Medicines reconciliation on admission: further issues

This is the second of two articles

from North Bristol NHS Trust. The first article focussed on ‘Improving Medicines Reconciliation on Admission’, this second article will focus on the related topics of:

During our work to improve medicines reconciliation on admission, one of the barriers to carrying out the process in a timely fashion was access to information. This next section highlights our experiences of using Connecting Care.

Connecting Care is a joint NHS and Social Care project and is a partnership between North Bristol NHS Trust, University Hospitals Bristol NHS Foundation Trust, Weston Area Health NHS Trust, Bristol City Council, North Somerset Council, GP practices in South Gloucestershire, North Somerset and Bristol, Bristol Community Health, North Somerset Community Partnership, South Gloucestershire Community Health Service, BrisDoc, Bristol Clinical Commissioning Group, North Somerset Clinical Commissioning Group, South Gloucestershire Clinical Commissioning Group and South West Commissioning Support.

Connecting Care is an IT solution that allows patients’ healthcare information to be recovered from multiple systems including GP systems and hospital systems as well as council and social services systems. A single portal can be used to log onto a patient’s health record providing easy access to vital healthcare information when a patient is transferred between care settings.

The pharmacy department at North Bristol NHS Trust (NBT) were issued with seven Connecting Care pilot licences in December 2013. These were issued to five pharmacists and two pharmacy technicians; we were then issued a further five licences in March 2013 issued to a further three pharmacists and two technicians.

When a patient is admitted to hospital it is vital that we obtain an accurate drug and medication history. Research has shown that when patients are transferred from one care setting to another between 30% and 70% of patients may have an unintentional change to their medication.¹ These unintentional changes can cause patient harm. At NBT, we have systems in place to ensure that we reconcile medicines within 24 hours of admission of a patient and this process is largely driven by pharmacy. Prior to the introduction of Connecting Care, patients’ medication history had to be requested from the GP surgery via a telephone request for a fax.

During the first three months of the pilot, pharmacy staff logged on to the system 1396 times representing a thorough testing of the system. Pharmacists with access to connecting care are now able to reconcile a patient’s drug history against their GP record without having to wait for a fax from the GP surgery.

Secure log on can be accessed from any Trust computer (see above.)

Searching by NHS number reduces risk of obtaining wrong patient details (see above).

Easy-to-use interface  includes GP medication history, allergy information, past medical history and much more.

The number of fax requests has now dramatically reduced saving both GP and pharmacy time. In December 2013 pharmacy was requesting approximately 800 faxes a month this reduced to approximately 200 faxes by April 2014 (Figure 1).

Connecting Care provides a fast and efficient way of accessing patients’ GP medication history. In comparison to ordering a fax, it provides multiple benefits in terms of time, efficiency and accuracy of information. Connecting Care provides seamless transfer of prescribing and medical history and will be a useful tool in preventing harm to patients.

The phase 2 roll out of Connecting Care will include sharing of much more information including discharge prescriptions, outpatient prescribing and chemotherapy prescribing. This is a big improvement to the current situation because outpatient and chemotherapy prescribing are often missing from GP records. The recent NHS England Patient Safety Alert³ highlights the patient safety issues that occur at discharge and asks for local initiatives to prevent patient harm. Connecting Care allows us to share a patient’s prescribing and medical history across care settings helping to avoid harm to the patient at admission and at discharge.

There is also a drive to roll out and integrate with community pharmacy systems. This will benefit community pharmacists who can follow patients up in the community and provide a targeted Medicines Use Review (MUR) service. It will also be beneficial for pharmacies that are offering advanced services such as the New Medicines Service.

