This site is intended for health professionals only!

by Tom Skelland MPharm MRPharmS PG DIP GPP; and Hannah Heales MPharm PG DIP GPP
Published on 20 March 2020

Share this story:
Twitter
LinkedIn

Electronic prescribing: Phased rollout vs ‘Big Bang’

Contracts have been signed and the team is ready to go, but what is the best approach when deploying a large-scale electronic patient record?

The pros and cons of implementing an electronic prescribing and medicines administration (ePMA) system into acute NHS Trusts has long been debated,1,2 and tips have been shared on successful deployments.3 At a time when many NHS Trusts consider how to approach such a deployment, with little to no impact on patients or their safety, this article seeks to define the timescales against which this can be done. There is also discussion around different methods for successful deployment, including pitfalls to avoid, provided by two ePMA and clinical informatics pharmacists who have successfully deployed ePMA systems into their Trusts using contrasting methods of implementation.

Definitions

  • Big Bang: Whole Hospital site goes live at once, that is, ‘Switch on’ at 9.00am Saturday morning for all wards – transfer to electronic solution at this time and paper is no longer used
  • Rolling Thunder: Deployment continues across the Hospital occurring over a 2–3-day period (that is, weekend or long weekend) – wards go live at various points throughout 48–72-hour period
  • Phased rollout: Large scale project planned over several weeks or months, where deployment occurs on a ward by ward basis – from one or two a week to one a day.
The sites

St George’s University Hospitals NHS Foundation Trust (SGH) and The Royal Free London NHS Foundation Trust (RFH) are both large teaching hospitals within greater London. Although both Trusts use Cerner Millennium as their Electronic Patient Record (EPR) solution, they chose different methods for implementing the ePMA solution component. But which was safer, more efficient and less costly?

SGH consists of ~1100 beds across two sites (Tooting and Queen Mary’s Hospital (QMH), Roehampton). At the main Tooting site, a Cerner solution was initially introduced for radiology and pathology orders in 2012. ePMA and electronic documentation deployment was started in 2014 and reached approximately 38% of inpatient beds, which included the first paediatric intensive care unit in Europe to move to an electronic system.4 Further deployment was halted until late 2018, when a large scale project was implemented, bringing the electronic system to a total of 89% of inpatient beds. Maternity areas were completed in June 2019, and the final service, neonatal intensive care, will go live in March 2020. QMH had complete inpatient, emergency department (ED) and outpatient service deployment in September 2019. Deployment to Tooting ED is planned for early 2020, with outpatient services to follow. 

RFH has 1770 beds across three sites: Barnet Hospital (BH; district general, 440 beds), Chase Farm Hospital (CFH; elective surgical site, 74 beds) and the Royal Free Hospital (RFH; specialist hospital, 830 beds), and in total over 30 locations where services are provided by the Trust. The EPR went live at BH, CFH and the maternity department at the RFH in November 2018. Roll out included the ED, urgent care centre, outpatients and theatres. ITU was out of scope. Further roll out is scheduled for the rest of RFH including ITU at both BH and RFH in 2021.

The approaches

SGH is considered to have introduced its electronic system in a ‘Phased Rollout’ fashion, particularly considering the 3–4-year break between deployments. In turn, both deployment periods can be further considered phased rollouts, as implementation occurred on one ward a day over several weeks. Engagement with medical teams was made in the run up to go live on each ward – doctors local to the area who were familiar with patients were encouraged to join their ward’s transcription team, supported by the Chief Clinical Information Officer and Divisional Clinical Information officers. Each ward was transcribed with the one-to-one support of an ePMA team member, or a highly trained pharmacist champion user; several of which were seconded for the project. This offered local medical teams the chance to receive some individual training, and time with an expert, which proved to be a valuable engagement tool. Transcription took place the afternoon preceding the morning ‘Go-Live’, with clear indication of what had been transferred using highlighters. On the day of Go-Live, charts were re-checked, and any new, non-highlighted medications added to the paper chart were transcribed.

RFH took a different approach. The majority of wards were prepped in the weeks preceding the go live date of 17 November 2018, at which point the EPR was switched on across Chase Farm, Barnet and Royal Free maternity at the same time. This approach is considered a ‘Big Bang’ method of implementation; however, the transcribing activity was staggered over 36 hours, so there is some debate as to whether this constitutes a true Big Bang or should be classified by another name: for example, ‘Rolling Thunder’.

