Volume 90, Issue 3 , Pages 373-380, March 2009
Self-Management Improves Outcomes in Persons With Limb Loss
Article Outline
Abstract
Wegener ST, Mackenzie EJ, Ephraim P, Ehde D, Williams R. Self-management improves outcomes in persons with limb loss.
Objective
To test the acceptance and effectiveness of a community-based self-management (SM) intervention designed to improve outcomes after limb loss. A priori hypothesis was that an SM intervention will be more effective than standard support group activities in improving outcomes.
Design
Randomized controlled trial.
Setting
General community.
Participants
Intervention (N=287) and control participants (N=235) with major limb loss.
Intervention(s)
Nine, 90-minute SM group sessions delivered by trained volunteer leaders. Retention rates at immediate postintervention and 6-month follow-up were 97% and 91% for the SM group.
Main Outcome Measure(s)
Primary outcomes were depression, positive mood, and self-efficacy. Secondary outcomes were improved functional status and quality of life.
Results
By using intent-to-treat analyses, the odds for being depressed are significantly lower for those in SM group, 50% less likely at treatment completion (95% confidence interval [CI]=0.3–0.9) and 40% less likely at the 6-month follow-up (95% CI=.03–1.1). Treatment completers have a 70% reduction in likelihood of being depressed at posttreatment (P<.01) and this persists at six months (P<.05). For those in the SM group, functional limitations were significantly lower at 6 months (P<.05), and general self-efficacy was significantly higher at immediate posttreatment (P<.05) and at 6 months (P<.05). Treatment completers have generally significantly larger effect sizes at all follow-up points. Pain intensity, self-efficacy for pain control, and quality of life were not significantly different between the groups. Subgroup analyses indicated the impact of the intervention was greater for participants who were less than 3 years postamputation, participants who were less than 65 years of age, or participants who showed at least 1 secondary condition at baseline.
Conclusions
The study provides evidence that SM interventions can improve the outcomes of persons with limb loss beyond benefits offered by support groups.
Key Words: Amputation, Depression, Rehabilitation, Self-care, Self-efficacy
List of Abbreviations: ACA, Amputee Coalition of America, CBT, cognitive-behavioral theory, CESD, Center for Epidemiologic Studies Depression Scale, CI, confidence interval, MFA, Musculoskeletal Function Assessment, PALS, Promoting Amputee Life Skills, SM, self-management, SWLS, Satisfaction with Life Scale
LIMB LOSS IS A POTENTIALLY disabling condition affecting nearly 1.9 million Americans.1, 2 With the increasing trend in the incidence of limb loss because of the aging of the population and the increase in prevalence of conditions leading to peripheral vascular disease, there is a growing interest in the development of programs aimed at the prevention of secondary conditions affecting those living with the loss of a limb. Secondary conditions of depression and pain are particularly common among persons with limb loss,3, 4, 5 with 65% to 75% of amputees reporting phantom limb, residual limb, or back pain 3 or more years postamputation.6, 7 Rates of major depressive disorder are reported to be as high as 34% to 35% for inpatient amputees and 21% to 35% for outpatient amputees.8, 9, 10, 11 When present, these secondary conditions can significantly impact function resulting in restrictions in activities and diminished quality of life.12 Theoretic models suggest, and data indicate, that both individual factors (eg, cognitive coping) and environmental factors (eg, social support) impact the development and severity of these secondary conditions.13
Traditionally, interventions to address these secondary conditions and improve the health and outcomes of persons with disability have focused on standard medical treatments such as medication or physical rehabilitation therapies, often to the exclusion of psychosocial interventions. More recently, however, attention has been directed at supplementing these more traditional approaches with individual and group-based SM programs that are rooted in the principles of the CBT and use 2 broad approaches: coping skills training and cognitive restructuring techniques. CBT-based SM interventions have shown promise for improving outcomes in some chronic conditions including rheumatologic diseases,14 fibromyalgia,15 and diabetes,16 although the clinical impact varies based on condition and endpoint chosen.17 SM interventions appear to achieve long-term reductions in pain and disability primarily through increases in self-efficacy18 and decreases in catastrophizing rather than via specific behavior changes.19
The goal of the project was to test the acceptance and effectiveness of a community-based SM intervention for improving outcomes after limb loss. Primary outcomes were reduced pain and depression and improved positive mood and self-efficacy; secondary outcomes were improved functional status and improved quality of life. We hypothesized that (1) the SM intervention will be more effective than standard support group activities in reducing pain and depression and improving positive mood, function, and overall quality of life; (2) improvements in pain, depression, function, and quality of life will be accompanied by improvements in self-efficacy; and (3) treatment gains observed at the end of treatment would be maintained at the 6-month follow-up. As the literature has indicated, younger people, people with shorter time since amputation, and those with identified secondary conditions may have different affective responses and needs for psychosocial interventions20; we also planned to determine if people in these subgroups would benefit more from the intervention.
