The growth and structure of the proprietary rehabilitation sector.

This article presents a summary of the high points of the first federally funded analysis of nationwide data on the structure, growth, and performance of rehabilitation in the private sector and is based on a 30-month study completed in 1987 by Berkeley Planning Associates and Harold Russell Associates. Commissioned by the Office of Planning, Budget and Evaluation, U.S. Department of Education, the study's design and conclusions were guided by a National Advisory Panel consisting of leaders from private rehabilitation firms, state vocational rehabilitation (VR) agencies, insurance firms, and prominent researchers in the fields of disability policy and rehabilitation.

One of the dramatic changes in the rehabilitation services field over the last decade has been the emergence, rapid growth, and increasing importance of private sector for-profit (or "proprietary") rehabilitation providers serving people with disabilities and often financed by insurance firms and self-insured employers. As more states began mandating rehabilitation benefits for injured workers under workers' compensation, the market for the services of these organizations has grown dramatically. But in an era where the "privatization" of traditionally public services has become promoted as a laudatory direction for public policy, there have also been calls for such private sector providers to play an increasing role in the provision of publicly funded vocational rehabilitation services.

The Berkeley Planning Associates/Harold Russell Associates study was designed to address the following overall questions:

* What is the current structure of vocational rehabilitation in the private sector; what factors appear to account for its growth; what practices characterize it; and what costs and client outcomes are associated with those practices?

* In what ways is rehabilitation in the private sector similar to and different from the state-federal VR program, particularly taking into account public-private differences in clients served and mandated goals of service?

* What lessons might the state-federal VR program learn from the private rehabilitation sector?

* How might the public and private sectors most effectively address the needs of all people with disabilities?

In order to answer these and related questions, information was gathered from many different sources, most importantly, from the following:

* The Provider Survey: a national mail survey of private rehabilitation firms, under the cooperative sponsorship of the National Association of Rehabilitation Professionals in the Private Sector (NARPPS). Responses were received from 329 firms which collectively handle about half of all cases served by the industry.

* The Insurance Survey: a national mail survey of 46 members of the Insurance Rehabilitation Study Group (IRSG) and five selected self-insured employers. The IRSG consists of insurance company managers, principally at the home office level, who are involved with rehabilitation issues.

The sample for this survey was selected to include those insurers with a known commitment to rehabilitation.

* A review of national and state level program data for the state-federal VR program.

* A review of literature pertaining to vocational rehabilitation in the public and private sectors.

* A field study conducted in six states: Arizona, California, Connecticut, Georgia, Michigan, and Texas. A total of 98 interviews were conducted with individual rehabilitation practitioners and representatives of private rehabilitation firms, insurance carriers, self-insured employers, state VR agencies, state workers' compensation agencies, and state workers' compensation funds.

* Analysis of program data collected from the Michigan Workers' Compensation program on 444 claimants served by proprietary firms and 84 claimants served by Michigan VR's free-based Insurance Program.

* Analysis of program data from the Arizona VR program comparing 106 public VR clients served by private rehabilitation firms with 164 clients served exclusively by VR counselors and nonprofit vendors.

Readers should interpret the study's findings knowing that data in the private rehabilitation sector had some inherent limitations: (1) wide variation in the definitions of rehabilitation terms; (2) differences in the ways costs and outcomes are measured; and (3) variation from state to state in workers' compensation laws and regulations. Respondents from nationwide rehabilitation firms and insurance companies also were required to generalize about company practices which varied widely among states. While about 30 percent of proprietary rehabilitation firms identified answered the mail survey, respondents included 7 of the nation's 10 largest rehabilitation companies. Thus, the respondent group includes firms serving the majority of clients who are served by proprietary rehabilitation firms.

Industry Variability (The Provider Survey)

The single most notable finding with respect to the rehabilitation providers who responded to the NARPPS-sponsored survey was the wide variation in their characteristics, practices, and outcomes. The differences among firms, even when serving similar populations, are far greater than is encountered in the public vocational rehabilitation sector. Though the results reported subsequently in this article represent central tendencies in the data, a great deal of diversity remains. The types of clients served by firms consistently proved to be far more powerful predictors of the firm's overall performance than other organizational characteristics or the firm's service specialization.

Industry Structure and Growth

Of the 329 private rehabilitation providers who responded to the survey, 27 percent were individual practitioners, 34 percent were single office firms with staff, and 39 percent were larger multi-office companies. Nearly 80 percent of all providers operated in only one state, and less than 5 percent provided services in 10 or more states. Overall, 12 companies--each accepting more than 1,000 cases in 1986'accounted for more than four-fifths of the service volume reported.