Twenty doctor and pharmacist transcribing teams were employed to transcribe every paper drug chart onto ePMA. Five consultant/senior pharmacist teams were assigned as quality assurance (QA) teams to check 10% of transcribed drug charts per ward. Table 1 highlights the main roles and responsibilities of staff involved in the transcribing activity. A ‘transcribing route’ was planned starting with the most acute wards and ending on the chronic long-stay wards. Ten teams across two wards at a time was considered the best approach, as this meant wards were transcribed quickly, limiting the risks associated with transcribing, such as new medications being added to paper drug charts once transcribed. Specialist wards such as paediatrics and cardiology were transcribed in smaller teams and these were completed by the specialist pharmacists and doctors who usually worked on those wards.

It is estimated that a team of two (one doctor and one pharmacist) transcribed at a rate of approximately three to four drug charts per hour. Transcribing commenced at 3pm on Saturday 17 November and paused at midnight. It resumed again at 8am and finished at midday on Sunday 18 November.

Other UK Trusts have taken a different approach to a ‘Big Bang’ deployment, whereby all preparation is completed in the preceding days, and the whole hospital is ‘switched on’ at a specific time, rather than ward by ward. 

One of the biggest risks at SGH was, and to a much lesser extent still is, the use of dual systems (paper and electronic). Consideration must be given to the transfer of care for patients who present to the ED where paper charts remain in use, and the downstream wards that prescribe electronically. The need to transfer between systems opens up the opportunity for error in transcription, and also the risk in missed doses if something is not prescribed in the correct manner. At RFH, this risk was only apparent during the switchover weekend, whereas a true Big Bang would mitigate this due to the simultaneous switch. 

Where to start?

For any Trust deciding to phase their rollout, or take the Rolling Thunder approach, contemplation of the ‘perfect ward’ to start on is imperative. 

A good tactic to take is to plan and reflect on the possible patient journey within the Trust.

Mapping the patient journey

ED

This is where the majority of patients present, and is  usually first or last to deploy in the project plan. As the ED is often a contained area, it can go live first, with the majority of patients being transcribed back onto paper if there is a decision to admit. Alternatively, the ED can remain on paper while the rest of the Hospital goes live, meaning as time goes on, more patients will need transcribing onto the electronic system as they move to electronic wards. Either way, specialty teams must be made aware of the possible need for dual prescribing during the transition period, to ensure patients have medications charted in the correct format and without delay. Similarly, potential reduction in productivity needs to be taken into consideration, and the time of year for ED deployment is imperative (that is, avoidance of Winter pressure months). 

Surgical wards 

Patients are likely to move to theatres and back. In this case it would be pertinent to deploy the EPR in theatres first, then move on to wards. Alternatively, if adopting the big bang approach, it might be sensible to agree once a patient is ‘Live’ on EPR that they remain ‘Live’.

Gynaecology wards 

Similarly, there might be a need to visit theatres.

Obstetrics patients

Labour theatre considerations, plus ante- and postnatal care wards. Will the system allow for antenatal patients to have their records shared with general practitioners, midwifery staff and other healthcare professionals who might need them? 

Specialist medicine wards 

Changes to medications can be frequent while correct regimens are being agreed upon.

Paediatric wards

A delicate group of patients with unlicensed and non-formulary options to be taken into consideration.

Intensive care areas 

Will these areas go live with medications, clinical documentation and charting of various necessary indicators all at once?

Ideally, a specific cohort of ‘stable’ patients would be the starting point, that is, those that have medications that do not regularly change, and who will not ‘touch’ other areas.

A proposed order of Go-Live for any phased deployment, would be to weigh up the pros and cons of starting with the chronic senior health medical ward(s) versus ED/acute medical wards, working through to the ‘eventual wards’ downstream of any surgical or other interventional procedures. It is essential that a ‘transcribing strategy’ is drawn up early and shared with the executive boards and senior teams so everyone is aware of the timescales and when their ward/area is expected to go live. 

Table 2 summarises the main topics of consideration, when planning such a project.