Methods
To test our hypotheses, we designed a trial in which pre-existing support groups were randomized to a treatment arm (participation in an SM program specifically designed for persons with limb loss) or control arm (participation in the existing support group activities). Given the evidence in the literature linking support group participation with reduced depression and distress,21, 22 increased knowledge,22, 23 and increased quality of life,24 we elected to use existing support groups as the control condition. Support group activities are widely available and represent the state-of-the-art care for persons with limb loss and other chronic conditions,25 and, thus, the goal was to determine if structured SM interventions improved outcomes compared with current care including support group attendance.
Participants and Inclusion Criteria
Eligible amputee support groups were identified from a list of support groups (n=91) registered with the ACA. Group inclusion criteria were meeting frequency of at least once a month and membership of 4 members or more. For each group, the registered support group leader was advised of the group's eligibility to participate in the study. Groups that agreed to participate in the study were randomly assigned in equal numbers to either the intervention or control arm of the study. After randomization, the groups received assistance in recruiting additional members through mailings to people in their region who were identified from an existing database of consumers of the ACA. Participants included were English-speaking people over 18 years of age who had limb loss.
The protocol was approved by the Institutional Review Board of the Johns Hopkins Bloomberg School of Public Health (Committee on Human Research). All participants provided written informed consent.
Interventions
The PALS SM Program was designed by using a participatory action research approach that included input from support group leaders and focus group participants. In a preliminary mailed survey, we obtained information from 93 support group leaders regarding the content, format, and timing of the PALS program. Subsequently, we used a 2-hour structured focus group format with one of the existing peer support groups to (1) refine the content, format, and timing of PALS; (2) identify qualifications for PALS group leaders; and (3) identify methods to overcome the barriers to recruitment and participation in PALS. The final development of the PALS SM was informed by input from an advisory committee and pilot testing.
The PALS intervention was designed to be delivered in a community setting by trained volunteer leaders. It is further characterized by (1) content developed specifically for persons with limb loss; (2) established SM principles (knowledge, problem solving, skill acquisition, and self-monitoring26); and (3) a focus on reducing secondary conditions and improving activity, participation, and quality of life. The intervention consisted of 8 weekly 90-minute sessions followed by a booster session 2 weeks later all delivered in a group format. Topics for the 8 sessions included (1) overview of self-management, (2) pain management, (3) building positive mood, (4) managing negative mood, (5) interacting with family and friends, (6) working with health care team/community resources, (7) building healthy habits, (8) relapse prevention, and (9) maintaining progress.
The PALS groups were led by 2 volunteers: a trainer and a facilitator. The trainer was an individual with experience in health care, experience teaching or human services activity, and experience working with amputees. The trainer was responsible for leading the session and content delivery. The facilitator provided assistance with logistics of the group meetings. At least one of the leaders was a person with limb loss. The trainers attended a 3-day training seminar that emphasized knowledge and practice of the manualized PALS SM program and training on group leadership. Facilitators were trained on a 2-hour conference call.
Control Group
Participants in the control group continued to attend the pre-existing, regularly scheduled, peer support group meetings led by volunteers. These groups met on average once a month over the 3-month treatment period. All study participants continued to receive their usual medical and prosthetic care.
Outcome Measures
Enrolled participants in both groups participated in a structured computer-assisted telephone interview at baseline, immediately after treatment, and 6 months after the intervention. The interview included the following measures of the primary outcomes: the Brief Pain Inventory,27 the CESD,28 Positive and Negative Affect Schedule,29 Positive States of Mind,30, 31 and a modified self-efficacy scale.19 Secondary outcomes were measured by using the Musculoskeletal Function Assessment–Short Form developed specifically to measure function and bothersomeness of functional limitations in persons with musculoskeletal impairments, including amputation32, 33 and the SWLS as a measure of global quality of life satisfaction.34 Participants also responded to questions regarding satisfaction and benefits of PALS program.