Approximately 22 percent of respondents commenced operations prior to 1979; 49 percent began during 1980-84; and 29 percent started after 1985. Nearly one-half of the individual respondents entered the field in 1985 or later. Interview data suggest that such firms were often formed by professionals who had gained their experience in larger firms. For the 154 firms in which growth (number of cases accepted in 1986 compared to 1983) could be measured, the median rate of growth was 57 percent.

Clients Served

The vast majority of clients (79 percent) were described as having an orthopedic impairment. Other primary disabilities include mental illness (7 percent), visual impairment (2 percent), mental retardation (2 percent) and hearing impairment (1 percent). Most were workers' compensation claimants. An increase in service for stress-related claims was reported. Three-fourths of the firms reported that fewer than 20 percent of their clients would be considered to have severe disabilities, as defined by the state-federal VR system.

Over four-fifths of clients were older than 24 years of age and half of that group were 45 or older. Over two-thirds were male, though the percentage of female clients appears to be increasing. Nearly 45 percent of all clients had not graduated from high school. Not surprisingly, given the focus on serving workers' compensation claimants, over 96 percent of clients had at least some work experience. A major change between 1983 and 1986 was the rising proportion of clients with recent work experience (within 18 months), suggesting that the average time from injury to referral has become shorter. Interview responses suggested that this decrease in delay prior to referral is an important factor in improving the effectiveness of rehabilitation services.

Service Practices

The provider mail survey, which asked about 22 different services, found considerable variation in the combinations of services. This occurred because of differences in state laws, definitions of services, and the service needs of referral sources. Specific observations about services were the following:

* Nearly two-thirds of the firms provided medical care coordination. This practice allows service providers to begin to explore the feasibility of return to work with the pre-injury employer, even before medical recovery is complete. There appears to be a growing demand for this service.

* Some vocational testing was provided directly by 65 percent of the private firms. The remainder relied on outside testing services. Firms which offered testing services tended to be associated with consistent growth during 1983-86.

* Physical capacities testing and "work hardening" (building strength and endurance) were offered directly by about 15 percent of firms, usually performed by physical and occupational therapists.

In addition to determining how frequently different services were offered, a factor analysis was performed to understand the ways in which "clusters" of services (services frequently offered in combination) help to explain variation in outcomes and costs. The six clusters identified were:

* Vocational/Placement;

* Testing;

* Medical Management/Job Modification Consultation;

* Skills Training Coordination;

* Catastrophic Case Management; and,

* Vocational Testimony.

Regression analyses of costs and outcomes took into account providers' patterns of service "specialization" as well as their client mix, size, age, and staffing patterns.

Outcomes, Cost, and Time in Process

The median number of weeks elapsed from date of case acceptance to date of closure was 26 weeks. Longer times were reported for companies providing vocational and training services and those serving high percentages of clients who were severely disabled, visually impaired, considered mentally ill, or under 25 years of age. Shorter times were reported by providers offering testing and medical management services and those serving larger numbers of hearing impaired clients and clients with less than a high school education. However, companies' performance appeared to be more strongly influenced by state regulatory structure and unmeasurable factors than by the types of services provided or by the characteristics of the individuals served.

When adjusted for service "specialization," as discussed above, different patterns of association emerged among services, costs, outcomes, and, most importantly, client characteristics. The average rehabilitation company reported that 62 percent of closed cases were returned to work in 1986. Of the remainder, nearly half of the cases were nonetheless considered successfully rehabilitated. Because these figures included reports from firms which provided only a single "specialized" service (and were not responsible for job placement), the reported percentage of clients returned to work may have been understated. Factors associated with high return to work rates included a concentration of medical management and training services, a large percentage of female clients, and the presence of a job developer on staff. Lower return to work rates were reported for firms operating in many states and those serving clients with severe disabilities, particularly those considered mentally ill or mentally retarded. Lower rehabilitation rates were also associated with clients who had been out of work for more than 18 months.

The median average cost (amount billed) per rehabilitation client was $1,602 during 1986. The highest average costs were reported by companies providing vocational and training services and those serving high percentages of clients with severe disabilities, attorney involvement, and no work history.

Staffing and Caseload Size

Over three-fourths of rehabilitation providers employed at least one rehabilitation counselor/specialist. In contrast, 34 percent of the firms have inhouse job developers; about the same number have nurses on staff. Field interviews indicated two distinct philosophies about staff specialization. Some firms required that staff specialize in medical management or vocational services. Other firms employed generalists who perform both functions. Firms following the specialist philosophy tended to hire job developers.

The median caseload carried by counselors/rehabilitation specialists was 26 clients. Job developers carried smaller caseloads--the median reported was 15 clients.

