 1.3+0.6 1.4 + 0.6 1.8 +__0.9
'Usable responses. Key: 1, often important; 2, occasionally important; 3, seldom important; 4, never important.
agreement being the lowest for John Doe No. 1. Fully one-third of the respondents (33%) felt that John Doe No. 1 did not need daily physical therapy, as compared with 25% for John Doe No. 2 and 3.3% for Jane Doe No. 1.
None of the demographic variables were significant for Jane Doe No. 1 or No. 2. In the case of John Doe No. 1, there were significant differences by age (Z2 9.6, df = 4 , P = 0.048) and staff/contract relationship (Z 2
Clinical decision making of physical therapists 3.1, df = 1, P = 0.079), and for John Doe No. 2 there were significant differences by age (X2 19.58, df= 4, P = 0.001) and number of years in the field (X2 24.52, df = 5, P = 0.000). Therapists under the age of 30 and those between the ages of 60 and 69 were significantly less likely to see the need for daily physical therapy in both cases than therapists in their thirties and forties. Therapists in practice from two to five years were significantly less likely to see the need for daily care in the case of John Doe No. 2 than therapists who had been in the field for ten or more years. And, therapists who were on contract to the facility were significantly less likely to recommend daily care in the case of John Doe No. 1 than staff therapists. DISCUSSION
The results of this study indicate that SNF physical therapists in Connecticut are very familiar with Medicare's normative standards and that these standards do affect their recommendations for care in simulated patient situations. Restoration potential and measurable progress (and, to a lesser degree, maintenance therapy) are rated 'often important' by the SNF physical therapists; they are among the criteria used to differentiate between levels of care in the openended responses; and one-quarter to one-third of the respondents do not recommend daily care when potential for improvement and recovery are questionable. Furthermore, younger therapists (and very old therapists, although the number of respondents in this category is small), contract therapists, and therapists with fewer years in the field are more likely to make recommendations based on fiscal criteria than older, more experienced, staff therapists. Nearly all of the SNF therapists (97%) recommend daily care when the medical condition is complex and rehabilitation potential is clear (Jane Doe No. 1). We
877
may infer from this that well-defined cases receive benefits and services without difficulty. The situation is more tenuous when improvement and recovery are questionable, because these are the standards that Medicare has imposed to limit payments. In the majority of these circumstances, SNF therapists also recommend daily care, but in a certain percentage of the cases they do not. When this occurs, access to benefits and services is severely curtailed. How do our findings compare with actual data on Medicare coverage? Most individuals who qualify for services are granted Medicare coverage at the level of the SNF (data from early 1989) [19]. On average, 67% qualify on technical grounds [19, p. 15] (i.e. they have a prior hospital stay of at least three days, there is a period of no more than 30 days between discharge from the hospital and admission to the nursing home, and they are receiving care for the same condition for which they were hospitalized), 36% have orders for physical therapy [19, p. 6], and, of those, 82% receive Medicare coverage [19, p. 34]. These figures are similar to the overall recommendations of the SNF physical therapists in the survey. What happens when these claims are forwarded to the FIs for review and payment? When covered physical therapy claims are reviewed by the FIs, on average, 93% are paid in full [20, p. 69]. There is variation across the three Connecticut FIs, however. The FI responsible for the most SNFs (52%), pays only 73% of its claims in full; the FI responsible for 31% of the SNFs pays 91% in full; and the FI responsible for the least number of SNFs (17%) pays nearly 100% in full [20, p. 67; 70]. This variation may produce uncertainty for some physical therapists and may affect their recommendations for care. Therapists who are reluctant to recommend daily care when recovery is questionable,
Table 4. Agreement between Connecticut S N F physical therapists and experts regarding standardized care Evaluations
Agreement (%)
Jane Doe No. l (n = 181) d Right hip fracture, independent at home, able to transfer and walk with a walker with assistance, goal to return home
96.7 a
John Doe No. I (n = 179) Above-the-knee amputation, independent at home, no prosthesis due to neurological condition, transfers with two people, no immediate plans to return home
67.0 b
John Doe No. 2 (n = 180) Left hip fracture, prior S N F resident, independent at S N F , paralysis left side due to old stroke, poor eyesight, transfers with assistance, refuses to walk at time of assessment
75.0
Jane Doe No. 2 (n = 181) Bronchitis, fairly independent at h o m e with son, long-term placement at S N F , stands, transfers and walks independently
84.5 ~
~The m e a n difference in scores between J a n e Doe No. ! and John Doe No. 1, John Doe No. 2, and Jane Doe No. 2 was significantly different from zero ( P = 0.0001). bThe mean difference in scores between John Doe No. I and John Doe No. 2 was not significantly different from zero ( P = 0.09). CThe m e a n difference in scores between Jane Doe No. 2 and John Doe No, I and John Doe No. 2 was significantly different from zero ( P = 0.0002 and P = 0.03, respectively). d Usable responses.