Self-reported information on sex, age, marital status, time since limb loss, level and etiology of limb loss, race/ethnicity, education level, household income and number in household, health insurance status, and comorbidities as measured by the Charlson Index35 was also collected at baseline.
The integrity of the PALS SM intervention was assessed by a review of session audiotapes and attendance records. A random sample of 10% of the sessions was reviewed by study investigators to ensure trainer adherence to the manualized PALS SM protocol.
Statistical Analysis
The primary analyses were performed by using an intention-to-treat method. We also examined treatment effects among those who completed treatment, defined as participants who attended 6 or more sessions of the intervention. For each dependent variable, except depressed mood, changes from baseline at immediate posttreatment and 6 months posttreatment were analyzed by using multivariate linear regression models. For depressed mood, logistic regression was used, and odds ratios are reported. The randomization of groups, rather than subjects, to control and treatment conditions raises the problem of increased similarity between subjects, decreased standard errors, and the related increased probability of type I error because of nested group effects. During the analyses, linear mixed effects models were constructed, and it was found that the within-group correlations were very small and nonsignificant and did not alter results. Multivariate regression analyses results are reported for ease of interpretation.
Power and sample size calculations were performed for the primary outcome pain. We determined that a sample size of 200 per group at treatment completion (effect size of 0.18 specifying an alpha level of 0.05) yielded a power estimate of 0.95. A sample size of 360 patients (N=180 in each group) completing the 6-month follow-up assessment allowed for statistical power of 0.93 to detect an effect size of 0.18 in group comparisons at 6 months.
Results
A total of 52 amputee support groups agreed to participate in the study and were randomly assigned to either the intervention (n=26) or control (n=26) arm of the study. A total of 502 people were enrolled in the study and completed the baseline interview (275 in the SM groups and 227 in the support groups, fig 1). Among those enrolled in the SM arm, 34 completed the baseline interview but never attended any of the SM classes. Group size ranged from 5 to 21 participants. A total of 487 participants completed the immediate postintervention interview (221 support group, 266 SM group), and 460 participants completed the 6 month postintervention interview (208 support group, 251 SM group). Resulting retention rates at immediate postintervention and 6-month follow-up were 97% and 92%, respectively, for the support group, and 97% and 91% for the SM group.
Fig 1.
Enrollment and scheduled follow-up in the PALS Self-Management Program. *Attended 1 or more of 9 self-management sessions, **total includes 3 subjects who did not complete a baseline interview, ***total includes 6 subjects who did not complete a baseline interview.
Outcomes
The groups were equivalent at baseline on the measured characteristics (Table 1, Table 2). At baseline, 28% of the support group and 32% of the SM group met criteria for probable depression as measured by the CESD.28 Pain levels for both groups were in the mild to low moderate range. The overall functional status for the sample as measured by the MFA–Short Form is characterized by mild to moderate limitations.32, 33 General self-efficacy scores at baseline were higher than those observed in other persons with chronic conditions.36 Satisfaction with quality of life was similar to that of people without limb loss.34
