The first 2 years of rheumatoid arthritis: The influence of acceptance on pain, physical limitation and depression

The influence of acceptance in the progression of pain, physical limitation and depression was explored in the first 2 years of rheumatoid arthritis. Latent growth curve models showed significant increases in pain, physical limitation and depression. Besides that, the levels of pain and physical limitation at the baseline were associated with acceptance but not its progression across time. Therefore, patients with higher scores of acceptance reported less pain and physical limitation. The progression of depression was associated with acceptance; higher acceptance patients had slower growth rates of depression across time, even when pain and physical limitations increased. The inclusion of pain acceptance in clinical practice is discussed.


Introduction
Rheumatoid arthritis (RA) is a chronic, painful, disabling disease that affects 0.35%-0.50% of the Portuguese population, where it has 40,000 diagnosed cases (Lipsky, 2000;Monjardino et al., 2011). Data also show that RA is the main reason for temporary disability, responsible for 40%-60% of cases of long-term disability. It is also responsible for 35%-41% of early retirement in the workforce (Barlow et al., 2001;Lipsky, 2000;Monjardino et al., 2011).
The relationship between persistent pain, stiffness, joint damage and physical disability has become well known. However, it seems that those symptoms also have important social and psychological implications (Keefe et al., 2002;Newman et al., 1996). Actually, the inability to perform daily life routines has been described as a stress-induced experience (Newman et al., 1996). Some of this research has shown that the most difficult emotional consequence of RA is The first 2 years of rheumatoid arthritis: The influence of acceptance on pain, physical limitation and depression the loss of control over one's life. Therefore, RA symptoms become a nuclear concern and individuals exclude other important aspects of life (McCracken et al., 2004a). Pain, stiffness and disability are major concerns for those with RA who look for medical assistance in order to relieve their pain and be able to perform their usual activities. Generally, RA patients struggle to control pain, without success, while being overwhelmed by their distressing thoughts, taking pain and related symptoms as barriers for living a meaningful life .
Research shows high prevalence of psychiatric disorders in RA individuals (Covic et al., 2006;Dickens et al., 2002;Riemsma et al., 2000;Sharp et al., 2001). Regier et al. (1988) refer that depression is two to three times more common in patients with RA than in general population. In addition, a Canadian Community Health Survey (n = 139,880) found that individuals with RA have twice the rate of major depression and suicidal ideation when compared to individuals without RA. This research also showed that individuals who reported chronic pain, limitations in daily activities and multiple health problems were more likely to become depressed (Bèland, 2002).
Research in the surgical literature, particularly related with the joint replacement, has found that depression was a significant predictor of both pain and physical limitation (Wylde et al., 2007), but it was also associated with persistent postsurgical pain (Wylde et al., 2011).
It is important to notice that RA is characterized by fluctuating symptoms and an unpredictable course (Dickens et al., 2002). Diagnosis may remove the uncertainty of unexplained symptoms but also adds other sources of stress, such as the knowledge and understanding to make informed decisions concerning treatment options and to carry out self-care activities (Barlow et al., 1999).
Within the new developments of cognitivebehavioural therapy (CBT), it has been emphasized that efforts to control pain may become problematic when it became central and disruptive and may dominate patient's life (McCracken et al., 2004a;McCracken and Yang, 2006). The combination of the processes of acceptance and mindfulness into cognitivebehavioural techniques (Hayes, 2004;Hayes et al., 1999) has highlighted the importance of contextual methods that emphasize the context, the function and the relationship that individuals establish with their thoughts and emotions Singh et al., 2008).
A large body of research has also reinforced that these emotional regulation processes may buffer the impact of disease and provide support in times of adversity Pinto-Gouveia, 2011a, 2011b). In the particular case of acceptance, a number of studies have shown that it is related not only with a better emotional, physical, social and work-related functioning, but also with less use of health-care resources, in both cross-sectional (McCracken, 1998(McCracken, , 2007 and longitudinal studies McCracken et al., 1999;Vowles, 2007, 2008;Nicholas and Asghari, 2006;Vowles et al., 2007).
Acceptance is emerging as a potentially valuable concept on how patients react and adapt to chronic pain (McCracken, 1998;McCracken et al., 2004aMcCracken et al., , 2004bViane et al., 2003). It is defined as a construct with two components, which concern the engagement in life activities while pain is being experienced, and the recognition that attempts to control or avoid pain are ineffective strategies (Kashdan et al., 2008;Kollman et al., 2009;McCracken and Eccleston, 2003;McCracken et al., 2004b). This emotional regulation process involves one being experientially open to the present moment, promotes a change in the way symptoms are experienced and influence individuals' behaviour (Baer and Krietemeyer, 2006;Hayes et al., 1999;McCracken, 1999McCracken, , 2005Vowles et al., 2008).
Besides that, there is no research about how these patients function over the course of the disease and what factors are associated with such changes when they occur. There is a lack of published studies on the relation of acceptance in the progression of pain, physical limitation and depression in a Portuguese sample of patients with RA. Therefore, the aims of this exploratory study are as follows: (1) to examine pain, physical limitation and depression across the quasi-naturalistic course of RA disease, (2) to explore the association of acceptance in the initial levels of pain, physical limitation and depression, and (3) to explore the association between acceptance and the increase of pain, physical limitation and depressive symptomatology across the first 2 years of RA disease progression.
Our research hypotheses are as follows: (1) the pain levels, perceptions of physical limitations and depressive symptomatology will increase significantly across the progression of RA on a 2-year period; (2) the degree of pain acceptance will be significantly related with physical and psychological constructs at the baseline assessment such that individuals with higher acceptance would report less pain, less perceptions of physical impairment and less depression; furthermore, we also hypothesize that (3) the degree of pain acceptance will be significantly related with the progression of physical (i.e. pain and physical limitation) and psychological variables (i.e. depression) with participants with higher levels of acceptance showing lower rates of pain and physical limitation and depression over time.

