Application of the Menninger Return-to-Work Scale among injured workers in a production plant.
Research on factors associated with RTW has been on-going since the 1960s. Developing a profile of the injured worker in terms of some of these factors offers the potential for predicting which workers will benefit most from rehabilitation assistance. Client profiles of older persons, without dependents or living alone, and who have little education and unskilled jobs, were found to have reduced probability of RTW (Foldspang, 1987; Gay & Wong, 1988; Milhous et al., 1989; Tate, 1992). Gay and Wong (1988) also associated personal characteristics of motivation and cooperation with client RTW.
Job satisfaction and perceptions of employer support were found to be significant indicators of positive rehabilitation outcomes (Allodi & Montgomery, 1979; Bear-Lehman, 1983; Bergquist & Larsson, 1977; Brewin, Robson, & Shapiro, 1983; Williams, 1991). Lancourt and Kettelhut (1992) found family stability to be a factor in RTW.
Wage replacement has been shown to have an inverse relationship to RTW (Brewin et al., 1983; Catchlove & Cohen, 1982; Guck, Meilman, Skultety, & Dowd, 1986; Sander & Meyers, 1984; Tate, 1992). Rehabilitation is positively influenced by the opportunity to be re-employed in the same job and with the existence of light-duty programs (Burton & Wilkinson, 1988; Deacon & Congdon, 1984; Gay & Wong, 1988; Williams, 1991).
The type and extent of injury also affects RTW. Them are various indices of severity, such as percentage of body affected (Helm, Walker, & Peyton, 1986), number of tests (Bear-Lehman, 1983; Milhous et al., 1989), whether surgery was performed (Beals & Hickman, 1972; Guck et al., 1986; Tate, 1992), complications (Gardner, Goodwill & Bridges, 1968), and persistence of pain symptoms (Bergquist-Ullman & Larsson, 1977; Catchlove & Cohen, 1982; Guck et al., 1986; Lancourt & Kettelhut, 1992). Regardless of the severity index used, these researchers have found a positive association between severity and amount of lost time.
The psychology of the disability syndrome has also been associated with increased time off from work. The disability syndrome is a condition in which subjective symptoms are not supported by objective findings, but the symptoms cause delayed recovery. Authors contend that a disabled individual's psychological reaction (compensation neurosis) to injury and the effect of secondary gains (reinforcement) which result, can be as disabling as the original injury (Burgel, 1986; Fitzler, 1983; Hanson-Mayer, 1984; Headley, 1989).
In 1986, researchers at the Menninger Clinic developed a scale using many of these factors to predict RTW and guide allocation of resources (Hester, Decelles, & Gaddis, 1986). The purpose of this study was to apply the Menninger Return-To-Work (RTW) Scale retrospectively to a population of workers who lost time due to work-related disabilities to determine the validity of the scale as a predictor of RTW in this population.
The Menninger Model
The Menninger Scale is a weighted scale from 0 to 100. A score greater than 65 assures RTW; a score of less than 39 equates to a 3% likelihood of RTW. According to Hester et al. (1986) scores provide direction for allocation of rehabilitation$resources. The Return to Work Center Project of the Menninger Research and Training Center of the Menninger Clinic focused on individuals who were gainfully employed, but due to disabling conditions were unable to engage in remunerative work for at least 5 months. Using data from 600 persons who were disabled five months to two years, investigators studied the association of ten variables to RTW following disability. The ten variables were: disability, age, sex, education, marital status, occupation, residence, employer, sources of support, and wage replacement.
There were three aspects of the Menninger study design that limit its applicability to other populations. The first aspect was inability to differentiate between work-related and non-work-related injuries. In the Menninger study 36% of the sample were disabled with injuries; but the proportion of work-related injuries was unknown. Research indicates that persons injured on the job who received workers' compensation remain disabled longer than those who suffer personal injuries (Brewin et al., 1983; Foldspang, 1987; Guck et al., 1986; Sander & Meyers, 1986; Sheikh, 1987; Walsh & Dumitru, 1988).
A second feature of the Scale was continuing to label a physically recovered person "disabled" until a job was found. Maximum medical improvement (MMI) did not alter eligibility for compensation benefits; this is different from state worker's compensation systems which terminate benefits at MMI. Therefore, time off work would be inflated in the Menninger study, and there was no way to determine the size of this group.
The final limiting characteristic of the initial study was classification of disability using the Merck Manual, which classifies by system of the body. Coding by system of the body results in combining chronic and acute diagnoses in one category. For example, both fractures and degenerative arthritis were classified as musculoskeletal problems; but it would not be expected that the length of disability from these problems would be the same, although both might be claimed as work-related disabilities.