Table 1. Baseline Characteristics in the Routine Support Group and Self-Management Group
| Routine Support Group (n=227) | Self-Management Group (n=275) | |
|---|---|---|
| Sex, n (%) | ||
 Men | 134(59) | 151(55) |
| Women | 93(41) | 124(45) |
| Age, mean±SD | 56.9±13.3 | 55.5±13.8 |
| Time since limb loss, median, y (range) | 5(0-65) | 4(0-74) |
| Level of limb loss, n (%) | ||
| Upper limb | 11(4.9) | 20(7.3) |
| Lower limb | 216(95.1) | 255(92.7) |
| Etiology of limb loss, n (%) | ||
| Dysvascular/diabetes | 79(34.8) | 102(37.1) |
| Trauma | 82(36.1) | 108(39.3) |
| Cancer | 24(10.6) | 15(5.5) |
| Limb deficiencies | 4(1.8) | 10(3.6) |
| Other | 38(16.7) | 40(14.5) |
| Self-reported race/ethnicity, n (%) | ||
| White, non-Hispanic | 179(78.9) | 215(78.2) |
| Black, non-Hispanic | 27(11.9) | 32(11.6) |
| Hispanic | 10(4.4) | 21(7.6) |
| Other | 11(4.9) | 7(2.6) |
| Education level, n (%) | ||
| Less than 12th grade | 12(5.2) | 14(5.1) |
| High school graduate/GED | 62(27.3) | 66(24) |
| Greater than 12th grade | 153(67.4) | 195(70.9) |
| Poverty status, n (%) | ||
| Not poor | 34(61.8) | 165(65.2) |
| Near poor | 51(23.5) | 57(22.5) |
| Poor | 32(14.8) | 31(12.3) |
| Health insurance, n (%) | ||
| Medicare | 103(45.4) | 125(45.5) |
| Medicaid | 13(5.7) | 16(5.8) |
| Private | 89(39.2) | 103(37.5) |
| Other | 11(4.8) | 14(5.1) |
| Uninsured | 11(4.9) | 17(6.2) |
Table 2. Primary and Secondary Outcome Measures at Baseline, Immediate Postintervention, and 6 Months Postintervention
| Variable | Baseline | Immediate Postintervention | 6 Months Postintervention | |||
|---|---|---|---|---|---|---|
| Routine Support Group (n=227) | Self-Management (n=275) | Routine Support Group (n=215) | Self-Management (n=263) | Routine Support Group (n=208) | Self-Management (n=251) | |
| Primary outcomes | ||||||
| Depressed mood (%) | 28.2 | 32.4 | 27.0 | 23.3 | 25.0 | 22.3 |
| Average pain | 3.1±3.8 | 3.5±3.4 | 3.5±3.6 | 3.8±3.2 | 4.4±2.4 | 4.2±2.5 |
| Positive affect | 23.6±9.1 | 23.4±9.9 | 23.7±9.4 | 25.2±10.1 | 23.5±9.0 | 24.6±9.2 |
| Positive states of mind | 17.8±2.9 | 17.4±2.9 | 18.0±2.4 | 17.9±2.7 | 17.6±2.8 | 17.8±2.6 |
| Self-efficacy | ||||||
| Pain | 5.7±2.5 | 6.3±2.3 | 6.1±2.4 | 6.7±2.4 | 6.2±2.2 | 6.7±2.4 |
| General | 7.7±1.9 | 7.7±1.9 | 7.7±1.8 | 8.1±1.8 | 7.8±1.8 | 8.2±1.7 |
| Secondary outcomes | ||||||
| Musculoskeletal function | ||||||
| Dysfunction index | 20.9±12.9 | 21.6±13.9 | 20.9±12.5 | 20.2±12.3 | 21.3±13.6 | 19.6±12.2 |
| Bothered index | 24.6±21.3 | 25.5±22.8 | 25.2±21.2 | 21.9±19.2 | 25.3±20.3 | 22.6±20.3 |
| SWLS | 21.8±8.2 | 21.9±8.3 | 21.9±8.0 | 22.8±8.6 | 22.4±7.8 | 23.2±8.0 |
Table 3 presents the effect sizes of the SM group compared with the support group using both an intention-to-treat analysis and an analysis based only on those who completed treatment (defined as the 166 or 63.8% of the participants who attended 6 or more sessions). In the intention-to-treat analyses, the odds of being depressed are significantly lower for those in SM group posttreatment, 50% less likely at treatment completion (95% CI=.0.3–0.9) and 40% less likely at the 6-month follow-up (95% CI=.03–1.1). Treatment completers have a 70% reduction in likelihood of being depressed at posttreatment (P<.01), and this persists at 6 months (P<.05). There was an initial trend toward improvement in positive affect (P<.10), but this trend was not maintained at 6 month follow-up. Completers showed significant improvement in positive cognitions as measured by the Positive States of Mind at 6 months (P<.01), an effect that was not present immediately posttreatment (P=.47). There were no significant differences in average pain intensity as measured by the Brief Pain Inventory.
