Readmission following discharge is relatively common in mental health services. In 2016, Donisi et al. reported that the overall 30-day unplanned readmission rate was 13 per 100 discharged patients for schizophrenia and 11 per 100 discharged patients for bipolar disorder. Unplanned hospital readmission figures are frequently used within healthcare as an indicator of quality, with rapid readmissions — particularly those within 30 days — being considered poor. According to data in a report prepared for the UK Department of Health in 2014, around 5–8% of unplanned hospital admissions were a result of medication issues. Furthermore, readmission within 30 days has been shown to be higher in those with a medication discrepancy, where the medication prescribed on admission unintentionally differs from the pre-admission medication.
Changes to medication during an inpatient episode, where a patient is admitted to hospital, are common, and timely communication between pharmacists and the GP after discharge is essential. National Institute for Health and Care Excellence guidance recommends that medicines-related information should ideally be shared within 24 hours of a patient being transferred or discharged.
The potential role of medication-related factors in readmissions was investigated in inpatient wards and across crisis resolution home treatment teams (CRHTTs) within the Central and North West London NHS Foundation Trust (CNWL).
Specific measures examined were:
- Prescription and collection of discharge medication;
- Discharge notification being sent to and received by the GP;
- Completion of post discharge follow-up on patients’ most recent admission.
Data were recorded on a locally designed data collection tool using information collated with Tableau, a data analytics tool. Patients’ electronic notes from SystmOne and current medication charts were reviewed, patients were interviewed and GP practices were contacted where appropriate.
Data collection was carried out by pharmacy staff over a two-week period from 22 June to 5 July 2020. Data were collected from 16 wards and 6 CRHTTs, from a total of 213 patients. Data included 99 new ward admissions and 114 newly accepted CRHTT referrals across the 4 hospitals in Jameson, CNWL’s central London mental health division. Where a patient’s care was transferred between the ward and CRHTT without a break in treatment, this was considered part of one admission.
For patients who had a previous admission to the mental health service, information was gathered from the electronic notes and patients were interviewed (where possible) to ascertain whether their current admission was related to medication as part of the medicines reconciliation process. Medication being a factor in this process was classified as:
- ‘Yes’, if medicine was considered the main reason for readmission, with no other contributory factors;
- ‘Partly’, if there were other significant contributory factors in addition to medicines (e.g. illicit drug use, social factors);
- ‘No’, if medication was not considered a factor (e.g. relapse owing to relationship or financial difficulties).
Data show that 59% of patients (n=126) had been previously admitted. In 33% (n=42) of cases, medication was found to be a significant factor for readmission and partly contributed in 32% (n=40) of cases. Patients’ time to readmission, defined as the number of days post discharge, ranged from 2 to 5,733 days; 12% (n=10) within 30 days, 10% (n=8) within 31 to 60 days and 78% (n=64) >61 days ago.
There was a notable contrast between the electronic records and patients’ responses. Excluding those patients who were away from the ward/CRHTT (n=39), only 3% (n=3) felt medication was a significant factor in their readmission, 7% (n=6) felt it partly contributed, 44% (n=38) felt it was not a factor, 31% (n= 27) refused to answer and 15% (n=13) were not sure.
Of the 82 readmissions identified as being either ‘significantly’ or ‘partly’ related to medication, 14 had no medication at discharge. Eight patients had not been given their discharge medication or returned to the ward to collect it. In these patients, this discrepancy either significantly (n=3) or partly (n=5) contributed to readmission with two readmitted within 60 days of discharge.
Only 86 (68%) patients had their most recent electronic discharge notification (eDNF) sent to their GP. Of the 40 (32%) that did not, medication was a significant factor to readmission in 17 cases and partly contributed in 11 cases. Of the patients where medication was either a significant or contributory factor, 3 were admitted within 30 days and 4 within 31–60 days, making up 39% (n=7) of the total medication-related readmissions within 60 days.
Of the 86 patients whose last eDNF was sent to their GP, the practices confirmed receipt in 95% (n=82) of cases. For the four that were not received, all patients had a medication-related readmission, with 1 readmitted within 60 days.
Where appropriate, a 3 or 7-day post discharge follow-up was completed in 65% of cases (n=69). Of the 24 who did not have a post discharge follow-up, medication was considered to be a factor to readmission in 15 cases; a significant factor in 9; and partly a factor in 6. Of these 15 patients, 4 were admitted within 30 days and 3 within 31–60 days, making up 39% (n=7) of the total medication related readmissions within 60 days.
Findings suggest that there is potential to reduce avoidable readmissions by adopting a multidisciplinary approach to discharge planning and continuity of care post-discharge. Findings also support recent work, which highlighted the importance of patients not being discharged without an adequate supply of their medicines.
The limitations of this review are that it did not consider classes of medication and only captured readmissions back within CNWL. Additionally, the method used to decide if medication was a factor to readmission was subjective, which could be avoided in future by using or developing validated tools.
Recommendations from this study aimed at pre- and post-discharge stages include:
- Trusts to capture information on medication-related readmissions, and identify patient and system-wide contributory factors to better support patients;
- Trusts to routinely record expected discharge date so discharge medications can be ordered in advance and patients are able to leave hospital with an adequate supply;
- Multidisciplinary team to embed medicines education, involving patients and carers at every opportunity to improve their understanding, insight and attitudes about the importance of medication;
- As part of follow-up post discharge, organisations to ensure medicines are discussed, including checking patients have an adequate supply and know where to obtain further supplies.
The authors have also set up a working group with patient and carer involvement to explore ways to support medication adherence and patient engagement with medication.
Yogita Dawda, lead pharmacist for mental health; Niina Ezewuzie, associate chief pharmacist; both at Central and North West London NHS Foundation Trust
- 1Donisi V, Tedeschi F, Wahlbeck K, et al. Pre-discharge factors predicting readmissions of psychiatric patients: a systematic review of the literature. BMC Psychiatry 2016;16. doi:10.1186/s12888-016-1114-0
- 2Exploring the cost of unsafe care. A report prepared for department of health. Frontier Economics. 2014.https://www.frontier-economics.com/media/2459/exploring-the-costs-of-unsafe-care-in-the-nhs-frontier-report-2-2-2-2.pdf (accessed Jul 2021).
- 3Uitvlugt EB, Janssen MJA, Siegert CEH, et al. Medication-Related Hospital Readmissions Within 30 Days of Discharge: Prevalence, Preventability, Type of Medication Errors and Risk Factors. Front Pharmacol 2021;12. doi:10.3389/fphar.2021.567424
- 4Medicines Optimisation: the safe and effective use of medicines to enable the best possible outcomes NICE Guideline 5. National Institute for Health and Care Excellence. 2015.https://www.nice.org.uk/guidance/ng5 (accessed Jul 2021).
- 5Uncollected discharge medicines on mental health wards and the impact on medicines-related readmissions. The Pharmaceutical Journal Published Online First: 2020. doi:10.1211/pj.2020.20208555