Secular trends in the incidence of anorexia nervosa: Integrative review of population-based studies

Objective and Method: Aggregating across retrospective cohort samples, this integrative review synthesizes the findings of 12 cumulative incidence studies (45 hypotheses) on anorexia nervosa secular trends. Results: (1) The female/male anorexia incidence rate ratio was estimated to be 8.20, 18.46 versus 2.25 cases per 100,000 per year, p < .05; (2) female teenagers experienced anorexia at a rate fivefold greater than other women, 50.82 versus 10.37 incident cases per 100,000 per year, p < .001; (3) no secular trend or change in the incidence of anorexia was observed among teenagers, while a near threefold increase was observed over the past 40 years among women in their 20s and 30s, 6.28 (1950–1964) versus 17.70 (1980–1992) cases per 100,000 per year, p < .05; and (4) the two cohort characteristics of age, and the age by year interaction accounted for nearly two thirds of the variability among anorexia incidence estimates, R² = .614, F(2,27) = 21.49, p < .001. After the two factors of age and the Age × Year interaction were accounted for, none of the other study characteristics, including study year(s), were found to be significantly associated with anorexia incidence, that is, a main effect of time was not observed. Discussion: The integrative evidence across the population-based epidemiologic studies covering 40 years in this field suggests strongly that, overall, the incidence of anorexia nervosa, particularly among those very young women at greatest risk of experiencing it, has not increased significantly. However, the risk does seem to have increased significantly among women in their 20s and 30s. © 1998 John Wiley & Sons, Inc. Int J Eat Disord 23: 347–352, 1998.     

Fractures in patients with anorexia nervosa,bulimia nervosa,and other eating disorders--a nationwide register study

OBJECTIVE: To study fracture risk in patients with anorexia nervosa (AN), bulimia nervosa (BN), or eating disorders not otherwise specified (EDNOS). METHOD: Cohort study including all Danes diagnosed with AN (n = 2,149), BN (n = 1,294), or EDNOS (n = 942) between 1977 and 1998. Each patient was compared with three randomly drawn age- and gender-matched control subjects. RESULTS: Fracture risk was increased in AN after diagnosis compared to controls (incidence rate ratio: 1.98, 95% CI: 1.60-2.44), but not before. The increased fracture risk persisted more than 10 years after diagnosis. A significant increase in fracture risk was found before diagnosis in BN (1.31, 95% CI: 1.04-1.64), with a trend towards an increase after diagnosis (1.44, 95% CI: 0.93-2.22). EDNOS patients had a significant increase in fracture risk before (1.39, 95% CI: 1.06-1.81) and after diagnosis (1.77, 95% CI: 1.25-2.51). DISCUSSION: The increased fracture risk many years after diagnosis indicates permanent skeletal damage.     

Time trends in the incidence of eating disorders: a primary care study in the Netherlands

OBJECTIVE: This Dutch epidemiological study used primary care-based data to examine changes in the incidence of eating disorders in the 1990s compared to the 1980s. METHOD: A nationwide network of general practitioners, serving a representative sample of the total Dutch population, recorded newly diagnosed patients with anorexia nervosa (AN) and bulimia nervosa (BN) in their practices during 1985-1989 and 1995-1999. RESULTS: The overall incidence of AN was stable (from 7.4 to 7.7 per 100,000). However, the incidence among 15-19-year-old females increased significantly (from 56.4 to 109.2 per 100,000). The incidence rate of BN decreased nonsignificantly from 8.6 to 6.1 per 100,000. CONCLUSION: The time trend of an increasing AN incidence among the high risk group continued to the end of the past century. The BN incidence did not rise as was expected.     

