Retrospective analysis of hypophosphatemia rates and other clinical parameters in patients with eating disorders

To retrospectively assess medical services of a specialist inpatient eating disorders (EDs) unit.

We retrospectively evaluated clinical parameters of 288 inpatients classified as 'moderately' or 'significantly' medically compromised between 1 January 2016 and 30 June 2019.

We analysed 288 patients (mean age 32.5 [SD = 11.4] years, 96% women, 76% with anorexia nervosa). Average length of stay was 38.4 (SD = 28.4) days. Average admission body mass index (BMI) was 14.8 (SD = 1.8) kg/m² , and 16.1 (SD = 1.9) kg/m² at 4 weeks. At admission, 82% of patients were considered significantly medically compromised, while 6% were deemed moderately compromised. Only 5% of patients required transfer to intensive care unit. Prevalence of hypophosphatemia was 17.7%; rates did not increase significantly between years despite more assertive re-feeding processes. There was no association between risk classification at admission and change in BMI at 4 weeks (F_(2,166) = 0.588, p = 0.557). BMI at admission was found to be significantly associated with clinical outcome (β = 0.92, p < 0.001).

Hypophosphatemia rates did not increase despite more assertive re-feeding over 3 years. Our results provide support for a model of treatment that simultaneously addresses the medical and psychiatric sequelae of patients with severe EDs.