Monday, June 3, 2013

Does Long-Term Rehab for Chronic Pain Help?

Integrative Pain ManagementChronic musculoskeletal pain conditions are becoming more prevalent worldwide and are significant causes of disability and work absenteeism. Integrative pain management programs — melding physical, psychologic, and socioeconomic factors in biopsychosocial models of rehabilitation — expectedly facilitate a return to functionality and work. A recently reported study described a lengthy, integrative rehabilitation program designed to aid patients on work-disability leave due to chronic pain; however, results seem somewhat disappointing.

Writing in the June 2013 edition of Pain Management Nursing, Arnhild Myhr, MS, and Liv Berit Augestad, PhD — from the University of Science and Technology, Trondheim, Norway — describe a 57-week multidisciplinary program expected to rehabilitate workers disabled due to chronic musculoskeletal pain [Myhr and Augestad 2013]. Study participants for final analysis included 191 patients (63 men, 128 women; mean age ≈48 years) with chronic pain (>6 months), absent from work due to their disorders for greater than 6-months, and without any manifest organic diseases. All patients had been referred to the rehab program by their respective healthcare providers.

The rehabilitation regimen — described as a “learning to cope program” — incorporated extensive physical, psychologic, and social-function assessments; individual and group counseling and exercise routines, training and education; and collaboration with employers. Participants attended the program 24 hours/week for the first 5 weeks, and from 6 to 18 hours/week during the subsequent 52 weeks. Measurements of pain (100mm Visual Analog Scale, or VAS) and mental health (Hospital Anxiety and Depression Scale, or HADS), as well as sociodemographic data, were obtained before and after the rehabilitation period.

The researchers were particularly interested in how mental health factors (anxiety/depression reflected in HADS scores) would affect pain outcomes. The following results were reported:

  • At baseline, men showed higher anxiety/depression (HADS) scores than women.

  • Both women and men reported significant improvements in pain and mental health (anxiety/depression) after the rehab program.

  • Compared with the general population in the region, study subjects scored higher on HADS both before and after the rehabilitation program.

  • Older men and men with low education levels showed less improvement in HADS scores.

  • In women, a significant association was found between change in pain and changes in both anxiety and depression. However, depression was a more important factor in women and anxiety was more important in men.

Myhr and Augestad conclude that most of the individuals with chronic musculoskeletal pain benefitted from the multidisciplinary rehabilitation program, with improvements in both pain and mental health status. Furthermore, results support the hypothesis that there may be an important association between chronic pain and psychologic factors, but there are gender differences both in reporting of the problems and in treatment effects.

There are a number of questions surrounding this study regarding its design, reporting, and efficacy of the intervention. The researchers suggest that study results should be considered within context of the Norwegian welfare system. Employees receive their full salary during sick leave for up to 1 year, after which time they may qualify for rehabilitation-related financial support. The system is complex, distinguishing between medical rehabilitation and work-related rehabilitation, with patients sometimes becoming overwhelmed by the many government departments and treatment specialists involved.

The 57-week rehabilitation program in this study was funded by the Norwegian social security office; although costs were not indicated, it might be assumed that it would be rather expensive on a per patient basis. The biopsychosocial model appeared to emphasize psychological aspects of chronic pain (ie, “learning to cope”) and this may reflect a bias toward accepting what has been described as a “psychogenic pain” diagnosis [discussed and criticized in a recent UPDATE here].

That is, the multimodal rehabilitation program in this study seems predicated on an assumption that physical pain is largely a manifestation of mental health disturbances — eg, anxiety and depression — and correcting those helps to ameliorate pain and restore the patient to functionality. In return-to-work programs this also may reflect stigmatizing suspicions that patients are malingering or feigning pain to seek refuge from work, and that it is most essential to address root causes of worker discontent and mental stress rather than the pain itself.

RehabStatsStudy results suggest that there were positive changes in a majority of subjects in terms of physical capacity, HADS (anxiety/depression), and pain [see Figure at right, from Myhr and Augestad 2013]. At the same time, however, there were significant proportions exhibiting negative changes, particularly increases in pain, even after 57 months of intensive rehabilitation. This may suggest that roughly one-third of patients needed a more medical rather than psychological approach to management of their pain conditions.

