Chest Pain Among Adults Presenting at the A&E Department of a Public Tertiary Health Care Institution During a 2-Year Period

1. Introduction

Chest pain is a frequent presentation in emergency departments (EDs) worldwide. In Trinidad, chest pain accounts for 8.5% of emergency admissions (Bahall, 2021). The principal clinical goal is prompt identification or exclusion of potentially life-threatening causes. Clinicians must also identify non‑cardiac causes (e.g., musculoskeletal, gastrointestinal, or psychological), which account for 44.7% (Wertli et al., 2019) to 55.1% (Khand et al., 2023) of all chest pain presentations. Many patients are treated and discharged, some are admitted, and the remainder are followed up as outpatients. Life-threatening conditions such as aortic dissection, pericarditis, pulmonary embolism, and myocardial ischemia (Kontos et al., 2010) are critical and must be identified early to prevent morbidity and mortality, treat and prevent complications (Hamm et al., 1997) and increase survival (Beck et al., 2015) However, misdiagnosis continues to occur, with acute myocardial infarction (AMI) missed in up to 2.1% of ED cases (Schull et al., 2006), pulmonary embolism in 7.5% of cases (Kwok et al., 2022) and acute aortic dissection in 14.1% (Zhan et al., 2012) to 33.8% (Lovatt et al., 2022) of cases. Misdiagnosis may result from physician-related factors such as limited experience, inadequate histories, or misinterpretation of ECGs, which further exacerbate these risks (Rusnak et al., 1989). Other contributors to misdiagnosis include atypical symptom presentation, lack of classic clinical features and presence of coexisting conditions (Lovatt et al., 2022). In low- and middle-incomecountries, suboptimal management is amplified: approximately: 55% of patients with chest pain in a major Pakistani ED underwent ECGs, and fewer than 5% received cardiac enzyme testing in some reported cohorts (Paichadze et al., 2015).  In addition, patients with a lowersocioeconomic status are reportedly ‘disproportionally affected, exhibiting longer wait times, less frequent use of diagnostic tests, and higher rates of revisits and adverse outcomes, even in universal healthcare systems (Herlitz et al., 2023).

Caribbean countries have similar barriers and challenges in ED settings, such as insufficient materials and resources, delayed flow of patients, increased workload, and shortage of staff and skills (De Freitas et al., 2020). Inadequate elicitation and documentation (Kachalia et al., 2007) of medical history and physical examinations can directly contribute to diagnostic errors and death. Other factors that compromise care includeincorrect interpretation of diagnostic tests, inadequate handoffs, excessive workloads (Kachalia et al., 2007) and inadequate knowledge/skill/reasoning (Newman-Toker et al., 2022). The necessity for proper evaluation may also prevent “avoidable admissions” without incurring significant costs to the patient or healthcare system (Groarke et al., 2013) or even avoid unnecessary admissions (Bhatti et al., 2019). The study investigated patients who presented to the emergency department with chest pain. The specific objectives were (1) to estimate the prevalence of symptom presentation; (2) to determine sociodemographic characteristics; (3) to determine the diagnosis, treatment, and dispatch of patients; and (4) to determine any associations between symptoms/diagnosis and sociodemographic factors.

2. Material and Methods

This study was retrospective in nature and included all adults presenting with self-diagnosed chest pain at the Accident and Emergency (A&E) Department of the largest public tertiary health care institute in Trinidad and Tobago during a 2-year period from 2015 to 2016.  No attempt was made to determine a sample because a sampling frame was unavailable for the random selection of participants.

Before data collection, all staff members (nurses, doctors, clerks, etc.) were briefed on the nature of the study and the importance of identifying and retrieving accurate information.  Data were obtained from a sample of patient files accessible from the A&E Department of the hospital. Records of all patients with chest pain—obtained from the registration book, which included patients’ names and diagnoses—were reviewed on a daily basis, except on weekends and public holidays. The inclusion criterion was an age of at least 18 years. This study had no exclusion criteria. Data on sociodemographic variables were collected by using a questionnaire (9) and included age, sex, ethnicity, marital status, employment status, monthly income, height, weight, waist and hip circumference, presenting symptoms and severity of chest pain (26), self-reported medical history (14), admission and time profile (7), ECG findings (1), blood investigations (13), other tests (4), ACS scores (3), A&E diagnosis (9), A&E treatment (9) and A&E dispatch (5).

Data were analyzed using the Statistical Package for the Social Sciences (SPSS) v23 via descriptive and inferential data analysis methods. Descriptive methods included frequency and percentage distribution tables, graphs, charts, and selected summary statistics. Inferential methods included 95% confidence intervals for the mean and proportion, hypothesis testing, including analysis of variance (ANOVA), and nominal and ordinal logistic regressions.

