Medication Without Harm is the World Health Organisation’s (WHO) Third Global Patient Safety Challenge. It aims to “reduce the level of severe, avoidable harm related to medication by 50% over 5 years, globally”1. Medication-related harm (MRH) includes harm from adverse drug reactions (ADR), non-adherence and medication errors2.
In a large, prospective UK study of 18,820 patients, the prevalence of an ADR-related admission was 6.5% with the average age of 76 years old and an estimated cost of £466 million annually3. In another observational study in Rotterdam, 24% of admissions in over 70year olds were due to ADRs4. A large, retrospective study of adverse events (AE) across 58 US hospitals identified 5077 cases of inpatient admissions in the over 65s from 99,628 emergency department visits with half occurring in those older than 805. A meta-analysis shows elderly patients are four times more likely to be admitted with an ADR compared to younger patients6. In a systematic review of MRH in older adults, between 17-51% patients experience MRH within 30 days post-discharge2,7. Our study team have shown that in the 8 weeks following hospital discharge, 37% of over 65s experienced MRH with an estimated cost of £400million annually8.
Factors contributing to MRH in older people are multiple including multimorbidity and polypharmacy9, age-related changes in pharmacokinetics and pharmacodynamics10 and non-adherence to medications11. Transition of care at hospital discharge is high-risk period for the occurrence of MRH with multiple contributory factors: the impact of acute illness, an inpatient stay and patient deconditioning12, medication discrepancies at admission or discharge13, patient/carer education on discharge medication regimens, the use of new medications14, and poor communication between secondary and primary care15,16.
The WHO identifies three key target areas to protect patients from MRH: high-risk situations, polypharmacy, and transitions of care1. The WHO defines high-risk situations as certain clinical circumstances in which the impact of MRH may be greater; this includes older individuals and those with hepatic or renal impairment17. Discharge of an older person is an especially high-risk situation encompassing many factors that might lead to MRH.
The NHS Discharge Medicines Service (DMS) is a newly commissioned community pharmacy service aiming to reduce avoidable post-discharge MRH based on an initiative led and delivered by the Academic Health Science Networks (AHSNs). The DMS is a system of electronic communication allowing hospital pharmacists to refer patients to community pharmacists to ensure that patients receive adequate support post-discharge. The evidence for the DMS is informed by several studies showing reduced hospital readmission rates and shorter hospital stay18–21.
Although effective, patient selection for the NHS DMS is based on hospital pharmacists’ judgement and not on any evidence-based risk stratification data22. NHS England supports the adoption of risk stratification using risk prediction models (RPMs) to identify individuals who will derive the most benefit from target interventions23. Risk stratification is increasingly important in a healthcare system challenged by an ageing population24, and an increasing prevalence of healthcare use due to MRH25.
Six RPMs exist: McElnay26, the BADRI Model27, the GerontoNet ADR risk score28, Trivalle29, the PADR-EC score30, and the ADRROP prediction scale31. These RPMs do not predict the risk of MRH occurring in the post-hospital discharge period. No impact studies of these tools have been published. They predict the risk of inpatient ADR/ADE27,29,31,32 or risk of an unplanned admission being due to ADR/ADE30,33.
Our study team has conducted the PRIME (Prospective study to develop a model to stratify the RIsk of Medication-related harm in hospitalised Elderly patients) study and developed and internally validated the first RPT to predict the absolute risk of an older person experiencing MRH in the 8-week post-discharge period34. The RPT was developed though a large multicentre, prospective observational cohort study. The PRIME RPT consists of eight variables routinely collected in hospital (age, gender, antiplatelet drug, sodium level, antidiabetic drug, past ADR history, number of medicines, and living alone). This tool considers demographic, medical and social factors in predicting the absolute risk of MRH. The PRIME tool calculates risk of a definite MRH; MRH that was classified as ’probable’ or ’possible’ was excluded.
Currently, there are no tools in clinical practice to target interventions to high-risk patients in the community following hospital discharge. The PRIME tool is the first RPT able to predict risk of MRH in older adults in the community in the 8 weeks post-discharge.
RATIONALE
Medication-related harm (MRH) for this study will include adverse drug reactions, medication errors, and a failure to take/receive medication, either following non-adherence or a failure in the supply chain.
The WHO identifies transition of care as a key area when it is necessary to protect patients from harm1. Previous interventions in the post-discharge period have been ineffective35–38. Published evidence for the NHS DMS has shown that the DMS can decrease readmission rate and shorten the length of hospital stay18–21. However, patient selection for DMS is based on hospital pharmacists’ judgement and not on any evidence-based risk stratification data22 . Our study team has shown that clinical judgement alone cannot predict MRH in older patients39.
NHS England recognises that risk stratification using RPMs can target those at greatest risk and those who are most likely to derive benefit from intervention23. The PRIME tool is the first tool to predict absolute risk of MRH in the 8 weeks post-discharge period and is a better predictor than clinical judgement 39. A medicine management plan has been developed by the research team in consultation with patients, carers, and healthcare professionals.
MRH risk prediction tools are not routinely used in clinical practice, as these tools have not been assessed for impact and implementation40 . The PRIME tool has been transparently developed and internally validated. To satisfy the next stage of risk-prediction model creation, the impact of the tool will be assessed on a new sample of older individuals. Targeted interventions at high-risk individuals may be a clinically and cost-effective intervention in reducing the rates of MRH in older adults. The PRIME team in collaboration with the AHSN-KSS will implement a risk-stratification approach linked with the NHS DMS. The study will recruit patients aged 65 and older discharged from 4 NHS trusts. The control arm will consist of NHS DMS care only. The intervention arm will consist of NHS DMS in addition to a specific medicines management plan (MMP) developed by the study team in consultation with patients and carers. The MMP will consist of:
- A copy of the hospital discharge summary
- Specific education about possible medication-related harm from discharge medications. Education will be delivered by the ward pharmacist and / or the ward doctor to the intervention participants (and family member/carers/friend if available) at the point of discharge.
- Clear guidance on who study intervention participants and their carer/family member/friend can contact (the GP or the community pharmacist) if patients experience any MRH.
- The name and contact of the community pharmacist will be provided by the ward pharmacist to those in the intervention arm/their family member/carer/friend.
- A copy of the percentage/ probability of harm from discharge medications calculated using the PRIME study RPT. The risk score will be presented in a visual analogue scale and will be offered to intervention patients and (if available their family member/carer/friend). See Appendix for document.
RESEARCH QUESTION/ AIM
Research Question
Will a Medicines Management Plan linked to the NHS DMS be more effective than the NHS DMS alone in reducing rates of MRH?
Objectives
- To measure and compare the rates of MRH in the two groups.
- To measure the costs of delivering the intervention and any associated MRH-related service use in the two groups in the 8-week study period, and to perform modelling to provide national cost estimates.
- To undertake a process evaluation.
Outcome
A clinical, economic, and service evaluation of NHS DMS alone compared to a MMP linked with the NHS DMS in reducing MRH.