POA ensures that patients are fully informed and patient safety and the clinical process is enhanced.⁵  Pre-operative clinics have traditionally being nurse-led; however, they are now becoming increasingly multi-disciplinary.⁵ With more than one third of patients over the age of 75 regularly taking four or more prescribed medicines and the rise in use of biologics and ‘high-risk’ medication, pharmacists have become increasingly involved.⁶

In a study (n=158) into the introduction of a pharmacist in pre-operative clinics, medication accuracy improved significantly with a 92% reduction in medication history inaccuracy post-pharmacist introduction. A reduction in medication related incidents on admission and an 80% reduction in surgery cancellation due to inappropriate use of medicines were also demonstrated following introduction of a pharmacist.⁷ Crucially, nursing and anaesthetic staff were highly in favour of the introduction of a pharmacist at the POA clinic, with 81% of those surveyed responding positively.⁷ The surgical team at NBT started to investigate ways in which we could have an impact upon the medicines reconciliation process peri-operatively.

During early 2009, audit data was collected on a group of elective surgical patients entering into the Enhanced Recovery After Surgery (ERAS) Programme. Interventions relating to the peri-operative management of medicines were identified. In an effort to try and resolve the identified problem areas, a four-week trial of pharmacist attendance at the pre-admission clinic was carried out in July 2009.⁸

The primary objective of the project was to demonstrate the benefits, in terms of time/cost savings and quality of care/patient safety in having a pharmacist present in pre-admission clinic. Desired outcomes were targeted to meet the Commissioning for Quality and Innovation (CQUINs), Safer Patients Initiative (SPI2), National Patient Safety Agency (NPSA) and Quality, Innovation, Productivity and Prevention (QIPP) requirements.

The trial patient numbers were low (n=15), but the outcomes were extremely positive and are summarised in relation to medicines reconciliation below.

Improved patient safety: No errors were made by the pharmacist compared to ten  errors made by doctors/nurses over a comparative time period; for example, perindopril recorded as 2mg OD instead of 4mg. Two incorrect doses were given on the ward and the patient’s BP was high.

Improved skill mix: Pharmacists are trained and highly skilled in this area. When questioned, pre-op assessment and ERAS specialist nurses felt that pharmacists’ skills are better suited to this role.

Time savings: Pharmacist obtained a medication list from the GP the day before clinic thus saving clinic consultation time. Pre-op assessment nurse time freed for other duties.

Anaesthetic time and cost savings: As experts in drug therapy, pharmacists are ideally placed to perform this role. A total of 37 interventions were made in the trial group of patients at pre-admission stage, many of which could be solved by the pharmacist without anaesthetic input.

Improve communication between patient/primary/secondary care: Pharmacists can provide consistent, single point of contact across this interface. As an example, an anticoagulated patient required post-clinic follow up with the GP to ensure correct pre-op preparation. The pharmacist did this. When questioned, multidisciplinary team members felt that pharmacist skills were more suited to this role.

Reduction in transcription errors: There were no errors in the trial group of patients compared to eight in non-trial group over a comparative time period, for example, patient’s regular beta-blocker (atenolol 100mg) was not transcribed on admission, so a dose was missed on the day of operation.

Time/cost savings of ward doctor/anaesthetist: Reconciliation of medication history and signing of drug chart is all that is required on admission. 80% of the trial group had the pre-prepared charts used on admission.

Reduction in administration errors peri-operatively: Drug charts were annotated at pre-op stage with important peri-op medicines management information. For example, written direction to omit ACE inhibitor on morning of operation was added to the chart. In a non-trial patient, this was missed on admission and dose was given.

Reduction in discharge prescription transcription errors: Pharmacists made no errors in the trial group compared to 12 in the non-trial group over a comparative time period. For example, four patients with aspirin withheld peri-operatively and not restarted on discharge and three patients eligible for extended thromboprophylaxis did not have enoxaparin prescribed on the discharge letter. In both situations, this information can be added to letter at pre-op stage.

In addition to the roles described above in the original trial, the funding enabled us to provide specialist input to the trial patients throughout the entire peri-operative period as summarised:

Data collection during the trial was focussed on length of stay, but we felt it was also important to try and assess improvements made in the medicines reconciliation process. Therefore, in December 2013, we carried out an audit over a two-week period to compare drug history taking errors by different health care professionals to show the nature of the errors found and to compare current practice to drug history-taking standards. Due to the small patient population seen at this time by the pharmacist in pre-op clinic, the patient number is low.

