Amy J Luck, John F MorganLuck, Fiona Reid, Aileen O'Brien, Joan Brunton, Clare Price et al
Luck A J, MorganLuck J F, Reid F, O'Brien A, Brunton J, Price C et al.
The SCOFF questionnaire and clinical interview for eating disorders in general practice: comparative study
BMJ 2002; 325 :755
doi:10.1136/bmj.325.7367.755
Validation of SCOFF questionnaire among pre-teenagers
Dear Sir,
During the last few years, the incidence of anorexia suffered a dramatic
increase. The onset of the disease occurs now at younger ages and among
both genders.(1) Experts are calling for identification of high risk
individuals. We have read with interest the article by Luck et al.
concerning diagnosis of anorexia with questionnaires.(2)
Two questionnaires are available. Thanks to its high validity, the Eating
Attitudes Test(EAT)(3)is one of the tests most frequentely used to
diagnose anorexia among teenagers and pre-teenagers. This test consists of
40 questions which may be answered in 6 different ways. Each question
scores from 0 to 3. Subjects whose total score is higher than 30 are
considered to be at high risk. In spite of its usefulness, this test is
not adequate for epidemiologic purposes because of its complexity and
length.
The SCOFF questionnaire,(4)for its validity,(2,5)its reduced number of
questions (only 5) and the simplicity of answers (yes/no) and scoring
system (0 or 1), is better adapted to the epidemiologic diagnosis.
However, this questionnaire has not been validated yet in pre-teenagers
(11-13 years).
We carried out a validation study, with cross-sectional design, of the
SCOFF questionnaire, using the EAT questionnaire as a gold standard. We
carried out this validation among 341 children of 10 schools of Santiago
de Compostela (North West Spain) in which we randomly selected one class
of children aged 11-13 years. Every child filled out both EAT and SCOFF
questionnaires.
We obtained complete information on 289 subjects (84.8%). The SCOFF
questionnaire maximizes the area under the ROC curve at value 2. Scores
below 2 indicate subjects with normal behaviour while scores of 2 or
higher indicate anorexia. Its sensitivity is 64.1% (95%CI: 56.7- 71.3) and
its specificity is 87.2% (95%CI: 85.1 - 89.3). In our population, the
positive predictive value is 43.9% (95%CI: 37.4 - 50.4) and the negative
predictive value is 94.0% (95%CI: 92.5 - 95.5).
The validity of the SCOFF questionnaire for diagnosis of anorexia among
pre-teens (11-13 years) is similar to that found among teens and
adults.(2) Because of the low proportion of false negatives, 6% (95%CI:
4.5; 7.5), we believe that our results endorse the use of SCOFF in
epidemiologic settings, in which the main goal is to identify subjects
with high risk of anorexia.
The SCOFF questionnaire could be used at school by teachers and educators
to identify those subjects with a higher risk of anorexia and forward them
to the psychologist of the centre.
1.Steinhausen HC. The outcome of anorexia nervosa in the 20th
century. Am J Psychiatry 2002;159:1284-93.
2.Luck AJ, Morgan JF, Reid F, O'Brien A, Brunton J, Price C, Perry L,
Lacey JH. The SCOFF questionnaire and clinical interview for eating
disorders in general practice: comparative study. BMJ 2002; 325: 755-6.
3.Garner DM, Garfinkel PE. The Eating Attitudes Test: an index of the
symptoms of anorexia nervosa.Psychol Med 1979; 9: 273-9.
4. Morgan JF, Reid F, Lacey JH. The SCOFF questionnaire: assessment of a
new screening tool for eating disorders. BMJ 1999; 319: 1467-8.
5. Perry L, Morgan J, Reid F, Brunton J, O'Brien A, Luck A, Lacey H.
Screening for symptoms of eating disorders: reliability of the SCOFF
screening tool with written compared to oral delivery. Int J Eat Disord
2002; 32: 466-72.
Competing interests:
None declared
Competing interests: No competing interests