Can we identify features associated with complex mental health difficulties in primary care electronic health records.

Problem

Complex mental health difficulties, such as personality disorder and dysthymia are common in general practice consulters. However diagnosing and coding of these disorders in electronic health records is much lower than expected from population surveys. We aimed to identify features in primary care records which may be useful in promoting greater recognition of complex mental health difficulties. This presented major methodological challenges given the number and range of possibly relevant features and the anticipated large number of uncoded cases.

Approach

We analysed data from the Connected Bradford database, an anonymised primary care database of approximately 1.15M citizens. We used multiple approaches to generate a large set of features representing multi-level collections of patient attributes across time and dimensions of healthcare. Feature sets included antecedent and concurrent problems (psychiatric, social and medical), patterns of prescription and service use and temporal stability of attendance. These were tested individually and in combination. We analysed the relationship between features and diagnostic codes using scaled mutual information.

Findings

We identified 3,420 records with a diagnostic code for personality disorder or other complex mental health related diagnosis. This was 0.3% of the population compared to an expected prevalence of 3-5%. We generated >500,000 features. The most informative feature was count of unique psychiatric diagnoses. Other features were identified, including binary features (e.g. presence or absence of prescription for antipsychotic medication), continuous features (e.g. entropy of non-attendance) and counts of features (e.g. concerning behaviours such as self-harm & substance misuse). Several of these showed odds ratios >=5 or <=0.2 but low positive predictive value. We suggest this is due to the large number of “cases” being uncoded and, thus appearing as “controls”.

Consequences

Complex mental health difficulties are poorly coded. We demonstrated the feasibility of using information theoretic approaches to develop a large set of novel features in electronic health records. Ideally these should be tested in a more tightly defined cohort. While these features are currently insufficient for diagnosis, several can act as prompts to consider further diagnostic assessment.