Baseline data collection

Baseline data will be collected by trained research nurses including demographic (e.g. age, gender, ethnicity), clinical (e.g. discharge diagnosis, co-morbidities, renal and hepatic function) and social indicators (e.g. care package received and living arrangements) using a form specifically designed to allow the data to be scanned into an electronic database for future analysis. Information relating to medication name, frequency, dose and use of compliance aids will be collected and coded according to the WHO-ATC code (http://www.whocc.no/atc_ddd_index/). In addition validated tools will be used to collect information relating to nutritional status (Malnutrition Universal Screening Tool), physical function (Barthel Activities of Daily Living Index, Hand Grip strength), cognitive function (Abbreviated Mental Test Score) and depression and anxiety (Patient Health Questionnaire-2, Generalised Anxiety Disorder scale-2). Some of these tool are routinely measured in elderly care wards in the UK and if this was not the case the research nurse would obtain the measurement. The hand grip strength of participants will be measured using the method described in the Southampton Protocol for Adult Grip strength Measurement using the JAMAR Hydraulic Hand Dynamometer [16]. The MUST score is routinely used on elderly care wards and is a five-step screening tool to identify adults, who are malnourished, at risk of malnutrition (undernutrition) or obese [17]. The Barthel ADL Index is a validated scale used to measure performance in activities in daily living (ADL) [18].

Data will be collected directly from the hospital records, following discussion with members of the care team and with patients and/or carers.

Following discharge the junior doctor from the discharging medical team will be asked to complete a questionnaire to determine their judgement of the likelihood of the patient experiencing MRH during the 8 week follow up period. This section of the data collection form is based on the National Patient Safety Agency (NPSA) Risk Model Matrix where the likelihood of an event is rated against the consequence [19]. The junior doctor will be asked to predict the likelihood that the patient will be readmitted or access healthcare in the community due to MRH in the ensuing 8 weeks post discharge (doubtful, possible, probable, definite) and will be asked to rate their confidence in this decision (a 6 point scale from ‘little or no confidence’ to ‘virtually certain’).