Lessons learned in Electronic Medication Management
Electronic medication management systems offer the promise of significant benefits, including reduced medical errors, better compliance, time savings, cost savings, and better drug safety.
However, early adopters have described a number of limitations and clinical risks associated with first generation systems. These range from making wrong drug selections from electronic pick lists, irrelevant alerts which reduce productivity, to the duplication of data entry across different departments.
Despite these problems, the adoption of medication management systems continues to gather pace. Few, if any, organisations have pulled the plug on new electronic medication management systems. While not always documented in the same detail, the benefits of these systems still outweigh the negatives. The reduction of medication errors is particularly difficult to measure because institutions do not have reliable statistics. This does not mean that serious medication errors do not occur, as pointed out in the seminal work ‘To Err is Human’ by the Institute of Medicine.
That said, it is important not to ignore the pain that early adopters have experienced. So, what are some of the important lessons that we have learned so far?
Set clear objectives
Almost all of our clients are currently implementing or have plans to implement electronic medication management capabilities, if they have not already done so. Among early adopters of InterSystems TrakCare, hospitals have reported reduced adverse allergy and drug interaction incidents, as well as savings through better inventory control and support for pharmaceutical substitution.
But medication management is a complex area involving many functional areas within a healthcare organisation. Having a functioning electronic system in itself — even a state-of-the-art one — does not guarantee good practice, or that all the expected benefits will be achieved.
It is better to implement a system meeting a limited set of objectives — and limited issues to resolve — to demonstrate clear benefits at the end of the day. This provides a solid platform to achieve additional objectives and benefits over time, as opposed to a large implementation — with a promise of ‘everything for everyone all at once’ — that remains mired in unresolved issues.
Initial objectives will vary between organisations. In some healthcare jurisdictions there are incentive payments for electronic prescribing, and an initial objective is simply accurate reporting to be able to receive these payments. Many hospitals want to better manage scheduled or dangerous drugs to increase safety, comply with regulations, and prevent losses through theft. This may not be a priority for a small day surgery, but it may be for a hospital with a large and complex case mix.
Close the loop
The term “electronic medication management” describes a system which includes electronic prescribing, dispensing, and medication administration. This reflects the current trend for healthcare organisations to seek benefits that are most easily achieved using a “closed loop” approach.
Medication errors are one of the biggest sources of preventable errors for most healthcare organisations. Electronic medication management systems can reduce errors associated with illegible handwriting or ambiguous nomenclature, uncertainty about drug dosing, and interactions between drugs.
But systems which are not connected hold less promise of reducing medication errors. An ePrescribing system may provide a care provider with access to a drug database and decision support system, but not connect with a pharmacy system. So the prescription may need to be printed, taken to the pharmacy and re-entered when it gets there, duplicating data entry and introducing the risk of transcription errors.
Even if the ePrescribing and pharmacy systems are connected through interfacing, they may not use the same information sources. The prescribed drug may have different names, or be listed by brand rather than generic description, multiplying the risk of picking errors. In addition, the pharmacy may not have access to the same electronic patient record. While the prescribing doctor knows all about the patient, the same clinical information may not be visible to or even available to the pharmacist. Both systems may generate contraindication alerts, but in one system the reason for over-riding the alert is visible, and in the other it is not. This may necessitate a phone call from the pharmacist to the prescribing doctor, wasting time. And the whole scenario may be repeated again when it is time for the nurses to administer the drug to the patient.
Reducing medical errors, achieving time savings through reduced data entry and unnecessary communications, and cost savings through better stock control and reduced wastage are all easier to achieve with a unified healthcare information system, which includes medication management functionality, with all data residing in a single repository.
Plan for adoption
Most healthcare organisations — particularly in advanced countries like Australia — face barriers in implementing electronic medication management. Most already use a number of specialised systems with loyal clinical user bases, many of which have a role in medication management. Standalone pharmacy systems are widely deployed, and ePrescribing and nursing systems also exist in various forms.
Organisations going down the closed loop path have some difficult decisions to make. Should they replace existing systems with a single healthcare information system? Or should they integrate a new system with, say, their existing ePrescribing or pharmacy system? If they decide to integrate, should they plan to phase out their legacy systems at a later stage? What are the trade-offs in terms of increased complexity with an interfaced or even integrated system, versus the capacity of users to adapt to a completely new system?
To complicate matters further, every organisation will have its own objectives, necessitating their own individual adoption plans. And, since staff can only cope with so much change at once, these adoption plans need to be staged and ongoing, and acknowledge that systems and requirements will continue to change over time.
