OUP user menu

Associations Between Spousal or Significant Other Solicitous Responses and Opioid Dose in Patients with Chronic Pain

Julie L. Cunningham PharmD, Sarah E. Hayes BA, Cynthia O. Townsend PhD, Heidi J. Laures BS, W. Michael Hooten MD
DOI: http://dx.doi.org/10.1111/j.1526-4637.2012.01434.x 1034-1039 First published online: 1 August 2012

Abstract

Objective. The primary aim of this study was to determine the effects of spouse or significant other solicitous responses on morphine equivalent dose among adults with chronic pain.

Design. Retrospective design.

Setting. Multidisciplinary pain rehabilitation center.

Patients. The cohort included 466 consecutively admitted patients who had a spouse or significant other and were using daily opioids.

Intervention. Three-week outpatient pain rehabilitation program.

Outcome Measures. Solicitous subscale of the Multidimensional Pain Inventory and morphine equivalent dose upon admission.

Results. The mean solicitous subscale score and morphine equivalent dose were 49.8 (standard deviation [SD] = 8.7) and 118 mg/day (SD = 149), respectively. Univariate linear regression analysis showed that greater subscale scores were associated with greater doses of opioids (P = 0.007). In a multivariate model adjusted for age, sex, ethnicity, years of education, employment status, pain duration, depression, and pain severity, the association retained significance (P = 0.007).

Conclusions. These findings suggest solicitous responses from a spouse or significant other may have an important influence on opioid dose among adults with chronic pain.

  • Opioids
  • Chronic Pain
  • Solicitous Response
  • Spouse or Significant Other

Introduction

Opioids are a widely recognized treatment for chronic non-cancer pain [1,2], and prescriptions for opioids have increased dramatically in the United States over the past 15 years [3–5]. Various demographic and clinical factors have been associated with the use of opioids for chronic pain including male sex, younger age, longer pain duration, greater pain intensity, and depression [6–11]. However, the clinical factors associated with the use of greater dosages of opioids have not been widely investigated. In a retrospective study, the use of ≥180 mg/day morphine equivalents was associated with having four or more pain diagnoses, and greater rates of comorbid medical, psychiatric, and substance use disorders compared with a cohort of adults using <180 mg/day morphine equivalents [12]. Further knowledge about the clinical factors related to the use of greater opioid dosages is important in that the use of ≥100 mg/day morphine equivalents has been associated with an increased risk of death due to unintentional overdose [13,14].

In addition to demographic and clinical factors previously associated with use of greater dosages of opioids, operant models of chronic pain [15] suggest that social interactions, particularly with spouses and significant others (S/SO), may positively reinforce various behaviors that could potentially influence opioid dose. Solicitous responses are verbal and behavioral social interactions that indicate concern or anxiousness about an individual's well-being, including the individual's experience of pain. Examples include frequently asking about pain-related symptoms, or excessive attentiveness to the individual's physical and emotional needs. Previous studies suggest that solicitous responses from S/SO influence several key features of chronic pain including pain intensity, physical disability, and pain behaviors [16–18]. However, the potential influence of S/SO solicitous responses on another key feature of chronic pain, specifically opioid dose, has not been previously reported. Knowledge of the association between S/SO solicitous responses and opioid dose could further highlight the important role of family members in the clinical course of adults receiving long-term opioid therapy for chronic pain.

The primary aim of this retrospective study was to determine the effects of S/SO solicitous responses on morphine equivalent dose among chronic pain patients enrolled in a 3-week pain rehabilitation program. Upon admission, the self-report baseline level of S/SO solicitous responses was assessed by patients using the solicitous subscale of the Multidimensional Pain Inventory (MPI).

Methods

Study Participants

All adult patients consecutively admitted to the Mayo Comprehensive Pain Rehabilitation Center from September 2003 to February 2007 with chronic nonmalignant pain were eligible for inclusion. During this period, 1,241 patients with chronic pain were admitted of which 466 (37.6%) had a spouse or significant other, and were using opioids on a daily basis. All patients provided consent for the use of their medical records for research purposes. The study protocol was approved by the Mayo Foundation Institutional Review Board. The current report represents a secondary analysis of an extensive dataset, portions of which have been previously published [19–23].

