Depression symptoms across cultures: an IRT analysis of standard depression symptoms using data from eight countries

Abstract

Purpose Prevalence estimates of depression vary between countries, possibly due to differential functioning of items between settings. This study compared the performance of the widely used Hopkins symptom checklist 15-item depression scale (HSCL-15) across multiple settings using item response theory analyses. Data came from adult populations in the low and middle income countries (LMIC) of Colombia, Indonesia, Kurdistan Iraq, Rwanda, Iraq, Thailand (Burmese refugees), and Uganda (N = 4732). Methods Item parameters based on a graded response model were compared across LMIC settings. Differential item functioning (DIF) by setting was evaluated using multiple indicators multiple causes (MIMIC) models. Results Most items performed well across settings except items related to suicidal ideation and “loss of sexual interest or pleasure,” which had low discrimination parameters (suicide: a = 0.31 in Thailand to a = 2.49 in Indonesia; sexual interest: a = 0.74 in Rwanda to a = 1.26 in one region of Kurdistan). Most items showed some degree of DIF, but DIF only impacted aggregate scale-level scores in Indonesia. Conclusions Thirteen of the 15 HSCL depression items performed well across diverse settings, with most items showing a strong relationship to the underlying trait of depression. The results support the cross-cultural applicability of most of these depression symptoms across LMIC settings. DIF impacted aggregate depression scores in one setting illustrating a possible source of measurement invariance in prevalence estimates.

How is depression experienced around the world? A systematic review of qualitative literature

“…Most previous research in this area has used quantitative data to compare differing patterns of symptom endorsement (e.g., Dere et al, 2013; Haroz et al, 2016). The reviews of qualitative studies that do exist have evaluated related topics, including perceived causes and preferred treatments for depression (Hagmayer and Engelmann, 2014), perceived barriers to accessing treatment for postpartum depression (Dennis and Chung-Lee, 2006), and perceived risk factors for postpartum depression in Sub-Saharan Africa (Wittkowski et al, 2014).…”

Measurement invariance of the Hopkins Symptoms Checklist: a novel multigroup alignment analytic approach to a large epidemiological sample across eight conflict-affected districts from a nation-wide survey in Sri Lanka

“…The bi-factorial structure (anxiety and depressive symptoms) has been supported by studies across diverse cultures, for example, for Southeast Asia [20] and Afghanistan [14]. A recent item response analysis conducted by Haroz and colleague [27] based on the HSCL-15 supported the cross-cultural equivalence of depression symptoms amongst ethnically and linguistically diverse conflict-affected populations from eight low-income countries (Colombia, Indonesia, Iraq, Rwanda, Kurdistan Iraq, Thailand, and Uganda). In addition, although all items showed some degree of differential item functioning (DIF), Indonesia being the only country where the prevalence estimate of depression could have been overestimated due to possible measurement variance [27].…”

Item and domain network structures of the Resilience Scale for Adults in 675 university students

“…Network analysis has also been applied to several psychological constructs, such as personality (Costantini et al, 2015), empathy (Briganti et al, 2018), attitudes (Dalege et al, 2017, intelligence (Van Der Maas et al, 2006) and self-worth (Briganti et al, 2019). Other studies used innovative methods, including networks to harmonize rating scales (Gross et al, 2018;Purgato et al, 2018;Haroz et al, 2016) Learning the network structure of a given construct (such as resilience) or mental disorder (such as PTSD) is particularly relevant in clinical practice since it highlights potential clinical target that may affect multiple symptoms or elements composing the network (Fried et al, 2018); for instance, intervening on the connection between two components of the network is likely to modify the clinical presentations of said components (such as symptoms). In the Running head: NETWORK ANALYSIS OF RESILIENCE specific case of resilience, which is considered a protection against mental disorders, learning the network structure of resilience components may highlight potential targets to strengthen the overall mental health of a given individual.…”