Sensory Sensitivity and Strategies for Coping With Pain

Participants

Participants were 116 healthy men and women. The majority were Australian and single, had completed high school, and were in paid employment. See Table 1 for participant demographics.

Procedure

This study (No. 2012000013) received ethical approval from the University of Queensland’s Behavioral and Social Science Ethical Review Committee. Data were collected for this study in 2012 and 2013 by two research students. Invitations to participate in the study were extended by means of email, word of mouth, and a social media private message to 162 potential participants. Participants were required to be age 18 yr or older and to have no prior history of chronic pain or current pain. Thirty-four adults either did not respond to the invitation or declined to participate for various reasons (most commonly time constraints).

The remaining 128 participants were invited to attend a testing session held at a mutually convenient place and time. Before commencing the experiment, each person was given a participant information sheet and consent form. Upon signing the consent form, participants completed several questionnaires relating to sensory preferences and affective states. They then participated in the cold pressor task (described later) in a private room. The cold pressor has been used extensively in experimental pain research, including research regarding pain sensitivity (Edwards et al., 2005). During the cold pressor task, the researcher was blind to participants’ affective states and sensory preferences.

Finally, participants completed a questionnaire related to coping strategies used during the cold pressor task. Twelve people who completed the protocol were excluded from the study because of their report of a current or past diagnosis of a chronic pain condition, resulting in a sample size of 116 (71.6%). No incentives were given to encourage participation in the study, and potential participants were informed that their participation was voluntary. Participant confidentiality was ensured through deidentification of all information.

Cold Pressor Apparatus

The cold pressor apparatus is a methodologically sound device for the temporary induction of nonharmful pain (Andrews, Meredith, & Strong, 2011; Meredith, Strong, & Feeney, 2006) and was used to provide a standard pain experience for all participants. The apparatus consisted of a container with two compartments: one containing water only, and the other filled with iced water. A pump was used to circulate water between the compartments when not in use. In this study, the water was maintained at 0–2°C, and the temperature was recorded before each use. Standard verbal instructions were provided to each person (Andrews et al., 2011) to submerge their nondominant hand and forearm into the water-only compartment for as long as possible. Although a 4-min ceiling was imposed on the cold pressor task to minimize risk of any adverse physiological effects, participants were not informed of this limit. At the conclusion of the cold pressor task, participants were debriefed. The procedure, in its entirety, took approximately 30 min per person.

Measures

Statistical Analyses

Statistical analyses were conducted using Stata Data Analysis and Statistical Software (Versions 11.2 and 13.0; StataCorp, College Station, TX). Initial data checking for normality identified several skewed continuous variables: age, stress, catastrophizing, coping self-statements, and hoping/praying. Catastrophizing, which was the only one of these variables used as an outcome variable, was transformed using square root transformation, successfully creating a normal variable. Preliminary correlation analyses were conducted to explore associations among variables and to screen for multicollinearity. As a result of the lack of normality of some variables, intercorrelations between variables were examined using Spearman’s correlation, with a significance level of p = .05.

To investigate associations between each AASP variable (low registration, sensation seeking, sensory sensitivity, and sensation avoiding) and each coping strategy (coping self-statements, catastrophizing, hoping/praying, ignoring pain, increasing behavioral activities, and diverting attention), 24 multiple regression models were constructed. In each model, one sensory processing variable acted as the independent variable; one coping strategy, as the dependent variable; and age, gender, and stress, as covariates.

Postregression diagnostics were conducted after each model was run, including tests for normality of residuals, homoskedasticity, outliers, and multicollinearity. Where the residuals were normally distributed, no transformations were necessary (Chen, Ender, Mitchell, & Wells, 2003). If postregression diagnostics revealed any violation of assumptions, the regression model was rerun with outliers removed. If this action failed to remedy the issue, a transformation of the dependent variable was conducted, and models were rerun using the transformed dependent variable. On the basis of this process, hoping/praying and coping self-statements were transformed on several occasions using square root and square functions, respectively. To control for the risk of Type I error, a Bonferroni adjustment was applied, with p < .002 adopted for regression analyses; however, because this study is exploratory in nature, ps of .05 are also reported.

Associations Between Passive Sensory Processing Styles and Coping Strategies

Associations Between Active Sensory Processing Styles and Coping Strategies

Associations Between Demographic Variables and Coping Strategies

Age was consistently and negatively associated with both catastrophizing and ignoring pain in all four sensory processing quadrants, although the association between age and ignoring pain in the sensation avoiding quadrant did not reach significance. In the quadrants of low registration, sensory sensitivity, and sensation avoiding, women were more likely to use increasing behavioral activities than men.

Active and Passive Coping

Although approaches to the classification of coping vary in the literature, for the purposes of this study, we adopted the dichotomy between active and passive coping styles (Peres & Lucchetti, 2010). This distinction is also theoretically consistent with the active and passive terminology used by Dunn (1997) in developing the Sensory Processing Model; however, similar coping dichotomies exist (e.g., approach [active] and avoidant [passive] coping). Passive coping is considered to be used when the person surrenders control or relies on external supports to cope with pain (Peres & Lucchetti, 2010). It has been characterized by feelings of helplessness (Snow-Turek et al., 1996) and has been associated with both psychopathology (e.g., depression; Brown & Nicassio, 1987) and poorer pain outcomes (Edwards et al., 2009). In contrast, Peres and Lucchetti (2010) noted that active pain coping involves an internal locus of control, whereby the person aims to control the pain through using his or her own resources (e.g., exercising, positive self-statements). Active coping has been inversely associated with psychopathology (Snow-Turek et al., 1996).