878
ROBIN MOREMENU1LI and ROSEMARYWOOD
may work in SNFs where the FI denies 27% of its claims. Nonpayment of services--particularly if one has to deal with the same FI repeatedly--is a powerful motivator to limit daily care to cases where progress and recovery are beyond dispute. There are other reasons why fiscal standards may affect recommendations for care. Most physical therapists are used to being dominated by physicians [1-3] and by the bureaucratic authority of the organizations that employ them [21]. Fiscal rules may represent one more constraint on their freedom. Moreover, the risks of noncompliance are fairly severe. If a claim is denied, the SNF may lose payment for services it has already rendered; if several denials accrue, the therapist's job or contract may be jeopardized. Some therapists may knowingly comply with fiscal rules in order to secure their positions in the medical hierarchy. It may also be the case that some therapists are unable to provide daily care. With the national shortage of physical therapists [22], some SNFs may share contract services with several other SNFs and may be forced to accept less than daily coverage. Likewise, some therapists may not be as willing or as able to discern between improvement as a 'goal' and improvement as a 'requirement' for payment. Nearly all therapists want their patients to improve, but, because time and resources are limited, some therapists may choose--actively or unwittingly--to concentrate their most intensive efforts on patients with the greatest potential for recovery. Similarly, there may be some therapists who do not provide daily care because they maintain stereotypically negative views [23] about an elderly person's ability to improve or regain lost function. Or, perhaps some patients are unable to tolerate daily treatment because they have complicated, fragile or unstable conditions. Given the narrow focus of the guidelines, how are most therapists able to secure benefits and services for their patients, even when recovery is uncertain? Perhaps, as Freidson explains, providers and patients "join together to get their own way" [3, p. 12]. Because third-party payers, such as Medicare, do not have direct access to patient-provider interactions, they must rely on the record afforded them by the provider. Because records are arbitrary constructs [3, p. 11], they can be made to serve the interests of the record-maker. Some SNF therapists may actually use their knowledge of Medicare's rules to obtain services for their patients. By incorporating the vocabulary of the rules into the patient's record, the therapist may construct an account of treatment that is more likely to survive the scrutiny of the FI's review process. Many elderly patients are capable of making slow progress in therapy, and careful wording of the patient's record may permit valuable time to maximize whatever potential is available. A phenomenon known as 'skimming' may also work in their favor. Providers, including nursing homes, often select patients that require little care,
because reduced care means lower costs for the facility [24-28]. If Connecticut SNFs routinely skim patients, they may be eliminating all but the 'best' rehabilitation candidates; patients that have no trouble complying with Medicare's rules and that bring increased federal revenues to the facility. According to the survey, younger therapists and therapists with little experience apply fiscal rules most frequently. Perhaps the youth and inexperience of these therapists makes them less confident about the types of patients that benefit from therapy; perhaps they are less knowledgeable about record keeping and Medicare's rules; or they may be more fearful of losing newly-acquired positions. Also, contract therapists appear to apply fiscal criteria more willingly than staff therapists. While staff therapists may be expected to pay more attention to facility policies and procedures, contract therapists may be more easily replaced if services are not paid by Medicare. In fact, bids for services may depend on the ability of the therapist to render reimbursable services. Staff therapists may be afforded greater leniency in this regard. The findings presented here only address the recognized importance of Medicare's guidelines by SNF physical therapists in Connecticut. Because Medicare is a federal program, it is important to test these findings with SNF physical therapists in other states and in regions beyond the northeast. It also would be useful to look at other guidelines, other practitioners and other settings.
CONCLUSION
Like many alternative providers, physical therapists seek greater autonomy, but they must pursue their ambitions in an environment that increasingly stresses the formalization of care. In the case of SNF physical therapists, one aspect of that environment consists of fiscal guidelines established by third-party payers such as Medicare. Physicians are wrestling with similar threats to their political and economic freedom, but continue to secure services for the majority of their patients. Physical therapists are not unlike physicians in this regard. However, they do differ in the degree of cultural authority they exercise. Medicine's authority is institutionalized in the policies of the state [3]; physical therapy exists on the margins, and, as such, is more vulnerable to policy changes. Consequently, as Congress entertains even deeper cuts in the Medicare program, and as health reform moves in the direction of managed competition, physical therapists must make every effort to maintain their position in order to continue to secure services for their patients. research was funded by the Health Care Financing Administration as part of the evaluation and monitoring of the Fox v. Bowen Federal Court decision in Connecticut Skilled Nursing Facilities and by NIMH training Grant MH15783-13. Acknowledgements--This
879
Clinical decision making of physical therapists REFERENCES
I. Freidson E. The reorganization of the medical profession. Med. Care Rev. 42, l l, 1985. 2. Freidson E. The medical profession in transition. In Applications of Social Science to Clinical Medicine (Edited by Aiken L. and Mechanic D.), p. 63. Rutgers University Press, New Brunswick, N J, 1986. 3. Freidson E. Power is Plural. Paper presented at the Annual Meeting of the American Sociological Association, Washington, DC, August 14, 1990. 4. Haug M. R. Deprofessionalization: an alternative hypothesis for the future. SocioL Rev. Monographs 20, 195, 1973. 5. Haug M. R. A re-examination of the hypothesis of physician deprofessionalization. The Milbank Q. 66, 48, 1988. 6. Light D. and Levine S. The changing character of the medical profession: a theoretical overview. The Milbank Q. 66, 10, 1988. 7. McKinlay J. B. On the professional regulation of change. Sociol. Rev. Monographs 20, 61, 1973. 8. Navarro V. Professional dominance or proletarianization?: Neither. The Milbank Q. 66, 57, 1988. 9. Stoeckle J. D. Reflections on modern doctoring. The Milbank Q. 66, 77, 1988. 10. Eisenberg D. M., Kessler R. C., Foster C., Norlock F. E., Calkins D. R. and DelBanco T. L. Unconventional medicine in the United States: prevalence, costs, and patterns of use. N. Engl. J. Med. 328, 246, 1993. 11. Adler M. and Brown C. C. Jr. Who reviews the reviewers? A retrospective study of Medicare determinations. N. Engl. J. Med. 302, 842, 1980. 12. Feder J. and Scanlon W. The underused benefit: Medicare's coverage of nursing home care. Milbank Memor. Fund Q./Hlth Soc. 60, 604, 1982. 13. Loeser W. D., Dickstein E. S. and Schiavone L. D. Medicare coverage in nursing h o m e s - - a broken promise. N. Engl. J. Med. 304, 353, 1981, 14. Smits H. L., Feder J. and Scanlon W. Medicare's nursing-home benefit: variations in interpretation. N. Engl. J. Med. 307, 855, 1982. 15. Estes C. L. The Reagan legacy: privatization, the welfare state and aging into the 1990s. In States, Labor Markets and the Future of Old Age Policy (Edited by Myles J. and Quadagno J.), p. 109. Temple University Press, Philadelphia, 1991. 16. Uili R. M. Resident selection in a Connecticut nursing home: a view from within. J. Aging Hlth, forthcoming 17. Tisdale S. Harvest Moon: Portrait of a Nursing Home, p. 11. Henry Holt and Company, New York, 1987. 18. SAS Institute Inc. SAS User's Guide: Statistics, Version 5 Edition, p. 799. SAS Institute Inc., Cary, NC, 1985. 19. Uili R. M. and Leaf P. J. An expansion of Medicare benefits: intended and unintended consequences, unpublished manuscript. 20. Leaf P. J. and Uili R. M. Outcomes of Fox v. Bowen: Final report of the Fox v. Bowen evaluation project. Submitted to Robert A. Burt, Special Master, Fox v. Bowen Federal Court Decision, State of Connecticut. Center for Health and Policy Research, Department of Epidemiology and Public Health, Yale School of Medicine, New Haven, CT, 1 August 1990. 21. Alexander J. A. and D'Aunno T. A. Transformation of institutional environments: perspectives on the corporatization of U.S. health care. In Innovations in Health Care Delivery: Insights for Organization Theory (By Mick S. S. and Associates), p. 53. Jossey-Bass Publishers, San Francisco, 1990. 22. Institute of Medicine. Allied Health Services: Avoiding Crises. National Academy Press, Washington, DC, 1989.
