van Eeden · The International journal of eating disorders 2023 · longitudinal epidemiological surveillance study · n=2,890,978 person-years

Increase in incidence of anorexia nervosa among 10- to 14-year-old girls: A nationwide study in the Netherlands over four decades.

Cited 22 times in the scientific literature.

Level 3 - non-randomized controlled study

Longitudinal nationwide primary care surveillance cohort

PubMed 37773004 · doi:10.1002/eat.24064 · record verified 2026-08-26

What was done

A nationwide network of general practitioners in the Netherlands (covering approximately 1% of the Dutch population) recorded newly diagnosed cases of anorexia nervosa (AN) and bulimia nervosa (BN) between 1985 and 2019, accumulating 2,890,978 person-years of observation. Diagnoses were standardized using DSM-IV criteria, with the same psychiatrist making final diagnostic determinations. Incidence rates (IRs) per 100,000 person-years were calculated for the total population, for females overall, and for females by 5-year age categories. Time trends in IRs were evaluated using JoinPoint regression analysis.

What was found

Among girls aged 10 to 14 years, the incidence of AN increased significantly from 8.6 to 38.6 per 100,000 person-years (average period percentage change = 56.7; 95% CI: 6.5 to 130.6). Across the total population (all ages), the incidence of AN remained stable, with IRs ranging from 6.0 (95% CI: 4.3 to 8.1) to 8.4 (95% CI: 6.4 to 10.8) per 100,000 person-years. The overall incidence rate of BN decreased from 8.7 (95% CI: 6.7 to 11.0) to 3.2 (95% CI: 2.0 to 4.9) per 100,000 person-years in the 2000s, before leveling off at 3.2 (95% CI: 2.0 to 4.8) in the 2010s.

Why it matters

This study provides long-term evidence of a substantial shift toward earlier onset or earlier detection of anorexia nervosa in young adolescent girls. It highlights an increasing need for pediatric-specific eating disorder resources, primary care screening, and targeted early intervention.

Limits

The surveillance network captures primary care visits for ~1% of the Dutch population, which may undercount individuals who never presented to a GP or received care entirely outside the public primary care framework. The abstract does not report the total absolute number of diagnosed cases, data on males, or demographic and clinical characteristics. Proposed drivers for the trend (such as social media use or changes in pubertal timing) were not directly measured in the study.

Cited by