A retrospective, descriptive study was used to characterize injured workers, assign scores and groups according to the Menninger Scale, and determine the validity of the Menninger Scale in predicting RTW and guiding the use of rehabilitation resources among a group of production workers. In order to categorize subjects, two steps were necessary. First, the subjects were scored and classified into seven groups according to the Menninger model. Secondly, the actual number of days lost was compared to expected days lost using actuarial data of specific injuries from the Disability Duration Guide (Nationwide, 1991) and the Medical Target Date Guide (Intracorp, 1991). Developed by the insurance industry, these guides contain estimated average length of disability by type of injury, based on actuarial data and experts in the field.
Operational Definitions of Variables
Demographic variables of age, sex, marital status, and years of education were recorded as they existed at the time of the injury. The employer, a large manufacturer in the Midwest, was the same for all workers; source of financial support was workers' compensation for all workers.
Definitions for the other variables were taken from the Menninger study. Type of disability was coded according to the Merck Manual: (03) cardiovascular, (04) pulmonary, (05) gastrointestinal, (10) musculoskeletal, (11) neurological, and (00) other. Occupations as characterized by the Menninger study were: (a) managerial/professional, (b) technical, (c) service, (d) craft, (e) assembly, and (f) farming. Residence was defined as rural (fewer than 76 people per square mile), urban (76 to 1499 people per square mile), or metropolitan (more than 1499 people per square mile). Wage was the ratio of the replacement amount to the regular amount earned.
The Menninger Scale determines "likelihood of RTW," which is the probability of RTW based on the total score. In the present retrospective study, 100% of the workers had returned to production work within the plant; therefore, "likelihood" was translated to "successful," which was defined as returning to work in less than or equal to the expected days away from work or in less than 30 days. Thirty days was selected as a lower boundary to prevent bias of two causes: varying levels of severity of injury and potential case management inconsistencies. If the worker was away from work longer than expected or more than 30 days, "unsuccessful RTW," a record was made indicating whether the worker had been provided with rehabilitation resources and if these services were provided within 30 days of the injury.
In this way, subjects were characterized by the score obtained according to predictor variables (0 to 100), and successful RTW recorded as a "yes" or a "no" for returning to work in an expected length of time, or within 30 days of the injury. Subjects who had not returned to work in the expected length of time, or 30 days, were also characterized by whether or not they had received rehabilitation.
The following projections were made using the Menninger model:
(1) 100% of the subjects with scores greater than 65, 96% of those with scores between 60-64, 86% with scores between 55-59, 69% with scores between 50-54, 24% with scores between 45-49, 6% with scores between 40-44, and 3% with scores less than 39 will RTW within the expected time with or without rehabilitation case management. (2) Those who do not RTW in the expected time or within 30 days of the injury, did not receive rehabilitation case management within 30 days of the injury.
The study was conducted at one plant location of a multi-site manufacturing company in the Midwest with 950 production employees. The sample was the total number of injured workers who lost time from work between 1987 to 1991, identified from a list of recipients of wage compensation for work-related injuries. The number of subjects in the study was 42; 32 men (76%), 10 women (24%). The age range was 22 to 66; average age was 42. Thirty-five were married (83%), five single (12%), two divorced (5%). All subjects had a high school education. All subjects were assembly workers. Thirty-five workers sustained musculoskeletal injuries (83%); five were burned (12%); one had carpal tunnel syndrome; one had a psychiatric diagnosis. Thirty-two lived in the metropolitan area (76%); ten lived in urban areas (24%); none lived in rural areas. Twenty-eight had wages replaced from 51% to 75% (67%); 14 had wages replaced from 26% to 50% (33%).
The range of scores on the Menninger RTW Scale among the study population was small compared to the Menninger study: 52-62. Seven subjects (17%) bad scores between 50-54, thirty-two (76%) had scores between 55-59, and three (7%) had scores between 60-64.
Twenty-one of the subjects (50%) were successful in returning to work in the expected time or within 30 days of the injury. Two of the seven subjects (28.6%) with scores between 50-54 were successful and five (71.4%) were unsuccessful in returning to work in the expected time. Of the 32 subjects with scores between 55-59, 17 (53.1%) successfully returned to work as expected and 15 (46.9%) did not. Two of the three subjects (66.7%) with scores between 60-64 experienced successful RTW.
The first premise in this study stated percentages of injured workers in each group that would RTW within the expected time with or without rehabilitation. These hypothesized percentages, using the Menninger Scale, were 69% of those with scores of 50-54, 86% of those with scores of 55-59, and 96% of those with scores of 60-64. The actual percentages among the subjects in this study were 28.6%, 53.1%, and 66.7% respectively. Therefore, the findings do not support the prediction. However, the findings of this study do show a pattern consistent with the scores. The percentage of "successful" workers varied directly with the scores.
