Patients With Atypical Chest Pain: Epidemiology and Reported Consequences

Background: Approximately a quarter of the global population experiences chest pain during their lifetime worldwide. Although largely non-life-threatening, many patients experience mental, physical, social, and financial consequences. Aim: This study aimed to describe and determine the epidemiology and consequences of patients presenting with atypical chest pain (ACP). Method: Data were obtained from 102 participants, from a desired sample size of 166. The target population was patients who presented with ACP at the Accident and Emergency Department of a Teaching Hospital in Trinidad during a two-year period, from January 2021 to December 2022. The data collection instrument used was a 34-item online questionnaire. Data were analyzed using both descriptive and inferential statistical methods. Results: Participants were predominantly women (63.7%; n = 65), between 31 and 50 years of age (74.5%; n = 76), in full-time employment (n = 58; 56.9%), who lived with at least one person (90.2%; n = 92) at the time of the episode. Overall, 61.8% (n = 63) reported having a stressful life. Hypertension (30.4%; n = 31) and diabetes (18.6%; n = 19) were the leading comorbidities. Participants experienced mild to severe anxiety (53.9%; n = 55), moderate to severe depression (25.5%; n = 26), moderate stress (65.7%; n = 67), and loneliness (25.5%; n = 26). A stressful life was associated with, and was a predictor of, both anxiety and loneliness. No sociodemographic variables were associated with depression or stress. The most common self-reported consequences were "fear as a result of the pain" (68.6%; n = 69), "interruptions to daily life" (60.8%; n = 61), "reduction in time spent on hobbies" (62.7%; n = 63), and costly diagnostic/investigative tests (62.7%; n = 64). The majority of patients (52.9%; n = 53) reported reduced quality of life. The most common treatment prescribed was paracetamol (53.9%; n = 55) and exercise (23.5%; n = 24). Conclusion: The study participants were mainly women, 31-50 years old, who had experienced anxiety, stress, or depression. They mainly experienced fear and self-reported a reduced quality of life.


Introduction
Non-cardiac chest pain (CP) is referred to as atypical chest pain (ACP) or chest pain of unknown cause.In the US emergency department (ED), "from 2006 to 2016, there were 42.48 million chest pain visits" [1].ACP accounts for almost 49-60% of emergency medical admissions presenting with acute central chest pain [2].Spalding et al. reported that 20-30% of all emergency medical admissions were attributable to acute central chest pain, with 49-60% of those categorized as atypical [2].Capewell and McMurray [3] found that more than 50% of patients presenting to the clinic with chest pain (CP) had a low risk for coronary heart disease.Due to uncertainty, the diagnosis is often recorded as atypical chest pain, non-cardiac chest pain, or "chest pain of unknown cause" [4], with its etiology attributed to "anxiety" or "panic."Despite the prevalence of ACP, Eslick [5] lamented the lack of clarity regarding the impact and natural history of non-cardiac chest pain in the community.This has tended to result in suboptimal management as well as compromised quality of life and overall productivity.This study aimed to identify the factors associated with and predictors of ACP and the consequences of ACP.

Materials And Methods
This study was a cross-sectional cohort of patients presenting with atypical chest pain (ACP) at the Accident and Emergency (A&E) Department of a Teaching Hospital in Trinidad and Tobago.The inclusion criteria were as follows: (i) at least 18 years of age, and (ii) a resident or citizen of Trinidad and Tobago.The exclusion criterion was difficulty communicating, not contactable, and missing data.Records from the A&E Departments of the two hospitals showed that, during the study period, 1085 patients presented with the criteria for non-cardiac chest pain.Of these, 17.3% (n = 188) were usable (could be located, no key missing data, no restricted access), and 9.4% (n = 102) of the target population were accepted for analysis (Figure 1).Ethical approval was obtained from the Ethics Committee of the University of the West Indies (UWI) and the Regional Health Authority (RHA), which had jurisdiction over each of the three teaching hospitals.Verbal consent to participate was obtained from all participants.

Psychosocial Factors
Instrument reliability/internal consistency: The reliability and internal consistency (Cronbach's alpha) of each of the four instruments used to measure anxiety, depression, loneliness, and stress are shown in Table 2.The reliability/internal consistency ranged from acceptable (0.754; UCLA_3) to good (0.891; GAD-7).The overall reliability was excellent (0.914; Table 2).