Table 3. Assessment of the Effect of Treatment at Immediate and 6 Months Posttreatment in Those Randomized (Intention to Treat) and Completing Treatment
| Model | Immediate Posttreatment | 6 Months Posttreatment | ||||||
|---|---|---|---|---|---|---|---|---|
| Completed Treatment (n=166) | Intention to Treat (n=260) | Completed Treatment (n=161) | Intention to Treat (n=249) | |||||
| Outcome Variable | Effect Size⁎ | P | Effect Size⁎ | P | Effect Size⁎ | P | Effect Size⁎ | P |
| Primary outcomes | ||||||||
| Depressed mood (OR) | .3⁎ | .002 | .5‡ | .03 | .3‡ | .03 | .6 | .12 |
| Average pain | .05 | .63 | .05 | .62 | .04 | .70 | .03 | .81 |
| Positive affect | .18† | .07 | .17† | .10 | .11 | .30 | .13 | .21 |
| Positive states of mind | .11 | .26 | .07 | .47 | .31‡ | .003 | .31‡ | .004 |
| Pain self-efficacy | .23‡ | .05 | .15 | .19 | .13 | .27 | .15 | .20 |
| General self-efficacy | .32‡ | .002 | .24‡ | .02 | .27‡ | .01 | .24‡ | .02 |
| Secondary outcomes | ||||||||
| MFA dysfunction | .25‡ | .02 | .13 | .19 | .33‡ | .002 | .21‡ | .04 |
| MFA bothered | .31‡ | .003 | .28‡ | .007 | .27‡ | .01 | .17† | .10 |
| SWLS | .05 | .66 | .04 | .68 | .09 | .39 | .06 | .55 |
⁎Multivariate regression model of mean change from baseline except in case of depression in which logistic regression was used (odds ratio [OR] reported). |
†P<.10. |
‡P<.05. |
Functional limitations were significantly lower for those in the SM group at 6 months (P<.05). There were significant reductions in bothersomeness of functional limitations immediate posttreatment (P<.01) with a diminution of effect size at 6 months, resulting in only a trend toward significance (P<.10). Examination of the treatment completers indicates these participants had significantly lower functional limitations and bothersomeness at all time points (P<.05) with effect sizes larger than those observed in the intention-to-treat analyses. Quality of life as measured by the SWLS was not significantly different between the groups. General self-efficacy, the person's confidence in his/her ability to manage the problems related to limb loss, was significantly higher for the SM group at immediate posttreatment (P<.05) and this gain was maintained at the 6-month follow-up (P<.05). For those who completed treatment, the effect size is larger at both immediate posttreatment and 6 months. Self-efficacy for pain control was not significantly different in the intention-to-treat analysis, and initial improvement in the treatment completers was not maintained at follow-up.
Overall, both support group and SM group participants perceived their respective activity as helpful with 75% and 79%, respectively, reporting a rating of 7 or more on a 0 to 10 scale. However, the SM participants rated the program as more helpful in managing pain (P<.01) and in increasing their confidence to improve their quality of life (P<.05). There was a trend toward rating SM as more helpful in the areas of managing mood (P=.06) and managing other limb loss–related problems (P=.06). Ninety-five percent of the SM participants would recommend the program to a friend. Among SM participants who had previously been members of a support group, 53% indicated that SM was much more helpful and 23% somewhat more helpful than support group attendance. The majority of both SM participants (58%) indicate benefits outweighed the effort with 32% seeing the benefits and efforts as equal. Three quarters (74%) indicated that the program was more important than other services received. SM participants also provided information on the timing of the SM intervention, with 42% indicating that SM training would be most helpful in the first 3 months after amputation, 30% in the 3- to 6-month timeframe, and 21% in 6 to 12 months.
Subgroup Analyses
Analyses were conducted to examine differential responses to the SM intervention. Specific to our hypotheses, we investigated the effect of the SM intervention among participants in 3 subgroups: (1) people less than 3 years postamputation (n=106 in the intervention group and n=78 in the control group); (2) people who had a significant secondary condition at baseline as indicated by an average pain intensity score equal to or greater than 4/10, a score of equal to or greater than 16 on the CESD, or a general self-efficacy score less than 6 (n=114 in the intervention group and n=149 in the control group); and (3) persons who were less than 65 years old (n=177 in the intervention group and n=130 in the control group). Before proceeding with these analyses, differences in baseline measures were compared between the support group and PALS SM group. Few differences were found, and any imbalance in covariates was addressed in the development of the multivariate models.
Table 4 summarizes the 6-month results from the intent-to-treat analysis (N differs slightly from total sample N above because of dropouts). Comparison of the effect sizes in these tables compared with those estimated for the entire population (see table 3) indicates that for all 3 subgroups, the impact of the SM intervention was substantially larger for these people when compared with the group as a whole as shown by the increased effect sizes. Participants less than 3 years postamputation in the SM group were 60% less likely to be depressed; had higher levels of positive cognitions (P<.05), functional abilities (P<.05), and general self-efficacy (P<.05); and, unlike the entire sample, this subgroup had higher levels of positive effect (P<.05). Participants who had at least 1 secondary condition in the SM group were 70% less likely to be depressed; had higher levels of positive cognitions (P<.001), functional abilities (P<.001), and general self-efficacy (P<.01); and, unlike the entire sample, this subgroup also had positive affect (P<.01). Participants less than 65 years of age in the SM group were 40% less likely to be depressed; had higher levels of positive cognitions (P<.05) and general self-efficacy (P<.05); and, unlike the entire sample, this subgroup had higher levels of positive affect (P<.05).