Excessive exercise in anorexia nervosa and bulimia nervosa: relation to eating characteristics and general psychopathology

OBJECTIVE: Excessive exercise is a well-known phenomenon in anorexia nervosa, but less is known about its role in bulimia nervosa. In addition, there is little evidence regarding the psychopathological processes that might act as predisposing, triggering, or maintaining factors for such exercise. The present study examined the presence of excessive exercise in different women with eating disorders, and its psychopathological correlates. METHODS: Case notes from 63 anorexia nervosa and 61 bulimia nervosa patients were examined. Two-way multivariate analyses of variance (diagnosis x use of excessive exercise) were used to determine the impact of the two factors upon eating characteristics (EAT-40 and BITE) and psychopathological symptoms (SCL-90-R). RESULTS: While high levels of depression were more likely among all patients who used excessive exercise, levels of anxiety and somatization were particularly high only among those anorexics who exercised excessively. DISCUSSION: Possible explanatory models are advanced to account for this pattern of findings, focusing on the possible use of exercise as an affect regulation strategy among anorexia nervosa patients. Further research is suggested to test and develop this model, and possible clinical implications are outlined.     

The clinical presentation of Japanese women with anorexia nervosa and bulimia nervosa: a study of the Eating Disorders Inventory-2

OBJECTIVE: The current study described the clinical presentation of anorexia nervosa (AN) and bulimia nervosa (BN) in Japan utilizing the Eating Disorders Inventory-2 (EDI-2) and assessed whether the clinical profile of eating disorder cases in Japan differs significantly from North American data. METHOD: Statistical comparisons of Japanese AN, BN, and non-eating-disordered (NonED) EDI-2 data were conducted across diagnostic groups and with the North American standardization sample. RESULTS: The Japanese diagnostic groups reported significant between-group differences on all eight EDI-2 subscales. Compared with the standardization sample, all the Japanese groups reported significantly greater maturity fears, the Japanese AN-restricting subtype group (AN-R) reported lower levels of drive for thinness and perfectionism, the Japanese BN group reported lower levels of drive for thinness, and the Japanese NonED Group reported lower levels of drive for thinness and perfectionism but higher rates of ineffectiveness. DISCUSSION: Differences between the Japanese and North American groups suggest that certain cultural differences exist in eating disorder profiles.     

A comparison of early family life events amongst monozygotic twin women with lifetime anorexia nervosa, bulimia nervosa, or major depression

OBJECTIVES: To investigate the differential profile of early family life events associated with lifetime anorexia nervosa (AN), bulimia nervosa (BN), and major depression (MD). METHOD: Only data from the monozygotic twins (n = 622) were examined from a community sample of female twins who had participated in three waves of data collection. Eating disorder and MD diagnoses were ascertained from the Eating Disorder Examination at Wave 3 and interview at Wave 2 respectively. Early family events were ascertained from self-report measures at Waves 1 and 3. Two case control designs were used, including a comparison of women: (1) who had lifetime AN, BN, MD, and controls, and (2) twin pairs discordant for either AN, BN, or MD (where the unaffected cotwin formed the control group). RESULTS: Across the two types of designs, compared to controls, both AN and BN were associated with more comments from the family about weight and shape when growing up. AN was uniquely associated with higher levels of paternal protection while BN was associated with higher levels of parental expectations. CONCLUSION: While some overlap among early life events was indicated, especially related to parental conflict and criticism, there was evidence to support some degree of nonoverlap among life events associated with AN, BN, and MD.     

Incidence of severe anorexia nervosa in Switzerland: 40 years of development

OBJECTIVE: The current study examined the development of the incidence of severe anorexia nervosa with five sampling periods covering the years 1956-1995 in a geographically defined region of Switzerland. METHOD: Applying the same methodology as in the earlier sampling periods, the medical records of all hospitals of the canton of Zurich were screened manually for first-time hospitalizations of female anorexia nervosa patients during the years 1993-1995. Incidence rates were compared with the previous sampling period (1956-1985). RESULTS: The incidence rate of severe anorexia nervosa in the total population and the rate in the population at risk (females 12-25 years old) did not differ significantly from the incidence rates of 1983-1985. DISCUSSION: The incidence of severe anorexia nervosa in the canton of Zurich rose significantly during the 1960s and 1970s. Since then, the incidence appears to have reached a plateau.