Somatic symptom disorder symptoms in individuals at risk for heart failure: A cluster analysis with cross-sectional data from a population-based cohort study

Abstract

Objective

Identifying whether experienced symptom burden in individuals with medical predisposition indicates somatic symptom disorder (SSD) is challenging, given the high overlap in the phenomenology of symptoms within this group. This study aimed to enhance understanding SSD in individuals at risk for heart failure.

Subjects and methods

Cross-sectional data from the Hamburg City Health Study was analyzed including randomly selected individuals from the general population of Hamburg, Germany recruited from February 2016 to November 2018. SSD symptoms assessed with the Somatic Symptom Scale-8 and the Somatic Symptom Disorder-12 scale were categorized by applying cluster analysis including 412 individuals having at least 5% risk for heart failure-related hospitalization within the next ten years. Clusters were compared for biomedical and psychological factors using ANOVA and chi-square tests. Linear regressions, adjusting for sociodemographic, biomedical, and psychological factors, explored associations between clusters with general practitioner visits and quality of life.

Results

Three clusters emerged: none (n = 215; 43% female), moderate (n = 151; 48% female), and severe (n = 46; 54% female) SSD symptom burden. The SSS-8 mean sum scores were 3.4 (SD = 2.7) for no, 6.4 (SD = 3.4) for moderate, and 12.4 (SD = 3.7) for severe SSD symptom burden. The SSD-12 mean sum scores were 3.1 (SD = 2.6) for no, 12.2 (SD = 4.2) for moderate, and 23.5 (SD = 6.7) for severe SSD symptom burden. Higher SSD symptom burden correlated with biomedical factors (having diabetes: p = .005 and dyspnea: p ≤ .001) and increased psychological burden (depression severity: p ≤ .001; anxiety severity: p ≤ .001), irrespective of heart failure risk (p = .202). Increased SSD symptoms were associated with more general practitioner visits (β = 0.172; p = .002) and decreased physical quality of life (β = −0.417; p ≤ .001).

Conclusion

Biomedical factors appear relevant in characterizing individuals at risk for heart failure, while psychological factors affect SSD symptom experience. Understanding SSD symptom diversity and addressing subgroup needs could prove beneficial.