Table 4. Results of Subgroup Intention-to-Treat Analyses PALS 6 Months Posttreatment
| Model | Persons <3 Years Postamputation (n=174) | Persons Increased Pain, Depressed Mood, or Low Self-Efficacy (n=248) | Persons <65 Years of Age (n=292) | |||
|---|---|---|---|---|---|---|
| Outcome Variable | Effect Size⁎ | P | Effect Size⁎ | P | Effect Size⁎ | P |
| Primary outcomes | ||||||
| Depressed mood (OR) | .4 | .17 | .3‡ | .003 | .6 | .16 |
| Average pain | .01 | .98 | .09 | .50 | .03 | .87 |
| Positive affect | .34‡ | .05 | .38‡ | .007 | .25‡ | .05 |
| Positive states of mind | .43‡ | .02 | .51‡ | <.001 | .38‡ | .003 |
| Pain self-efficacy | .34† | .07 | .15 | .29 | .22 | .12 |
| General self-efficacy | .41‡ | .02 | .41‡ | .003 | .30‡ | .02 |
| Secondary outcomes | ||||||
| MFA dysfunction | .37‡ | .04 | .47‡ | <.001 | .22† | .08 |
| MFA bothered | .13 | .44 | .26† | .06 | .21† | .09 |
| SWLS | .21 | .23 | .25† | .08 | .13 | .30 |
⁎Multivariate regression model of mean change from baseline except in case of depression in which logistic regression is used (odds ratio [OR] reported). |
†P<.10. |
‡P<.05. |
Discussion
To our knowledge, this is the first randomized controlled trial investigating the effectiveness of an SM program for persons with limb loss. The study provides evidence that SM interventions can improve the outcomes of persons with limb loss beyond any benefits offered by support group participation. This suggests that the structured, skill-based activities of the SM approach are beneficial beyond the potential nonspecific benefits including modeling, social support, and informal information sharing, which may accrue from peer support groups.
The study adds to the growing literature that supports SM as a strategy to achieve patient-centered care as called for by the Institute of Medicine in their report on Crossing the Quality Chasm.37 The SM approach reifies the patient-centered model because it widens the focus from the patient's medical needs to the needs of the patient as a whole; places a premium on patient involvement; and highlights the need for improved patient education, skills, family involvement, and emotional support. PALS and other SM interventions respond to current changes in health care that require patients and their families to assume greater responsibility for their health care.
Strengths of PALS SM program are the high level of retention, consumer satisfaction, and potential for sustainability. The high level of retention (>90%) and consumer satisfaction are attributed to use of a participatory action research approach in the design and implementation of PALS. Groups of persons with limb loss and members of the ACA played an integral role in the development of both the intervention and study design. At least one of the PALS leaders in each group was a person with limb loss, which provided credibility and opportunity for learning through peer modeling. The program has potential for sustainability because collaboration with a national consumer organization provides a mechanism to recruit and train PALS group leaders and distribute the program nationwide.
The literature indicates that SM interventions in the context of regular medical care have improved outcomes in several conditions with retention of improvements in self-efficacy, self-management behaviors, symptom reduction, and medical utilization.14, 16, 18, 38, 39, 40, 41 Although these overall results are encouraging, a recent meta-analyses of SM interventions for older adults indicates that the magnitude of improvement varies depending on condition and endpoint chosen.17 In this study, the effect sizes observed in the total sample intention-to-treat analyses for depressive symptom risk reduction are of considerable clinical significance. The effect sizes in the other primary and secondary outcomes are statistically significant but clinically modest and are similar to those reported by other investigators in other musculoskeletal conditions. In part, this limited impact is caused by the limited room for improvement in the areas of pain, positive affect, and quality of life, which were relatively positive at baseline.
However, the subgroup analyses indicate that even greater benefit was detected for persons who were either less than 3 years postamputation, had a secondary condition at baseline, or were less than 65. These findings are consistent with observations in other musculoskeletal impairments in which early cognitive-behavioral–based interventions within a secondary prevention model have been effective in reducing pain, distress, and disability.42, 43 The potential for greater differential benefits for persons less than 65 is also supported by the literature. Although not universally replicated,44 when studies do find a relationship between age and adjustment postamputation, findings indicate that older persons tend to be less depressed after amputation.20, 45 Horgan and MacLachlan20 suggest that this relationship between age and adjustment is mediated by activity restrictions, with restrictions because of pain or illness being relatively more distressing to younger persons.
