Community subgroups in dyspepsia and their association with weight loss.

Abstract

A link between dyspepsia symptoms and weight loss is controversial. We aimed to determine whether or not weight loss is a marker of dyspepsia. Independent community-based cross-sectional studies. Subjects were randomly selected from the general population in Sydney, Australia. All subjects completed validated community health surveys. Two distinct data collections were used; the first as a training sample (N = 888) and the second as a validation sample to confirm the findings of the first (N = 2,907). The study was focused on weight loss, which was categorized as (a) any weight loss, (b) substantive weight loss (> or =3 kg), and (c) weight loss expressed as percentage of body weight. All dyspepsia symptoms studied were positively associated with weight loss although the strength of association did vary. Nausea and vomiting were most strongly associated with weight loss as were meal-related complaints such as postprandial fullness. Similarly, clusters formed based on symptoms were strongly differentiated in terms of weight loss with clusters characterized by nausea, vomiting, and early satiety/postprandial fullness reporting 25-30% weight loss prevalence over the previous 3 months compared with around 10% prevalence in clusters characterized by low dyspepsia symptom burden. Weight loss > or =3 kg followed a similar pattern but with a prevalence approximately half that of any weight loss, while weight loss expressed as percentage of body weight followed the same pattern. Dyspepsia symptoms are clearly and, in some cases, strongly associated with weight loss, both any loss of weight and substantive weight loss. Weight loss should be considered a warning symptom of dyspepsia.