Subgrouping patients with comorbidity between functional dyspepsia and irritable bowel syndrome: application of multidimensional item response theory and latent profile analysis

Abstract

AbstractBackgroundComorbidity between different subtypes of functional gastrointestinal disorders(FGIDs) is of high prevalence in clinical practice. Heterogeneity of clinical appearances has led to difficulty in individualized diagnosis and comprehensive management of FGIDs.AimsTo discover the hidden clinical patterns of patients with comorbidity between functional dyspepsia(FD) and irritable bowel syndrome(IBS).MethodsIn a retrospective cross-section study, a self-report questionnaire that consist of items indicating 5 different assessing domains including gastrointestinal discomforts, systemic discomforts, psychological disorders, and environmental aggravated factors was used as basic instrument for clinical assessment. With item response theory, the theoretical framework of assessment was evaluated, and latent traits of patient were quantified in the simulated computerized adaptive testing. Latent class analysis was used for uncovering the hidden patterns over the heterogenous clinical appearances. And differences among the profiles were compared referring to the spectrum of clinical appearances and the clinical diagnosis.ResultsWith 996 patients enrolled in the study, the validity and reliability of the instrument were evaluated as adequate (Cronbach’s alpha indices =0.72, Split-half reliability =0.84). The construct validity was also evaluated to be adequate with Chi-square/df=3.45, CFI=0.92, GFI=0.96, RMSEA=0.05, TLI=0.90, RMR=0.02. The 7-profile model was evaluated to be with better fitness with Entropy=0.98, Lo-Mendell-Rubin likelihood ratio test-pvalue<0.01, Bootstrap likelihood ratio test-pvalue<0.01. And the patterns detailed the heterogeneity of clinical appearance of FGIDs patients either in general condition or discomfort in specific dimension.ConclusionsWith application of multidimensional variable analysis, this article summarized the hidden patterns beneath the heterogenous clinical features. And quantitative approaches helped equip clinician with individualized and comprehensive tool in the management of complex diseases such as FGIDs.