Pain coping and social support as predictors of long-term functional disability and pain in early rheumatoid arthritis
Introduction
Rheumatoid arthritis (RA), a chronic inflammatory disease affecting the joints, is one of the most costly musculoskeletal disorders, primarily due to the impairment of daily activities and loss of work as a consequence of functional disability and pain (Allaire, Prashker and Meenan, 1994, Yelin and Callahan, 1995). In the first year of the disease, impaired functioning and elevated pain levels are already comparable to that of patients with longstanding RA (Evers, Kraaimaat, Geenen and Bijlsma, 1997, Evers, Kraaimaat, Geenen and Bijlsma, 1998, Meenan, Kazis, Anthony and Wallin, 1991), about 40% of the patients face occupational disability, and 75% suffer from limitations affecting their leisure time and social activities (Albers, Kuper, van Riel, Prevoo, van’t Hof, van Gestel and Severens, 1999, Fex, Larsson, Nived and Eberhardt, 1998, van Jaarsveld, Jacobs, Schrijvers, van Albada-Kuipers, Hofman and Bijlsma, 1998). In addition, functional impairment in the first year of RA predicts future loss of employment, a worse long-term prognosis and mortality (Corbett, Dalton, Young, Silman and Shipley, 1993, Eberhardt, Larsson and Nived, 1993, Fex, Larsson, Nived and Eberhardt, 1998, Rasker and Cosh, 1989, Sherrer, Bloch, Mitchell, Young and Fries, 1986). Consequently, medical treatment is increasingly geared toward more aggressive treatment in the initial stage of the disease to decrease unfavorable long-term disease outcomes (van Jaarsveld, Jacobs, van der Veen, Blaauw, Kruize, Hofman, Brus, van Albada-Kuipers, Heurkens, ter Borg, Haanen, van Booma-Frankfort, Schenk and Bijlsma, 2000, van de Putte, van Gestel and van Riel, 1998).
Irrespective of the influence of biomedical factors, chronic pain research has supplied relatively convincing evidence that psychosocial factors can affect the course of functional disability and pain in patients with RA and other chronic pain disorders (see for reviews e.g., Keefe, Smith, Buffington, Gibson, Studts and Caldwell, 2002, Linton, 2000, Turk and Okofuji, 2002). Specifically, fear-avoidance models have received wide attention and provided supportive evidence that pain-related avoidance factors, such as avoiding activity and catastrophic pain cognitions, are linked to future pain outcomes in chronic pain patients (e.g., Lethem, Slade, Troup and Bentley, 1983, Linton, 1985, Philips, 1987, Vlaeyen, Kole-Snijders, Boeren and van Eek, 1995, Vlaeyen and Linton, 2000). Fear-avoidance and the associated avoidance of activity are supposed to result in enhanced physical impairment, e.g. due to physical deconditioning processes, decreased muscular endurance and strength. Cognitive responses consisting of fearful, catastrophizing beliefs about pain are thought to bring about a preoccupation with bodily symptoms and avoidance of activity, which in turn exacerbate pain and functional disability. Besides cognitive-behavioral factors, social resources are assumed to have an impact on long-term chronic pain outcomes, including social networks and perceived support from close others, which may inhibit avoidance of physical and social activities, and have a beneficial impact on functional disability and pain (e.g., Cohen and Wills, 1985, Keefe, Smith, Buffington, Gibson, Studts and Caldwell, 2002, Uchino, Cacioppo and Kiecolt-Glaser, 1996).