Table 1 and Figure 2 summarise the audit standards used and the compliance with those standards by each health care professional audited.

The results clearly show that pharmacists met the audit standards more than the other health care professionals in all but one of the eight standards.

As discussed, we have been working hard to embed a consistent medicines reconciliation pro-forma into all NBT admissions paperwork. Figure 3 supports the importance of this by demonstrating the significant increase in error rate of drug history taking by healthcare professionals when a pro-forma is not used.

These results have supported our long-term commitment to providing pharmacist input to the pre-operative process.

In an audit carried out at the Avon Orthopaedic Centre (AOC) at NBT, drug charts were completed at the pre-admission stage for only 18 (42%) of the patients audited, with variable quality of the sources used to complete the pre-operative drug history. The author therefore recommended the adoption of a dedicated specialist pharmacist for accurate drug reconciliation and completion of drug charts prior to admission.⁹

The success of the projects to date has resulted in a long-term commitment to funding of 1.5 WTE Band 8a pharmacists to work as part of the pre–operative assessment team at NBT from May 2014. This is an exciting time for the surgical team as we are involved in the new centralised pre-operative service in the new hospital.

There are various areas we would look to develop in the future, especially as independent prescribers. For example:

Reducing the use of blood could be added into the POA assessment as there is growing evidence that giving blood carries both risk and financial implications, and other treatments could be considered as first line. POA prescribing pharmacists would be ideally placed to drive forward anaemia clinics in POA. There may be potential for savings from the process of the use of iron infusions in an outpatient area rather than the use of blood.

This role is currently filled by the anaesthetist so we would be contributing to a time and therefore cost saving if we took on this role and efficiencies in the clinics where the anaesthetist is not present.

The pharmacist could perform this role, which would be a time saving for the pre‑op nurse who currently carries out this role.

Patients would be assessed at the pre-op stage and the most appropriate therapy chosen. For high-risk patients this could result in a decreased length of stay if PONV is prevented.

We successfully worked on embedding a surgery-specific medicines reconciliation pro-forma into the POA booklet and are now working towards developing the same in the Day Surgery assessment paperwork. Consistency across the Trust is essential in ensuring the medicines reconciliation process is safe and accurate.

We have worked with nurses on improving their skills using the MLE package “South West Medicines Management Training Scheme: Standalone Module 4 – Medicines Reconciliation: Handbook Module 4 Handbook v3 January 2012” developed in conjunction with the South West Regional Medicines Information and Training (SWMIT). This package is currently under review and our work is still ongoing.¹⁰

Medicines reconciliation is audited at NBT but with a focus on the safer patient initiative of medicines reconciliation within 24 hours of admission. This audit sought to calculate the cost avoidance achieved through pharmacist led medicines reconciliation.

The Karnon Report,¹¹ a National Institute for Health and Care Excellence (NICE)-sponsored cost effectiveness review from the University of Sheffield, used a meta-analysis tool to calculate the cost incurred by preventable drug errors, and the cost of a pharmacist employed to prevent these errors from happening. Based on the assumption that each patient admitted to hospital is prescribed five medicines, the average cost avoidance calculated from this report is £5.52 per patient.

The overall aim of this audit is to localise the findings of the Karnon report to NBT, proving that the cost avoidance per patient within this Trust through medicines reconciliation is at least £5.52.

The standards defined by the Karnon report are:

Following a small, but successful pilot study, the data collection week was launched across the Trust (two hospital sites) with the inclusion criteria of all patients admitted to hospital during that week. There were 440 replies, which is in keeping with expectation based on previous key performance indicator (KPI) data.

The results showed that Band 6 pharmacists take on average 6.81 minutes per medicines reconciliation, whilst both Band 7 and Band 8a pharmacists take 10.14 minutes.

