Cannabis use is a global public health issue associated with increased risks of developing mental health disorders, especially in young people. We aimed to investigate the relationships between cannabis exposure and risks of receiving mental illness diagnoses or treatment as outcomes.

A population based, retrospective, open cohort study using patients recorded in ‘IQVIA medical research data’, a UK primary care database. Read codes were used to confirm patients with recorded exposure to cannabis use who were matched up to two unexposed patients. We examined the risk of developing three categories of mental ill health: depression, anxiety or serious mental illness (SMI).

At study entry, the exposed cohort had an increased likelihood of having experienced mental ill health [odds ratio (OR) 4.13; 95% confidence interval (CI) 3.99–4.27] and mental ill health-related prescription (OR 2.95; 95% CI 2.86–3.05) compared to the unexposed group. During the study period we found that exposure to cannabis was associated with an increased risk of developing any mental disorder [adjusted hazard ratio (aHR) 2.73; 95% CI 2.59–2.88], also noted when examining by subtype of disorder: anxiety (aHR 2.46; 95% CI 2.29–2.64), depression (aHR 2.34; 95% CI 2.20–2.49) and SMI (aHR 6.41; 95% CI 5.42–7.57). These results remained robust in sensitivity analyses.

These findings point to the potential need for a public health approach to the management of people misusing cannabis. However, there is a gross under-recording of cannabis use in GP records, as seen by the prevalence of recorded cannabis exposure substantially lower than self-reported survey records.

Clinical audit methodology was used to compare the treatment of alcohol misusers at risk of Wernicke's encephalopathy in an acute medical setting, and to assess the impact of providing information about best practice to prescribing doctors. All patients prescribed thiamine during an admission to an acute hospital trust over a 6-month period were identified, and data about their treatment episode were collected retrospectively. Hospital pharmacists then provided all prescribers with a flowchart summarising current prescribing guidelines, and prescribing patterns were re-audited 6 months later.

Over two audit periods, half of the patients prescribed thiamine whose case notes we examined had symptoms suggestive of Wernicke's encephalopathy, and another 30% were at high risk. Prescribing adhered to hospital guidelines only in 14% of cases, with the pharmacy-led intervention associated with a small but significant increase in the number of patients receiving adequate treatment for Wernicke's encephalopathy.

Wernicke's encephalopathy is relatively common in alcohol-dependent individuals admitted to hospital, and it is easily and cheaply managed. However, even when potential cases are identified, prescribing guidelines are followed in a minority of cases, even with prompting by a hospital pharmacist. This may be related to the limited research base concerning the optimum dosing schedule of thiamine, or fears about possible anaphylaxis when using parenteral preparations.

A 4-month pilot was conducted to assess whether the clinical librarian model, which has been successfully used within acute hospitals, would work in a mental health setting. A librarian attended weekly clinical team meetings in two community mental health teams to help generate clinical questions. A summary of the evidence on each topic was then presented the following week. An evaluation of the pilot was carried out using a questionnaire survey, focus groups and interviews.

Results suggest that the project had produced a positive impact within the teams and begun the process of embedding evidence-based information within clinical practice.

With some adaptations, the clinical librarian model can be an effective method of implementing evidence-based practice and addressing continuing professional development needs within mental health clinical teams.