Involve stakeholders and communicate benefits
In healthcare organisations, resistance to change is almost inevitable. Clinicians may be reluctant to migrate away from a standalone system they are loyal to, or resent the effort involved in data entry with a new system.
That is why it is important to involve stakeholders from the early stages. With a flexible and configurable platform, organisations can create solutions which are easier to use, more functional, and with access to far more information than the legacy systems that clinicians are familiar with. Organisations also need to invest in training. The system may be easy to use, but clinicians are busy and usually respond best to task‑based and incremental learning opportunities.
While users will initially focus on their own environment, they will also become open over time to valuing the benefits achieved by upstream and downstream users. Additional data entry may be required. But if it is seen as benefiting others by driving additional functionality elsewhere in the cycle, users will be more accepting of the changes involved.
Organisations can accelerate this process by measuring the benefits achieved and regularly communicating them with stakeholders. Rather than initiating this after implementation, measuring benefits against objectives should be planned from the outset so that they can be demonstrated as early as possible. Healthcare information systems with embedded analytics make it easier to automate the measurement of benefits and report against KPIs. That way, users of the system can see their own roles as part of the overall effort of delivering improvements in medication management, rather than working within isolated units.
Maintain good practice
Modern healthcare information systems contain an ever‑expanding array of sophisticated clinical functionality designed to make life easier for care givers. So it is understandable that users expect to be able to rely upon the support of electronic medication management systems.
The problem is that systems are not a replacement for clinical knowledge or clinician engagement with patients. Implementing a good system does not excuse organisations from doing the hard work of determining exactly how that system should maintain and improve upon existing good practice.
Most systems come with a pre‑configured set of rules that users need to modify to suit their own requirements. This may include determining the minimum information to be collected by the system, configuring the kinds of alerts that different users will receive, management of the formulary and standardisation of measurement units, and enabling appropriate security for users.
Often organisations need to implement more complex rules to, for example, reflect the clinical pathway for a certain condition, such as diabetes. In addition to decision support systems based on standard drug databases, this may rely on data held in the patient record about age, weight and test results. Building these rules is much simpler in a unified environment based on common data sources.
Make the system friendly and manage alerts
Making the system easy to use for all groups involved in medication management is a major focus for systems developers. Different users have different functional requirements, and the system should allow them to navigate across different areas without losing track of the task at hand. Information in an electronic medication management system should be expressed in a way that is appropriate to each person’s role, even though the underlying data is identical.
The management of safety alerts is probably the number one ease-of-use issue within medication management systems. Distracting and unnecessary alerts have been the subject of numerous complaints from clinical users of early systems. Alerts about contraindicated medications, for example, are a powerful tool that can cut down on medication errors. But clinicians have complained that they are constantly having to deal with more alerts than they need to do their jobs.
Systems need to have the capacity to filter alerts, and this may need to go beyond a user’s specific role to also take into account their level of experience or seniority. Recent InterSystems’ development provides the prescriber with a unified view of alerts and a friendly way to action them. The display of alerts and the need to record an override reason can be constrained by severity of the risk. This important information provided by the prescriber is available to the dispensing pharmacist and administering nurse, so they are fully informed and assured that any changes to the patient condition between prescription and administration time are taken into account.
Medication management represents a significant part of InterSystems’ current development efforts, with advancements planned in the areas of workflow, user interfaces, medication charts, and the management of alerts.
Clinical users are standing on the doorstep of the potential for information overload, which has long been a part of other work environments. Improved visibility of clinical information also creates the capacity for increased stress. One of the challenges for systems developers is to effectively filter information to include just the parts that clinical users need to work efficiently and safely.
Systems developers, implementers, and healthcare organisations all have a part to play in taking on board the feedback of early adopters of electronic medication management systems. If we get it right, not only will later adopters be spared from suffering from the same problems, but their success will pave the way for further advances in healthcare information systems involving clinical users.
Product Manager, InterSystems
Joan Ostergaard is a Product Manager at InterSystems, focused on the strategic product direction of InterSystems TrakCare™. Her role includes market, business and technology analysis, and vendor relationship management. Previously she has worked at Kodak in positions including Clinical Consulting and Education Director for Asia Pacific, and Applications Manager across multiple geographies. She is a qualified Radiation Therapist and holds an Advanced Certificate in Management.
- Linda T. Kohn, Janet M. Corrigan, and Molla S. Donaldson, Editors, Committee on Quality of Health Care in America, Institute of Medicine. To Err Is Human: Building a Safer Health System (2000). National Academy Press, available from: http://www.nap.edu/catalog.php?record_id=9728
Posted in Australian eHealth