In addition to experimental and cross-sectional research, prospective studies have demonstrated the relevance of pain-related avoidance factors and social resources in various chronic pain populations, including RA patients (see e.g., Jensen, Turner, Romano and Karoly, 1991, Keefe, Smith, Buffington, Gibson, Studts and Caldwell, 2002, Linton, 2000, Turk and Okofuji, 2002). For example, in one of the first prospective studies conducted by Brown and Nicassio (1987), passive coping with pain, including the restriction of activities and catastrophic pain cognitions, predicted functional disability and pain in RA patients after 6 months (Brown & Nicassio, 1987). Behavioral and cognitive factors also independently predicted future RA outcomes: avoidance of activity was related to increased functional disability after one year (Evers, Kraaimaat, Geenen and Bijlsma, 1998, van Lankveld, Naring, van’t Pad Bosch and van de Putte, 1998), while worrying and catastrophic pain cognitions predicted functional disability and pain after six months (Keefe, Brown, Wallston, & Caldwell, 1989) and functional disability after one year (Evers, Kraaimaat, Geenen & Bijlsma, 1998). In contrast, active coping with pain by ignoring pain sensations, using distraction or continuing activities in spite of pain have only incidentally been linked to more favorable future outcomes (e.g., Brown & Nicassio, 1987), suggesting that not using passive coping strategies may be more crucial than the use of specific active strategies (see e.g., Jensen, Turner, Romano and Karoly, 1991, Turk and Rudy, 1992). Apart from pain coping, there is also increasing evidence that social support, such as qualitative aspects of perceived social support and quantitative aspects of the size of social networks, affect future functional limitations and pain in chronic pain patients. For example, lower levels of perceived support have been shown to be prospectively related to more interference in daily activities in RA patients after one year (Smith & Wallston, 1992) and increased pain after one year (Waltz, Kriegel, & van’t Pad Bosch, 1998), while less extended social networks predicted functional disability after one year (Evers, Kraaimaat, Geenen & Bijlsma, 1998).
In view of the empirical evidence on RA and comparable findings in other chronic pain disorders (see e.g., Jensen, Turner, Romano and Karoly, 1991, Keefe, Smith, Buffington, Gibson, Studts and Caldwell, 2002, Linton, 2000), there seems to be relatively clear support that pain coping and social resources affect future outcomes in chronic pain patients. However, prospective studies among RA and other chronic pain patients have generally assessed the impact of these factors on future outcomes over a relatively short period of time (no longer than one year), which can be considered rather short-term outcomes in the realm of chronic conditions. Fear-avoidance models usually propose that once avoidance mechanisms have been established they result in increasing detrimental effects over time and affect long-term outcomes (Lethem, Slade, Troup and Bentley, 1983, Philips, 1987, Vlaeyen, Kole-Snijders, Boeren and van Eek, 1995). However, relevant variables could differ for short-term and long-term outcomes. For example, solely the combination of cognitive-behavioral and social factors — and not single factors — could have an impact on long-term outcomes, e.g. for patients with more passive pain coping and less social support. In addition, personality characteristics and biomedical factors may have long-term modifying effects. For example, the personality characteristics of neuroticism and extraversion and patients’ clinical status have been shown to possibly modify the relationship of pain coping and social resources to outcomes in chronic pain patients, suggesting that the detrimental effects of passive pain coping or fewer social resources might only occur in patients with more neuroticism, less extraversion and a worse clinical status (e.g., Brown, Wallston and Nicassio, 1989, Phillips and Gatchel, 2000, Vlaeyen, Seelen, Peters, de Jong, Aretz, Beisiegel and Weber, 1999, Wade, Dougherty, Hart, Rafii and Price, 1992). Finally, pain coping and social resources have been shown to affect first year outcomes (Evers, Kraaimaat, Geenen, & Bijlsma, 1998), and correspondence with findings for longstanding RA suggests that the same mechanisms are involved both early in the disease and later on. Since pain coping and social resources may be a focus of early intervention, it is particularly relevant to show whether these factors affect long-term functional disability and pain in RA patients at the earliest point in time for intervention—at diagnosis.
