A latent profile analysis of the typology of bulimic symptoms in an indigenous Pacific population: evidence of cross-cultural variation in phenomenology

Abstract

Background: Previous efforts to derive empirically based eating disorder (ED) typologies through latent structure modeling have been limited by the ethnic and cultural homogeneity of their study populations and their reliance on DSM-IV ED signs and symptoms as indicator variables.

Method: Ethnic Fijian schoolgirls (n=523) responded to a self-report battery assessing ED symptoms, herbal purgative use, co-morbid psychopathology, clinical impairment, cultural orientation, and peer influences. Participants who endorsed self-induced vomiting or herbal purgative use in the past 28 days (n=222) were included in a latent profile analysis (LPA) to identify unique subgroups of bulimic symptomatology.

Results: LPA identified a bulimia nervosa (BN)-like class (n=86) characterized by high rates of binge eating and self-induced vomiting, and a herbal purgative class (n=136) characterized primarily by the use of indigenous Fijian herbal purgatives. Both ED classes endorsed greater eating pathology and general psychopathology than non-purging participants, and the herbal purgative class endorsed greater clinical impairment than either the BN-like or non-purging participants. Cultural orientation did not differ between the two ED classes.

Conclusions: Including study populations typically under-represented in mental health research and broadening the scope of relevant signs and symptoms in latent structure models may increase the generalizability of ED nosological schemes to encompass greater cultural diversity.

Figure 1

Relative endorsement of eating disorder symptom indicators in the BN-like class (represented by the solid line) and herbal purgative class (represented by the dashed line) Note. For dichotomous variables (vomiting, laxative use, fasting, and herbal purgative use), relative endorsement is plotted as the percentage of participants in each class who endorsed that symptom. For ordinal variables (overvaluation, binge eating, and driven exercise), relative endorsement is plotted as the mean of the three ordered categories (assigned point values of 0, 1, and 2 for increasingly greater frequencies) divided by two, so as to fit on a 0–1 scale.