Pain and Disability in Families of Chronic Noncancer Pain Patients CARLA RODGERS, M.D., LESLIE LEVIN, M.D. MARY ROBESON, M.S., ELISABETH
SHAKIN KUNKEL, M.D.
The authors conducted a cross-sectional study to determine the prevalence of chronic noncancer pain or disability in first-degree relatives ofpatients who had sustained a physically traumatic injury with resultant chronic pain. A control group ofgeneral internal medicine patients was also interviewed about existence in their family of chronic pain or of disability, either pain-related or related to nonpainjul chronic illness. None of the control patients were themselves experiencing acute or chronic pain. There was no statistically significant difference between the two groups in the prevalence of chronic pain or disability in family members. The study sfindings suggest that chronic noncancer pain patients should not be viewed as a uniform group with regard to learned behaviors and/or familial aggregation. (Psychosomatics 1996; 37:476-480)
n general, certain pain behaviors may be more likely to recur or persist because they are reinforced, whereas other pain behaviors may occur less frequently or less strongly in the absence of reinforcing events. I A number of authors have postulated that pain patients develop or sustain their symptoms in response to the behavior of family members. 2-4 Furthermore, Pilowski and Spence have found attitudes toward illness linked to ethnic background, suggesting a cultural component to illness behaviors. s Several studies have reported increased levels of pain and distress in family members of . chronic noncancer pain patients. The studies include retrospective analyses of familial incidence of pain 4.6 and a determination of the effect of chronic pain on the psychological well-being of spouses, with a noted difference of increased depression among wives of pain patients over husbands. 7 Children also seem to experience more pain and missed school days if relatives in their families have painful iIIness. 8.9 These studies included heterogeneous groups of patients 476
with regard to etiology of their pain complaints. The present study was conceived when the psychiatric consultant to a university-based pain center noted that chronic noncancer pain patients she evaluated reported family histories generally negative for chronic pain or disability, which was the opposite of traditional expectations. Since most of our pain patients were receiving disability as well, either worker's compensation or social security disability, we decided also to ask about family members on disability. We were interested to see if family patterns of behavior might influence applying for disability,
Received September I. 1994; revised November 21. 1994; accepted March 15. 1995. From the Department of Psychiatry and Human Behavior. the Center for Research in Medical Education and Health Care, and the Department of Psychiatry and Medicine. Jefferson Medical College; and St. Christopher's Hospital. Philadelphia. Pennsylvania. Address reprint requests to Dr. Rodgers. Jefferson Medical College. Department of Psychiatry and Human Behavior, 1020 Sansom Street, Suite 1652, Thompson Building, Philadelphia. PA 19107-5004.
Rodgers et al.
if not the actual report of pain. It is unclear in the studies that have looked at disability whether it is a result or cause of chronic pain. lo The possibility, of course, exists that those who are in greater pain are those who apply for disability.lo One cannot discuss the existence of chronic pain without mentioning the possibility of depression. In their comprehensive review of the literature, Romano and Turner ll have stated that studies of pain patients with depression vary from a prevalence of 10% to 100%. Conversely, patients with a primary diagnosis of depression voice pain complaints from 30% to 100% of the time, depending on the investigation. Although the current study does not focus on depression, we will make note of the percentage of patients in our chronic noncancer pain group who met DSM-III-R criteria for major depression or adjustment disorder with depressed mood. The work we did differs somewhat from previous investigations. It is cross-sectional in design, includes a control group, and contains a fairly homogeneous group of patients with regard to etiology of the pain, specifically physical trauma. As Flor et al. stated in their extensive review on pain and families, studies of pain patients that include control groups are rare. 6 One previous study has looked at familial models of pain complaints with a control group. These workers found that there were considerably more pain complaints in the families of pain patients,12 but their cohort also included patients of diverse etiologies. METHODS The psychiatric consultant for a universitybased pain center evaluated 52 consecutive patients at the pain center over a 2-year period (1991-1993) for noninvasive pain management. In each of these cases, the referring pain center anesthesiologist had determined that the patient was not suitable for nerve blocks or steroid injections, or the anesthesiologist had attempted a series of blocks or steroid injections without successful alleviation of pain. Of the 52 patients, 39 had experienced physical trauma at VOLUME 37 • NUMBER 5 • SEPTEMBER - ocrOBER 1996
their workplace, in motor vehicle accidents, as a result of surgery, or falls. The duration of the pain in these patients, by their own report, was between 1.5 and 5 years. Table I summarizes how the patients were injured. Table 2 correlates the nature of the patients' complaints with the diagnosis at the time of referral to the psychiatric consultant. Of particular interest are the four surgical cases. The two persons with postoperative thoracic outlet syndrome had had shoulder surgery. The two with postoperative neuromas had had foot surgery. In all four cases, these operations were initially considered routine. It was not clear from a review of the records whether any untoward operative event had occurred. To determine if the family histories of the pain patients were different from the patients without pain, we compared them with a group of 30 general internal medicine patients, I of whom we excluded from the study when it was discovered that the person was the offspring of a previously interviewed control subject. The control subjects were all patients in the private practice of a general internist at the same university medical center where the pain center was located. The control subjects were chosen randomly and consecutively from persons who consented to participate. The internist allowed the investigators to interview the control patients in the practice conference room after their visits were completed and any bloodwork was done. None of these persons had a current or previous complaint of chronic noncancer pain. The authors obtained information on the age, race, and gender of both the pain patients and control subjects. The resulting control group of 29 patients was comparable in age with the pain group, with a mean age of 44 years 9 months TABLE I.