23. Cockerham W. C. This Aging Society, p. 131. Prentice Hall, Englewood Cliffs, N J, 1991. 24. Harrington C. Public policy and the nursing home industry. In Readings in the Political Economy t~f Aging (Edited by Minkler M. and Estes C. L.), p. 144. Baywood, Farmingdale, NY, 1984. 25. Relman A. S. The new medical-industrial complex. N. Engl. J. Med. 303, 963, 1980. 26. Relman A. S. Selling to the for-profits: undermining the mission. Hlth Progr. 66, 81, 1985. 27. Steinwald B. and Neuhauser D. The role of the proprietary hospital. Law Contemporary Problems 35, 817, 1970. 28. Stevens R. In Sickness and in Wealth: American Hospitals in the Twentieth Century, p. 342. Basic Books, New York, 1989.
APPENDIX Jane Doe No. 1 was a patient who was admitted to the SNF following hospitalization for a right hip fracture. Prior to her fall, she had been independent at home. At the time of the evaluation she was able to transfer from one surface to another with the minimal to moderate assistance of one other person and walk 15 ft with a walker with the minimal to moderate assistance of one other person. Her long-term goal was to walk without assistive devices and return home to independent living. Her treatment plan included progressive resistive exercises, bed mobility and transfer training, and gait training. John Doe No. l was a patient who was admitted to the SNF following an above-the-knee amputation due to severe peripheral vascular disease. The physician had determined that prosthetic training was unrealistic due to a neurological condition that the patient had had since birth. Although the patient had been independent at home prior to the amputation, he did not have immediate plans to return home. At the time of the evaluation, he was able to stand and transfer between surfaces with the maximal assistance of two people. His long-term goal was to become independent in a wheelchair. His treatment plan included therapeutic exercises to all four extremities and his trunk, sitting balance exercises, transfer training, stand-pivot activities and wheelchair mobility training. Jane Doe No. 2 was a patient who had been admitted to the SNF after being hospitalized for bronchitis. Prior to her hospitalization, she had lived fairly independently at home with her son. She was a long-term placement in the SNF at the time of her admission because her son had taken ill. She was able to stand, transfer between surfaces and walk by herself independently, although she had a flat-footed gait on level surfaces and some shortness of breath when climbing three stairs. Her goals were to maintain her level of independence on the unit and her treatment plan consisted of nursing supervision for wandering. As she was a long-term placement, stairs were not a major concern. John Doe No. 2 was a patient who had been a resident of the SNF for six years. He was admitted to the hospital and readmitted to the SNF after an operation for a left hip fracture. He also was paralyzed on the left side due to an old stroke. Prior to his hip fracture, he had been able to walk by himself about the facility. He had poor eyesight and, since the fracture, was fearful of falling. At the time of the evaluation, he could stand and transfer between surfaces with the moderate to maximal assistance of one additional person, but he refused to walk at that time. His long-term goal was to achieve independent transfers and supervised walking with a hemi-walker in the facility. His treatment plan consisted of therapeutic exercises to his lower extremities, transfer training, wheelchair mobility instruction, gait training with a walker and staff inservicing on proper positioning.
0277-9536(94)00221-5
Soc. Sci. Med. Vol. 40, No. 7, pp. 873-879, 1995 Copyright © 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0277-9536/95 $9.50 + 0.00
THE EFFECT OF THIRD-PARTY PAYERS ON THE CLINICAL DECISION M A K I N G OF PHYSICAL THERAPISTS ROBIN MOREMENUILIj and ROSEMARYWOOD 2 IDepartment of Sociology, Northern Illinois University, DeKalb, IL 60115, U.S.A. and 2Department of Physical Therapy, Quinnipiac College, Mt. Carmel Avenue, Hamden, CT 06518, U.S.A. A~tract--According to Freidson, third-party payers have eroded the political and economic authority of medicine. To what extent is this also true for alternative practitioners such as physical therapists? The effect of Medicare's restrictive guidelines on physical therapy practice in skilled nursing facilities (SNFs) is examined. SNF physical therapists in Connecticut were surveyed (response rate 99%) using a mixture of open-ended and fixed-alternative responses. Results indicate that SNF physical therapists recognize Medicare criteria and view them as important. Twenty-fiveto 33% of SNF. therapists recommend care based on the guidelines.Younger therapists, therapists with fewer years in the field, and contract therapists are more greatly influenced by the guidelines than older, more experienced, staff therapists (P < 0.08). Those who recommend care based on the guidelines may do so because of possible nonpayment for services already rendered, or because they fear loss of their positions. The majority of Connecticut SNF patients who qualify do receive therapy and Medicare coverage. Therapists may use their knowledge of the guidelines to secure services for their patients, or SNFs may be selecting patients that have the best chances for recovery. Like physicians, SNF physical therapists are under pressure from third-party payers to economize and rationalize, but most continue to secure services for their patients.
