This site is intended for health professionals only

Picking up the pieces after discharge

Dr Clive Henderson explains how his pilot scheme reduced costly prescribing errors after patients left hospital

It is widely known among GPs that patients have drugs stopped, started or changed once admitted to hospital – without adequate explanation to either the patient or their practice. Studies have shown the potential seriousness of these medication errors, but primary care often lacks the resources to do anything about it [1,2,3].

Medication errors are not only damaging to patients, but also potentially costly in wasted drugs and hospital admissions and a drain on practice time. We did not realise how serious the situation was in our area until we gained funding to do a pilot scheme to proactively improve post-discharge care and collect evidence on how often errors were made.

Getting funding

In order to improve the situation in our practice and quantify the problem, we successfully applied for funding for a one-year pilot scheme from our PCT.

Funding was for a resident pharmacist in our practice to support patients in managing and understanding their medicines after discharge from hospital. The pharmacist was also responsible for validating information about medications provided by local hospitals at discharge and comparing this with practice records.

A business case was submitted to our local practice-based commissioning group and funding for a part-time pharmacist was secured. The funding came from our PCT's locality commissioning fund (£5,500) and savings from a prescribing incentive scheme (£5,300).

What we did

After securing adequate funds, we focused our efforts on patients discharged from hospital after an admission for a cardiovascular or diabetes-related illness and complex elderly medical discharges.

The pharmacist checked the list of medications on the discharge letter against the practice's repeat medication list and identified any discrepancies.

The pharmacist telephoned each patient if discrepancies were found. The patient or carer was asked to list all the medications the patient was taking. Any concerns identified were discussed with the patient or carer to see if they knew why medicines had been stopped, started or changed.

Where there seemed to be discrepancies in the medication information, or any hint of confusion or lack of understanding from the patient, the pharmacist arranged to see the patient in surgery or at home.

Apparent errors, often as omissions, were resolved by the pharmacist by liaising with either the primary or secondary care team. Often this involved time-consuming calls to the hospital to track down the consultant involved or the patient's notes. Where necessary, a GP review of the patient was arranged.


The results indicated a significant problem had existed in our area. We have recently finished the pilot and have yet to fully analyse the results, but to date the pharmacist has reviewed 171 discharge letters and there were inaccuracies in 98 (a rate of 57%). See the box below for the results from a preliminary analysis, looking at 96 records.

The inaccuracies were mainly dictation and typing errors, so, for example, one patient had carvediol 3.125mg on a discharge letter, but 6.25mg on a repeat screen. Talking to the consultant revealed the 3.125mg dose was a dictation error.

Other medications initiated in hospital were completely missed off the information given to the practice.

Patients were often not told, or were confused, about what medications they should be taking and why. Also, they were often not told to stop taking certain medications – for example, clopidogrel – by the hospital and so continued with unnecessary drugs.

Impact of the audit

It is difficult to translate these figures into avoided morbidity, mortality or re-admission, but the study demonstrates practical lessons that need to be learnt.

We have been able to give feedback to the PCT about the problems we identified and which hospitals were hardest to deal with. Our pilot results were concerning enough to make the PCT look to find a solution.

The PCT has included this work as part of its quality, innovation, productivity and prevention agenda to prevent ill health and re-admission. We are now working with the PCT medicines department to get standardised discharge letters instituted at local acute trusts. These will detail all medications and how long they should be taken for and will list those which have been stopped and why. We have a meeting scheduled to discuss this soon.

We are also looking at how this work can be incorporated into future commissioning decisions via the Commissioning for Quality and Innovation funding arrangement, so if there is expenditure as a result of medication errors on discharge, then money is withheld from the quality premium paid to secondary care.

Our funding for this pilot has not been renewed – this could only happen if the pilot was financially self-sustaining. As things stand, errors still occur but we lack the resources to fully clarify them.

But this pilot did show that it would be prudent to have a pharmacist check patient understanding of medications prior to discharge, or at least to fund a pharmacist scheme in primary care.

I will push for these changes through my local GP commissioning consortium.

Dr Clive Henderson is a GP in York. Pharmacists Tracy Robinson and Debbie Needham undertook the reviews

A pharmacist in the practice can help reconcile medication after discharge Preliminary analysis of pilot scheme

A sample of 96 records was reviewed by a pharmacist, with the following results:
• 12% of patients were seen acutely after discharge by a doctor as a result of the pharmacist's review.
• 50% of patients had their repeat medication changed.
• 21% of patients were found to be having particular problems with complying and understanding their medication.
• 5% of patients had medication stopped or reduced due to side-effects.


Visit Pulse Reference for details on 140 symptoms, including easily searchable symptoms and categories, offering you a free platform to check symptoms and receive potential diagnoses during consultations.