Derivation of a prediction model for a diagnosis of depression in young adults: a matched case-control study using electronic primary care records.

Background
Approximately 80,000 children and young people in the UK suffer from severe depression but many are untreated due to poor identification of early warning signs and risk factors.

Aims
Derive and investigate discrimination characteristics of a prediction model for a first diagnosis of depression in young people aged 15-24 years.

Method
A matched case control study, using electronic primary care records. Stepwise conditional logistic regression modelling investigated 42 potential predictors including symptoms, co-morbidities, social factors, drug and alcohol misuse.

Results
Of the socioeconomic and symptomatic predictors identified, the strongest associations were with depression symptoms and other psychological conditions. School problems and social services involvement were prominent predictors in males aged 15 to 18 years, work stress in females aged 19 to 24 years.
Conclusion
Our model is a first step in the development of a predictive model identifying early warning signs of depression in young people in primary care.