Participants
Participants from three Portuguese health units participated in the study (N = 55). Participants were 44 women and 11 men, with age ranging from 18 to 86 years (M = 55.28, standard deviation (SD) = 17.91) and a mean of 6.58 years of education (SD = 4.10). Inclusion criteria included the following: (1) being 18 years old or older and (2) being diagnosed with RA between 3 and 6 months after the diagnosis (according to the American College of Rheumatology). Exclusion criteria included the following: (1) an identified terminal illness, (2) presence of severe psychopathology measured by psychiatric evaluation and (3) attending to psychological interventions.

Measures
All measures used in this study were translated and adapted to Portuguese by a bilingual translator. Conceptual and lexical similarities of both original and Portuguese versions were verified through reverse translation procedures.
Each participant completed a set of instruments that included several self-report questionnaires. Demographic variables were assessed with a general checklist including patient gender, age, marital status, profession, years of education and clinical diagnosis.
The Short-Form McGill Pain Questionnaire (MPQ: Melzack, 1987; translation and adaptation: Melzack, 2005) is a 15-item adjective checklist rated on a 4-point intensity scale (from 0 = None to 3 = Severe) (i.e. sensory and affective dimensions) as well as two single-item measures (i.e. visual analogue numerical scale and pain intensity). The MPQ has both a total score and partial scores, where high scores mean higher levels of pain. For the purposes of this study, only the adjective checklist was used. The internal consistency estimates for both sensory and affective dimensions were .78 and .76, respectively (Melzack, 1987).
The Arthritis Impact Measurement Scale-2 (AIMS2: Brandão et al., 1995;translation and adaptation: Costa and Pinto-Gouveia, 2005) is a 78-item self-report scale that assesses health status in a multidimensional fashion using specific scales. The scales of AIMS are scored in an inverse way, where low scores indicate better health status. The nine original scales of AIMS could be combined into three or five component models of health status. For the purpose of this study, only the physical function was used.
The Depression, Anxiety and Stress Scale (DASS; Lovibond and Lovibond, 1995;DASS-42: Pais-Ribeiro et al., 2004) is a 42-item selfreport scale with three subscales: depression, anxiety and stress. A 4-point rating scale (1 = Did not apply to me; 4 = It happened most of the time) is used in each of the 42 items. Each subscale score may range from 0 to 42, where high scores mean higher negative emotional states. In the Portuguese adaptation, DASS-42, internal consistency was high (Cronbach's alpha ranged from .83 to .93; Pais-Ribeiro et al., 2004).
The Chronic Pain Acceptance Questionnaire (CPAQ; McCracken et al., 2004b;translation and adaptation: Costa and Pinto-Gouveia, 2009) is a 20-item self-report questionnaire rated on a 7-point Likert-type scale (from 0 = Never to 6 = Always) and measures acceptance to chronic pain. The questionnaire comprises two subscales: pain willingness and activity engagement. This questionnaire has both a total score (range from 0 to 156) and partial scores (range from 0 to 54, for pain willingness; 0 to 66, for activity engagement); higher scores mean high pain acceptance. Cronbach's alpha was .82 and .78 for both pain willingness and activity engagement. In addition, both scales showed a modest correlation: r = .36 (McCracken et al., 2004). The Portuguese version of CPAQ showed a Cronbach's alpha of .89 for activity engagement and .83 for pain willingness. In addition, moderate correlations between psychopathology, self-compassion, experiential avoidance and rumination were observed (Costa and Pinto-Gouveia, 2009).