The object of the present study was to examine the long-term effects of pain coping and social support on functional disability and pain in patients with early RA. This study was conducted as a follow-up to the previously reported effects of pain coping and social support (Evers, Kraaimaat, Geenen & Bijlsma, 1998), showing that passive pain coping strategies, and to a lesser degree lower levels of social support, assessed at the time of diagnosis, predicted the course of functional disability in the first year. Additional analyses revealed that these factors did not predict the one-year pain outcome in this sample. Our current study aims at follow-up results, studying the effects of active and passive pain coping and social support at the time of diagnosis on the course of functional disability and pain after three and five years. In line with the literature supporting short-term effects, less active and more passive pain coping and lower levels of social support were expected to predict a less favorable long-term course of functional disability and pain. In addition to these main effects, it was examined whether the personality characteristics of neuroticism and extraversion account for the relationship between pain coping, social support and the long-term outcome. The possible moderator effects of personality characteristics and clinical status on pain coping and social support were also explored, assuming that the effects of less active and more passive pain coping and lower levels of social support on functional disability and pain would be greater in patients with personality characteristics of more neuroticism and less extraversion and in patients with a worse clinical status at the time of diagnosis. Finally, the moderator effects of social support on pain coping were exploratively examined, predicting that the detrimental effects of less active and more passive pain coping are increased in patients with lower levels of social support.
Section snippets
Sample and procedure
The sample consisted of outpatients with recently diagnosed RA from five hospitals in the Netherlands. All patients participated in one of two medical trials for second-line antirheumatic drugs (van Everdingen, Jacobs, van Reesema and Bijlsma, 2002, van Jaarsveld, Jacobs, van der Veen, Blaauw, Kruize, Hofman, Brus, van Albada-Kuipers, Heurkens, ter Borg, Haanen, van Booma-Frankfort, Schenk and Bijlsma, 2000). Inclusion criteria for the trials were a minimum age of 18 years, diagnosis according
Change in clinical status during the study period
During the five-year period, there was a significant mean decrease in disease activity (F(3,73)=22.2, p<0.001 and F(3,73)=10.6, p<0.01 for the ESR and joint score, respectively). In addition, pain and one indicator of functional disability significantly decreased within five years after diagnosis (F(3,75)=9.6, p<0.01 for pain; F(3,75)=14.5, p<0.001 for grip strength). Post-hoc tests indicated that this improvement in clinical status was most obvious in the first year of the disease: all
Discussion
The role of pain coping and social support was examined in relationship to the long-term outcome of functional disability and pain in early RA. In line with previous results at the one-year follow-up (Evers, Kraaimaat, Geenen & Bijlsma, 1998), passive pain coping assessed at the time of diagnosis still had a detrimental effect on functional disability at the three-year follow-up, but not at the five-year follow-up. In addition, social support consistently predicted a less unfavorable course of
Acknowledgements
Preparation of this article was supported in part by grants from the Dutch Arthritis Association (”Nationaal Reumafonds”). We thank G.A. van Albada-Kuiper, A.H. Bakker, I. van Booma-Frankfort, E.J. ter Borg, R. Brons, A.A. van Everdingen, H.C.M. Haanen, A.H.M. Heurkens, D.M. Hofman, R. Huisman, C.H.M. van Jaarsveld, A.W.J.M. Jacobs-van Bree, A.A. Kruize, H. van Mourik, I. Nuver-Zwart, Y. Schenk, D.R. Siewertsz-van Reesema, M.J. van der Veen, and S. van Wijk for collecting study data.
References (59)
- et al.
Development of a questionnaire for the assessment of active and passive coping strategies in chronic pain patients
Pain
(1987) - et al.
Psychosocial predictors of functional change in recently diagnosed rheumatoid arthritis patients
Behaviour Research and Therapy
(1998) - et al.
Cognitive, behavioral and physiological reactivity to pain as a predictor of long-term pain in rheumatoid arthritis patients
Pain
(2001) - et al.
Tailored cognitive-behavioral therapy in early rheumatoid arthritis for patients at risk: A randomized, controlled trial
Pain
(2002) - et al.
The psychobiology of chronic pain
Advances in Behaviour Research and Therapy
(1990) - et al.
The role of spouse reinforcement, perceived pain, and activity levels of chronic pain patients
Journal of Psychosomatic Research
(1987) - et al.
The influence of family support on chronic pain
Behaviour Research and Therapy
(1990) - et al.
Coping with chronic pain: a critical review of the literature
Pain
(1991) - et al.
Coping with rheumatoid arthritis pain: Catastrophizing as a maladaptive strategy
Pain
(1989) - et al.
Outline of a fear-avoidance model of exaggerated pain perception. I
Behaviour Research and Therapy
(1983)