Ethical approval was granted on April 28, 2015, by the Clinical Governance and Ethics Committee of the South-West Regional Health Authorityof Trinidad and Tobago.

3. Results

3.1 Demographics

Sex was nearly equally divided between male (n = 1006, 48.6%) and female (n = 1058, 51.2%) and patients ranged in age from 18 years to over 75 years (Table I). In particular.  Specifically, the percentage of patients 75 years or older (n = 211; 10.2%) was nearly twice the percentage of patients aged 18–24 years (n = 111; 5.4%). More than one-half of patients were of Indo-Trinbagonian descent (n = 1174; 56.8%; Table 1).

The mean age was 53.4 years (SD = 16.76 years), with the exception of individuals aged 75 years and older. The number of persons presenting increased with increasing age category. The ANOVA showed no difference between the mean ages of male and female participants (p = 0.304) or among any of the four ethnic groups (p = 0.991).

3.2 Medical History     

3.2.1 Comorbidities of Chest Pain

Hypertension (n = 941, 45.5%), previous heart attack (n = 694, 33.6%), diabetes (n = 658, 31.8%), and ischemic heart disease (IHD) (n = 541, 26.3%) were the most comorbid conditions (Table 2). 

NA: not applicable

**Self-reported history of heart attack may be inaccurate as some individuals equate a previous hospital admission for chest pain with having had a myocardial infarction, despite the absence of a confirmed diagnosis. Some patients may similarly believe that they have hypercholesterolemia simply because they are prescribed cholesterol-lowering therapy. 

3.2.2 Lifestyle and Exercise Habits

Smoking (n = 274; 13.3%) was the leading self-reported lifestyle-related risk factor, followed by alcohol consumption (n = 181, 8.8%) and abdominal obesity (n = 9, 0.4%). More than four-fifths of patients (n = 1737; 56.1%) reported not consuming fruit and vegetables daily; the remaining 331 patients did not respond.

3.3 Presenting Symptoms

Most patients (n= 875, 42.3%) presented with shortness of breath, followed by nausea (n = 417, 20.2%), palpitations (n = 364, 17.6%), light-headedness (n = 245, 11.9%), and vomiting (n = 207, 10.0%). Nearly one-third of patients (n = 694 ;33.6%) reported having experienced at least one previous heart attack (Table 3).

3.4 Associations Between Sociodemographic Characteristics and Health Conditions

Chi-square tests revealed the following significant associations (Table 4):

1.      Sex and palpitations (p = 0.011)

2.      Age group and nausea (p ≤ 0.001)

3.      Age group and fever (p = 0.002)

4.      Sex and fatigue (p = 0.048)

5.      Age group and sweating (p = 0.032)

6.      Age group and pleuritic pain (p = 0.026)

7.      Age group and coughing (p = 0.015)

8.      Age and shortness of breath (p = 0.001)

Table 4 shows that age was associated with six symptoms, including nausea, fever, and sweating, whereas sex was associated with palpitations and fatigue only.

3.5 Predictors of Presenting Conditions

Chi-square analysis and subsequent logistic regression identified the predictors of presentation at the A&E with at least one symptom: palpitations, fever, or cough (Table 5).

The following were revealed through corresponding odds ratios (ORs):

Sex was a predictor of palpitations (p =0.001). In particular, women were almost twice as likely to present with this condition than men.

Age group was predictor of presenting with fever:

Among patients in the 18- to 24-year-old and 25- to 34-year-old age groups, compared with patients 75 years or older, the odds of presenting with fever decreased with increasing age.

Age group was predictor of presenting with coughing.

In particular, among patients 18–24 and 25–34 years old. compared to patients aged 75 years or older, the odds presenting with coughing decreased with increasing age.

Age group was predictor of presenting at the A&E with shortness of breath.

The odds of shortness of breath among patients 45–54 years of age were significantly higher than those of patients 75 years of age or older

3.6 A & E Diagnoses

The diagnostic status was not documented in 1899 (91.8%) of persons who presented at A&E with chest pain.  Table VI.  Muscular pain (n = 86, 4.2%) was the leading diagnosis, followed by gastroesophageal reflux disease (GERD) (n = 48, 2.3%) and Unstable Angina (n = 43, 2.1%). Pulmonary Embolism (n =1, 0.1%) was the least prevalent. None of these patients had AD (Table 6).

NB. Persons may have ≥1 diagnosis; therefore the total may be >2068 and >100%.

*May include patients with atypical chest pain.

Figure 1: Emergency medical diagnosis

A&E, accident and emergency; STEMI, ST-elevation myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; GERD, gastroesophageal reflux disease

3.7 Treatment/Triage Time

Table 7 shows the length of time (in hours) from arrival at A&E to the start of triage for the 228 patients for whom both figures were recorded by age group, male sex, female sex, and the combined sex.

