Latent profile analysis of biopsychosocial measures in older patients with (un)explained persistent somatic symptoms

Abstract

Patients with persistent somatic symptoms (PSS) often receive either somatic or psychiatric care, depending on whether symptoms are respectively medically explained (MES) or unexplained (MUS). This separation may not be as clinically relevant as previously assumed; however, research on data-driven subgroups within cohorts of older patients with PSS is not available. Our goal is to identify more clinically relevant homogeneous subgroups beyond the distinction of MUS and MES among older patients with PSS by using a data-driven approach. We performed two Latent Profile Analyses (LPAs), one focused on 6 somatic health measures, the other on 6 psychosocial measures, using data from a case-control study with participants (>60 years) with MUS (N = 118) or MES (N = 154), recruited from the general public, general practices and secondary healthcare. We identified two somatic-health based (strong, vulnerable) and four mental-health based (strong, vulnerable, lonely, non-acceptance) profiles. We found no statistically significant overlap between the somatic – and mental health profiles (κ = 0.019). Health related quality of life negatively associated with the mentally – and somatically vulnerable profiles. We conclude that singular focus on MUS/MES distinction in the care for older PSS patients may lead to undertreatment of the most vulnerable patients. Integrated care is recommended when treating persistent somatic symptoms in later life, regardless of the (un)explained origin of the symptoms.