| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychology (A.M.N., C.M.N., T.A.C., J.F., M.L.), Drexel University, Philadelphia, Pennsylvania; Department of Medicine (A.M.N., C.M.N., D.J.), Drexel University College of Medicine, Philadelphia, Pennsylvania; Healthy Weight Program (M.S.X.), Children's Hospital of Philadelphia, Philadelphia, Pennsylvania.
Address correspondence and reprint requests to Arthur M. Nezu, Department of Psychology, MS 515, Drexel University, 245 N. 15th Street, Philadelphia, PA 19102-1192. E-mail: amn23{at}drexel.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Key Words: noncardiac chest pain • social problem solving • stress • coping
Abbreviations: CVD = cardiovascular disease; FAS = Framingham Anger Scales; HADS = Hospital Anxiety and Depression Scale; MANOVA = multivariate analysis of variance; MPI = myocardial perfusion imaging; NCCP = noncardiac chest pain; PSS = Perceived Stress Scale; PST = problem-solving therapy; RA = research assistant; SPS = social problem solving; SPSI-R = Social Problem-Solving Inventory—Revised.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Persons with NCCP have been found to be more likely to have psychiatric problems than either the general public or those with chest pain due to organic causes (4–6). One-year follow-up studies of patients with NCCP indicated that about 75% of patients continued to have recurrent chest pain (5,6). Maintaining factors for NCCP seem to include the belief that chest pain is due to cardiac problems and the presence of a psychiatric history (7). In addition, patients with NCCP are more likely to be characterized by ineffective coping strategies when faced with chest pain as compared with patients with other disorders, such as coronary heart disease and irritable bowel syndrome (8,9). Research has also indicated that such psychosocial problems predate, as well as coexist or result from, NCCP (10).
The present study sought to investigate the role that social problem solving (SPS) plays with regard to NCCP. SPS is a multidimensional psychosocial variable that has been repeatedly identified as an important factor in the pathogenesis of both mental health and health problems resulting from poor adaptation to stress (11,12). SPS has been defined as the cognitive-behavioral process by which a person attempts to identify or discover effective or adaptive solutions for stressful problems encountered during the course of everyday living (13). In this context, it involves the process whereby individuals attempt to direct their coping efforts at altering the problematic nature of a situation itself, their reactions to such situations, or both. SPS refers more to the meta-process of understanding, appraising, and adapting to stressful life events, rather than representing a singular coping strategy or activity.
Problem-solving outcomes in the real world are hypothesized to be largely determined by two general, but partially independent processes: a) problem orientation and b) problem-solving style (14). Problem orientation involves a set of generalized thoughts and feelings concerning problems in living as well as one's ability to successfully resolve them. It can either be positive (e.g., viewing problems as opportunities to benefit in some way; perceiving oneself as able to solve problems effectively), which serves to enhance subsequent problem-solving efforts, or negative (e.g., viewing problems as a major threat to one's well being; overreacting emotionally when problems occur), which functions to inhibit attempts to solve problems.
Problem-solving style refers to specific cognitive-behavioral activities aimed at coping with stressful problems. They can be either a) adaptive, leading to successful problem resolution, or b) dysfunctional, leading to negative consequences, such as psychological distress. Rational problem solving is the constructive style geared to identify an effective solution to the problem and involves the systematic application of specific problem-solving tasks (e.g., accurately identifying why a given situation is a problem, generating alternative solutions, conducting a cost-benefit analysis of the potential outcome of implementing the various alternative solutions, monitoring the actual consequences of an implemented solution). Dysfunctional problem-solving styles include a) impulsivity/carelessness (i.e., impulsive, hurried, and incomplete attempts to solve a problem); and b) avoidance (i.e., avoiding problems, procrastinating, and depending on others to solve one's problems).
Important differences have been identified between individuals characterized as "effective" versus "ineffective" problem solvers. In general, when compared with effective problem solvers, persons characterized by ineffective problem solving report a greater number of life problems, more health and physical symptoms, more anxiety, more depression, and more psychological maladjustment (12). In addition, a negative problem orientation has been found to be associated with negative moods under routine and stressful conditions in general, as well as pessimism, negative emotional experiences, and clinical depression (13). Further, persons with a negative orientation tend to worry and complain more about their health (15).
Relevant to the present discussion, researchers have also addressed the association between SPS and pain. Specifically, SPS has been found to be significantly associated with premenstrual and menstrual pain (16), psychosocial impairment and distress among patients entering pain rehabilitation programs (17), functional impairment among adults experiencing chronic low back pain (18,19), pain intensity, depression, and disability among patients with chronic pain (20), and pain intensity and quality of life among veterans with spinal cord injuries (21).