Study Limitations
There are several study limitations that must be considered. In comparison to national data on persons with limb loss, the study sample includes a higher percentage of people with amputations from trauma and a lower percentage of nonwhite participants. Thus, we should be cautious about generalizing these results to the entire population of persons with limb loss. A second limitation is that the study design randomly assigned groups to the control or intervention arm of the study. The impact of this randomization approach in comparison to random assignment of people is uncertain. The recruitment of the intervention group from those already attending a support group leads to a sample biased toward those who are actively engaged in, and predisposed to participate in, psychosocial interventions. This type of intervention may be more effective in this type of person. However, the use of support group participation as the control condition set a high standard for showing the impact of the SM intervention. Another limitation to the current study is the potential impact of increased contact time in those who received the intervention. The PALS groups met 9 times over the 3-month period, whereas the support groups met on average 3 times. Although the level of contact time may have an influence, our goal was to determine if structured SM interventions improved outcomes compared with current care including support group attendance. Support groups do not generally meet weekly so it was not reasonable to equalize contact between the groups. Finally, when multiple outcomes measures are used, there is the potential of increased type I errors because of multiple statistical tests. Although there are several approaches to managing the issue including use of a Bonferroni correction, given the fact that there are very few published studies in this area, we elected to report the effect sizes and P values so that readers are able to make their own judgments as to the impact of the intervention.
Conclusions
Several lines of research suggest approaches that may enhance the benefits of SM. There is growing recognition that utilization of peer mentors may be helpful in assisting people with new impairments with successful adaptation.4 Motivational interviewing techniques have been developed and shown to be efficacious in increasing participation in a variety of health behaviors.46 Training health care providers in the importance of SM and how to better support people in their use of these techniques should increase enrollment in SM.47 Finally, using a secondary prevention model that emphasizes early treatment after initial impairment is thought to be effective in reducing secondary conditions and improving “downstream” health in terms of positive outcomes of activity and participation.48, 49 Although these approaches are promising, studies are needed to determine if the use of these methods in concert with SM programs improves outcomes for persons at risk for disability.
Acknowledgments
We thank Brendan Klick, ScM, Johns Hopkins University School of Medicine for statistical analyses assistance; Amputee Coalition of America collaborators: Paddy Rossbach; Pat Isenberg; Susan Tipton; the PALS Advisory Committee; and PALS trainers, facilitators, participants, and Advisory Board members.
References
- . Current estimates from the National Health Interview Survey, 1996 (Centers for Disease Control and Prevention/National Center for Health Statistics). Vital Health Stat. 1999;200:1–203
- . Estimating the prevalence of limb loss in the United States: 2005-2050. Arch Phys Med Rehabil. 2008;89:422–429
- . Environmental barriers experienced by amputees: the Craig Hospital Inventory of Environmental Factors-Short Form. Arch Phys Med Rehabil. 2006;87:328–333
- . Phantom pain, residual limb pain, and back pain in amputees: results of a national survey. Arch Phys Med Rehabil. 2005;86:1910–1919
- Depressive symptoms and mental health service utilization among persons with limb loss: results of a national survey. Arch Phys Med Rehabil. 2005;86:650–658
- Chronic phantom sensations, phantom pain, residual limb pain, and other regional pain after lower limb amputation. Arch Phys Med Rehabil. 2000;81:1039–1044Aug
- . Phantom limb, residual limb, and back pain after lower extremity amputations. Clin Orthop Relat Res. 1999;29–38Apr(361)
- . Depression among amputees. J Clin Psychiatry. 1983;44:256–258
- . Depression in men with traumatic lower part amputation: a comparison to men with surgical lower part amputation. Mil Med. 2003;168:106–109
- . Back pain as a secondary disability in persons with lower limb amptuations. Arch Phys Med Rehabil. 2001;82:731–734
- . Social and psychological factors in adjustment to limb amputation. J Soc Behav Pers. 1994;9:249–268