With this in mind, we expected sensory sensitivity and low registration (passive sensory approaches) to be associated with passive coping strategies (catastrophizing and hoping/praying), as stated in Hypothesis 2. Moreover, we expected sensation avoiding and sensation seeking (active sensory approaches) to be associated with active coping strategies (ignoring pain, coping self-statements, diverting attention, and increasing behavioral activities), as stated in our third hypothesis.

As noted earlier, both sensory sensitivity and low registration were initially related to catastrophizing; however, contradicting Hypothesis 2, the association was lost for low registration after controlling for age, gender, and stress. In addition, sensation avoiding, an active sensory pattern, was associated with catastrophizing. This finding is more consistent with the conceptualization of avoidance as a passive coping strategy and challenges the conceptualization of sensation avoidance as an active sensory pattern. This result is also consistent with the view that it is the low sensory threshold dimension (sensory sensitivity and sensation avoiding) rather than the active or passive nature of the coping approach (low registration and sensation seeking) that accounts for the association between sensory perceptions and pain catastrophizing (i.e., participants with lower sensory thresholds were more likely to catastrophize). The other passive coping style, hoping/praying, was not found to be associated with any particular sensory pattern. This result may not be surprising given the debate about whether hoping/praying, or religious coping in general, should be considered passive (Peres & Lucchetti, 2010). In contrast to expectations, sensory sensitivity was linked with diverting attention, which is typically viewed as an active coping strategy.

In partial support of Hypothesis 3, sensation seeking (an active sensory approach) was significantly associated with coping self-statements and, to a lesser extent, with increasing behavioral activities and ignoring pain (active coping strategies). This finding is consistent with expectations and with previous evidence of a lack of association between sensation seeking and both catastrophizing (Engel-Yeger & Dunn, 2011b) and negative mood (Engel-Yeger & Dunn, 2011a, 2011c). It suggests that, in the context of a painful experience, sensation seeking may constitute a protective factor.

As noted earlier, and in contrast to expectations for Hypothesis 3, sensation avoiding (an active, low sensory threshold category) was associated only with catastrophizing. One possible explanation for this finding is the evidence, reported in prior literature (Engel-Yeger & Dunn, 2011a, 2011c), of associations between sensation avoiding and adverse mood states, which may influence a person’s tendency to choose a passive coping strategy. However, this result was retained when controlling for stress. As discussed in relation to sensory sensitivity, people with low sensory thresholds may have heightened sensitivity (low thresholds) to pain and therefore be more likely to be overwhelmed and respond with maladaptive coping strategies. Given that both sensory sensitivity and sensation avoiding (Engel-Yeger & Dunn, 2011a, 2011c), and maladaptive coping (Edwards et al., 2009; Karsdorp et al., 2012), have been associated with negative affect, this result is theoretically consistent. These findings suggest that high scores on sensation avoiding or sensory sensitivity (low sensory threshold) may constitute vulnerability factors and potentially be associated with poorer pain outcomes.

Note that only one of the two active sensory categories, sensation seeking, was associated with active pain coping strategies. This result suggests that the combination of both high sensory threshold and a more active behavioral approach was required to support use of active pain coping strategies. Either high sensory threshold or active behavioral coping alone was not sufficient.

Implications of Stress, Gender, and Age

Research examining the relationships among stress, coping, and sensory approach is limited. However, stress or symptoms of pathological stress (e.g., posttraumatic stress disorder [PTSD]) have been cited as influential factors in acute pain coping (Hekmat, Staats, & Staats, 2005) and pain perception (Geisser et al., 1994). Although not formulated as a hypothesis, we did expect that stress would be consistently and positively associated with catastrophizing. For example, PTSD symptoms have been found to be associated with catastrophizing (Horsham & Chung, 2013). In the past, distress has been linked to diverting attention from pain (Crisson & Keefe, 1988), particularly when the person feels helpless during the pain experience.

Evidence has also shown clear gender differences in pain perception and coping. For example, Keogh and Herdenfeldt (2002) noted that, although men were more likely to use behavioral distraction strategies, women tended to use more social support and positive self-talk. Results of the current study, however, suggested that female gender was consistently associated with increased use of behavioral (active) strategies. The reason for this finding is unclear, but it may be related to nuances of the current study sample. For example, women in this study were uncharacteristically more likely to report being low registration and sensation seeking, both sensory patterns that have a high sensory threshold. Given that both of these sensory patterns have been associated with more adaptive functioning, the gender links with behavioral strategies found in this study appear intuitive. Further investigation into the impact of demographic variables on associations between sensory patterns and the pain coping strategies used by women could provide clarity for this finding.

It is noteworthy that age was negatively related to several coping strategies. Specifically, older age was linked with less use of all pain coping approaches, particularly catastrophizing and ignoring pain. Although some evidence suggests that older people may cope better with pain by using greater impulse control, more positive appraisal of painful experiences (Diehl, Coyle, & Labouvie-Vief, 1996), and higher levels of stoicism (Cook & Chastain, 2001), the associations between age and coping are complex. For example, Watkins, Shifren, Park, and Morrell (1999) found that in a sample of 121 people with rheumatoid arthritis aged 34–84 yr, all reported using more active coping strategies when experiencing mild pain and more maladaptive coping strategies when in severe pain, regardless of age. The authors recommended that “research regarding illness and coping is most informative when it captures the individual–situational interaction of dealing with stressors such as chronic pain” (p. 217). Of note, older people in the current study were also more likely to be male and less likely to be sensation seeking or sensory sensitive.