Procedure
The study was conducted with the formal approval and authorization required by the participating institutions. If participants expressed interest, the researchers asked them to sign the informed consent form and gave them the questionnaire package. The evaluation took place in a physician's office. Accordingly with ethical requirements, it was emphasized that participant's cooperation was voluntary and that their answers were confidential.
Participants were evaluated at three different times: 3-6 months, 1-year and 2-year intervals after the first doctor's diagnosis. Each evaluation lasted between 60 and 120 minutes. Time between each evaluation was chosen based on the disease progression. Evaluations were conceptualized in order to understand how individuals interpret and understand their own symptoms, what theories they develop regarding the causes and treatments, and also what they are able to do about their symptoms and the progression of the disease (Newman et al., 1996).

Data analysis
This study has a longitudinal design with selfreported measures. Kolgomorov-Smirnov tests were used to inspect the normal distribution of all variables. Although some variables showed a statistically significant deviation from the normal distribution, close inspection of the skewness and kurtosis values (all within the (−0.5, 0.5) interval) showed that this deviation was not problematic for further inferential analysis (Kline, 2005;Marôco, 2010;Tabachnick and Fidell, 2007).
The presence of multivariate outlier was assessed with the Mahanalobis Distance -DM 2 . No observations were deleted from the data set, because they could contain important information related to the studied phenomena.
Latent growth curve models (LGMs) were fitted to the longitudinal data, in order to study the progression of the pain and physical limitation perceptions and depression after accounting for individual differences in the disease status at baseline. A conditional LGM was also used to evaluate the impact of acceptance in the progression of pain, physical disability and depression.

Data reliability
Spearman correlations showed that educational background was not correlated with the scores from the variables under study. In addition, the Cronbach's alphas obtained in this study show the suitability of the measures (MPQ = .78; AIMS2 -Physical Function = .90; DASS-42 -Depression = .97; CPAQ total = .86).
Basal levels of pain differed significantly between participants around a mean level of 6.934 (SD 2 (basal levels) = 23.542; Z = 2.252; p = .024). From the baseline to the following 2 years, the pain levels increased linearly at a mean rate of 1.224/month (Z = 2.48; p = .013). This increase rate was homogenous among participants as indicated by the no significant variance around the mean growth (SD 2 (Slope) = 8.122; Z = 1.777; p = .076). These results showed that the evolution of pain did not differ significantly among the participants. The rate of pain growth showed tendencies towards negative significance with the baseline pain levels (r = −.401, p = .293).
A conditional model was fitted to data in order to evaluate the impact of acceptance on the evolution of pain symptoms (see Figure 1). Acceptance showed a significant negative impact on the baseline pain levels (β = −.859; p < .001), indicating that individuals with higher acceptance did report less pain. However, the rate of pain growth between baseline and the following 2 years was not significantly associated with acceptance (β = .015; p = .947). Therefore, acceptance was not a significant predictor of the individual differences in pain growth among individuals.
Basal levels of physical limitation differed significantly between participants around a mean level of 10.777 (SD 2 (basal levels) = 3.448; Z = 2.688; p = .007). The physical limitation levels increase linearly at a mean of .746 per month (Z = 4.155; p = .000) between baseline and after 2 years. This increase rate was not homogenous among participants as indicated by the significant variance around the mean growth (SD 2 (slope) = 1.292; Z = 3.320; p < .001). These results showed that the evolution of physical limitation differed significantly between participants. In addition, the rate of physical limitation growth was negatively correlated with the baseline physical limitation levels (r = −.612; p = .045). This negative correlation suggests that participants with higher physical limitation levels at the baseline increased their physical limitation at a lower rate than participants with lower physical limitation levels. A conditional model was fitted to the data to evaluate the impact of acceptance on physical limitations.
Acceptance showed a statistically significant negative impact on the baseline physical limitation levels (β = −.943; p < .001). This result shows that individuals with higher scores of acceptance did report less physical limitation. In contrast, the physical limitation growth between baseline and 2 years after was not significantly associated with acceptance (β = −.054; p = .806). Acceptance did not explain why physical limitation increased differently between participants.
Basal levels of depression were homogenous between participants (SD 2 (basal levels) = 15.897, Z = 1.247; p = .213). The depression levels showed a tendency to increase linearly between the baseline and after 2 years at a mean rate of 1.154/month (Z = 1.880; p = .060). This increase rate was not homogenous among participants as shown by the significant variance around the mean growth (SD 2 (Slope) = 11.031; Z = 2.533; p = .011). These results showed that the evolution of depression differed significantly between participants. The rate of depression growth showed a tendency to positive correlation with the baseline depression levels (r = .705, p = .120). The conditional model showed that acceptance had a statistically significant negative effect on the basal depression levels (β = −.398; p = .052) (see Figure 1). Thus, participants with higher levels of acceptance did report less depression levels. Furthermore, acceptance had a statistically significant negative effect associated with the rate of depression growth between the baseline and 2 years after (β = −.674; p < .001) indicating that participants with higher levels of acceptance reported a lower rate of depression growth across time.