Summary Statistics: Attribute Scores
Table 4 shows selected summary statistics of anxiety, depression, loneliness, and stress scores.GAD-7, PHQ-9, and UCLA-3 scores were skewed left; while PSS-10 scores were normally distributed.

Analysis of Variance (ANOVA): Total Attribute Scores
Table 5 shows the p-values obtained from the application of ANOVA methods to the total score for each of the four attributes.As seen, depression was the only attribute for which p ≤ 0.005.In other words, there were no statistically significant differences between, or among, mean depression scores for any of the sociodemographic variables of interest to the study.

TABLE 5: P-values for analysis of variance (ANOVA).
Multiple comparison methods (Tukey's test) showed multiple associations, especially between anxiety and at least one comorbidity, social support, and less stress.For GAD-7 (anxiety), patients with at least one comorbidity had a significantly higher mean anxiety score than those without comorbidities (p = 0.045).For PHQ-9 (depression), there were no differences between or among the mean depression scores for any sociodemographic variables.For UCLA-3 (loneliness), the mean loneliness score of the patients with hypertension was significantly higher than that of the patients without hypertension (p = 0.028).Patients with no social support had a higher mean score than those with social support (p = 0.027); and patients with at least one comorbidity had a higher mean score than those with no comorbidities (p = 0.017).For PSS-10 (stress), the mean stress score of patients receiving social support was significantly lower than that of patients not receiving social support (p = 0.029).
II. Depression (PHQ-9): Figure 6 shows that levels of depression ranged from moderate to severe (25.5%; n = 26).There were no associated factors among the sociodemographic variables; hence, there were no predictors of depression.IV.Stress (PSS-10): Figure 7 shows that almost two-thirds of the patients (65.7%; n = 67) had moderate stress and that the prevalence of high stress was 9.8% (n = 10).However, the Chi-square analysis showed that there were no associated variables.

Psycho-Social-Financial
Overall, "fear as a result of the pain" (68.6%; n = 69) was the most reported psycho-social consequence, and "fear of having to go to the hospital" (51.0%; n = 52) was the least reported (Figure 8).
The social consequence most reported was interruptions to daily life (60.8%;n = 61), followed by having little time for other activities (54.9%; n = 56), while the least reported was exclusion from family activities (43.1%; n = 44) (Figure 9).

Predictors of Financial Consequences
Binary logistic regression showed that of the three demographic variables associated with financial consequences, monthly income was a predictor of both costly investigative tests and costly treatment modalities, and there were no predictors of reduced attendance at work (Table 8).AOR: adjusted odds ratio; CI: confidence interval.

95% CI for AOR
The other reported consequences are shown in Figure 10.The least reported consequence was "frequent occurrence of pain" (5.7%; n =16), and the most reported was "reduced quality of life" (52.9%; n = 53).
Monthly income was associated with reduced quality of life (Chi-square: 11.480, df = 5, p = 0.043); however, it was not a predictor.

Socio-Demographics and Lifestyle
Patients were noted to be primarily women (63.7%) and of 31-50 years of age (43.1%), corroborating with Eslick et al.'s study [10], which found that 52.0% of patients with ACP were women.The study also found that the majority of patients with ACP had either secondary or tertiary school education (87.3%) and were employed full-time (56.9%) at the time of ACP onset.This study further showed a correlation between income and ACP, with the prevalence being highest in the salary bracket of TT $2500-$5000 (23.5%) and lowest in those earning a monthly income of more than TT $10000 (6.9%).This corroborates the findings of Wong et al. [11], who showed that prevalence declined with increasing income.
The leading comorbidities were a medical history of hypertension (30.4%) or diabetes (18.6%).These findings contrast with the results of Lau et al. [12], who observed that there was an absence of diabetes, hypertension, or prior heart disease in 51.7% of CP patients.However, this correlates with the prevalence observed among patients with chest pain in the ED, as the majority had hypertension (44.5%) or diabetes (33.3%) [4].
The prevalence of lifestyle factors revealed some regular alcohol consumption (8.8%), smoking cigarettes (16.7%), and recreational drugs (6.9%).Smoking prevalence was low in this study compared with a study by Wilhelmsen et al. [13], who found that 41% of participants who engaged in smoking also presented with ACP.Furthermore, in a study by Karlaftis et al. [14], the results showed that 23% of patients with ACP consumed alcohol.However, our study revealed that only 8.8% of the study population engaged in alcohol consumption.Additionally, 66.7%, 73.5%, and 52.9% of the sample did not engage in regular exercise, consume fruits daily, and consume vegetables daily, respectively.