General areas for deliberation

Once the Go-Live plan has been decided, the finer details can be taken into consideration. A summary of valuable hints and tips from SGH and RFH Go-Lives can be seen in Table 3.

Support staff

Support staff in the form of champion users and super users are imperative for success. Clinical on-call super users, available 24/7 in the weeks after Go-Live is the ideal, along with extended IT service desk opening hours. To ensure these super users know the specific nuances to the individual Trust build, it would be ideal to offer secondments to existing Trust staff and provide extensive training. Once secondments are complete, the Go-Live knowledge, along with extended system experience and familiarity is retained within clinical areas. If this is not feasible and there is dependence on floor walkers, it is recommended they too have local specific training to ensure they are aware of the differences to other Trust builds they may have experienced. This prevents the development of ‘bad habits’ from the outset.

Reductions in services 

Consider reducing elective services in the weeks surrounding the Go-Live date.

Annual leave 

A reduction in annual leave over the Go-Live period for project teams to ensure the most knowledge is on site for deployment and post-deployment support.

Development of a robust escalation plan for any issues 

Ensure people know who to contact in local areas, and in turn that those people know how to escalate any issues to the right person, or up the chain of command. Have strong communications between the teams and feedback responses to any issues to people on the ‘shop floor’ so they are aware concerns are being acted upon. Whatsapp™ and other instant messaging tools are fantastic here. 

Creation of a transparent transcribing plan 

Share the planned route at Bed meetings and try to update during cutover as often as possible to the Executive Board. An EPR Go-Live can cause lots of anxiety among senior leadership, so having a clear, transparent and up to date plan can help ease this. RFH created a live EPR transcribing website and app, so all wards could see the expected and actual Go-Live times. SGH demonstrated a dedicated webpage that followed the rollout and shared good news stories.

Maintain the workflow in pharmacy 

Do not forget about what happens within the dispensary, and that new systems may result in an impact on ‘to take out’ turnaround times and other key performance indicators. Discharge summaries will take longer to write, therefore times of peak dispensary work may alter, but also interventions, changes, and dispensing times may be impacted. Consider extending the pharmacy hours and employing extra staff to answer the inevitable increase of phone calls to the dispensary and medicines information lines. 

Some Trusts may be moving from an established EPR to a new one. Generally, the same issues must be taken into consideration, however for ease of transcription, it may be necessary to follow downtime procedures and move all patients onto paper the weekend before, to allow for a seamless transition to the new system.

Conclusions

This article outlines some of the considerations of ‘Big Bang’ versus ‘Phased’ ePMA rollout. Each method has different challenges, prerequisites, and benefits, but the decision as to which approach best suits a single hospital should be made based on the individual Trust’s requirements, and by looking at the project and stakeholders as a whole. The main take home messages, despite the method of roll out chosen, are: use of motivated, local champion super users, senior management oversight of a well communicated cutover plan, and a robust training strategy. 

References

  1. Ranji SR, Rennke S, Wachter RM. Computerised provider order entry combined with clinical decision support systems to improve medication safety: a narrative review. BMJ Qual Saf 2014;23:773–80.
  2. Franklin BD, Puaar S. What is the impact of introducing inpatient electronic prescribing on prescribing errors? A naturalistic stepped wedge study in an English teaching hospital. Health Informatics J 2019; 18 Mar:1460458219833112.
  3. Wilkinson E. E-prescribing in hospitals: ten ways to make it a success. Pharm J. www.pharmaceutical-journal.com/news-and-analysis/features/e-prescribing-in-hospital-ten-ways-to-make-it-a-success/20206912.article?firstPass=false (accessed January 2020).
  4. St George’s goes live with first-of-type Cerner critical care system. www.ehealthnews.eu/industry/4126-st-george-s-goes-live-with-first-of-type-cerner-critical-care-system (accessed January 2020).

Acknowledgements

The authors thank Corinne Hooper, Senior Knowledge Leader, Medications Management, Cerner, and the solution deployment teams and staff at SGH and RFH Hospitals.



Most read




Latest Issue

Be in the know
Subscribe to Hospital Pharmacy Europe newsletter and magazine
Share this story:
Twitter
LinkedIn