It is well known that non-adherence to prescribed opioid regimens can decrease both the safety and effectiveness of this analgesic therapy. A new study examined factors associated with such noncompliance and found that somatization strongly influenced over- and underuse of medication. This may suggest strategies for more effective opioid prescribing, but there were some important limitations of this research.
Reporting in the May 2011 edition of the Journal of Pain, researchers affiliated with the Veterans Affairs Palo Alto Health Care System and Stanford University School of Medicine, Menlo Park, California, examined the relationship between somatization and non-adherence to prescribed opioid regimens [Trafton et al. 2011]. Subjects included 191 veterans who had received an opioid prescription during the prior 12 months; they were predominantly male (92%) Caucasians (73%) with a mean age of 62.5 years. Most (84%) had chronic pain lasting 6 months or longer, and 60% reported receiving an opioid prescription for more than 90 consecutive days.
Each participant completed a 1.5-hour interview that incorporated 15 assessments, including depressive symptoms, medication side effects, opioid pain medication usage, and somatization. The constructs of particular interest were:
- Underuse — taking opioid medication in less dosage than prescribed and reporting inadequate pain relief or impaired functioning.
- Overuse — taking more of an opioid analgesic than specified by the prescriber.
- Appropriate use — taking opioid medication as prescribed (either as-needed or within the daily maximum) and/or suitably discontinuing the opioid if it was no longer needed.
- Somatization — level of distress experienced from somatic symptoms, such as stomach pain or indigestion, dizziness, nausea, and trouble sleeping (determined via the Patient Health Questionnaire-15, or PHQ-15).
The researchers found that, overall, 80% of study participants reported some level of somatization during the prior month, and 40% indicated moderate to severe levels of distress about physical symptoms. At the same time, a majority of all patients (72%) had taken their opioid medication appropriately; of the 28% who were non-adherent 19% underused and 9% overused.
Somatization played a major role in non-adherence (see graph below). Compared with patients not experiencing somatization, the percentage of patients underusing their opioids (blue bars) increased as the level of somatization increased, but this remained fairly significant and stable throughout low-moderate-severe levels. Increases in medication overuse (red bars) were directly related to increasing levels of somatization, but reached the greatest percentage of overuse at the severe level.
Another way of looking at these data is that approximately 92% of patients reporting no somatization effects used their opioid medication appropriately (green bars). Whereas, only 75%, 66%, and 46% of those reporting low, moderate, and severe levels of somatization, respectively, were appropriate users.
Consistent with prior studies of medication adherence in general, increased depression and medication side effects were associated with greater non-adherence to opioid regimens. However, sophisticated statistical analyses conducted by the investigators determined that somatization was the primary factor driving the opioid overuse and underuse outcomes that were observed. Therefore, they concluded that pain management treatment plans utilizing opioid therapy may be optimized by addressing patient distress about physical symptoms.
COMMENTARY: An important finding of this study was that somatization may be a separate and critical factor affecting medication compliance, rather than merely serving as a proxy for depression and troublesome side effects. The clinical implication is that patients’ distress about somatic symptoms should be carefully considered by practitioners, rather than merely attempting to ameliorate depression or side effects with additional pharmacotherapy or other adjunctive medical treatments.
Unfortunately, the report authors do not adequately define somatization in ways that clinicians can put into practice; nor do they speculate as to why/how this distress motivates noncompliance. Somatization has been generally defined by others as a patient’s “tendency to experience and communicate somatic [physical] distress in response to psychosocial stress and to seek medical help for it." More than merely being the “kvetching” — that is, complaining or whining — of an unhappy patient, somatization also has been linked by psychologists to mental disorders, such as anxiety, depression, or somatoform disorder (physical symptoms without a medically explained cause).
In some cases, somatization has been associated with hypochondriasis [discussed in an UPDATE here]. And, then, there also is catastrophizing to be considered; the tendency to think the worst and worry excessively about symptoms or potential symptoms. Thus, somatization might be multidimensional and difficult to identify during typical patient-practitioner interactions.
Misunderstandings of these concepts could backfire at the point of care. For example, a healthcare provider might be uncomfortable prescribing opioid analgesics to a patient whose distress about physical symptoms is perceived as exaggerated or severe — and respond by unduly minimizing dosing. On the other hand, if somatization is properly recognized, skills training for the patient — cognitive behavioral therapy, relaxation techniques, etc. — could help to facilitate more adequate dosing while averting medication noncompliance.
Another failing of the present study, which the authors acknowledge, is that evidence of noncompliant behaviors was collected in a retrospective fashion via interview, at the same time the level of somatization was assessed; therefore, although somatization appeared to be directly associated with non-adherent opioid use it cannot be reliably concluded that somatization is predictive of this outcome. A prospective trial, determining levels of somatization at various points in time and correlating this with the degree of compliant medication-use behavior would be more definitive.
Finally, an important limitation of this current study was its focus on a population of male, white, older veterans. It seems that a great deal of recently reported research in the pain management field has come from U.S. Veterans Affairs Health Care Systems, which is probably convenient and expedient for the researchers, but may have no external validity for the population at large. Therefore, the results of this study are of interest, and may be of significant merit, but they need confirmation by better designed, broader scale investigations.