Given the above context, the following hypotheses were advanced: 1) consistent with previous research, patients with NCCP, as compared with a matched control group of patients with chest pain testing positive for cardiac disease, would report higher levels of psychological distress (i.e., depression, anxiety, anger, and perceived stress); 2) patients with NCCP, again in comparison with the control group, would be characterized by ineffective problem solving above and beyond that which might be attributable to differences related to negative affectivity and prior history of cardiac disease; 3) SPS would be significantly associated with pain among patients with NCCP, above and beyond the variance which might be accounted for by such confounding variables; and 4) the relationship between SPS and pain would be greater in patients with NCCP than patients with cardiovascular disease (CVD).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
To be able to identify a group of individuals who tested positive for underlying cardiac disease, 150 individuals were initially sought for participation. Potential patients were provided with a brief overview of the research protocol. Interested and willing persons were then provided with a more complete description of the nature and purpose of the study by a research assistant (RA). Informed consent was then obtained from individuals willing to complete a series of questionnaires. This procedure occurred before MPI testing. Therefore, both the patient and RA were blinded to the actual outcome of the cardiac evaluation. In other words, all participants, at the time of the psychological testing, were unaware as to whether they would test positive or negative for heart disease and completed all inventories within that context.
Of the 150 individuals who consented to participate, based on the results of the MPI, 55 patients (30 men; 25 women) were eventually found to test positive for underlying CVD or ischemia that could be linked to the symptoms of chest pain. Interpretations of MPI results were conducted independently by a cardiologist specifically trained in this protocol. Of these 55 individuals, one male patient's questionnaires were incomplete; therefore, his results were dropped from the study. Of the remaining 95 individuals, five persons had unclear perfusion results, thus leaving a pool of 90 individuals with identified negative results for cardiac disease. This cohort of patients thus constituted the NCCP patient group. The overall percentage of NCCP patients in this sample was close to 63%, which is consistent with other studies (22,23).
To conduct a meaningful evaluation of the differences between chest pain patients with and without underlying heart disease, an attempt was made to match the 54 patients with identified CVD with NCCP persons regarding three major demographic characteristics—gender, self-identified ethnic background, and age. Among the CVD group, 16 were white, 36 were black, and two were Hispanic. Therefore, a similar constellation of NCCP persons were sought. In addition, for each of these participants with CVD, an attempt was made to identify a subject with NCCP who, in addition to being the same ethnicity, was born within a range of ±5 years. If more than one such individual existed in the NCCP group that was "matched" on gender, ethnic background, and age range, a random selection was conducted to identify an appropriately matched NCCP person. Because a match was not possible for one of the 54 subjects with CVD, her data were not included in the final analysis, thus resulting in final cohorts of 53 CVD patients and 53 NCCP patients.
Measures
The self-report inventories described below were administered to all study participants.
The Hospital Anxiety and Depression Scale (HADS) is a reliable and valid measure for medical patients (24–26) and includes an anxiety scale and a depression scale, both of which contain seven items. More recently, the HADS has been validated for specific use with patients who present with NCCP (27). Based on an assessment of the specificity and sensitivity of the HADS, it has been recommended that a score of 8/9 on the anxiety scale can be used as a cut-off for panic disorder for this population, whereas a cut-off score of 4/5 on the depression scale is useful when screening for depressive episodes among individuals with NCCP.
The Perceived Stress Scale (PSS) was designed to measure the degree to which situations in one's life are appraised as stressful (28). The 10-item version used in this study contains questions regarding the degree to which respondents find their lives unpredictable, uncontrollable, and overloading. Analyses by Cohen et al. (28) suggested that the PSS is represented by strong internal consistency (Cronbach's 
estimates range between 0.75 and 0.86) as well as test-retest reliability (0.85).
The Framingham Anger Scales (FAS) were developed for the Framingham study of coronary heart disease risk to assess how anger is expressed when it is felt (29). In addition to assessing internalization (anger-in) and externalization in an attacking or blaming way (anger-out), the FAS measure somatization (physical symptoms) and discussion of anger (anger-discuss).
Pain Intensity and Frequency
Using a rating scale of 1 (no pain) to 7 (severe pain), the subjects were asked to indicate how intense their chest pain was at its worst during the past month. Using a scale of 1 (not at all frequent) to 7 (very frequent), they were also requested to rate the frequency of their chest pain during the past month regardless of its intensity at any given time.