Study Setting

The Mayo Clinic pain rehabilitation program is an outpatient intervention of 3-week duration. The clinical setting has been previously described [24]. To briefly summarize, the program is based on a cognitive behavioral model where the primary treatment goal is functional restoration. Patients were involved in daily physical and occupational therapy, and all patients attended daily educational group sessions related to management of depressive symptoms, relaxation and stress management training, the use of pain medications for chronic pain, sleep hygiene, moderation of daily activities, and the elimination of pain behaviors.

Determination of Morphine Equivalent Dose

Upon admission, all medication use was determined by self-report and a detailed medication history interview, as previously described [19]. Specifically, a detailed pharmacist interview with the patient and examination of all medication containers was completed at admission to determine the specific dosage and interval of use for each medication. In patients taking daily opioids, the opioid dose was converted to a daily oral morphine equivalent dose using an equianalgesic conversion software program [25] that has been used extensively at our treatment center [9,19,21–23].

Measures

Baseline demographic and clinical characteristics were collected at admission including age, sex, duration of pain, marital status, years of education, employment status, primary pain site, and medication use.

The MPI has been widely used to assess the psychosocial impact of chronic pain [26]. This self-report questionnaire has proven reliability and construct validity [27]. The 52-item self-report questionnaire contains 12 subscales, and raw scores were converted to standardized T-scores with a mean of 50 and a standard deviation (SD) of 10 (range 0–100) [28]. The solicitous subscale comprised six questions (e.g., “How supportive or helpful is your spouse or significant other to you in relation to your pain?”) and provides a measure of solicitous responses from S/SO where higher scores signify greater levels of solicitous responses. The pain severity subscale quantifies pain intensity and pain-related suffering based on responses to the following three questions: 1) “Rate the level of your pain at the present moment”; 2) “On average, how severe has your pain been during the last week”; and 3) “How much suffering do you experience because of your pain?” The higher scores on the pain severity subscale indicate greater pain and suffering as assessed by the patient.

The Center for Epidemiologic Studies-Depression Scale is a measure of the severity of depressive symptoms [29]. The 20-item self-administered questionnaire has established reliability and validity [30,31]. Total scores range from 0 to 60 with higher scores indicating higher levels of depression.

Data Analyses

Demographics (age, sex, ethnicity, educational status, and employment status) and clinical characteristics (pain duration, primary pain site, depression, pain severity, solicitous response, and morphine oral equivalent dose) were summarized for the study cohort. Mean and SD were reported for continuous variables, and count and proportion were reported for categorical variables. Univariate and multivariate linear regression analyses were performed with morphine equivalent dose as the dependent variable. Univariate regressions were first performed with morphine equivalent dose as the dependent variable, and the independent variables included age, sex, ethnicity, years of education, employment status, pain duration, depression, and pain severity. These variables have been previously shown to influence opioid dose among adults with chronic pain [6–11,32]. Multivariate linear regression was then performed with morphine equivalent dose as the dependent variable, and adjusted for all the above listed factors. The level of significance for all statistical tests were set at P < 0.05, and all analyses were completed using PASW (Version 18.0, IBM, Inc., Chicago, IL, USA).

Results

Sample Characteristics

The study cohort included 466 adult patients with chronic pain who had an S/SO and were using opioids daily. Table 1 contains a summary of baseline demographic and clinical characteristics. The mean solicitous subscale score and daily morphine equivalent dose for the study cohort were 49.8 (SD = 8.7) and 118 mg/day (SD = 149), respectively.

View this table:
Table 1

Demographic and clinical characteristics

VariablePatients (N = 466)
Age, mean ± standard deviation (SD)48.2 ± 13.2
Female sex, No. (%)325 (70)
Pain duration (years), mean ± SD9.9 ± 10.4
Years of education, mean ± SD14.8 ± 2.8
Currently working, No. (%)101 (22)
Ethnicity, No. (%)
  Caucasian454 (97)
  African American4 (1)
  Hispanic6 (1)
  Other2 (<1)
Primary pain site, No. (%)
  Low back pain152 (33)
  Fibromyalgia75 (16)
  Headache32 (7)
  Neck28 (6)
  Generalized50 (11)
  Abdominal37 (8)
  Pelvic21 (4)
  Upper extremity27 (6)
  Lower extremity32 (7)
  Face9 (2)
  Chestwall3 (<1)
Depression,* mean ± SD27.7 ± 12.2
Pain severity,* mean ± SD49.9 ± 8.9
  • * Center for Epidemiologic Studies-Depression Scale; Multidimensional Pain Inventory pain severity subscale.

