Unexplained versus explained symptoms: The difference is not in patients' language use. A quantitative analysis of linguistic markers

Inge Stortenbeker; Tim Olde Hartman; Anita Kwerreveld; Wyke Stommel; Sandra van Dulmen; Enny Das

doi:10.1016/j.jpsychores.2021.110667

Unexplained versus explained symptoms: The difference is not in patients' language use. A quantitative analysis of linguistic markers

J Psychosom Res. 2021 Nov 8:152:110667. doi: 10.1016/j.jpsychores.2021.110667. Online ahead of print.

Authors

Inge Stortenbeker¹, Tim Olde Hartman², Anita Kwerreveld³, Wyke Stommel³, Sandra van Dulmen⁴, Enny Das³

Affiliations

¹ Centre for Language Studies, Radboud University, Nijmegen, the Netherlands. Electronic address: inge.stortenbeker@ru.nl.
² Radboud University Medical Center, Radboud Institute for Health Sciences, Department of Primary and Community Care, Nijmegen, the Netherlands.
³ Centre for Language Studies, Radboud University, Nijmegen, the Netherlands.
⁴ Radboud University Medical Center, Radboud Institute for Health Sciences, Department of Primary and Community Care, Nijmegen, the Netherlands; NIVEL (Netherlands institute for health services research), Utrecht, the Netherlands; Faculty of Health and Social Sciences, University of South-Eastern Norway, Drammen, Norway.

PMID: 34775157
DOI: 10.1016/j.jpsychores.2021.110667

Abstract

Objective: Patients with medically unexplained symptoms (MUS) are believed to have a deviant way of talking about complaints. This study systematically compared linguistic markers in symptom presentations of patients with MUS and medically explained symptoms (MES).

Methods: This content analysis (cross-sectional study) conceptualized relevant linguistic markers based on previous research about MUS communication. Linguistic markers included negations ("not"), intensifiers ("very"), diminishers ("a little"), first or third person subject ("I" vs. "my body"), subjectivity markers ("I think") and abstraction ("I'm gasping for breath" vs. "I'm short of breath"). We also coded valence, reference to physical or mental states, and consultation phase. We compared 41 MUS and 41 MES transcribed video-recorded general practice consultations. Data were analyzed with binary random intercepts models.

Results: We selected and coded 2752 relevant utterances. Patients with MUS used less diminishers compared to patients with MES, but this main effect disappeared when consultation phase was included as predictor. For all other linguistic variables, the analyses did not reveal any variation in language use based on whether patients had MUS or MES. Importantly, utterances' valence and reference to physical or mental state did predict the use of linguistic markers.

Conclusion: We observed no systematic variations in linguistic markers for patients who suffered from MUS compared to MES. Patients varied their language use based on utterances' valence and reference to physical or mental states. Current ideas about deviant patient communication may be based on stigmatized perceptions of how patients with MUS communicate, rather than actual differences in their talk.

Keywords: General practice; Linguistic analysis; Observational coding; Unexplained symptoms.