Social Problem-Solving Inventory Revised (SPSI-R)—Short Version (30)
The short version of the SPSI-R is a 25-item multidimensional measure of social problem-solving ability derived from a factor analysis of the original theory-driven Social Problem-Solving Inventory (31). In addition to a total score, it consists of five scales that measure two constructive dimensions (positive problem orientation, rational problem solving) and three dysfunctional dimensions (negative problem orientation, impulsivity/carelessness style, avoidance style). Respondents are asked to rate items (e.g., "I go out of my way to avoid having to deal with problems in my life," "Before I try to solve a problem, I set a specific goal so that I know exactly what I want to accomplish.") on a 5-point Likert-type scale ranging from 0 (not at all true of me) to 4 (extremely true of me). Higher scores on a given scale represent higher levels of that particular problem- solving dimension. Research has found the SPSI-R to have strong internal consistency (range of 
coefficients of 0.79–0.95 across the five scales) and test-retest reliability (estimates of 0.93 and 0.89 for the total score over a 3-week period among two different samples), as well as strong structural, concurrent, predictive, convergent, and discriminant validity (30). It has also been found to be sensitive to the effects of treatment (32).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
2) analyses revealed a lack of differences between groups regarding all of these demographic parameters with the exception that significantly more patients with CVD had a history of cardiac problems as compared with NCCP individuals (p < .001).
|
A multivariate analysis of variance (MANOVA) included four cardiac indices obtained during a physician's clinical examination and cardiac testing initially yielded a significant omnibus Wilks' Lambda F(4,101) = 12.96, p < .001. As Table 2 denotes, significant differences were found between CVD and NCCP subjects regarding both their left ventricular ejection fraction, as would be expected, as well as their baseline heart rate level. No differences, however, were found regarding either type of blood pressure index at baseline.
|
Pain Parameters
An analysis was conducted initially to determine if differences exist between groups regarding the two pain indices. Results of the MANOVA indicated an overall lack of differences between the two groups, omnibus F(2,103) = 1.01, p = .37, regarding both pain intensity and pain frequency (Table 3).
|
Depression, Anxiety, Perceived Stress, and Anger
A second MANOVA was conducted next to determine potential differences between groups with regard to the two HADS scales, the PSS, and the four anger scales (Table 3). Initial results revealed a significant omnibus Wilks' Lambda estimate for the main effect for group, F(7,98) = 2.52, p = .02. As Table 3 shows, subsequent individual contrasts revealed that patients with NCCP reported significantly higher levels of depressive symptomatology, higher levels of anxiety symptoms, higher levels of perceived stress, and significantly higher scores on both the anger symptoms and anger-out scales, as compared with their CVD counterparts. No differences were identified between groups with regard to either the anger-in or the anger-discuss scales.
Social Problem Solving
The next analysis addressed the major hypothesis predicting differences in SPS between the two groups. However, given the strong relationship generally found between SPS and psychological distress (12,13), in conjunction with the consistent findings that NCCP is also correlated with psychological distress (as was already identified in this investigation), it is important to assess the uniqueness of SPS beyond the contribution of general negative affectivity. As such, a multivariate analysis of covariance was conducted next whereby the five scales of the SPSI-R served as dependent variables, whereas the two HADS measures, the PSS, and the anger symptom scale of the FAS served as covariates. Initial results revealed that as a block, the various covariates served a significant influence on SPS, F(5,96) = 2.67, p = .02. Moreover, after partialling out such an effect, differences were found between subjects with NCCP and their matched counterparts regarding the overall block of the five SPSI-R scales, F(5,96) = 3.96, p = .003. As Table 4 indicates, subsequent individual contrasts revealed significant differences between groups across three of the five SPSI-R scales. Specifically, patients with NCCP, in comparison to their counterparts with CVD, reported significantly greater levels of a negative orientation, significantly lower positive problem orientation, and lower rational problem-solving style scores. No significant differences between groups were identified for the two maladaptive problem-solving styles, i.e., impulsivity/carelessness or avoidance.
|
Effects of Prior CVD History
In conducting the above analyses, it was assumed that all patients were equally likely to perceive their pain to be cardiac related, particularly because they presented for testing at a cardiac imaging facility. However, if one has a history of CVD, it is possible that such a belief is stronger as compared with the person without such a background. Therefore, CVD patients in this study without previous cardiac events may be thought of as more similar to those individuals also without such a history, but who eventually would be categorized as an NCCP patient. As such, it is conceivable that prior history of CVD may have an effect on one's pain experience. However, subsequent tests excluding all 20 individuals with a history of CVD that repeated the above analyses revealed no differences in results, indicating a lack of effect on pain in this study due to prior history.
