Methods
For purposes of this study, the psychological variables of interest included measures of the perceptions of mood, anxiety, fatigue and sleep disturbances. Organizational outcomes were addressed in measures of employee job satisfaction, work-related injuries and absenteeism. While there are limited data on the effect of massage on the physiological and psychological factors of employees, the effect of massage on organizational factors, such as job satisfaction and work-related injuries, has not been reported.
 
The study design was a randomized, controlled experimental investigation. Male and female health-care workers (n=100) between the ages of 25 and 60 were recruited to participate in this study. Anticipating a moderate effect size, power analysis (power=0.8, p=0.05) indicates this was an adequate sample for a clinically relevant effect.15 Nursing personnel in one of two units that were supervised by the same manager were randomized by unit to participate in either the treatment or control groups. The massage therapy group received WSAM (work-site acupressure massage), twice weekly for 20 minutes, during break times, while the control group did not receive any specific intervention during break times of equal duration.
Approval for the study was obtained from the hospitals institutional review board. All participants gave written, informed consent prior to entering the study. Staff were recruited after meeting with members of this research project. The gender and ethnic composition of the sample reflect a diverse population.
 
Subject inclusion criteria were: agreement to no change in lifestyle during the eight-week period of the protocol. This included no changes in normal patterns of exercise, diet, rest or alternative therapies. Subject exclusion criteria were: 1) presence of an infection during the previous two weeks; and 2) employees who receive regularly scheduled massage therapy.
In order to control for hormonal effects associated with menstruation, all pre-menopausal women in the control and treatment group began the study between menstrual cycle days five and 11.

WSAM Treatment Procedure
The WSAM was provided by certified massage therapists, who were trained in the specific protocol. Each subject in the experimental group received a 20-minute massage, twice weekly, for a period of eight weeks. The massage was scheduled in the later half of the employees shift. Subjects were fully clothed, and the massage consisted of a blend of traditional massage, acupressure and reflexology. 
 
The massage protocol included bodywork on the upper to middle back, back of the neck, the face, upper chest, shoulders and feet. Deep-tissue massage was not used during any part of this protocol.
 
The first area to be worked was the back of the neck, the shoulders, and the region below the shoulder blades. Treatment consisted of applying light-to-medium pressure in each area for one minute in a circular motion. The face and upper chest area were worked using acupressure, which consisted of light finger and thumb pressure. Each point was worked for a slow count of 10. Foot reflexology was used to reduce stress by applying medium pressure in an alternating pattern starting just below the toes, and working down to the heel and back to the base of the toes. In addition, each toe was worked, again using medium pressure. 
 
Subjects in the control group were expected to take a 20-minute break twice weekly without WSAM treatments. A quiet room was provided where subjects in the control group could remove themselves from the work environment. Investigators met with the control group at least every other week to monitor the subjects compliance with study procedures.

Outcome Measures
Baseline demographic and health-history data for each subject were obtained prior to initiating the treatment. In addition, at the beginning of the study, and on completion of the study, subjects completed the General Well Being Scale, Profile of Mood States, State-Trait Anxiety Inventory, and Multidimensional Fatigue Inventory, General Sleep Disturbance Scale, and Index of Work Satisfaction questionnaires. Data also was collected on work-related injuries and absenteeism.
 
Limitations
Limitations in the study design are related to A) Generalizability; and B) Sampling. A convenience sample of subjects from a large university teaching hospital located on the West Coast was used. Results from a Robert Woods Johnson Foundation survey found that subjects in the West were more likely than those in the East to utilize alternative therapies.16 Therefore, these results may not be generalizable to employees in other areas of the United States who are less accepting of alternative therapies. In addition, increased managed-care changes in the health-care environment may have contributed to a degree of stress that is greater than elsewhere. This, too, may affect the generalizability of the study findings to other health-care environments.
 
In order to control for the effect that differences in management style might have, subjects were recruited into units managed by the same supervisor. As a result, there may be differences between the general population and the sample population, which may affect the findings.
Data analysis was carried out using SPSS, version 9.0 (for computerized statistical analysis). Descriptive statistics, including frequencies, means, and standard deviations, were analyzed. Inferential statistics included one-tailed t-tests to determine pre- and post-study differences. For purposes of this study, a p value less than .05 was considered significant.

Demographic Data
Figure 1 shows the demographic and health data for the study participants in the experimental and control groups. A total of 100 subjects participated in the study, but not all post-study instruments were completed. Reasons given for noncompliance included being too busy, forgetting or misplacing some of the instruments. Compliance was poorest among those subjects in the control group, even though, as an incentive, a one-hour massage was offered to the control group upon completion of the study. There were 89 subjects who completed all of the pre- and post-study instruments.

The mean age for the sample was 41.6 plus or minus () 8.1 years. Ten of the 89 subjects were smokers. Sixty-five subjects reported that they exercised regularly. Chi-square analysis showed no significant differences between the control group and treatment group for gender ( p=.15), marital status (p=.80), ethnicity (p=.27), overall health (p=.26), smoking (p=.45) or personal use of alternative therapies (p=.26). No significant differences in demographic data were noted in subjects who did not complete the study. 
 
Health-care workers at this institution generally work 12-hour shifts. Thirty-seven of the participants worked a 12-hour day shift, while 24 participants worked a 12-hour night shift. Fourteen of the subjects worked an eight-hour day shift. For subjects working a 12-hour shift, the treatment occurred between the seventh and ninth hours of their shift. For subjects working an eight-hour shift, the treatment occurred during the later half of their shift. All participants in the experimental group completed a minimum of 14 of the 16 scheduled massage therapy sessions.

Continue