Evaluation of Staff Performance and Billing Practices

Sixty-five percent of the firms held staff accountable for meeting productivity objectives. Fifty-nine percent rewarded staff for productivity: of these, the major types of rewards used were salary bonuses (45 percent); salary increases (31 percent); recognition (13 percent); and paid time off (11 percent). Bonuses were most often based on billable hours worked. The use of billable hours as a basis of measuring performance and providing rewards was reported in interviews to be controversial, with some large providers using promotions instead, lest the emphasis on billable hours create pressure to overstate the time spent on cases and overcharge referral sources.

Almost all (88 percent) provider firms billed referral sources on a "per professional staff hour" basis, though some of those firms billed on a different basis for specific services, such as for evaluation. Firms that billed a flat rate per unit of service (17 percent) and those that charged on a per client basis (13 percent) provided testing and other services of standard duration.

Referral and Payment Sources

For most firms, insurance carriers provide the majority of case referrals. Onehalf of the respondents received at least 65 percent of their referrals from this source. Self-insured employers represented the second most common referral source--far less frequent, but increasing in importance with a median referral rate of 5 percent. An examination of referral sources by company type revealed that larger firms received a slightly greater percentage of their clients from insurance companies and self'insured employers than one-office firms or individual providers. Conversely, claimant attorneys referred a higher percentage of cases to individual providers.

Workers' compensation insurance paid for the vast majority of clients in private sector rehabilitation. For the majority of firms, workers' compensation covered 90 percent of the clients. Long-term disability and auto liability insurance paid for a small but growing percentage of clients.

Providers were asked under what circumstances they would be willing to serve cases referred by state VR agencies. While many respondents (particularly smaller companies) expressed interest in such referrals, many stated that a flat per capita payment rate (43 percent) or a "payment contingent on success" (78 percent) would pose barriers to serving such cases. Respondents reported willingness to serve severely disabled clients without recent employment histories, but most believed that state VR reporting and oversight procedures would be burdensome.

The Perspective of Referral Sources (The Insurance Survey)

Nearly three-fourths of the insurance survey's respondents (72 percent) reported offering workers' compensation insurance, with 70 percent using rehabilitation services in conjunction with that coverage. Other forms of insurance offered were general liability (58 percent), auto liability or no-fault (56 percent), long-term disability (44 percent), and health and accident (48 percent). Use of rehabilitation was far less frequent for coverage other than worker compensation; rehabilitation use was reported by between 24 percent and 40 percent of respondents for other coverages.

Three-fourths of all respondents employed at least one rehabilitation director or coordinator. While 56 percent employed only one such person, another 20 percent employed more than one. Approximately the same percentage (58 percent) employed one or more rehabilitation nurses. A smaller percentage (42 percent) employed rehabilitation counselors.

When asked what types of information were available and were used to monitor services and evaluate the effectiveness of providers, a significant percentage of firms did not monitor such items as: time between disability onset and referral (36.7 percent); time between referral and initial contact (39.6 percent); types of services provided (28.6 percent); and wage at placement (43.8 percent).

Field interviews confirmed survey reports that insurance company monitoring of rehabilitation providers focused largely on total rehabilitation costs, costs for an individual case, claims payments, and provider bills. While many companies were in the process of developing computerized programs to monitor certain process indicators, such as time intervals and costs of individual services provided, few had systems currently in place to monitor cost savings related to specific service providers or specific services. The process of evaluating providers was characterized as largely subjective, with more emphasis on good communications, responsiveness, and outcomes than on vendor costs. Of firms which did calculate cost savings, nearly half (47 percent) used such information in order to monitor provider performance.

When asked who actually provided the rehabilitation services, 47 percent of insurers and employers stated that they handled at least some cases using their own staff exclusively. Two-thirds of insurers sent the majority of cases to private vendors. By contrast, 58 percent of the insurers indicated no referrals to state VR, and 36 percent indicated that between 1 percent and 10 percent of cases were sent to VR. The use of non-profits was focused on assessment functions and work hardening.

Public-Private Sector Comparisons

A major policy concern from the outset of the study was to understand the similarities and differences between the private rehabilitation sector and state-federal VR programs. However, comparisons of costs and outcomes are difficult because of several differences between public and private rehabilitation. Examples are:

* the private sector objective of restoring a worker's pre-injury earning capacity versus the state-federal goal of maximizing the VR client's vocational potential;

* a prevalence in the private sector of clients with some work history versus a preponderance of state VR clients who have little or no work experience and have severe disabilities;

* widely varying recordkeeping practices among private firms versus consistent and publicly available data on state VR services, expenditures, and outcomes; and

* a lack of common definitions or recognized categories of private sector service and outcome measures versus use of federally developed definitions by state programs.