The second premise stated those who did not RTW in the expected time or within 30 days of the injury, did not receive rehabilitation case management within 30 days of the injury. Of the twenty-one who were "unsuccessful" in returning to work, only one had rehabilitation within 30 days of the injury. Therefore, this prediction was supported.
Table 1 Comparison of Menninger "Likelihood RTW" and the Present Study "successful RTW" Present Study Menninger Study SCORE n Successful RTW Likelihood RTW [less than]39 0 na 3% 40-44 0 na 6% 45-49 0 na 24% 50-54 7 29% 69% 55-59 32 53% 86% 60-64 36 7% 96% [greater than]65 0 na 100%
When the Menninger Scale was applied to a sample population of 42 injured workers in a production plant in the Midwest, the RTW Scale was unable to correctly predict percentages of workers who would successfully RTW. Therefore, this study did not validate the exact percentages of the Menninger instrument; however, the scores did have a pattern: workers with lower scores were less "successful" returning to work than those with higher scores. This pattern of scores suggests potential validity in the method of scoring certain factors for directing rehabilitation resources.
This study expected to find greater percentages of subjects successfully returning to work than the Menninger study for three reasons. One, subjects were included in the present study from the first day of injury instead of waiting five months as in the Menninger study; and the longer a worker is out of work, the greater the probability of not returning (Beals & Hickman, 1972; Lancourt & Kettelhut, 1992; Milhous et al., 1989; Williams, 1991). Secondly, chronic diseases are uncommon in workers' compensation claims (Bureau of National Affairs, 1992), but the Menninger study included chronic health conditions, which are associated with difficult recovery. The third reason for expecting rapid, successful RTW for workers in the present study was availability of modified jobs as soon as medical stability was reached. Subjects in the Menninger study who were recovered but did not have specific jobs continued to be labeled disabled. Although expecting more successful RTW among the present study population than in the Menninger study, only 50% of the subjects had "successful" RTW; and the percentages of those returning in each group fell significantly short of the Menninger predictions. The reason for fewer successes in the present sample is largely explained by the small sample size and the stringent definition of success.
Due to the small sample size and homogeneity of the group with respect to many of the variables, the scores fell in a narrow range. Variables that Hester et al. (1986) found most predictive of RTW (i.e., source of support and education) were identical for all subjects in the present study. To be "successful" the worker had to meet a target date determined by diagnosis or within 30 days of injury; in the Menninger study, to be successful, workers could RTW anytime within two years, regardless of diagnosis.
Even though the success rate in the present study was not as high as expected using Menninger scores, there was a relationship between scores and successful RTW; and 100% of the workers in this retrospective study returned within two years. This raises the issues of the role and timing of rehabilitation. In the present study, five of the 21 "unsuccessful" workers had rehabilitation case management; two others were offered services, but refused. In only one case, rehabilitation services were offered within 30 days of the injury. This worker, with a back strain, participated in a six-week work hardening program and returned to work within five months of the injury. In another case, rehabilitation services were initiated within one year. Three of the workers who received formal rehabilitation services were not entered into the programs until more than two years after the injury. The rehabilitation programs were work hardening and work simulation. In all five cases, RTW occurred immediately or shortly after completion of the rehabilitation program. Would the success rate have been improved if all 21 "unsuccessful" workers had received timely rehabilitation services? The Menninger study suggests evaluation for rehabilitation for persons with scores from 45-59. This approach would have directed attention to 20 of the 21 unsuccessful subjects, and possibly increased timely RTW (success rate) for these subjects.
Recommendations for Future Study
Future studies are needed to learn if other variables can be added to the Menninger Scale to increase its predictability; and to learn optimal timing for rehabilitation case management. One such study could use the existing data, adding certain other variables, then performing an analysis of variance. Variables such as length of employment with the present employer and attorney involvement could be added. Type of injury should be more definitive than "musculoskeletal." Adding and defining variables could clarify which variables contribute most strongly to RTW.
Another study could be experimental research designed as a prospective, longitudinal study in which some workers would receive rehabilitation case management from the first day of injury, some would receive rehabilitation services based on "scores," and the control group would not receive rehabilitation services. This type of research would help answer the question of when to begin rehabilitation case management. Prospective research could also evaluate the aspect of job satisfaction. Much of the literature emphasizes the importance of job satisfaction as a predictor of early RTW. This could not be measured in the Menninger study nor in the present (retrospective) study; however, it would be a key part of a longitudinal study.