Associations Between Solicitous Response and Opioid Use

Univariate regression analysis demonstrated that greater solicitous subscale scores were associated with greater morphine equivalent dose (P = 0.007), where every one-point increase in the solicitous subscale score was associated with a 2.0 mg/day increase in opioid dose (Table 2). In addition, younger age (P = 0.018) and male sex (P < 0.001) were significantly associated with greater morphine equivalent dose. In a multivariate model adjusted for age, sex, ethnicity, years of education, employment status, pain duration, depression, and pain severity, the association between the solicitous subscale score and the opioid dose retained significance (P = 0.007) where every one-point increase was associated with a 2.1 mg/day increase in morphine equivalent dose. In this multivariate analysis, younger age (P = 0.006) and male sex (P = 0.026) were significantly associated with greater morphine equivalent dose.

View this table:
Table 2

Linear regression analyses with baseline morphine equivalent dose as the outcome variable

VariablesB Coefficient in Univariate Analysis (95% CI)P ValueB Coefficient in Multivariate Analysis (95% CI)P Value
Solicitous responses2.02 (0.52, 3.54)0.0072.12 (0.59, 3.66)0.007
Age−1.02 (−1.86, −0.17)0.018−1.49 (−2.44, −0.42)0.006
Male sex47.98 (21.95, 74.01)<0.00133.28 (4.06, 62,49)0.026
Ethnicity−15.79 (−31.69, 0.11)0.052−15.06 (−38.36, 8.23)0.205
Years of education−1.01 (−5.11, 3.09)0.6282.05 (−2.85, 6.95)0.412
Currently employed9.99 (−19.00, 38.99)0.49918.73 (−15.18, 52.64)0.278
Pain duration0.08 (−1.08., 1.24)0.8940.64 (−0.69, 1.96)0.344
CES-D score0.58 (−0.43, 1.58)0.2590.36 (−1.42, 1.80)0.543
MPI pain severity0.89 (−0.49, 2.28)0.2060.19 (−1.42, 1.80)0.817
  • CES = Center for Epidemiology Studies-Depression Scale; MPI = Multidimensional Pain Inventory pain severity subscale; CI = confidence interval.

Discussion

The primary finding of this study was that higher scores on the solicitous subscale of the MPI were associated with greater daily morphine equivalent dose among a community sample of adults with chronic pain. In both univariate and multivariate linear regression analyses, every one-point increase in the solicitous subscale score was associated with a 2.0 mg/day and 2.1 mg/day increase in daily morphine equivalent dose, respectively. The multivariate analysis further showed that the association between the solicitous response subscale and morphine equivalent dose retained statistical significance after adjusting for other demographic and clinical characteristics including age, sex, ethnicity, years of education, employment status, pain duration, depression, and pain severity. In this multivariate model, younger age and male sex were also significantly associated with greater morphine equivalent dose, which is consistent with previous observations [6].

