Although research has identified women's prodromal and acute myocardial infarction (MI) symptoms, diagnosing coronary heart disease in women remains challenging. Knowing how individual symptoms cluster by race and other characteristics would provide key diagnostic information. We performed a secondary analysis to: (a) generate naturally occurring symptom clusters based on prodromal and acute MI symptom scores separately, (b) examine the association between women's characteristics and symptom clusters, and (c) describe the percentage of women who reported experiencing the same symptoms in both prodromal and acute MI phases.
The database contained retrospective self-reported data obtained by telephone survey from 1270 women (43% black, 42% white, 15% Hispanic) with a confirmed MI recruited from 15 geographically diverse sites. Data included frequency and severity of 33 prodromal symptoms, intensity of 37 acute MI symptoms, and comorbidities/risk factors. We used both bivariate and multivariate analyses to examine associations between cluster assignment and characteristics/risk factors. Because of the possibility of complex interactions, we explored nonlinear interactions with recursive partitioning.
Cluster analysis yielded 3 naturally occurring clusters for each of the prodromal and acute symptom sets. Each cluster contained women who reported increasing frequency and severity of symptoms. Six characteristics (age, race, body mass index, personal history of heart disease, diabetes, smoking status) were strongly associated with the clusters. Body mass index was the most important factor in classifying prodromal symptoms, whereas age was for acute symptoms.
Black women younger than 50 years were more likely to report frequent and intense prodromal symptoms, whereas older white women reported the least. Younger, obese, diabetic black women reported the most acute symptoms, whereas older nonobese, nondiabetic white women reported the fewest. Symptom clusters and characteristics of women in these clusters provide valuable diagnostic information. Further research with a control group is needed.
Jean C. McSweeney, PhD, RN, FAHA, FAAN Associate Dean for Research, College of Nursing, University of Arkansas for Medical Sciences, Little Rock.
Mario A. Cleves, PhD Professor, College of Medicine, University of Arkansas for Medical Sciences, Little Rock.
Weizhi Zhao, MS Biostatistician, College of Medicine, University of Arkansas for Medical Sciences, Little Rock.
Leanne L. Lefler, PhD, APN Assistant Professor, College of Nursing, University of Arkansas for Medical Sciences, Little Rock.
Shengping Yang, MS Statistical Analyst, Department of Statistics, St Jude Children's Research Hospital, Memphis, Tennessee.
The National Institute of Nursing Research supported this work with 2 grants, 1 RO1 NR04908 and 1 R01 NR05265.
Correspondence Jean C. McSweeney, PhD, RN, FAHA, FAAN, College of Nursing, 4301 W Markham St, Slot 529, Little Rock, AR 72205 (firstname.lastname@example.org).