Key words--medical dominance, physical therapy, third-party payers, Medicare
INTRODUCTION
Freidson [1-3] concedes that formalized, bureaucratic controls are chipping away at the dominance of the medical profession. Unlike others [4--9], he insists that medicine maintains significant cultural authority, but he recognizes that a "small, heavily concentrated group of public and private 'third-party' payers" has reduced the political and economic power of the profession [3, p. 7]. He suggests that third-party payers mediate the doctor-patient relationship by laying down such "restrictive rules and conditions" that doctors and patients often are prevented from getting the services they want and need [3, p. 12]. Payers impose restrictions on the resources that are available to doctors and patients by routinely reviewing and challenging claims for reimbursement [3, p. 10; 91. If third-party payers are challenging the clinical freedom of physicians, what effect are they having on practitioners outside of the mainstream? Many unconventional therapies are not even recognized or reimbursed by payers such as the government and insurance companies [10]. But, of those that are, how have the economizing and rationalizing efforts of third-party payers affected the decisions these practitioners make about the care they recommend? We use physical therapy to address these concerns. Physical therapists are reimbursed by third-party payers in all 50 states. Traditionally, the payment
of physical therapy services is subject to physician authorization, however, recent legislative efforts have challenged these institutionalized practices in some states. As of May 1994, 30 states passed direct access legislation that permits physical therapists to practice independent of physician referral [conversation with official of the American Physical Therapy Association (APTA), 5/24/94]. As physical therapists seek greater autonomy, what effect do the rules of thirdparty payers have on their clinical freedom? Physical therapists practice in a wide variety of settings. For the purposes of this study, we choose to look at physical therapy in skilled nursing facilities (SNFs). The primary third-party payer of physical therapy services in SNFs is Medicare. Using a sample of SNF physical therapists in Connecticut, we examine the effect of Medicare's Part A rules on their clinical decision making.
Medicare rules Part A Medicare was designed to pay for room, board and therapy services in a SNF when physical therapy is needed on a daily basis (daily, as defined by the government, is five days per week). A complex set of government rules has evolved to determine 'need' [11-14]. Theoretically, the Medicare statute is the source of these rules [42 CFR Chap. IV]; however, a more restrictive set of guidelines has been imposed on practitioners (Health Insurance Manual [131).
873
874
ROBIN MOREMENUILI and ROSEMARYWOOD
While the guidelines are intended to clarify the statute, they are decidedly more limited. Where the statute says: The restoration [i.e., recovery] potential of a patient is not the deciding factor in determining whether skilled services are needed. Even if full recovery or medical improvement is not possible, a patient may need skilled services to prevent further deterioration or preserve current capabilities [42 CFR sec. 409.32(c), 1987], the guidelines state that skilled physical therapy is needed only when there is the "expectation that the condition will improve significantly in a reasonable (and generally predictable) period of time [Health Insurance Manual 13, Section 3101.8B(c)]". Progress must be measurable and swift (usually within two weeks); if not, the services are considered 'unreasonable and unnecessary' and payment is denied. The guidelines have become the government's normative standard for Part A Medicare coverage in SNFs because, in times of tightening resources, they limit reimbursement and reduce federal expenditures. Moreover, the government does not review claims itself; it delegates this task to fiscal intermediaries (FIs) who review and pay (or deny) claims on its behalf. Fiscal intermediaries make decisions about payment retrospectively. This permits FIs to determine compliance with the guidelines after care has been given. In the 1980s, the policies of the Reagan administration made SNF Medicare coverage for physical therapy even more difficult to obtain. The administration's politically-defined economic crisis resulted in cost containment and federal funding cuts [15] that put increasing pressure on the FIs to deny even more Part A Medicare claims. Feder and Scanlon [12] discuss how Medicare restricted coverage through exclusions for custodial care, lack of significant improvement and maintenance therapy. Loeser, Dickstein and Schiavone [13] contend that the FIs went beyond their own guidelines to further limit Part A coverage, implementing an arbitrary and unpredictable set of rules. Smits, Feder and Scanlon [14] look at the claims review process of eighteen FIs and find tremendous variation in their decision making. They conclude that coverage is based on an arbitrary set of criteria that "often were implicit, unwritten, and not available for perusal or comment" [14, p. 860]. Adler and Brown [11] examine the cases of 61 people who were denied Part A Medicare coverage for physical therapy and conclude that the fiscal intermediaries did so without objective evidence (i.e. there was no medically valid reason for denying payment). The alleged purpose of these 'arbitrary and restrictive' standards is cost containment; fiscal criteria are being imposed on medical decision making in order to curtail federal expenditures. The overall effect is to deny elderly sick people access to Medicare-covered services. Individuals of modest means are hardest hit because they have neither wealth nor the welfare
system to protect them. Instead, they pay out-ofpocket for services that often amount to thousands of dollars a month [13]. Some choose to defer physical therapy treatment because it represents an additional expense they can ill afford [1 I]. The negative consequences of deferring treatment are profound; without rehabilitation, few recover sufficiently to be able to return home [16, 17]. Insurance rules place SNF physical therapists in an awkward position. How are they to get services for their patients and still get paid by Medicare? To what extent are therapists even aware of Medicare's normative standards? Do these standards limit the provision of physical therapy services? How do these standards affect physical therapists' recommendations for care? The purpose of this study is to clarify the effect of Medicare's Part A rules on the clinical decision making of SNF physical therapists.
