How insomnia classification influences prevalence estimates and clinical comparisons in chronic musculoskeletal pain: a cross-sectional study

Background
Insomnia is common in people with chronic musculoskeletal pain (CMP). However, insomnia is often operationalised in various way, impairing the comparability between studies and limiting clinical integration of findings.
Objectives
This secondary exploratory analysis of data from a pre-registered cross-sectional survey aimed to investigate how varying insomnia classifications affect insomnia prevalence and the outcomes of comparisons of clinical parameters between insomnia and non-insomnia groups in individuals with CMP.
Methods
Data from 1322 individuals with CMP were used for the analyses. Insomnia was classified in six different ways (Strictest classification: modified DSM-5-based classification – most inclusive classification: any nighttime symptoms). Prevalence rates were examined across the different classifications, and statistical comparisons of clinical parameters (e.g., pain-, mood-, and sleep-related measures, each assessed with a single item) between insomnia and non-insomnia groups were conducted, with standardised mean differences estimated to assess how classification choices influence effect sizes.
Results
Stricter classifications led to substantially lower prevalence rates compared to more inclusive classifications with differences of up to 25 percentage points (53.33% vs. 78.59% for the strictest and most inclusive definitions). Statistically significant differences between insomnia and non-insomnia groups were found for all variables and classifications. However, for 25 of 27 variables, at least one disagreement in effect size category was observed between classifications, with less stringent classifications consistently yielding larger effect sizes.
Conclusions
The results of this study highlight that clinicians and researchers should carefully consider used classification criteria when interpreting insomnia prevalence rates and related clinical outcomes in the context of CMP.