Veterans Walk to Beat Back Pain
Sarah L. Krein PhD RN
VA Ann Arbor Healthcare System, Ann Arbor, MI
Ann Arbor, MI
Funding Period: October 2008 - March 2012
Chronic pain, and especially back pain, is common among VA patients, affecting approximately 60% of veterans using VA primary care services. Chronic pain is associated with increased disability, poorer health status, anxiety and depression, decreased quality of life, decreased employment, and increased health services use and costs. Current evidence suggests that exercise is one of the most effective strategies for managing chronic pain. However, there are few clinical programs that use generally available tools and a relatively low-cost approach to help patients with chronic back pain initiate and maintain an exercise program.
The primary objective of this study was to assess the efficacy of an Internet-mediated, pedometer-based intervention designed to increase walking and improve pain-related function among individuals with chronic back pain. The specific aims of this randomized, controlled trial were: 1) to determine whether an internet-mediated, pedometer-based intervention would reduce pain-related functional interference among patients with chronic back pain in the short term (at 6 months) and over a 12-month timeframe; 2) to assess the effect of the intervention on walking (measured by step counts), quality of life, pain intensity, pain-related fear, and self-efficacy for exercise among patients with chronic back pain; and 3) to identify factors associated with a sustained increase in walking over a 12-month timeframe among patients randomized to the intervention.
We conducted a parallel group, randomized controlled trial of an Internet-mediated, pedometer-based intervention to promote walking and improve pain-related function among patients with chronic back pain compared to enhanced usual care. Participants were followed for 12 months to investigate the efficacy of the intervention in assisting patients with initiating and maintaining a regular walking program and improving pain-related function. We recruited patients with chronic back pain (primarily low back pain) receiving care at one VA health care system. Study patients were randomly allocated in a 1:1 ratio by a computer program to one of two groups: 1) enhanced usual care, or 2) the Internet-mediated, pedometer-based intervention. All participants attended an educational program for individuals with chronic back pain. Study participants randomized to the intervention were given an enhanced pedometer and access to a study website that provided step goals and feedback, targeted motivational messages, and social support through an e-community. Those in the control group received a pedometer but did not have access to the website. Both groups completed on-line survey assessments at baseline, 6, and 12 months, and were asked to report adverse events on a regular basis. The primary outcome was changes in pain-related functional interference, as measured using the Roland-Morris Disability (RMD) Index, a back pain specific measure, and the Medical Outcomes Study pain effects measure, which assesses the effect of pain in general on mood and behaviors as well as the severity of pain over the past 4 weeks. Secondary outcomes included changes in average daily steps as measured using the study pedometer, health-related quality of life, pain intensity, pain-related fear, and self-efficacy for exercise. The analysis was conducted based on the original randomized treatment assignment regardless of participation (an intent-to-treat analysis) and included both a complete case analysis as well as an all case analysis using a linear mixed-effects model. Between-group differences in change scores (6 months or 12 months) were estimated after adjusting for baseline values of the outcome variables.
A total of 229 Veterans with chronic back pain were randomly assigned to the intervention (n = 111) or the control group (n = 118). Participant follow-up was approximately 90% in both groups at 6 (intervention: n = 101; control: n = 106) and 12 (intervention: n = 102; control: n = 105) months. Study participants were predominantly male (89% in the intervention and 86% in the control group), with an average age of about 51 years and average body mass index of 31. Over 40% of patients in both groups rated their general health status as fair or poor, and many also reported taking narcotic medications for pain relief (41% in the intervention group and 49% in the control group). There were, however, no significant differences in baseline characteristics between the two study groups. Among those completing the assessments at 6 months, intervention patients reported significantly less back pain-related disability compared to controls (RMD: 7.2 vs. 9.2, adjusted difference: 1.53, p = 0.02) as well as less general pain-related functional interference (37.6 vs. 43.6, adjusted difference: 3.60, p = 0.09), lower pain scores (4.7 vs. 5.2, adjusted difference: 0.48, p = 0.06) and greater exercise self-efficacy (6.4 vs. 5.7, adjusted difference: 0.81, p = 0.006). There was no between-group difference in fear avoidance scores (13.2 vs. 14.0, adjusted difference: 0.63, p = 0.42). Average step counts increased in the intervention group by nearly 800 steps compared to only 5 steps for the control group (adjusted difference: 725.5, p = 0.12). The adjusted differences in the all case analysis also suggested more improvement among those in the intervention compared to the control group, although some of the differences were no longer statistically significant: RMD adjusted difference 1.24, p = 0.09; general pain effects adjusted difference 2.56, p = 0.25; average pain score adjusted difference 0.51, p = 0.06; exercise self-efficacy adjusted difference 0.66, p = 0.03; and, average step counts adjusted difference 724, p = 0.08. At 12 months, both the complete case and all case analyses continued to show positive improvements and slightly better scores for those in the intervention group compared to the control group. However, the observed differences were no longer statistically significant for any of our key outcome measures. In light of this and the variable level of engagement among intervention participants, additional strategies to keep people active may also be needed. In addition, analysis of both qualitative and quantitative data are ongoing, to identify patient factors or systematic issues that might be associated with, or help to improve, engagement and sustainability.
Improving pain management is a VHA national priority. Exercise is an integral part of managing chronic back pain, but to be effective requires that patients actively participate as part of the management process. Veterans in the Internet-mediated intervention, compared with those in the control group, reported a greater decrease in back pain-related disability (i.e., better function) and increased their physical activity (walking) in the 6 months following study enrollment. Intervention patients also reported less pain and improvements in general pain-related function, and had better exercise self-efficacy, compared to control patients. However, even though intervention patients continued to report greater improvements in pain-related function, the differences between groups were no longer statistically significant at 12 months. A facilitated walking intervention that uses an enhanced pedometer and the Internet may be an important strategy for helping patients with chronic back pain engage in exercise and improve outcomes, at least in the short term. Additional strategies may be needed, however, to ensure sustained or continuing improvements over a longer timeframe. Nonetheless, such intensive but primarily automated interventions can be used to deliver care with broad reach and could be an efficient way of delivering or supplementing care provided through traditional facility-based programs.
DRA: Aging, Older Veterans' Health and Care, Acute and Combat-Related Injury, Musculoskeletal Disorders
DRE: Treatment - Observational, Prevention, Treatment - Efficacy/Effectiveness Clinical Trial
Keywords: Back pain, Behavior (patient), Exercise, Functional Status, Pain
MeSH Terms: none