Cause or the traumatic injury per patient report
Motor vehicle accident
Surgical sequelae Falls
Chronic Noncancer Pain Patients
Pain patient's chief complaint and physical diagnosis
Job-related injuries: Patient Complaint 15 with low back pain 2 with shoulder pain 5 with upper extremity pain
6 fibromyalgia 9 neuropathic pain 2 impingement syndrome
2 with neck pain
3 with lower extremity pain
2 with lower extremity pain
2 with upper extremity pain
2 with lower extremity pain I with upper extremity pain
Race African Caucasian American Hispanic Asian Pain patients
The control subjects were interviewed by either the pain center psychiatrist or one trained designee, a fourth-year medical student. Both interviewers used the same questions and recorded the patients' answers at the time of the interview. RESULTS
Physician Ox 2RSD ITOS
and 41 years 8 months, respectively. The pain patients and control subjects were dissimilar in distribution by sex, with three times as many men being present in the pain group. Additional demographic information is contained in Table 3. Both the control group and pain patients were asked the following questions and given the following definitions: I) Chronic noncancer pain is defined as lasting more than 6 months on an almost daily basis. It is caused by medical problems other than cancer. First-degree relatives include parents, siblings, and children. Do any of your first-degree relatives have a history of chronic pain which is not caused by cancer? 2) If the answer is yes, then which family member is it? 3) Do you know what kind of pain it is? 4) Do you know what causes the pain? 5) How long has it lasted? 6) Does that family member or any other first-degree relative receive disability payments? 478
Pain patients Control subjects
Physician Ox 3RSD 2TOS
Falls: Patient Complaint
Gender Male Female
4 thoracic outlet syndrome (TOS)
5 with upper extremity pain
Surgical sequelae: Patient Complaint
Distribution of pain patients and control subjects by gender and race
I reflex sympathetic dystrophy (RSD)
Motor vehicle accidents: Patient Complaint
Of the 39 pain patients, 4 reported that a firstdegree relative experienced some form of chronic noncancer pain; and 4 reported that a first-degree relative was on disability. None reported that a first-degree relative with chronic pain was receiving disability payments. Of the 29 general medicine controls, 7 reported firstdegree relatives with chronic noncancer pain, and 2 reported that they had a relative who was receiving disability payments and did not have chronic pain. One control subject reported a relative receiving disability who was experiencing chronic pain. Table 4 compares the results of our interviews with both the pain patients and control subjects. By using the results of the total number of pain patients and control subjects, we determined the statistical significance. A chi-square test was done comparing the case and control subjects on the presence or absence of pain and/or disability. The contingency table of patients, pain vs. control, by presence of pain and/or disability in relati ves, yes or no, is shown in Table 5. The results were not significant
Rodgers et al.
Comparison between first-degree relatives of pain patients and control subjects Disability or Pain
Total Pain patients Control subjects Pain patients (female only) Control subjects (female only) Pain patients (male only)
Control subjects (male only)
Disability and Pain
Note: This summary compares the number of relatives experiencing pain and/or disability. It also compares families by gender distribution.
ses on results by race and gender distribution because the numbers were too small, especially with regard to race. An additional inference can be made from this data, however. Both the female pain patients and the female control subjects reported more first-degree relatives with chronic pain and disability than the male patients in either group. We will address this issue later in the discussion. Finally, 7 of the 39 pain patients met DSMIII-R criteria for either major depression or adjustment disorder with depressed mood, for a total of 18%. The control subjects were not questioned about symptoms of depression. DISCUSSION As stated earlier, the purpose of this study was to verify the anecdotal observations by a pain center psychiatric consultant that a certain group of pain patients did not appear to have an increased prevalence of pain or disability in TABLE S.