SPS as a Predictor of Pain
The next analysis addressed the hypothesis positing that SPS would be significantly related to pain above and beyond the variance accounted for by negative affectivity. Two sets of hierarchical multiple regression analyses were therefore conducted next for all original 90 NCCP individuals. One set focused on pain intensity as the dependent variable, whereas pain frequency served as the dependent variable in the second set of analyses. For each regression, a block of "control" variables was entered first (i.e., HADS-depression, HADS-anxiety, PSS, and anger symptoms), followed by one of the SPSI-R scale scores (i.e., positive problem orientation (PPO), negative problem orientation (NPO), rational problem solving (RPS), impulsivity/carelessness scale (ICS), avoidance scale (AS)) resulting in a total of ten regression analyses. Table 5 contains the zero-order correlations among these variables.
|
Pain Intensity
Results of the first set of analyses (Table 6) indicated that the block of covariates representing negative affectivity was significantly related to ratings of pain intensity, R2 = .14 (adjusted R2 = .10), F(4,85) = 3.51, p = .01. Next, adding a specific SPSI-R scale to the regression analysis one at a time indicated that the NPO and RPS variables each led to a significant increase in R2 above and beyond that due to negative affectivity in predicting pain intensity, addition to R2 = .16, F(1,84) = 19.83, p < .001, addition to R2 = .08, F(1,84) = 9.16, p = .003, respectively, whereas the remaining three problem-solving factors (i.e., PPO, ICS, AS) did not.
|
Pain Frequency
Results of the second set of analyses further indicated that the block of covariates representing negative affectivity was also significantly related to ratings of pain frequency, R2 = .14 (adjusted R2 = .10), F(4,85) = 3.52, p = .01 (Table 7). Adding each of the five SPSI-R scale scores, one at a time, to the regression analysis revealed that the NPO, RPS, and AS variables each led to a significant increase in R2 above and beyond that due to negative affectivity in predicting pain frequency, addition to R2 = .07, F(1,84) = 7.56, p = .007, addition to R2 = .08, F(1,84) = 9.07, p = .003, addition to R2 = .06, F(1,84) = 6.39, p = .01, respectively, whereas the remaining two problem-solving variables (i.e., PPO, ICS), did not.
|
SPS and Pain: Specific to NCCP?
To evaluate whether the relationship found between problem solving and pain is actually unique to NCCP, a comparison of the zero-order correlations between problem solving and pain by patient group was conducted. Specifically, the significance of the z-score representing the difference in magnitude of Pearson r values for NCCP versus CVD patients was assessed. NPO and RPS were found to be significantly related to ratings of pain intensity among both patient groups (Table 8). For NCCP individuals, ICS and AS scores were additionally found to be significantly correlated with pain intensity. In assessing the significance of the magnitude of differences in these correlations, only a trend was identified for such differences and only with regard to the association regarding NPO and pain intensity (i.e., this relationship was marginally stronger among NCCP individuals as compared with the CVD sample).
|
As shown in Table 9, none of the zero-order correlations between the problem-solving measures and ratings of pain frequency was found to be significant among the 54 individuals with CVD. For the NCCP group, their NPO, RPS, ICS, and AS scale scores were significantly related to the measure of pain frequency. In assessing the difference in magnitude of these similar correlations between patient groups, once again, only a trend was found, where the relationships between both the NPO and AS scales and pain frequency was marginally stronger among NCCP patients as compared with CVD individuals.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
More importantly, in support of the major hypotheses, NCCP patients, in contrast to their CVD counterparts, were characterized by less effective social problem solving with specific regard to both types of problem-orientation dimensions (i.e., less positive and more negative), as well as with regard to the lowered use of a rational problem-solving style (i.e., the systematic and planful approach to solving life's problems). Moreover, such differences were not simply a function of generalized negative affectivity or negative self-reports. Further, certain SPS variables were found to be significantly associated with pain intensity and frequency among a larger group of NCCP patients, again, above and beyond the influence of general negative affectivity. Finally, in an analysis to determine whether this relationship is unique or specific to NCCP, only a trend was identified whereby the NCCP and CVD groups differed in their relationship between certain problem-solving variables and pain indices before covarying negative affect (i.e., only marginally stronger relationships were identified for the NCCP patients with specific regard to negative problem orientation and pain intensity and negative orientation and avoidance with regard to pain frequency).