Allodi, F., & Montgomery, R. Psychological aspects of occupational injury. Social Psychology, 14, 25-29.
Bergquist-Ullman, M., & Larson, U. (1977). Acute low back pain in industry: A controlled prospective study with special reference to therapy and confounding factors. Acta Orthopaedica Scandinavica, 170, 1-117.
Beals, R. K. & Hickman, N. W. (1972). Industrial injuries of the back and extremities. Journal of Bone and Joint Surgery, 54-A, 1593-1610.
Bear-Lehman, J. (1983). Factors affecting return to work after hand injury. American Journal of Occupational Therapy, 37, 189-194.
Brewin, C. R., Robson, M. J., Shapiro, D. A. (1983). Social and psychological determinants of recovery from industrial injuries. Injury: The British Journal of Accident Surgery, 14, 451-455.
Bureau of National Affairs. (1992). New BLS statistics show 6.8 million workplace injuries and illnesses recorded in 1990. Winter, 1992. Washington, DC: Author.
Burgel, B. J. (1986). Disability behavior: Delayed recovery in employees with work compensable injuries. American Association Of Occupational Health Nurses Journal, 34, 26-30.
Burton, W. N., & Wilkinson, F. (1988). Cost management of short term disability. American Association of Occupational Health Nurses Journal, 36, 224-227.
Catchlove, R., & Cohen, K. (1982). Effects of a directive return to work approach in the treatment of workman's compensation patients with chronic pain, pain, 14, 181-191.
Deacon, S. P. & Congdon, G. J. (1984). Rehabilitation after illness and injury - A study of temporary alternative work arrangements. Journal of Sociological Occupational Medicine, 34, 46-49.
Fitzler, S. L. (1983). The disabled employee: Physical, psychological, and social changes. Occupational Health Nursing, 31, 9-15.
Foldspang, A. (1987). Standardized performance tests and their impact on the decisions determining the type of rehabilitation program. Scandinavian Journal of Social Medicine, 15, 253-60.
Gardner, D. C., Goodwill, C. J., & Bridges, P. K. (1968). Absence from work after fracture of the wrist and hand. Journal of Occupational Medicine, 10, 114-117.
Gay, D. A. & Wong, D. W. (1988). Predicting rehabilitation outcomes from clinical and statistical data: a probability model. International Journal of Rehabilitation Research, 11, 11-19.
Guck, T. P., Meilman, P. W., Skultety, F. M., & Dowd, E. T.(1986). Prediction of long-term outcome of multi-disciplinary pain treatment. Archives of Physical Medicine and Rehabilitation, 67, 293-296.
Hanson-Mayer, T. P. (1984). The worker's disability syndrome. Journal of Rehabilitation, 50(3), 50-54.
Headley, B.J. (1989). Delayed recovery: Taking another look. Journal of Rehabilitation, 55, 61-66.
Helm, P. A., Walker, S. C., & Peyton, S. A. (1986). Return to work following hand burns. Archives of Physical Medicine and Rehabilitation, 67, 297-298.
Hester, E. J., Decelles, P. G., & Gaddis, E. L. (1986). Predicting which disabled employees will return to work: The Menninger RTW Scale. (Available from Menninger Clinic, Return to Work Center, 700 Jackson, 9th floor, Topeka, KS, 66603.)
Intracorp (1991). Target Date Guide. Philadelphia: Author.
Lancourt, J. & Kettelhut, M. (1992). Predicting return to work for lower back pain patients receiving worker's compensation. Spine, 17, 629-640.
Milhous, R. L., Haugh, L. D., Frymoyer, J. W., Ruess, J. M., Gallagher, R. M., Wilder, D. G., & Callas, P. W. (1989). Determinants of vocational disability in patients with low back pain. Archives of Physical Medicine and Rehabilitation, 70, 589-593.
Nationwide Insurance (1991). Disability Duration Guide. Columbus, OH: Author.
Sander, R. A., & Meyers, J. E. (1986). The relationship of disability to compensation status in railroad workers. Spine, 11, 141-143.
Sheikh, K. (1987). Occupational injury, chronic low back pain and return to work. Public Health, 101, 417-425.
Tate, D. G. (1992). Workers' disability and return to work. American Journal of Physical Medicine and Rehabilitation, 71, 92-96.
Walsh, N. E. & Dumitro, D. (1988). The influence of compensation on recovery from low back pain. Occupational Medicine, 3, 109-121.
Williams, J. R. (1991). Employee experiences with early return to work programs. American Association of Occupational Health Journal, 39, 64-69.
Karen J. Martin, RN PhD, College of Nursing and Health, University of Cincinnati, 3110 Vine Street, Cincinnati, OH 45221-0038.