ANOVA showed no statistically significant differences between or among the mean times to triage of male and female patients (p = 0.692), among the seven age groups (p = 0.704,); or among the four ethnic groups (p = 0.704).

3.7.1 Treatments

Post-triage treatments administered at A&E included the three leading treatments: aspirin (n=482, 23.3%), clopidogrel (n=450, 21.8%), and heparin (n= 456, 22%; Table 8).

3.8 A&E Dispatch

Dispatch information was unavailable for 1179 (57.0%) patients presenting to the A&E department with chest pain (Fig.2). Most. 24.4% (n = 505%) patients were dispatched to the medical ward, 11.3% (n =234) were discharged fully, and less than 1% each were sent home with review and with instructions to attend the clinic, respectively.

Figure 2: Patient Triage Dispatch

4. Discussion

The study participants were primarily female (n = 1058, 51.2%), between 45 and 64 years (n = 876, 42.3%), and Indo-Trinbagonian (n=1174, 56.8%). These demographics were similar to other studies conducted in Pakistan which showed that most patients were male (60%), and the mean age was 42 years (Paichadze et al., 2015). Another study reported a male-to-female ratio of 2:1 (Fouda et al., 2023). The overall mean age of the participants was 53.4 (SD = 16.76), which was similar to the findings of previous studies which reported mean ages of 53.2 years (Chanda et al., 2023) and 56.7 years (Jayasekera et al., 2024).

Patients were primarily hypertensive (n = 941, 45.5%), diabetic (n = 658, 31.8%), and had a history of ischemic heart disease (IHD) (n = 541, 26.3%). Chanda et al. (2003) similarly found the prevalence of hypertension, diabetes and coronary artery disease was 35.5%, 30.5%, and 19.5, respectively.  Mateen et al. (2024) found the prevalence of hypertension, diabetes and family history of IHD among patients with cardiac chest pain was 67%, 40%, and 46%, respectively. Approximately one-third of the participants (n = 694, 33.6%) in our study reported experiencing at least one previous heart attack. However, such self-reported history may be inaccurate as some individuals equate previous hospital admission for chest pain with having had a myocardial infarction. The Caribbean region has the highest prevalence of elevated blood pressure in the Americas, with estimates ranging from 20.9% to 27.1% (Caribbean Public Health Agency, 2021). In particular, studies conducted in Trinidad and Tobago, report hypertension rates exceeding the regional average (Alvarado, 2023) and diabetes mellitus in approximately 14.5% of the population (Office of the President of the Republic of Trinidad and Tobago, 2019). Lifestyle factors include abdominal obesity (n = 9; 0.4%), smoking (n = 274; 13.3%), and alcohol consumption (n = 181; 8.8%). Compared with our study, a similar investigation of the clinico-epidemiological profile of patients presenting with acute chest discomfort at the ED in India reported a higher smoking prevalence (48.5%) but a lower rate of alcohol consumption (3.5%) (Chanda et al., 2023). None of the participants reported daily consumption of fruits and vegetables, 1733 (83.8%) participants reported that they did not do so daily, and; the remaining 334 (16.2%) patients did not respond.

4.1 Presenting Symptoms

Of the cohort of patients who presented with chest pain, atypical and classical chest pain presentations were documented in 1.1% (n = 22) and 5.0% (n = 103) respectively. Other studies reveal prevalences of 39.5% and 60.4% for atypical and classical/typical chest pain, respectively (Tuyyab et al., 2017).  Clinicians may not have assigned a specific diagnosis because they lacked sufficient clinical information or were reluctant to commit to a diagnosis without certainty. Undocumented cases may include many of these cases as well. Similar findings such as a lack of adequate information or simply did not wishing to commit to a diagnosis have been reported in several studies (Carlton et al., 2015; Sagristà-Sauleda & Barrabés, 2020; Swap & Nagurney, 2005).

A large proportion of patients presented with shortness of breath (n= 875, 42.3%) followed by nausea (n = 417, 20.2%) and palpitations (n = 364, 17.6%).  Strangely certain symptoms are associated with sex and age. Sex was associated with palpitations (p = 0.011) and fatigue (p = 0.048). Similar findings have been reported were (Milner et al., 1999; O’Donnell et al., 2012). Age was associated with nausea (p < 0.001), fever (p = 0.002), sweating (p = 0.032), pleuritic pain (p = 0.026), cough (p = 0.015), and shortness of breath (p = 0.001).  However, these associations have not been previously reported.