- An analysis of outcomes of reconstruction or amputation after leg-threatening injuries. N Engl J Med. 2003;347:1924–1931
- . Linking clinical variables with health-related quality of life (A conceptual model of patient outcomes). JAMA. 1995;273:59–65
- Evidence suggesting that a chronic disease self-management program can improve health status while reducing hospitalization: a randomized trial. Med Care. 1999;37:5–14
- . Self-management of fibromyalgia: the role of formal coping skills training and physical exercise training programs. Arthritis Care Res. 1998;11:432–447
- . Effectiveness of self-management training in type 2 diabetes: a systematic review of randomized controlled trials. Diabetes Care. 2001;24:561–587
- Meta-analysis: chronic disease self-management programs for older adults. Ann Intern Med. 2005;143:427–438
- . Arthritis self-management studies: a twelve-year review. Health Educ Q. 1993;20:17–28
- . The beneficial outcomes of the arthritis self-management course are not adequately explained by behavior change. Arthritis Rheum. 1989;32:91–95
- . Psychosocial adjustment to lower-limb amputation: a review. Disabil Rehabil. 2004;26:837–850
- . Professional and paraprofessional group treatments for depression: a comparison of cognitive-behavioral and mutual support interventions. J Consult Clin Psychol. 1999;67:491–501
- . Meta-analysis of psychosocial interventions for caregivers of people with dementia. J Am Geriatr Soc. 2003;51:657–664
- . Social support in diabetes: a systematic review of controlled intervention studies. Patient Educ Couns. 2005;59:1–12
- . Effects of psychosocial interventions on quality of life in adult cancer patients: meta analysis of 37 published controlled outcome studies. Patient Educ Couns. 2003;50:179–186
- . Who talks? (The social psychology of illness support groups). Am Psychol. 2000;55:205–217
- . Self-management education: history, definition, outcomes, and mechanisms. Ann Behav Med. 2003;26:1–7
- . Pain assessment: global use of the Brief Pain Inventory. Ann Acad Med Singapore. 1994;23:129–138
- . The CES-D Scale: a self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401
- . Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol. 1988;54:1063–1070
- . Additional validation of a scale to assess positive states of mind. Psychosom Med. 1998;60:26–32
- . A scale for measuring the occurrence of positive states of mind: a preliminary report. Psychosom Med. 1988;50:477–483
- . Short musculoskeletal function assessment questionnaire: validity, reliability, and responsiveness. J Bone Joint Surg Am. 1999;81:1245–1260
- . Musculoskeletal Function Assessment instrument: criterion and construct validity. J Orthop Res. 1996;14:182–192
- . The Satisfaction with Life Scale. J Pers Assess. 1985;49:71–75
- . A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–383
- . Effect of a self-management program on patients with chronic disease. Eff Clin Pract. 2001;4:256–262
- . Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: Institute of Medicine; 2001;
- . Three-year follow-up of bibliotherapy for depression. J Consult Clin Psychol. 1997;65:324–327
- . Self-help treatment for migraine headaches: a controlled outcome study. Headache. 1981;21:196–200
- . Group education for patients with rheumatoid arthritis. Patient Educ Couns. 1993;20:177–187
- . Evidence suggesting that health education for self-management in patients with chronic arthritis has sustained health benefits while reducing health care costs. Arthritis Rheum. 1993;36:439–446
- . A cognitive-behavioral approach to the prevention of chronic back pain. In: Turk DC, Gatchel RJ editor. Psychological approaches to pain management: a practitioner's handbook. New York: Guilford Press; 2002;p. 317–333
- . Prevention of chronic pain after whiplash. Emerg Med J. 2002;19:526–530
- . Social discomfort and depression in a sample of adults with leg amputations. Arch Phys Med Rehabil. 1992;73:1169–1173
- . Psychological response to amputation as a function of age and time since amputation. Br J Psychiatry. 1984;144:493–497
- . Motivational interviewing: a systematic review and meta-analysis. Br J Gen Pract. 2005;55:305–312
- . Exploring oncologists' beliefs about psychosocial groups: implications for patient care and research. J Psychosoc Oncol. 1994;12:77–87
- . Healthy People 2010: understanding and improving health. 2nd ed.. Washington, D.C.: U.S. Government Printing Office; 2000;
- . Early identification and intervention in the prevention of musculoskeletal pain. Am J Ind Med. 2002;41:433–442
Supported by the Centers for Disease Control and Prevention (grant no. R04/CCU322981).
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.
PII: S0003-9993(08)01700-0
doi:10.1016/j.apmr.2008.08.222
© 2009 American Congress of Rehabilitation Medicine. Published by Elsevier Inc. All rights reserved.
Volume 90, Issue 3 , Pages 373-380, March 2009