The average patient admitted to NBT was actually found to bring in 5.99 items each, more than the 5 set by the original standards. The cost of the faxing process at NBT incurs a cost of £1.07 per form, less than the £2.45 found by the Karnon report. Finally, the cost avoidance achieved by pharmacist led medicines reconciliation was found to be, on average, £13.64 per patient, ranging from £10.82 for Band 6 pharmacists and £18.40 for the more cost effective Band 8 specialist pharmacists (Figure 4).

The project was limited in terms of judging actual clinical impact by the difficulties encountered with recording of intervention information. There could also be further work done using data from money saved by using patient’s own drugs (PODs) on admission to hospital to give a clearer overall picture.

Use of Band 8 pharmacists in this role shows how cost avoidance can be achieved most effectively. There would also be a benefit in educating clerking prescribers who admit patients, minimising errors at the first possible opportunity).

The Medication Safety Thermometer is a national tool that was developed by a steering group led by Dr David Cousins (who at the time was Senior Head of Patient Safety for Safe Medication Practice and Medical Devices, NHS England). The steering group involved senior nurses, medical directors, pharmacists, safety policy leads and improvement and measurement specialists and was initially developed in Manchester.

The purpose of the tool is to help understand medication errors and resulting harm. This is achieved by engaging frontline staff in collecting data to establish a baseline and then measure improvement. Staff involved collate a snapshot on one day each month and  this included: information on medication omission (‘missed doses’), medicines reconciliation, allergy status and identifying harm from high-risk medicines (in line with Domain 5 of the NHS Outcomes Framework).

The tool was initially piloted from July 2013 and this was extended to larger scale testing prior to national implementation. NBT became involved in September 2013 and data was initially collected on 20 patients on one day a month; it was then planned to increase the number of patients to all patients on five medical/five surgical wards. By January 2014, data was being collected on 150 patients a month. However Pharmacy had identified the need to engage the help of nursing staff with this monthly audit in order to cover all the patients. By June 2014, approximately 36 Trusts submitted data.

As NBT moved to a new hospital in May 2014, the configuration of patient areas has changed and ‘gates’ are larger and can include mixed specialties covered by more than one pharmacist. The practicalities of auditing all patients in one area have not yet been resolved and as the thermometer aims to audit all patients in one area; this is one issue we have raised. There are monthly tester’s WebEx discussions to discuss various issues experienced by the test sites as well as national meetings where Trusts can discuss how other Trusts are dealing with specific issues and any future developments.

The Trust Development Authority that oversees transitions into a Foundation Trust expects use of the Medication Safety Thermometer to be in progress so it is vitally important that at NBT we carry on developing this work. The vision is for nursing staff to provide the initial STEP 1 data collection on a drug round and the pharmacists to cover STEP 2, and discussions with senior nursing staff at NBT regarding nursing involvement are underway. STEP 3 has now also been introduced.

All data is now hosted on the NHS Safety Thermometer website (www.safetythermometer.nhs.uk/) and the project is hosted and run by ‘Haelo’ (www.haelo.org.uk) and the current Project Manager is Kurt Bramfitt.

Information is regularly available via:

– Participation log for each Trust:

– Numbers of patients surveyed each month

– WebEx attendance

– Downloads (data collection forms etc.)

– Tutorials and Learning Resources

– Dashboard which displays:

– Proportion of patients with reconciliation started within 24 hours of admission

– Proportion of patients who have had an omitted dose within the last 24 hours

– Proportion of patients with medicine allergy status documented

– Proportion of patients with an omission of a critical medicine

– Proportion of patients receiving a high risk medicine within the last 24 hours

– Proportion of patients that trigger an MDT referral.