METHODS
In April 1989 we surveyed SNF physical therapists in Connecticut (n = 184) as part of a federally-funded evaluation of admissions procedures in all Medicarecertified SNFs in Connecticut. The surveys were designed to understand how SNF physical therapists made treatment decisions, using a mixture of openended and fixed-alternative questions. The response rate was 99%. (The evaluation was ordered by the federal court in Connecticut; the court's authority enabled us to achieve such a high response rate; only two surveys were not returned.) Measures Direct measures. We included three direct measures of Medicare's normative standards. They were 'restoration potential', 'measurable progress' and 'maintenance therapy'. These measures came from the guidelines. The guidelines indicated that the FIs were to make payment decisions based on the restoration potential of the patient and the patient's ability to make measurable progress in therapy. Patients with good potential for improvement and recovery were entitled to daily physical therapy and Medicare benefits; maintenance therapy was a type of therapy reserved for patients whose improvement had plateaued or for whom full recovery was not possible (they usually received it less than daily, i.e. three times per week). We included these measures to establish the therapists' familiarity with Medicare's normative standards of care. Each measure was assessed on a four-point scale ranging from one (often important) to four (never important). A mean ranking was derived for each measure. Levels of care. The SNF physical therapists also were asked to list three criteria they used when deciding about different levels of care (daily physical therapy, less than daily physical therapy and no
875
Clinical decision making of physical therapists physical therapy treatment). We were interested in the extent to which normative fiscal standards figured into their open-ended responses. All criteria were entered into a database and sorted alphanumerically by field, line and word. The resulting lists were independently sorted by the two investigators and items associated with each level of care were grouped according to similarity of content. For example, 'potential to reach goals', 'potential for home discharge', 'good potential for rehab', 'realistic potential for positive results', 'potential for recovery' and 'restorative potential' were grouped together to comprise the following indicator for daily physical therapy: 'Rehab/Restorative Potential: Good/Excellent'. The top five indicators for each level of care were retained. Physical therapy evaluations. A third measure consisted of SNF physical therapists' responses to real patient situations. We deliberately chose cases where prospects for improvement and recovery varied. We wanted to see whether or not therapists made recommendations based on the patient's rehabilitation potential and ability to make measurable progress in therapy. Two of the four cases had complex medical problems, but presented with complications that severely affected their prospects for improvement and recovery (John Doe No. 1 had an amputation, but also had a neurological condition since birth that prevented him from using a prosthesis; John Doe No. 2 had a hip fracture, but also had poor eyesight and an old stroke on the same side as the hip fracture; neither had plans to return home). The third case had a medical problem similar to the first two, but presented with a good prognosis for improvement and recovery (Jane Doe No. 1 had a hip fracture with no complications, was beginning to walk and had definite plans to return home), and the final case was chosen as a distractor (Jane Doe No. 2 was independent and did not require the skills of a physical therapist). Please refer to the Appendix for detailed descriptions of the four cases. The case studies were taken from actual assessments of SNF residents in the larger study. A standardized admission evaluation form is used by SNF physical therapists in Connecticut. We included standardized forms for each of the four cases in the survey. There was no identifying information on the forms. Each form contained subjective and objective findings, a list of patient problems, short- and longterm goals and a recommended treatment plan. At the end of each form, the SNF physical therapists were asked to indicate whether or not daily, less than daily or no physical therapy treatment was needed, based on the information contained in the form. We asked three physical therapists with extensive long-term care experience and no involvement in Connecticut SNFs to serve as experts and to define standards of care for each patient. They were instructed to base their recommendations on the need for therapy services and not on fiscal or payment
criteria of any kind. They concluded, with 100% agreement, that Jmae Doe No. i, John Doe No. ! and John Doe No. 2 all needed daily physical therapy to achieve their short- and long-term goals and to carry out their treatment plans. They also concluded that no physical therapy treatment was needed in the case of Jane Doe No. 2. The responses of the SNF physical therapists in the survey were compared with the standards set by the experts and the percentage agreement was calculated in each of the four cases. Where there was agreement, the respondent was assigned a score of one; where there was disagreement, a score of zero was assigned. The scores for each case were averaged and the average scores were converted to percentages. The extent of agreement between the four cases was tested for significant differences. A paired-comparisons t-test was used. A new variable was created that contained the difference between each pair of scores, and a Student's t and probability value were calculated to test whether the mean difference was significantly different from zero [18]. The sample was then divided into two groups: those that agreed with the experts and those that did not. The two groups were compared for significant differences by sex, ethnicity, age, years in the field, staff/contract relationship and type of physical therapy training. RESULTS
Demographic data Seventy-nine percent of the SNFs in Connecticut are proprietary. The average facility size is 109 beds. The primary payer of SNF services is Medicaid (59%), followed by Private Payment (30%) and Medicare (8%) (approx. 3% of the beds are empty). Table 1 provides demographic information on the sample of SNF physical therapists that responded to our survey. Based on the information contained in the table, the prototypical respondent was a white woman in her thirties who had basic baccalaureate training in physical therapy and who had been out in the field for quite some time. She was more likely to be on contract to the facility than a staff member and the facility where she practiced did not have a particularly strong rehabilitation focus.
Direct measures Table 2 indicates a high degree of familiarity with Medicare's standards. 'Restoration potential' and 'measurable progress' were rated often important and 'maintenance therapy' was rated occasionally important by the sample.
Criteria for levels of care The top five open-ended responses for each level of care are listed in rank order in Table 3. As Table 3 indicates, restoration potential and the ability to make progress or improve in physical therapy figure
876
ROBIN MOREMEN UILI a n d ROSEMARY WOOD Table 1. Demographic characteristics of SNF physical therapists 1989 (n = 182) (%)
Measure Sex
Female Male
82 18
Table 3. Rank ordering of criteria used by Connecticut physical therapists for determining level of care needs (n = 177)~
Criteria .for daily physical therapy 1. 2. 3. 4. 5.
Rehab/restoration potential: good/excellent Diagnosis: acute/recent/new Prior level of function Ability to progress/improve: good/excellent Need for skilled services
Criteria Jbr less than daily physical therapy
Ethnicity African american Asian Caucasian Hispanic Other
1.0 0.5 97.5 0.5 0.5
I. 2. 3. 4. 5.
Ability to progress/improve: slower/none/plateaued Self/others carry out/assist with program Rehab/restoration potential: fair/limited/poor Diagnosis: acute/chronic/old Prior level of function
Criteria for no physical therapy treatment
Age Under 30 yr 30-39 yr 40-49 yr 50-59 yr 60-69 yr
13 41 26 13 7
Number of years in the field <2 2-5 6-9 10-14 15-19 >19
5 11 I1 21 15 37
Rehab focus of facility Limited Moderate Strong
22 38 40
Relationship to facility Contract P.T. Staff P.T.
