Background
Chest pain is a frequent cause of visits in medical emergency and cardiology settings [
1‐
4]. In approximately 50% of cases, patients present with noncardiac chest pain (NCCP), that is, chest pain in the absence of identifiable cardiac etiology [
4‐
9]. Even though NCCP is often medically benign, its negative impact on quality of life is long-lasting and comparable to that of cardiac disease [
10‐
15].
NCCP is associated with a high prevalence of psychiatric comorbidity (41–88%) [
12,
16‐
21]. The two most common psychiatric disorders in patients with NCCP visiting an emergency department are panic disorder (PD; 14–50%) and generalized anxiety disorder (GAD; 6–33%) [
19,
21‐
27]. These psychiatric comorbidities are associated with a less favourable NCCP presentation and have a serious impact on the patient’s quality of life [
21,
24,
28].
Indeed, in patients with NCCP, PD is associated with increased NCCP frequency and severity, increased risk of NCCP recurrence [
21,
28‐
31] and lowered health-related quality of life [
28‐
30]. However, only the physical component of health-related quality of life of the 12-item Short-Form Health Survey Version 2 (SF-12v2) appears to be significantly affected in patients with NCCP who present with comorbid PD [
28]. However, these results need to be interpreted with caution, as some are from cross-sectional studies and have yet to be confirmed prospectively [
21,
30].
The impact of comorbid GAD on NCCP and health-related quality of life has yet to be assessed. However, the presence of at least one psychiatric disorder is associated with elevated pain severity and life interference due to NCCP [
21]. Moreover, GAD has been linked to lowered quality of life in primary care patients, especially with regard to emotional health [
32‐
34]. In fact, the quality of life of patients with GAD has also been found to be similar to that of patients with major depressive disorder and chronic medical conditions, such as arthritis [
35‐
37]. Based on these data, it is likely that the presence of comorbid GAD negatively affects the presentation of NCCP and its consequences on health-related quality of life.
Even if PD and GAD are both anxiety disorders, they are independent diagnostic entities that are likely to have a differential impact on NCCP severity and health-related quality of life in patients with NCCP. Currently, it remains unclear how patients with NCCP and PD compare to patients with NCCP and GAD in terms of patterns of symptoms and health-related quality of life. Therefore, the objectives of the present study were to describe and prospectively compare patients with NCCP, with or without comorbid PD or GAD, in terms of (1) NCCP severity; and (2) the physical and mental components of health-related quality of life. Assessing each component of health-related quality of life is essential in order to understand how PD and GAD respectively and differentially affect patients and to guide clinical decision-making.
Considering that PD is more specifically characterized by interoceptive fear and hypervigilance than GAD [
38,
39], it was expected that patients with PD would report greater NCCP severity than those with GAD. Based on the literature, it was also hypothesized that patients with PD would report a lower physical quality of life, while patients with GAD would report a lower mental quality of life. Overall, it was expected that patients with NCCP and comorbid PD or GAD would present with higher NCCP severity and lower health-related quality of life than those with NCCP without comorbid PD or GAD [
21].
Discussion
The first objective of this study was to describe and compare NCCP severity in patients with or without comorbid PD or GAD at baseline and at the six-month follow-up. Overall, the patients still reported episodes of NCCP at the six-month follow-up, which corroborates the well-documented persistence of these symptoms in the literature [
1,
7,
30,
48]. A time effect of moderate size was found for all the patients (η
p2 = 0.162), as average NCCP severity tended to decrease from baseline to the six-month follow-up, which supports the results of Dammen and colleagues [
29]. This improvement in symptom severity could be explained by some form of reassurance obtained during the emergency department visit. Still, average NCCP severity was higher in the patients with comorbid PD or GAD at both time points (η
p2 = 0.029). This result is in line with those indicating that comorbid psychiatric disorders are associated with increased NCCP severity [
21,
28‐
31]. Surprisingly, no significant differences were found between the patients with PD and those with GAD in terms of average NCCP severity. This does not support the primary hypothesis, namely, that patients with PD would report more severe NCCP. The greater-than-expected association of GAD with NCCP severity may be explained by intolerance to uncertainty. The impact of medical uncertainty associated with diagnoses of exclusion, such as NCCP [
2,
24], might be underestimated in patients with GAD, who are already well-known for their intolerance to uncertainty. Indeed, intolerance to uncertainty is associated with pain severity through the catastrophizing of pain in patients with chronic pain [
49,
50]. Therefore, it is possible that the medical uncertainty associated with NCCP, coupled with the intolerance to uncertainty of patients with GAD, increases anxiety significantly, leading these patients to catastrophic interpretations and an amplified perception of bodily symptoms, as experienced by patients with PD as well. Moreover, tolerance of negative emotional states could also explain the strong association of GAD with NCCP severity. Indeed, tolerance of negative emotional states has been independently associated with chronic pain severity, while being closely linked to avoidance of internal experiences and difficulties in emotion regulation found in patients with GAD [
51,
52].
Another surprising result is that the presence of both PD and GAD was not associated with greater NCCP severity than the presence of either disorder. This result is not consistent with that of White and colleagues [
21], namely, that patients with more psychiatric disorders report greater NCCP severity. However, this apparent contradiction results may be explained by the inclusion of a greater range of disorders (e.g., all anxiety and mood disorders, substance-related disorders) in that study compared to this one.