In our previous article,⁴ Claire Howard, Deputy Chief Pharmaceutical Officer NHS England updated that: “NHS England now plans to incorporate medicines reconciliation rates by NHS Trust in their forthcoming Medicines Optimisation prototype dashboard. The dashboard will highlight to Clinical Commissioning Groups (CCGs) and Trusts which Trusts are routinely contributing to the Medication Safety Thermometer and if so, their reported medicines reconciliation rate. This dashboard will help commissioners and providers to think about the use of medicines in general terms rather than just a focus on their cost.”

The Medicines Optimisation dashboard was launched nationally on 12 June 2014. This links the information on Missed Doses and Medicine Reconciliation from the Medication Safety Thermometer and Safety information from NHS England.

The launch was to all NHS Trusts and there will be pressure from CCGs on Trusts not participating so this again raises the profile of Medicines Reconciliation, Missed Doses and other risk issues and makes these national targets.

From the NHS England launch it was stressed that the dashboard is intended to allow local NHS organisations to develop improvements. It is not intended as a performance measurement tool and there are no targets. It is hoped that CCGs and Trusts review this information together and agree how to use it locally. It was also highlighted that it will be important for Local Professional Networks, Strategic Clinical Networks and Academic Health Science Networks to use the indicators in their collaboration with patients, CCGs, Trusts and the Pharmaceutical Industry to support local improvement.

The four key principles of medicines optimisation areshown in Figure 5.

“NHS England recognises that we will need to engage with patients and the public in a way that few of us have to date. We will need to work more closely with patients to better understand their issues around medicines use and to co-develop solutions that better support them with their medicines-taking.”

NHS England stresses that the dashboard is a prototype and is the first of a series of iterations aimed at moving the focus towards understanding how well patients in a CCG area are supported to get the most from their medicines and thereby use them as intended, suffer no harm and ultimately derive the most benefit that medicines have to offer.

There is guidance on how to use the dashboard and each Indicator should be viewed in conjunction with the specification.

In our experience through the Safer Patients Initiative (SPI2) it became clear that specifying what has to be measured and how it has to be measured is key in producing more comparable results.

During SPI2 some Trusts were stating that they had 100% reconciliation but on closer probing in one example, the Pharmacists were carrying out the reconciliation and then the technician followed half an hour later to measure this. As with any data collection, the process needs to ensure that bias is reduced in order to accurately reflect the outcome of the process.

The key to improvement work is to measure your own performance and to use that measurement to improve your processes. This therefore needs accurate definition of the criteria of what is being measured: what population of patients, what time period and how the data is collected, using what information, when and by whom? As long as this is standardised then improvements can be demonstrated and then if the criteria are changed, the run chart can be annotated to show this.

For example, many Pharmacy departments do not offer a weekend clinical service and so it is reasonable to measure what you can influence and improve – so measuring Sunday – Friday admissions only is acceptable and changed when the service is extended. (Although there may not be a Sunday clinical service, patients admitted on a Sunday can have their reconciliation carried out on a Monday)

However if benchmarking is in operation, then it is more important that standardised methods and definitions are used otherwise benchmarking will not compare like with like. This is why the Quality, Innovation, Productivity and Prevention (QIPP) benchmarking scheme asked for details from all Trusts participating as to how data collection was carried out and the numbers of patients that were recorded.

During the initial stages of work to pilot the NHS Medication Safety Thermometer we had six pharmacists involved and even with this had a variety of interpretations as to what to record. It is important to ensure clear definitions, ongoing reviews and training of new staff to ensure that data collection remains consistent in interpretation.

SPI2 used an ongoing run chart, which took measurements from five patients per ward per week. This produces very different results to the snapshots for the Medication Thermometer of all patients on specified wards on one day a month.

These two sets of data are different and provide different perspectives.

Ongoing data can be time consuming to maintain but gives you an accurate picture of the ongoing process and it is easier to see changes (positive or negative) and these can be investigated, and if significant the run chart can be annotated.

Quality improvement methodology tells us that regular sampling, which is founded on clear definitions when used by trained motivated staff, is better than snapshot data as it provides a more reliable picture of clinical processes and procedures in operation. This means it is better as an early warning and enables earlier intervention.