The potential clinical mechanisms by which higher self-report scores on the MPI solicitous subscale contribute to the use of greater quantities of prescribed opioids warrant further consideration. As previously noted, operant theories of chronic pain suggest that social interactions positively reinforce behaviors and, in turn, may influence several key clinical aspects of chronic pain [15]. More specifically, S/SO solicitous responses have been associated with greater measures of pain, but the association may be influenced by sex and marital status [18,33–35]. The associations between S/SO solicitous responses and pain may be partly influenced by marital satisfaction in that the association between solicitous responses and pain were stronger among couples who were martially satisfied [36]. Additionally, S/SO solicitous responses have been associated with greater measures of physical disability, pain interference, and more frequent use of pain behaviors [16–18,37–39]. Similar to the associations between solicitous responses and greater levels of pain, the association between S/SO solicitous responses and measures of functionality were stronger among couples who reported greater levels of marital satisfaction in some, but not all, studies [17,18]. Regarding the influence of marital satisfaction within the framework of an operant model, previous investigators have suggested that S/SO solicitous responses may be “more reinforcing” to the individual with chronic pain when the “quality of the relationship is good”[40]. Collectively, these observations suggest that S/SO solicitous responses, particularly among martially satisfied couples, positively reinforce several key characteristics of chronic pain that are routinely assessed in clinical practice. Consequently, greater pain intensity, poorer levels of physical dysfunction, and the more frequent occurrence of pain behaviors in this particular group of patients may secondarily predispose clinicians to prescribe greater quantities of opioids to manage pain symptoms.

The observations from this study have clinical implications with regard to the cognitive-behavioral treatment model of chronic pain. Previous research suggests that involvement of the spouse in cognitive-behavioral-based coping skills training can have favorable effects on pain and functionality in randomized trials [41–43]. In general, spouse-assisted coping skills training refers to the systematic involvement of the spouse as the patient receives training in attention-diversion skills (relaxation, imagery, and distraction), cognitive training strategies (cognitive restructuring), and paced-activity training (activity-rest cycling and pleasant activity scheduling) [43–45]. In addition, the couple is encouraged to develop a “menu of couples skills” including mutual goal setting, behavioral rehearsal, and home-based practice [43]. The findings from these studies, and the observations from our current study, support the supposition that incorporation of the spouse in the cognitive-behavioral treatment plan could potentially mitigate the deleterious effects that solicitous responses have on clinically assessed pain-related functioning, and secondarily mollify the association between S/SO solicitous responses and opioid dose.

This study has limitations. The demographics of the study cohort could limit the generalization of the results. For example, the mean age of the cohort was 48 years, and majority of study participants were Caucasian women with an average educational level 2 years beyond high school. However, the demographic characteristics of our study population were similar to the characteristics of a community-based sample (N = 3,575) of adults with chronic pain derived from the surrounding area of our institution [46]. In this study, the mean age of study participants was 54 years, the majority of individuals were self-identified as “white” and “female,” and 77% had greater than 12 years of education [46]. Despite these demographic similarities, the clinical characteristics of patients receiving long-term opioid therapy and referred for multidisciplinary pain rehabilitation at a tertiary care medical center could be different from the clinical characteristics of a random population-based sample of adults with chronic pain. Thus, the risk of referral bias cannot be excluded in that the association between S/SO solicitous responses and daily morphine equivalent dose among our patients may not be applicable in other populations of adults with chronic pain. Previous studies from our pain treatment center have found a significant association among pain severity, depression, and opioid dose [20,23,47]. The absence of these associations herein could have been due, in part, to selection bias in that the study cohort was limited to patients receiving daily opioid therapy and who were self-identified as having an S/SO. Given the important associations between marital satisfaction and S/SO solicitous responses, the inclusion of a measure of marital satisfaction in our statistical models could have influenced the study findings. Additionally, the correlational study design limits the ability to infer a direct causal link between S/SO solicitous responses and opioid dose despite the inclusion of multiple variables in the data analyses.

In conclusion, the findings from this study suggest that higher patient self-report scores on the solicitous subscale of the MPI were associated with greater daily morphine equivalent dose. This observation is consistent with operant theories of pain and highlights the important influence that S/SO solicitous responses can have on the temporal course of chronic pain. Future research that incorporates S/SO involvement in cognitive-behavioral treatment protocols could further elucidate the clinical mechanisms by which S/SO solicitous responses influence opioid dose.

Footnotes

  • Conflict of interest/disclosure: All authors had full access to the data, analysis, and manuscript. We declare that all authors listed on the manuscript have no financial disclosures to declare.

  • The research was conducted at the Mayo Pain Rehabilitation Center, Department of Psychiatry and Psychology, and the Translational Research Unit for Chronic and Acute Pain, Department of Anesthesiology, Mayo Clinic, Rochester, MN.

References

View Abstract