The second objective of the present study was to describe and compare health-related quality of life in patients with NCCP, with or without comorbid PD or GAD, at baseline and at six-month follow-up.
Physical quality of life improved over time for all the patients, regardless of the group, which is consistent with previous studies of patients with NCCP with or without PD [
8,
29]. This limited improvement (η
p2 = 0.014) could be explained by some form of reassurance obtained by all patients during the emergency department visit. However, one could argue that, in the long term (e.g., one-year or two-years follow-up), certain groups of patients may be at greater risk of negative trajectories than others once the reassurance effect has worn off. Of note, the presence of PD, with or without comorbid GAD, was significantly and meaningfully (η
p2 = 0.260) associated with a diminished physical quality of life over time, which is in line with the primary hypothesis and results of Bull Bringager and colleagues [
28]. The results also show a lower physical quality of life in the patients with PD than in those with GAD. This observation corroborates reports that PD is the anxiety disorder that affects the physical functioning subscale the most in patients without NCCP [
32,
34].
As regards mental quality of life, a time x group interaction effect of small size (η
p2 = 0.018) was obtained. Indeed, the presence of GAD, alone or in comorbidity with PD, was significantly associated with a lower mental quality of life than PD alone, this effect being greater at the follow-up than at baseline. These results support the primary hypothesis of a greater association of GAD with lowered MCS scores and are also consistent with the positive association found by White and colleagues [
21] between the number of disorders and the impairment reported by patients. Moreover, the absence of a significant difference between the patients with PD and those with no comorbidities at the six-month follow-up also suggests that PD has a significant but limited impact on mental quality of life.
In summary, this study highlights the considerable negative impacts of GAD on NCCP and health-related quality of life of patients, which confirms that they represent a particularly vulnerable subgroup of patients, just like patients with NCCP and PD. Therefore, GAD should also be identified early in the process so that patients can benefit from appropriate treatment or referral. Some brief self-report instruments have been shown to be useful in screening for PD in patients with NCCP, such as the Revised Panic Screening Score [
53,
54] and the Psychiatric Diagnostic Screening Questionnaire [
55,
56]. A similarly brief instrument, such as the GAD-2, could be used to screen for GAD during a patient’s initial visit for NCCP [
57]. This study also highlights the differential impact of PD and GAD on health-related quality of life, which reinforces the relevance of assessing specific domains of quality of life because they appear to vary across anxiety disorders. These findings shed light on specific needs of these patients that could be targeted in order to improve their quality of life.
Currently, psychological interventions for NCCP are offered to patients to prevent pain from becoming chronic [
58]. While a Cochrane review recommends cognitive-behavioural therapy for these patients, its benefits appear to be modest and largely limited to the first three months after the intervention [
58]. As for offering psychological treatment for the comorbid anxiety disorder itself in patients with NCCP, studies show that cognitive-behavioural therapy is effective in reducing PD severity, but little is known about its impact on NCCP [
59‐
61]. With regard to GAD, no study to date has assessed the effectiveness of cognitive-behavioural therapy in patients with NCCP. Based on the findings of the present study, the current intervention for patients with NCCP only is likely to be insufficient to treat patients with NCCP and comorbid PD or GAD.
The primary strengths of this study are its prospective nature, the consecutive sampling, the high retention rate at the six-month follow-up (80.2%) and the large sample size, especially for the assessment of NCCP severity, thanks to the high measure completion rate. Another strength of this study is the application of a standardized psychiatric interview and the high diagnostic reliability achieved. Moreover, the use of generalized linear mixed models helped minimize potential type I errors, as dependence between time points for each patient was considered. Finally, this study was able to address a current knowledge gap by prospectively assessing NCCP severity and health-related quality of life in patients with NCCP and GAD. It also prospectively compared NCCP severity and health-related quality of life according to the patients’ anxiety disorder profile.
This study has some limitations that should be taken into consideration when interpreting the results. First, one should bear in mind that the presence of PD and GAD was not reassessed at follow-up. Considering the chronic course of these disorders over time [
31,
62], a potential impact on the results appears unlikely. Second, certain analyses were conducted on small subgroups of patients, and a lack of statistical power could explain some of the negative results. However, group effects were obtained by combining certain subgroups, which made it possible to draw relevant conclusions. Third, potential implications for the findings’ internal validity and generalizability of the results need to be addressed. On one hand, patients who refuse to enrol in the study might have been different from those who participated in the study with respect to some sociodemographic characteristics or general health. On the other hand, significant differences between patients in the final sample and those who were not included in the study should be acknowledged. The patients in the final sample (n = 915) for the first objective (NCCP severity) were more educated, more likely to be married or in a common-law relationship, and had a higher family income. Consequently, one could argue that the patients in the final sample might have had a better understanding of research and better health, which might have resulted in an underestimation of NCCP severity in the present study. The patients in the final sample (n = 434) for the second objective (health-related quality of life) were older and less likely to be working than those who did not complete the measure at either time point. The age difference may mean that the PCS scores were lower than they should have been, while the MCS scores were higher than they should have been [
34]. The difference regarding employment status may have led to a lowered health-related quality of life in the study since the patients who were not working might have had poorer health.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.