60 40
Type of P.T. training Bachelors program Certificate program Entry-level masters program Post professional masters program Doctoral program
83 7 2 7 1
heavily in each of the levels of care. Also important are the patient's diagnosis, his or her prior level of function, the need for skilled services and the availability of other personnel. When the top five indicators are used to create 'ideal-typical' patients for each level of care, the following characterizations emerge. Daily physical therapy. The SNF therapists in the survey felt that daily physical therapy was most appropriate for a person with a recent, acute medical problem (such as a hip fracture) who had 'good' or 'excellent' potential for recovery and the ability to make 'good' or 'excellent' progress toward a prior level of function, and for whom skilled physical therapy was necessary to achieve those goals.
1. 2. 3. 4.
Self/others carry out program/patient independent Rehab/restoration potential: poor/none Diagnosis: long-term/chronic/old/unstable/terminal Ability to progress/improve: plateaued/none, or goals met/ optimal benefit 5. No need for skilled services
aUsable responses.
Less than daily physical therapy. The respondents felt that less than daily physical therapy was appropriate for a person with an acute or chronic problem who had 'fair' or 'limited' potential for recovery, whose progress toward a prior level of function was slower or had plateaued, and for whom skilled services may or may not be required (i.e. other staff members or the patient her- or himself might be able to carry out or assist with the program). No physical therapy. The SNF physical therapists felt that no physical therapy treatment was appropriate for a person with a chronic long-term problem, or for a person who already was independent. If the problem was chronic or long-term, there usually was 'little' or 'no' potential for progress or recovery and no need for skilled physical therapy because other staff members (i.e. nurses, aides) could provide the needed care. If the patient was independent, the independence was present on admission or was achieved earlier in the stay. Physical therapy evaluations The majority of SNF physical therapists in the survey agreed with the standards set forth by the experts (i.e. that Jane Doe No. 1, John Doe No. 1 and John Doe No. 2 all needed daily physical therapy to achieve their goals and that Jane Doe No. 2 did not require care) (see Table 4). The extent of agreement varied, however, and significant differences were noted. There was significantly more agreement about the need for daily services for Jane Doe No. 1 than for either John Doe No. 1 or John Doe No. 2, with
Table 2. Physical therapists' mean rankings on the degree of importance of restoration potential, measurable progress and maintenance therapy Measures Restoration potential ( n = 179)a Measurable progress (n = 180) Maintenance therapy (n=177)
Mean R a n k i n g + S D 1.3+0.6 1.4 + 0.6 1.8 +__0.9
'Usable responses. Key: 1, often important; 2, occasionally important; 3, seldom important; 4, never important.
agreement being the lowest for John Doe No. 1. Fully one-third of the respondents (33%) felt that John Doe No. 1 did not need daily physical therapy, as compared with 25% for John Doe No. 2 and 3.3% for Jane Doe No. 1.
None of the demographic variables were significant for Jane Doe No. 1 or No. 2. In the case of John Doe No. 1, there were significant differences by age (Z2 9.6, df = 4 , P = 0.048) and staff/contract relationship (Z 2
Clinical decision making of physical therapists 3.1, df = 1, P = 0.079), and for John Doe No. 2 there were significant differences by age (X2 19.58, df= 4, P = 0.001) and number of years in the field (X2 24.52, df = 5, P = 0.000). Therapists under the age of 30 and those between the ages of 60 and 69 were significantly less likely to see the need for daily physical therapy in both cases than therapists in their thirties and forties. Therapists in practice from two to five years were significantly less likely to see the need for daily care in the case of John Doe No. 2 than therapists who had been in the field for ten or more years. And, therapists who were on contract to the facility were significantly less likely to recommend daily care in the case of John Doe No. 1 than staff therapists. DISCUSSION
The results of this study indicate that SNF physical therapists in Connecticut are very familiar with Medicare's normative standards and that these standards do affect their recommendations for care in simulated patient situations. Restoration potential and measurable progress (and, to a lesser degree, maintenance therapy) are rated 'often important' by the SNF physical therapists; they are among the criteria used to differentiate between levels of care in the openended responses; and one-quarter to one-third of the respondents do not recommend daily care when potential for improvement and recovery are questionable. Furthermore, younger therapists (and very old therapists, although the number of respondents in this category is small), contract therapists, and therapists with fewer years in the field are more likely to make recommendations based on fiscal criteria than older, more experienced, staff therapists. Nearly all of the SNF therapists (97%) recommend daily care when the medical condition is complex and rehabilitation potential is clear (Jane Doe No. 1). We
877
may infer from this that well-defined cases receive benefits and services without difficulty. The situation is more tenuous when improvement and recovery are questionable, because these are the standards that Medicare has imposed to limit payments. In the majority of these circumstances, SNF therapists also recommend daily care, but in a certain percentage of the cases they do not. When this occurs, access to benefits and services is severely curtailed. How do our findings compare with actual data on Medicare coverage? Most individuals who qualify for services are granted Medicare coverage at the level of the SNF (data from early 1989) [19]. On average, 67% qualify on technical grounds [19, p. 15] (i.e. they have a prior hospital stay of at least three days, there is a period of no more than 30 days between discharge from the hospital and admission to the nursing home, and they are receiving care for the same condition for which they were hospitalized), 36% have orders for physical therapy [19, p. 6], and, of those, 82% receive Medicare coverage [19, p. 34]. These figures are similar to the overall recommendations of the SNF physical therapists in the survey. What happens when these claims are forwarded to the FIs for review and payment? When covered physical therapy claims are reviewed by the FIs, on average, 93% are paid in full [20, p. 69]. There is variation across the three Connecticut FIs, however. The FI responsible for the most SNFs (52%), pays only 73% of its claims in full; the FI responsible for 31% of the SNFs pays 91% in full; and the FI responsible for the least number of SNFs (17%) pays nearly 100% in full [20, p. 67; 70]. This variation may produce uncertainty for some physical therapists and may affect their recommendations for care. Therapists who are reluctant to recommend daily care when recovery is questionable,
Table 4. Agreement between Connecticut S N F physical therapists and experts regarding standardized care Evaluations
Agreement (%)
Jane Doe No. l (n = 181) d Right hip fracture, independent at home, able to transfer and walk with a walker with assistance, goal to return home
96.7 a
John Doe No. I (n = 179) Above-the-knee amputation, independent at home, no prosthesis due to neurological condition, transfers with two people, no immediate plans to return home
67.0 b
John Doe No. 2 (n = 180) Left hip fracture, prior S N F resident, independent at S N F , paralysis left side due to old stroke, poor eyesight, transfers with assistance, refuses to walk at time of assessment
75.0
Jane Doe No. 2 (n = 181) Bronchitis, fairly independent at h o m e with son, long-term placement at S N F , stands, transfers and walks independently
84.5 ~
~The m e a n difference in scores between J a n e Doe No. ! and John Doe No. 1, John Doe No. 2, and Jane Doe No. 2 was significantly different from zero ( P = 0.0001). bThe mean difference in scores between John Doe No. I and John Doe No. 2 was not significantly different from zero ( P = 0.09). CThe m e a n difference in scores between Jane Doe No. 2 and John Doe No, I and John Doe No. 2 was significantly different from zero ( P = 0.0002 and P = 0.03, respectively). d Usable responses.