Discussion
This study examined pain, physical limitation and depression across the quasi-naturalistic progression of disease in a sample from the Portuguese population with RA. Furthermore, this study intended to explore the potential association between acceptance and the initial levels of pain, physical limitation and depression. Finally, it also intended to explore the association between acceptance and the progression of pain, physical limitation and depression across the first 2 years of the RA disease.
In accordance with our first hypothesis, we found that pain, physical limitation perceptions and depression increase significantly over the first 2 years of the RA progression. Current data were consistent with previous research showing that individuals with RA face the prospect of a life with increasing pain and disability (Newman et al., 1996). It is also well known that in the first year of the disease, functional disability is determined and progressively deteriorates with the disease progression (Evers et al., 2003;Newman et al., 1996;Repping-Wuts et al., 2008).
LGM results showed that evolution of physical limitation perceptions differed significantly between participants. So, individuals with higher physical limitation scores at the baseline assessment were those who present a smaller progression of these symptoms over time.
Taking into account the data presented in this study and previous research findings that suggest acceptance as an important process involved in the way individuals react and adapt to chronic pain (McCracken, 1998;McCracken et al., 2004b;Viane et al., 2003;Vowles et al., 2007), conditional LGM were used to further investigate the hypothesis that acceptance is positively associated with these variables (i.e. pain, physical limitation and depression).
Acceptance showed a significant negative association with the baseline scores of pain and physical limitation, but the rate of pain and physical limitation growth from the baseline to the following 2 years was not significantly related with acceptance. This means that participants with higher scores of acceptance in the baseline reported lower scores of pain and physical limitation, compared to those with lower levels of acceptance. However, acceptance did not explain why pain and physical limitation increased differently between the participants.
LGM findings from both group and individual perspectives gathered in this study are in accordance with previous research in chronic pain with other statistical approaches, suggesting a promotion of physical functioning in individuals who adopt an open attitude and observe unwanted experiences without trying to modify, control or avoid them (McCracken, 1998;McCracken et al., 2004a;McCracken and Yang, 2006;Viane et al., 2003).
Our results also suggest that an attitude of willingness to face unwanted internal experiences at the early stages of the RA influences the level of pain and physical limitation reported at this time of RA progression. Individuals who recognize that they have pain and physical limitation, but act as if these unwanted internal experiences do not necessarily imply disability, seem to be those who live a more fulfilling life in accordance with their goals and values (McCracken, 1998;McCracken et al., 2004a;McCracken and Yang, 2006;Viane et al., 2003). It is important to notice that the association between acceptance and the baseline reports may lead to a higher effect size of acceptance in the first evaluation (3-6 months after RA) than in the following evaluation moments.
Simultaneously, we also hypothesized that depression would increase significantly over the 2 years progression of RA. Again, our findings were consistent with our hypothesis: LGM results showed different levels of growth between the baseline and 2 years later, despite no differences in baseline scores of depression were reported.
Results also showed that participants with higher scores of willingness to perform daily activities and who recognize the ineffectiveness of attempts to control or avoid unwanted internal experiences reported lower depressive symptoms at the baseline assessment. Moreover, the negative association of acceptance and the rate of depression growth showed that participants with higher levels of acceptance between the baseline and the following 2 years presented a lower rate of depression growth over time.