4.2 A&E Diagnosis/Dispatch

In our study, the prevalence of ACS was 3.7% (n=773)%, and that of STEMI, NSTEMI, and unstable angina was 0.8% (n =16), 0.9% (n = 18), and 2.1% (n = 43),) respectively, among patients presenting with chest pain in the ED which was much lower than that of other similar studies: 13% (n=26) (Fouda et al., 2023) and 49.7% (n= 566) (Asma et al., 2018).  Muscular pain (n = 86; 4.2%) was the leading cause of chest pain, followed by GERD (n = 48; 2.3%). This finding contrasts a study by Khan et al. (2023), which reported higher prevalence of musculoskeletal and gastrointestinal chest pain as 22.9% and 15.7%, respectively.  These low figures may have resulted from the large number (n = 1899, 91.8%) of patients who were undocumented. One patient presented with pulmonary embolism. This number is quite small when compared to other similar studies which reported life-threatening diagnoses such as APE (0.61%), aortic dissection (0.18%), and pericarditis/tamponade (1.23%) (Asma et al., 2018). 

Dispatch information was unavailable for 1179 patients (57.0%) presenting to the A&E department with chest pain. In the absence of these records, it was assumed that the patients were discharged. Among individuals for whom data were recorded, 24.4% (n = 505) and 1.7% (n = 35) were dispatched to the medical ward and cardiac ward, respectively, and 11.3% (n = 234) were allowed to return home without further follow-up. Less than 1% were sent home with a follow-up review and instructions to attend the clinic. Leite et al. (2015) found that in a cohort of emergency department chest pain presentations, 81.1% of patients (n = 189) were discharged home, while 18.9% (n = 44) were admitted to the hospital for further management. Chanda et al. (2023) similarly found that 49.7% were warded and 40.2% were discharged from the ED. 

4.3 Treatment/Timings

Emergency triage time (i.e., time between arrival and the initiation of triage) was completed in only 228 patients (11.0%).  The patient’s triage time ranged from almost immediately upon presentation at the A&E (i.e. time to triage = 0 h) to 24 h. This time contrasts with the expected standard of care (Byrne et al., 2023). Srivatsa et al. (2025) found that 43.4% of patients with acute chest pain met the <10 min ECG goal. Based on the European Society of Cardiology (ESC) guidelines, chest pain requires require prompt and early assessment, focused medical history, physical examination, vital sign monitoring, ECG, blood investigations, and necessary medication (Byrne et al., 2023). 

Our study revealed that 23.3% (n = 482) of patients received aspirin, 21.8% (n = 450) received clopidogrel, and 22% (n = 456) received heparin. These data strongly suggested that a substantial proportion of patients, approximately 23%, were managed as though they had acute coronary syndrome (i.e., STEMI, NSTEMI, or unstable angina), despite the absence of explicit documentation or formal documentation. These figures are lower than that found in other studies which revealed that aspirin use was 41.9% and heparin was 32.9% in patients with suspected ACS (Khursheed at al., 2015).

4.4 Documentation

Our study revealed lack of documentation regarding triage timing (n = 1840, 89%), medical diagnosis (n = 1899, 91.8%), and dispatch (n = 1179, 57%). This finding is similar to that of other studies that showed suboptimal data quality, with data completeness being only 57.3% in emergency medical records (Paichadze et al., 2015); incomplete documentation contributed significantly to missing or insufficient clinical information.

4.5 Design/Study Limitations

The limitations include the use of nonrandom sampling, lack of proper documentation, and failure to document provisional emergency diagnoses.

4.6 Conclusion and Recommendation

A major problem is documenting of diagnosis, treatment, and dispatch. The most commonly documented diagnoses were muscular pain and ACS, with antiplatelets being the most commonly prescribed drug. Proper guidelines, preferably in areas dedicated to chest pain, are mandatory to optimize patient care and prevent complications. Data collection templates and improved surveillance and data management techniques, including continuous training and quality control, are needed.

Author Contributions:   "Conceptualization, M.B.; Methodology, M.B; Software, G.L; Validation, M.B; Formal Analysis, G.L; Investigation, M.B; Resources, MB; Data Curation, G.L; Writing – Original Draft Preparation, M.B; Writing – Review & Editing, M.B, G.L; Visualization, M.B; Supervision, M.B; Project Administration M.B.”

Funding: Not applicable.

Competing interests: The authors declare that they have no competing interests.

Informed Consent Statement/Ethics approval: This study was approved by the Ethics Committee of the South-West Regional Health Authority and University of the West Indies. The requirement for informed consent was waived.

Data Availability Statement: The data supporting the findings of this study are available from the corresponding author upon request.

Acknowledgements: We would like to thank premedical students Rajiv and Brittany, who assisted in data collection, and Dr Ashmanie Ramjit,who assisted in the preparation of the manuscript.

Declaration of Generative AI and AI-assisted Technologies: This study has not used any generative AI tools or technologies in the preparation of this manuscript.