878
ROBIN MOREMENU1LI and ROSEMARYWOOD
may work in SNFs where the FI denies 27% of its claims. Nonpayment of services--particularly if one has to deal with the same FI repeatedly--is a powerful motivator to limit daily care to cases where progress and recovery are beyond dispute. There are other reasons why fiscal standards may affect recommendations for care. Most physical therapists are used to being dominated by physicians [1-3] and by the bureaucratic authority of the organizations that employ them [21]. Fiscal rules may represent one more constraint on their freedom. Moreover, the risks of noncompliance are fairly severe. If a claim is denied, the SNF may lose payment for services it has already rendered; if several denials accrue, the therapist's job or contract may be jeopardized. Some therapists may knowingly comply with fiscal rules in order to secure their positions in the medical hierarchy. It may also be the case that some therapists are unable to provide daily care. With the national shortage of physical therapists [22], some SNFs may share contract services with several other SNFs and may be forced to accept less than daily coverage. Likewise, some therapists may not be as willing or as able to discern between improvement as a 'goal' and improvement as a 'requirement' for payment. Nearly all therapists want their patients to improve, but, because time and resources are limited, some therapists may choose--actively or unwittingly--to concentrate their most intensive efforts on patients with the greatest potential for recovery. Similarly, there may be some therapists who do not provide daily care because they maintain stereotypically negative views [23] about an elderly person's ability to improve or regain lost function. Or, perhaps some patients are unable to tolerate daily treatment because they have complicated, fragile or unstable conditions. Given the narrow focus of the guidelines, how are most therapists able to secure benefits and services for their patients, even when recovery is uncertain? Perhaps, as Freidson explains, providers and patients "join together to get their own way" [3, p. 12]. Because third-party payers, such as Medicare, do not have direct access to patient-provider interactions, they must rely on the record afforded them by the provider. Because records are arbitrary constructs [3, p. 11], they can be made to serve the interests of the record-maker. Some SNF therapists may actually use their knowledge of Medicare's rules to obtain services for their patients. By incorporating the vocabulary of the rules into the patient's record, the therapist may construct an account of treatment that is more likely to survive the scrutiny of the FI's review process. Many elderly patients are capable of making slow progress in therapy, and careful wording of the patient's record may permit valuable time to maximize whatever potential is available. A phenomenon known as 'skimming' may also work in their favor. Providers, including nursing homes, often select patients that require little care,
because reduced care means lower costs for the facility [24-28]. If Connecticut SNFs routinely skim patients, they may be eliminating all but the 'best' rehabilitation candidates; patients that have no trouble complying with Medicare's rules and that bring increased federal revenues to the facility. According to the survey, younger therapists and therapists with little experience apply fiscal rules most frequently. Perhaps the youth and inexperience of these therapists makes them less confident about the types of patients that benefit from therapy; perhaps they are less knowledgeable about record keeping and Medicare's rules; or they may be more fearful of losing newly-acquired positions. Also, contract therapists appear to apply fiscal criteria more willingly than staff therapists. While staff therapists may be expected to pay more attention to facility policies and procedures, contract therapists may be more easily replaced if services are not paid by Medicare. In fact, bids for services may depend on the ability of the therapist to render reimbursable services. Staff therapists may be afforded greater leniency in this regard. The findings presented here only address the recognized importance of Medicare's guidelines by SNF physical therapists in Connecticut. Because Medicare is a federal program, it is important to test these findings with SNF physical therapists in other states and in regions beyond the northeast. It also would be useful to look at other guidelines, other practitioners and other settings.