Snapshot data is more time consuming, but only on one day, but enables you to see a more complete picture of more factors contributing to the overall service and harm in the system on that day. Single isolated snapshots on a monthly basis may provide the quantity of data equal to weekly smaller sampling but currently provide a greater depth

The ideal situation would be an ongoing thermometer but due to staffing and time constraints this will not happen and so the combination of the two sets is an improvement of one set alone.

At NBT our strategic objectives include moving towards as near to real time data collection as possible. This is seen as a valuable aid for early intervention to prevent quality and system failures. This is dependent on various factors including the introduction of electronic prescribing.

As above, some Trusts have just measured on one ward whereas others are measuring in multiple areas. This needs to be clearly stated so that results can be interpreted.

We originally analysed all our admissions to find that at least one patient a year was admitted on nearly all of our wards (approximately 50). Our decision was to focus on wards with over 2% of the total numbers of admissions – which covered 80% of admissions.

In visiting another Trust whose results were not as good as they wanted, it became apparent that they only collected data from five patients per month. One poor result can therefore have a 20% impact, rather than a 5% impact if they had collected data from 20 patients per month.

If admission numbers are low, then it may not be possible to collect five patients per week (20 per month) and this can skew the data. We are currently reviewing how we can differently measure and collect and display data from wards with very low admissions.

In discussing data collection with different pharmacy departments, there are two main areas of difference and these need further agreement and clarification to give meaning to the benchmarking process.

Medicines reconciliation can be broken down into two stages as defined by the National Prescribing Centre:¹²

– Basic reconciliation (Stage 1)

– Full reconciliation (Stage 2)

Some Trusts have decided to measure Stage 1 data and others Stage 2 data. Our work with SPI2 promoted using the easiest method of data collection. Knowing when stage 2 is complete is easier for Trusts with electronic prescribing systems (eP). Some Trusts without eP are taking valuable time in measuring completion rather than progressing the medicines reconciliation process.

There are also various interpretations of this. Exact timing can again only easily be measured by Trusts with eP. For the many Trusts without this, a single date of admission covers patients admitted from one minute past midnight through to one minute to midnight – a difference of 23 hours and 58 minutes.

Some Trusts have different ways of measuring this.

Howard et al¹³ state that “Within 24 hours = measured from the first date on the in-patient drug chart and completed by the end (5pm) of the following day”.

Information from the NHS Medication Safety Thermometer guidance booklet on the website indicates an exact 24 hours, which most Trusts cannot know: “Was medicines reconciliation for all medicines undertaken (started) by the pharmacy team within 24 hours of the admission to this care setting?”

We are currently reviewing our data to see the relevance between the difference in our results from ongoing and snapshot data.

In summary, data collection needs to be explicit and different data may not be comparable.

For the future, the Medication Safety Thermometer and the Medicines Optimisations Dashboard have both contributed to highlighting the importance of medicines reconciliation and medicines safety.

The recent NICE guidance2 (CG183: 4 September 2014) Drug allergy: diagnosis and management of drug allergy in adults, children and young people” also refers to the Medicines Reconciliation process:

“1.2.8 Carry out medicines reconciliation for people admitted to hospital in line with recommendations in Technical patient safety solutions for medicines reconciliation on admission of adults to hospital (NICE patient safety solutions guidance 1).”

For us at NBT, medicines reconciliation on admission is now fundamental to our way of working. Since the end of May 2014, we have provided a more extensive Clinical Pharmacy service to 64 beds on the emergency zone, working from 8am–7pm Monday–Friday and 10am–3pm on Saturdays and Sundays. Having access to Connecting Care has enabled us to improve our medicines reconciliation processes.

We are keen to progress this nationally and over the past two years have been disseminating and sharing our work through workshops, articles, awards and poster applications.

For the future, we want to maintain our performance in processes on admission but also start to improve our processes on discharge. This work will also link to the recent NHS England Patient Safety

Alert.³