Chi-square comparing first-degree relatives of pain patients and control subjects on presence of either disability or pain' Positive Pain or Disability Present
Pain patients (n =39) Control subjects (n =29) aX2
Negative Pain or DisabiUty
= 1.67. df = I. NS.
VOLUME 37. NUMBER 5. SEPTEMBER - OCTOBER 1996
their families. The authors expected that firstdegree relatives of pain patients would have a higher prevalence of pain and/or disability than first-degree relatives of general medicine patient control subjects. The results of our work, however, showed no statistical difference in family pain or disability history between the two groups. One of the controversies regarding pain assessment relates to the accuracy of observing certain behaviors such as grimacing as a reliable indicator of pain. 13 Fordyce has stated that how patients verbally describe their pain does not always match their behaviors. 14 While this issue has not been resolved, it appears that most of our pain patients knew their relatives were in pain as a result of verbal report. Perhaps pain patients who are concerned with their own discomfort would be less likely to pick up on the nonverbal cues of their relatives than persons who are not in pain. Possible other explanations for our results include the relatively small sample size on which this work was based, although other authors have found a significant difference in family history between pain and control groups when using a small cohort. 12 Another possibility may relate to the gender of patients and their relatives. At least one other study noted gender differences in pain reporting attributed to the gender of the listener, that is, both men and women were more likely to report pain sensation to a woman than to a man.'5 In our study, the number of men in the pain group was three times that of the control group. As we noted in the results section of this report, more women 479
Chronic Noncancer Pain Patients
in both the pain and control groups reported positive family history for pain and disability than did the men in either group. If both men and women are more likely to report pain sensations to women, the larger percentage of the women patients in the control group might have increased the total report of pain by their firstdegree relatives. A further possibility may have to do with the fact that there has been a reported relationship between depression and chronic pain, as previously noted. Some authors have shown that primary affective illness affects the presentation of chronic pain. 16 It is unclear, though, how the existence of depression in a patient might affect his or her understanding of another family member's pain.
A final possibility that might account for our results is that our group of pain patients is more homogeneous than in other studies that looked at family members. Perhaps patients who are injured are different than patients who experience headaches or have other causes of chronic pain. Other authors have shown that certain groups of chronic noncancer pain patients, like pelvic pain sufferers, must be studied separately from other pain patients with regard to etiology and psychosocial variables.) In light of our results, this concept may have to be extended to other subsets of pain patients as well.
The authors thank Wade Berrettini. M.D., Ph.D., for his help in designing this study.
References I. Fordyce WE. Roberts AH. Sternbach RA: The behavioral management of chronic pain: a response to critics. Pain 1985; 22: 113-125 2. Turk DC. F10r H. Rudy TE: Pain and families. I. Etiology. maintenance. and psychosocial impact. Pain 1987; 30:3-27 3. Toomey TC. Hernandez JF. Gittelman DF. et al: Relationship of sexual and physical abuse to pain and psychological assessment variables in chronic pelvic pain patients. Pain 1993; 53: 105-109 4. Minuchin S. Baker L. Rosman B. et al: A conceptual model of psychosomatic illness in children. Family organization and family therapy. Arch Oen Psychiatry 1975;32:1031-1038 5. Pilowsky I. Spence ND: Ethnicity and illness behavior. Psychol Med 1977; 7:447-452 6. F10r H. Turk DC. Rudy TE: Pain and families. II. Assessment and treatment. Pain 1987; 30:29-45 7. Romano JM. Turner JA. Clancy SL: Sex differences in the relationship of pain patient dysfunction to spouse adjustment. Pain 1989; 39:289-295 8. Lavigne JV. Schulein MJ. Hahn YS: Psychological as-
pects of painful medical condition in children. II. Personality factors. family characteristics. and treatment. Pain 1986; 27: 147-169 9. Dura JR. Beck SJ: A comparison of family functioning when mothers have chronic pain. Pain 1988; 35:79-89 10. Dworkin RH: Compensation in chronic pain patients: cause or consequence (letter). Pain 1990; 43:387-388 II. Romano JM. Turner JA: Chronic pain and depression: does the evidence support a relationship? Psychol Bull 1985; 97: 18-34 12. Violon A. Giurgea D: Familial models for chronic pain. Pain 1984: 18:199-203 13. Keefe FJ. Dunsmore J: Pain behavior: concepts and controversies. American Pain Society Journal 1992; 1:92-100 14. Fordyce WE: Behavioral Methods for Chronic Pain and Illness. CV Mosby, St. Louis. MO, 1976 15. Maixner W. Humphrey C: Gender differences in pain and cardiovascular responses to forearm ischemia. Clin J Pain 1993; 2:16-25 16. Blumer D. Heilbronn M: Chronic pain as a variant of depressive illness. J Nerv Ment Dis 1982; 170:381-406