Furthermore, while improvements in all areas were statistically significant (p<0.05 or better), the reported effect sizes for those changes (Cohen’s d) were only small to medium — ranging from 0.26 to 0.62. For example, the average change in pain perception (VAS scores) among women from baseline to 57-weeks was only 14% (d=0.53), and it was 12% in the men (d=0.26), neither of which are clinically significant improvements. [Interpretation of effect sizes was discussed in an UPDATE here].

In both cases, average VAS pain scores were still in the moderate range (56-60mm) following the long rehabilitation program. This might raise questions as to overall program effectiveness and its cost justification. There were additional unresolved questions in this regard due to limitations of the study’s design and reporting:

  • There is no mention in study data of the specific chronic musculoskeletal pain conditions among program participants, and whether the disorders were work-injury related or associated with other etiologies. The appropriateness of the psychological coping approach versus a more medical orientation cannot be determined.

  • The researchers do not mention the use of adjunctive medications for pain, anxiety, or depression by subjects. Assuming that at least some patients also were receiving medication therapy this could be a confounding factor; although, any analgesic pharmacotherapy received was probably inadequate judging from persistently moderate post-rehab pain scores.

  • There was no randomization to a control group for comparison (eg, wait list, alternative therapy, etc.), so effects of natural pain-disorder progression or remission could not be taken into account.

  • Outcomes data were self-reported by study participants and prone to biases or placebo effects that, without a control group, could not be adequately detected and assessed. There also was potential for the so-called “Hawthorne Effect”; that is, merely being part of a long-term study can influence how subjects respond, and this usually facilitates favorable intervention effects.

  • Most important, the researchers do not report on how many participants were able to return to work as a result of the lengthy rehabilitation program. This was either a serious oversight or, perhaps, this particular outcome was unfavorable and intentionally omitted — there is no way of knowing which is the case. Still, the researchers concede that, “…the most effective measure of success in rehabilitation of work-disabled individuals is reentry into the work environment.”

Myhr and Augestad suggest that the development of pain should be seen as a dynamic process, which includes both neurophysiologic pain regulation and learning as well as cognitive, emotional, and behavioral conditions. “It is reasonable that people suffering mental illness also may have increased attention toward their complaints and consequently the pain is of greater worry,” they state. “Many participants had had their complaints for many years and had been absent from work for a considerable period of time. The low level of education combined with a long absence period from work may indicate that the participants had different challenges in daily life in addition to their chronic pain.”

All of this rhetoric seems slanted toward a psychological perspective — a “psychogenic” model — which tends to belittle the medical etiologies and manifestations of chronic pain. The researchers do admit, “it is of interest to see that about one-third of the subjects had more pain and about one-fourth had more mental problems after the rehabilitation.” Their only recommendation is that future research should examine failings of the rehabilitation program and determine what characterizes these subjects — making them resistant to change.

At the same time, Myhr and Augestad note that, since the rehabilitation program was multidisciplinary, it was not possible to differentiate effects (of lack of effects) associated with individual elements of the program. Additionally, the program was customized for each participant, making the approach less standardized and difficult to compare with results of other studies.

Controversies surrounding integrative pain management approaches were described in a previous UPDATE [here]. Most recently, in an UPDATE [here], we described two uncontrolled, relatively short-term integrative programs addressing chronic widespread pain and chronic back pain. Psychological approaches (eg, “coping skills” training) had no impact on pain and little benefit for depression or global effects. Whereas, a program oriented more toward medical interventions offered significant benefits for pain relief and functional improvement, at least short term.

There is increasing interest in integrative — interdisciplinary and multimodal — approaches to chronic pain management using biopsychosocial models, but such programs can be expensive and difficult for many patients to access. Too often, they seem focused on avoiding pharmacotherapy and other medical interventions in deference to presumptive psychological etiologies and processes.

So far, high quality evidence appears to be lacking in support of strictly behavioral health or psychological approaches to chronic pain. Clearly, more research is needed to determine balanced and standardized integrative strategies that demonstrate long-term efficacy for benefitting both pain and mental health factors, as well as cost-effectiveness.

REFERENCE: Myhr A, Augestad LB. Chronic Pain Patients—Effects on Mental Health and Pain After a 57-Week Multidisciplinary Rehabilitation Program. Pain Management Nursing. 2013(Jun);14(2):74-84 [abstract here].

eNotifications Don’t Miss Out. Stay Up-to-Date on Pain-Topics UPDATES!
Register [here] to receive a once-weekly e-Notification of new postings.