272 Treating Eating Disorders
treatment would seem to be effective in terms of weight gain and mortal-
ity. However, Crisp et al. (1991) reported the results from a randomized
control trial which allocated patients to receive inpatient treatment,
outpatient-based family and individual therapy, outpatient group therapy,
or one outpatient assessment. Of the 80 patients recruited into the
trial, only 59 patients took up their treatment, and the results showed no
differences between the four groups in terms of global measures of out-
come. Similarly, Gowers et al. (2007) compared inpatient treatment with
specialized outpatient care and reported poor outcomes in terms of adher-
ence to inpatient treatment and patient outcomes. McKensie and Joyce
(1992) also reported data from their 5-year follow-up study of 112 patients
in New Zealand which showed that 48 percent were readmitted to hospital
on more than one occasion after their first admission, and that those
aged under 16 were even more likely to be readmitted. This has been labeled
the revolving door syndromeas some patients gain weight, leave hospital,
then lose weight and are readmitted. Steinhausen et al. (2008) explored the
frequency and predictors of readmittance in a clinical sample from five
European cities. The results showed that nearly 45 percent of patients
required at least one episode of readmission. Further, readmission was
predicted by a number of variables including low weight at both first
admission and first discharge, paternal alcoholism, and eating disorder
in infancy. In addition, those with repeated admissions showed poorer
outcomes. In a similar vein, Castro-Fornieles et al. (2007a) explored the
predictors of weight maintenance following discharge from inpatient
treatment. Forty-nine patients were followed up after 9 months, and the
results showed that weight maintenance by this time was predicted by a
higher BMI at admission and lower abnormal eating attitudes, depression,
and higher motivation to change at discharge. Weight maintenance has
also been associated with lower levels of perfectionism at first admission
(Nilsson, Sundbom, and Hägglöf, 2008). These results suggest that inpa-
tient treatment can bring about substantial weight regain, enabling the
patient to be discharged. A large minority, however, then lose this weight
and end up being readmitted. Those who benefit the most from inpatient
treatment seem to be less severe in terms of a number of parameters at
their first admission, suggesting that those whose eating disorder is caught
earlier and/or is less entrenched respond most positively to treatment.
Without inpatient treatment, even if patients are readmitted, it is possible
that the mortality rates would be even higher and that even if patients relose
the weight, they are being kept alive until a time when they are able to embark
upon a full recovery.