Overall Comparison of Clients and Services

The differences noted above were reflected in the following overall observations about the public and private clients and services:

* Disability type among private sector clients was heavily concentrated in the orthopedic category (nearly four-fifths of all cases), while mental illness and mental retardation cases together made up nearly half of all VR cases.

* There were more men in the typical private caseload (68 percent) than in the VR caseload (60 percent).

* Private clients were less likely than state VR clients to receive diagnostic and testing services, because the emphasis was on return to a former job, if possible. Testing was used primarily if there were a need to determine the client's transferable skills.

* The private sector was often serving clients during their medical recovery, while VR tended to become involved with clients after they were medically stable.

* Private sector services were more likely to emphasize job analysis and placement and less likely to emphasize training than state VR.

Implications of the Findings for Public Policies

The study, in its interviews with proprietary sector counselors and managers and with the National Advisory Panel, continually sought suggestions for public policy and vocational rehabilitation practice based on the proprietary sector's experiences. Among the suggestions that the public sector might consider in its efforts to improve performance for clients of VR programs are the following:

* Direct contact with employers pays off. Such contact may take the form of identifying and negotiating job modifications, outlining tax incentives, and reassuring anxious employers about their often unjustified fears of hiring people with disabilities.

* Smaller caseloads (20-40) allow the counselor to provide frequent, ongoing attention to clients and allow more time for placement efforts.

* Early intervention pays off. Taking advantage of opportunities to intervene early with clients, physicians, and employers (especially when an illness or injury is not covered by insurance) could result in preventing subsequent disability. Similarly, early notification of the client and referral to rehabilitation when insurance benefits are about to expire (e.g., auto no-fault cases) would enhance the chances of returning a person to work.

* The focus on achieving the VR client's maximum potential may not always be in the client's interest. Private sector providers in Arizona who were contracted by the state's vocational rehabilitation agency to serve VR clients with public funding and who did so with success reported having to help their state-referred clients scale down their expectations and develop more realistic objectives than those laid out in the original VR plan. State agencies might consider adopting the attitude that a client's "first job" may provide meaningful work experience and serve as a first step toward achieving maximum potential. Counselors could then reassess their clients' progress over time and decide to make additional investments in training or education to help upgrade clients' employment potential.

* Public relations and marketing should get more attention. VR agencies should actively inform employers about what they can do and how well they are doing. Furthermore, VR agencies should constantly stay in touch with employers to make sure they understand and respond to employer needs.

* Even large organizations can provide personalized services. Like small private firms, large firms emphasized personal contact between rehabilitation specialists and referral sources to facilitate responsiveness and accountability. Likewise, insurers often request individual specialists when making case referrals. VR agencies could do well to emphasize this personal approach.

During the study, a number of larger policy issues were also raised, analyzed, and reviewed. Comparative performance analysis of the public VR sector with the proprietary sector revealed that each sector was capable of achieving considerable success with the types of clients in which the other sector more typically specialized. The proprietary sector was able to serve successfully VR program client caseloads with their more frequent severe disability, limited work experience, and poor education. The public VR agencies proved fully capable in some states of competing with the proprietary sector in serving workers' compensation claimants, if only given the chance. Knowing that one needed to perform in order to compete and achieve future referrals and service "contracts" appeared to be an important factor in stimulating counselors and managers in both sectors to their best performance.

Also, interviewees emphasized the importance of downstream impacts for the public VR program when the workers' compensation system does fail (or not even try) to achieve vocational rehabilitation. Those clients failed by the workers' compensation system appeared to have the public VR program ultimately as their provider of last resort. After the financial benefits of settlements were spent out, workers' compensation clients were reported in a number of states as turning to the taxpayer-funded VR program for a rehabilitation service effort. (It is impossible using the state-federal program data to determine how many public clients have previous workers' compensation involvements in their personal histories.) After such an extended break in employment, such clients would pose especially difficult challenges to VR counselors. The public VR program thus has a major stake in the quality and performance of the proprietary sector rehabilitation industry, even if that industry continues to focus primarily on the workers' compensation market.

This study was funded by the U.S. Department of Education, Office of Planning, Budget and Evaluation, under Contract No. 300-85-0141. The full report is available in two volumes from the U.S. Department of Education and from Berkeley Planning Associates, 440 Grand Ave., Suite 500, Oakland, CA 94610, and is entitled Private Sector Rehabilitation: Lessons and Options for Public Policy (Dec., 1987). Dr. Collignon served as the Principal Investigator for the study and Ms. Vencill was the Project Director. An additional emphasis of the research was a comparison of the performance of the public and private rehabilitation sectors when serving public VR program clients and when serving workers' compensation clients.

Dr. Collignon is President, and Ms. Barker and Ms. Vencill are Vice Presidents, Berkeley Planning Associates, Oakland, CA.