Towards precision in the diagnostic profiling of patients: leveraging symptom dynamics in the assessment of major depressive disorder

Abstract

Major depressive disorder (MDD) is a heterogeneous mental disorder. International guidelines present overall symptom severity as the key dimension for clinical characterisation. However, additional layers of individual differences may reside within severity levels related to differences in how symptoms interact with one-another in a given patient, referred to as symptom dynamics. We investigate these individual differences by estimating the proportion of patients that display differences in their symptom relationship patterns while sharing the same overall symptom severity. Methods:Patients with MDD recruited at four centresin the Netherlandsbetween 2016-2018 rated their baseline symptom severity using the Inventory for Depressive SymptomatologySelf-Report (IDS-SR).Momentary indicators for symptoms were collected throughEcological Momentary Assessmentsscheduled to measure each patient fivetimes per day for 28 days.Each patient’s symptom dynamics were estimated using dynamicnetworks based on thegraphical vector autoregressive model. Individual differences in these symptom relationshippatterns in groups ofpatients sharing the same symptom severity levels were estimated using Individual Network Invariance Tests, before the overall proportion of patients that displayed differential symptom dynamics while sharingthe same symptom severity was calculated. To compute95% bootstrapped confidence intervals around this proportion, 10,000 re-estimations followingrandom draws with replacement of the symptom severity groups were performed. A supplementary simulation study was conducted to investigate the accuracy of our methodology by identifying its average false positive detection rate in a simulated scenario where no individual differences should be present. 3ResultsOut of 74patients, 73 were analysed (Mage= 34.57[SD13.12]; 56.16% females; 63.01% employed) and 8,395 observationswere conducted across the 28-day period (average completed assessments per person: 115; SD16.81).23 severity levels (IDS-SR values) were observed in the sample, enabling the investigation of individual differences in 23 groups of MDD patients(between 2 to 6 patientsper group)sharing the same severity. Differential symptom dynamics were identified across 63.01% (95% bootstrapped CI 40.98, 82.05)ofpatients displaying the same severity. The simulation study revealedour method’saverage false detection of individual differencesin our scenarioto be 2.22%.ConclusionsThemajority of MDD patients sharing the same symptom severity displayed differential symptom dynamics. Our findings wererobust against false positive conclusions. Examining symptom dynamics provides information about the person-specific psychopathological expression of patients beyond severity levelsby revealing how symptoms aggravate each other over time. These resultssuggest thatsymptom dynamicsmay serve as a promisingdimension for clinical characterisation, warranting replication in independent samples.Toinformpersonalisedtreatmentplanning,a next step concerns linking different symptom relationship patterns totreatment response andclinical course,including patterns related to spontaneous recovery andforms of disorder progression