Psychosocial Factors Prevalence and Associations
Our study revealed that the majority of participants experienced at least mild anxiety, and a significant number experienced moderate depression, loneliness, and stress.McDevitt-Petrovic et al. further asserted a correlation between psychological difficulties such as anxiety, depression, panic disorders, and chest pain [15].Most patients had mild to severe anxiety (53.9%), and moderate to severe anxiety occurred in 27.4% of participants.This contrasts with a study by Wertli et al. [16], which highlighted that anxiety disorders are rarely considered a possible cause of ACP by physicians.However, this is notably different from a study by Demiryoguran et al. [17], which stated that 70% of patients had a high Hospital Anxiety and Depression Scale or anxiety score.In this study, a strong association was observed between anxiety and perceived stress.
Our results indicated a prevalence of moderate to severe depression of 25.5% was lower than that reported by Fagring et al. [18] and Lin et al. [19], who found that depression was self-reported in 35% and 50% of the men and women, and in 61.5% of the sample, respectively.This contrasts with the much lower depression levels reported by Eslick et al. [10], who documented the prevalence of clinical depression as 7%.The prevalence of loneliness in this study was 25.5%.Previous studies have demonstrated a correlation between loneliness and ACP [20].Paul et al. [21] posited that loneliness is more common in women and those with lower socioeconomic status.This study showed that 65.7% of the patients with ACP had moderate stress, and 9.8% had high stress.A previous study by Bahall et al. [4] among patients with chest pain at the ED showed that only 2% self-reported having a stressful life.In the present study, stress was significantly associated with anxiety and loneliness.Kim et al. reported a positive correlation between stress [22] in patients with recurrent chest pain.

Psycho-Social-Financial
Overall, participants were psychologically, socially, and financially affected by ACP.In our study, there was a general fear of pain, knowing or not knowing the diagnosis, and visiting the hospital.Participants' greatest fear was "as a result of the pain" (68.6%), and the least but still considerably high was "fear of having to go to the hospital" (51.0%).Similar fear has been reported in other studies [23].The most commonly reported social consequences of ACP were interruptions to daily life (60.8%),followed by having little time for other activities (54.9%), reduced socialization (50.0%), and exclusion from family activities (43.1%).This correlates well with a study by Eslick and Talley, who reported interruptions in daily activities in 63% of participants [24].In this study, there was also a significant reduction in time spent on hobbies (62.7%).Due to uncertainty and attempts to clarify their diagnosis, patients reported costly treatment (56.9%) and costly diagnostic/investigative tests (62.7%) as the most common financial consequences.This finding was similar to that of Eslick et al., who reported a high cost of treatment [25].We found that monthly income was a predictor of costly investigative tests and treatment modalities.The majority of participants reported "reduced quality of life" (52.9%), similar to the low quality of life reported by patients with ACP found by Mourad et al. [26].Another study by Eslick et al. found that 36% of patients with ACP reported much lower quality of life levels [25].

Treatment
The majority of patients used paracetamol (53.9%), and 23.5% used exercise as a post-episode mechanism for self-management of their condition.Other studies have reported that commonly used drugs include proton pump inhibitors, muscle relaxants (nitrates, calcium channel blockers), and pain modulators [27].

Limitations
This study had multiple limitations.This included inappropriate coding of hospital files leading to the rejection of many patients and communication difficulties because of outdated patient phone numbers in the files.Much of the data were based on recall and subjective responses, which carry a certain amount of bias.One significant disadvantage of the study is that it did not include certain anthropometric chest measurements, which have a substantial correlation with ACP [28].

FIGURE 10 :
FIGURE 10: Reported consequences of chest pain episode.

Table 7
shows patients' ethnicity, monthly income, whether or not they had social support, and the association of these factors with at least one of the financial consequences.