These associations of acceptance have never been tested in Portuguese RA patients with LGM. However, these results are consistent with previous cross-sectional studies, suggesting the importance of acceptance-based approaches in several chronic pain conditions (McCracken, 1998(McCracken, , 1999(McCracken, , 2005McCracken et al., 2004a). The higher levels of depression in RA individuals than in the general population (Covic et al., 2006;Creed, 1990;Dickens et al., 2002;Katz and Yelin, 1993;Mindham et al., 1981;Pincus et al., 1996;Regier et al., 1988;Sharp et al., 2001) has suggested important mechanisms between depression and acceptance.
One possible mechanism of the association between depression and acceptance might be the fact that acceptance promotes the contact with the present moment, allowing the individual to take better advantage of its positive processes such as the interactions with family members. According to Costa and Pinto-Gouveia (2013), acceptance seems to play a significant buffer effect on the relationship between social support 1 year after RA diagnosis and depression reported 2 years after. According to Costa and Pinto-Gouveia (2013), it is mainly in individuals with low levels of acceptance and low social support that depression levels are greater, because they are fused to the evaluations of themselves as a disabled, limited or dependent person. In addition, Kool et al. (2013) suggested that the promotion of health in RA individuals requires that social support must be considered as a main goal in clinical intervention.
Surgical literature particularly related with joint replacement has also found that depression is a significant predictor of persistent pain and physical limitation (Wylde et al., 2011). However, the co-occurrence of multiple-site persistent pain leads to the identification of a possible chronic pain syndrome, based on a generalized vulnerability to pain (Wylde et al., 2007).
Results gathered in this research have important clinical implications and are encouraging for further developments of acceptance-based approaches in the RA context. In general, individuals who actually engage in positive and functional everyday activities despite of pain and recognize that avoiding or controlling it are ineffective strategies were those who used the health-care system less, who were least distressed and disabled by their condition and were also the ones most likely to be working (McCracken et al., 2004b). The change in the relationship that individuals have with unwanted internal experiences enable them to act effectively and to promote a broader repertoire of response to the experience of pain and associated symptoms (Dahl et al., 2004;. The findings presented here are not free of some methodological limitations. First, the use of such a complex statistical procedure with a small sample has been involved in controversies, although Monte Carlo simulations have demonstrated that the LGM holds up well with relatively small samples (Muthén and Muthén, 2002). Moreover, the measures used in this study showed good psychometric properties, there were no missing data and the modelled variables were based on the sum of the items from the respective scales. Second, our data were gathered through patients' self-reports.
The indirect nature of this assessment method allows for a number of factors to contribute to the patients responses on the questionnaires. A possible solution to address the limitations of retrospective self-report and to reduce extraneous influence on the data is the use of self-monitoring, laboratory methods or methods in treatment contexts that measure patient behaviour in the setting and at the time that it occurs in future research.

Conclusion
Taking into account the results of this study, the depression rate of growth can be predicted by the patient's levels of acceptance. Therefore, patients with high acceptance present smaller growth rates of depression across time than patients with low acceptance, even when pain and physical limitations are increasing. The road to psychological health, specifically the case of RA, seems to involve consistently orienting these patients towards living a more valued life, helping them to let go of the struggle with disease and to start living here and now.

Funding
Preparation of this manuscript was supported in part by the Individual Post-Doctoral Fellowship (SFRH/ BPD/78227/2011) of the second author (Joana Costa) provided by Fundação para a Ciência e Tecnologia, a Portuguese fundation which promots the advancement of scientific and technological knowledge.