Overall, these findings are in keeping with an SPS model of psychological stress and distress (12–15), which suggests that much of what is viewed as "psychopathology" can be understood as ineffective or maladaptive coping behavior and its consequences, where individuals are unable to adequately cope with stressful circumstances in their lives and their ineffective attempts to do so leads to negative consequences, such as anxiety, depression, anger, physical symptomatology (including chest pain), and the creation of new stressful problems (13,34). According to this model, such "symptoms" can also serve as problems by themselves that further require individuals to engage in additional coping efforts. As such, significant chest pain can be viewed as the direct or indirect outcome of ineffective coping (e.g., a symptom of stress) (22), a stimulus that engenders other types of psychological distress, such as depression and anxiety, as well as new stressful problems (e.g., how to treat the pain, how to deal with physical limitations), or as a result of depression itself (35). Over the course of time, it is likely that such factors interact in a reciprocal fashion (22). However, because the present investigation was not able to test these various pathways directly, such an analysis remains speculative at the present time. In any case, it seems that SPS plays an important role within this framework, regardless of the actual sequence of events, and therefore, represents a potentially important treatment target for psychosocial interventions to address among patients with noncardiac chest pain.
The importance of teaching individuals effective problem-solving principles (i.e., problem-solving therapy or PST) to decrease maladaptive behavior and psychological distress has been supported by scores of empirical studies (13,15). PST has been identified as a key psychosocial strategy for coping with chronic illness (36), an effective approach for enhancing one's quality of life for various medical patient populations, such as individuals diagnosed with cancer (32,37,38), and has been found to foster patient adherence to other forms of psychosocial and medical treatment (39,40). More relevant to the present study, for example, van den Hout et al. evaluated whether PST provided a significant added value to a behavioral graded activity protocol in treating patients with nonspecific low back pain with regard to work-related disability (41). Results indicated that in the second half year after the intervention, patients receiving both graded activity and PST training had significantly fewer days of sick leave than their counterparts who received graded activity plus group education. Further, work status was more favorable for the participants receiving PST in that more employees had a 100% return-to-work and fewer patients received disability pensions 1 year post treatment. Given this context, a potentially fructuous avenue of intervention research in the future is to evaluate the efficacy of PST for NCCP.
Our findings should be interpreted with caution, in part, due to the cross-sectional nature of this investigation, which prevents a more definitive evaluation of causal relationships. Therefore, future research should use a prospective design to better assess the directionality of these associations. In addition, only patients who sought medical consultations to determine the cause of their chest pain were included in this investigation, which may have inadvertently served to bias the internal validity of the results. For example, Bunde and Martin (42) recently found that experiencing depression during the 2 weeks before hospitalization for acute myocardial infarction symptomatology, such as chest pain, increased the delay time in seeking treatment. In the present study, the amount of time between when participants first experienced chest pain and when they completed the inventories (i.e., presented to the nuclear imaging laboratory) is unknown. Therefore, it is difficult to ascertain how any subject coped with (i.e., "problem solved") the stressful nature of experiencing chest pain (e.g., asking help from family, seeking treatment options, avoidance), further questioning whether pain serves as an initial stimulus for further psychological distress or represents the negative outcome of ineffective coping with other stressful life events. As such, future studies should include an assessment of the influence of SPS on the persistence of NCCP (i.e., the continued experience of chest pain in spite of evidence that such pain is noncardiac related).
Future research in this area might additionally focus on differential reactions to being provided with the actual outcome of the cardiac testing. In other words, assuming that in part the high levels of distress reported by all participants (although NCCP patients reported significantly higher levels of distress than CVD individuals) were in response to concerns regarding the possibility that one has CVD, an important aspect of the overall picture would be to determine whether differences exist between these groups in the manner in which they react to the ultimate news about their cardiac status. Given the literature regarding "hot" versus "cold" affective states on health judgments (43), it would be interesting to note whether differences in distress between NCCP and CVD patients would emerge being provided with the results of their MPI testing, whether the manner in which such test results are presented to NCCP patients by medical professionals affects such differences (44), and relevant to the present discussion, whether SPS serves to mediate such differences (45).
We would like to thank the following individuals for their assistance in data collection and scoring: Melissa Clark, Aimee Kim, Austen Krill, Steve Marshall, and Tatiana Vasilevskaia.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
DOI:10.1097/PSY.0b013e31815a995a
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. A. Shelby, T. J. Somers, F. J. Keefe, S. G. Silva, D. C. McKee, L. She, S. J. Waters, I. Varia, Y. B. Riordan, V. M. Knowles, et al. Pain Catastrophizing in Patients with Noncardiac Chest Pain: Relationships With Pain, Anxiety, and Disability Psychosom Med, October 1, 2009; 71(8): 861 - 868. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