CONCLUSION
Like many alternative providers, physical therapists seek greater autonomy, but they must pursue their ambitions in an environment that increasingly stresses the formalization of care. In the case of SNF physical therapists, one aspect of that environment consists of fiscal guidelines established by third-party payers such as Medicare. Physicians are wrestling with similar threats to their political and economic freedom, but continue to secure services for the majority of their patients. Physical therapists are not unlike physicians in this regard. However, they do differ in the degree of cultural authority they exercise. Medicine's authority is institutionalized in the policies of the state [3]; physical therapy exists on the margins, and, as such, is more vulnerable to policy changes. Consequently, as Congress entertains even deeper cuts in the Medicare program, and as health reform moves in the direction of managed competition, physical therapists must make every effort to maintain their position in order to continue to secure services for their patients. research was funded by the Health Care Financing Administration as part of the evaluation and monitoring of the Fox v. Bowen Federal Court decision in Connecticut Skilled Nursing Facilities and by NIMH training Grant MH15783-13. Acknowledgements--This
879
Clinical decision making of physical therapists REFERENCES
I. Freidson E. The reorganization of the medical profession. Med. Care Rev. 42, l l, 1985. 2. Freidson E. The medical profession in transition. In Applications of Social Science to Clinical Medicine (Edited by Aiken L. and Mechanic D.), p. 63. Rutgers University Press, New Brunswick, N J, 1986. 3. Freidson E. Power is Plural. Paper presented at the Annual Meeting of the American Sociological Association, Washington, DC, August 14, 1990. 4. Haug M. R. Deprofessionalization: an alternative hypothesis for the future. SocioL Rev. Monographs 20, 195, 1973. 5. Haug M. R. A re-examination of the hypothesis of physician deprofessionalization. The Milbank Q. 66, 48, 1988. 6. Light D. and Levine S. The changing character of the medical profession: a theoretical overview. The Milbank Q. 66, 10, 1988. 7. McKinlay J. B. On the professional regulation of change. Sociol. Rev. Monographs 20, 61, 1973. 8. Navarro V. Professional dominance or proletarianization?: Neither. The Milbank Q. 66, 57, 1988. 9. Stoeckle J. D. Reflections on modern doctoring. The Milbank Q. 66, 77, 1988. 10. Eisenberg D. M., Kessler R. C., Foster C., Norlock F. E., Calkins D. R. and DelBanco T. L. Unconventional medicine in the United States: prevalence, costs, and patterns of use. N. Engl. J. Med. 328, 246, 1993. 11. Adler M. and Brown C. C. Jr. Who reviews the reviewers? A retrospective study of Medicare determinations. N. Engl. J. Med. 302, 842, 1980. 12. Feder J. and Scanlon W. The underused benefit: Medicare's coverage of nursing home care. Milbank Memor. Fund Q./Hlth Soc. 60, 604, 1982. 13. Loeser W. D., Dickstein E. S. and Schiavone L. D. Medicare coverage in nursing h o m e s - - a broken promise. N. Engl. J. Med. 304, 353, 1981, 14. Smits H. L., Feder J. and Scanlon W. Medicare's nursing-home benefit: variations in interpretation. N. Engl. J. Med. 307, 855, 1982. 15. Estes C. L. The Reagan legacy: privatization, the welfare state and aging into the 1990s. In States, Labor Markets and the Future of Old Age Policy (Edited by Myles J. and Quadagno J.), p. 109. Temple University Press, Philadelphia, 1991. 16. Uili R. M. Resident selection in a Connecticut nursing home: a view from within. J. Aging Hlth, forthcoming 17. Tisdale S. Harvest Moon: Portrait of a Nursing Home, p. 11. Henry Holt and Company, New York, 1987. 18. SAS Institute Inc. SAS User's Guide: Statistics, Version 5 Edition, p. 799. SAS Institute Inc., Cary, NC, 1985. 19. Uili R. M. and Leaf P. J. An expansion of Medicare benefits: intended and unintended consequences, unpublished manuscript. 20. Leaf P. J. and Uili R. M. Outcomes of Fox v. Bowen: Final report of the Fox v. Bowen evaluation project. Submitted to Robert A. Burt, Special Master, Fox v. Bowen Federal Court Decision, State of Connecticut. Center for Health and Policy Research, Department of Epidemiology and Public Health, Yale School of Medicine, New Haven, CT, 1 August 1990. 21. Alexander J. A. and D'Aunno T. A. Transformation of institutional environments: perspectives on the corporatization of U.S. health care. In Innovations in Health Care Delivery: Insights for Organization Theory (By Mick S. S. and Associates), p. 53. Jossey-Bass Publishers, San Francisco, 1990. 22. Institute of Medicine. Allied Health Services: Avoiding Crises. National Academy Press, Washington, DC, 1989.
23. Cockerham W. C. This Aging Society, p. 131. Prentice Hall, Englewood Cliffs, N J, 1991. 24. Harrington C. Public policy and the nursing home industry. In Readings in the Political Economy t~f Aging (Edited by Minkler M. and Estes C. L.), p. 144. Baywood, Farmingdale, NY, 1984. 25. Relman A. S. The new medical-industrial complex. N. Engl. J. Med. 303, 963, 1980. 26. Relman A. S. Selling to the for-profits: undermining the mission. Hlth Progr. 66, 81, 1985. 27. Steinwald B. and Neuhauser D. The role of the proprietary hospital. Law Contemporary Problems 35, 817, 1970. 28. Stevens R. In Sickness and in Wealth: American Hospitals in the Twentieth Century, p. 342. Basic Books, New York, 1989.
APPENDIX Jane Doe No. 1 was a patient who was admitted to the SNF following hospitalization for a right hip fracture. Prior to her fall, she had been independent at home. At the time of the evaluation she was able to transfer from one surface to another with the minimal to moderate assistance of one other person and walk 15 ft with a walker with the minimal to moderate assistance of one other person. Her long-term goal was to walk without assistive devices and return home to independent living. Her treatment plan included progressive resistive exercises, bed mobility and transfer training, and gait training. John Doe No. l was a patient who was admitted to the SNF following an above-the-knee amputation due to severe peripheral vascular disease. The physician had determined that prosthetic training was unrealistic due to a neurological condition that the patient had had since birth. Although the patient had been independent at home prior to the amputation, he did not have immediate plans to return home. At the time of the evaluation, he was able to stand and transfer between surfaces with the maximal assistance of two people. His long-term goal was to become independent in a wheelchair. His treatment plan included therapeutic exercises to all four extremities and his trunk, sitting balance exercises, transfer training, stand-pivot activities and wheelchair mobility training. Jane Doe No. 2 was a patient who had been admitted to the SNF after being hospitalized for bronchitis. Prior to her hospitalization, she had lived fairly independently at home with her son. She was a long-term placement in the SNF at the time of her admission because her son had taken ill. She was able to stand, transfer between surfaces and walk by herself independently, although she had a flat-footed gait on level surfaces and some shortness of breath when climbing three stairs. Her goals were to maintain her level of independence on the unit and her treatment plan consisted of nursing supervision for wandering. As she was a long-term placement, stairs were not a major concern. John Doe No. 2 was a patient who had been a resident of the SNF for six years. He was admitted to the hospital and readmitted to the SNF after an operation for a left hip fracture. He also was paralyzed on the left side due to an old stroke. Prior to his hip fracture, he had been able to walk by himself about the facility. He had poor eyesight and, since the fracture, was fearful of falling. At the time of the evaluation, he could stand and transfer between surfaces with the moderate to maximal assistance of one additional person, but he refused to walk at that time. His long-term goal was to achieve independent transfers and supervised walking with a hemi-walker in the facility. His treatment plan consisted of therapeutic exercises to his lower extremities, transfer training, wheelchair mobility instruction, gait training with a walker and staff inservicing on proper positioning.