Medication reconciliation is a method for reducing medication errors, patient harm and costs. In this study we showed that insufficient knowledge of care professionals, unclear task allocation, and a lack of collaboration within, and between, inpatient and outpatient settings are important barriers from the perspective of the health care professionals. On the other hand, health care professionals highlight drivers, such as a good implementation plan, patient empowerment, and obligation by the government, as benefiting the implementation. The barriers and drivers we identified can help to develop strategies for improving the implementation of medication reconciliation.
Our study found several barriers to the implementation of medication reconciliation. Firstly there was a lack of awareness as well as insufficient knowledge of health care professionals. Noticeable was the lack of awareness about the health care problem. Professionals do not know how many medication errors are made and the impact these can have on the wellbeing of the patient. Knowledge of the bundle and how best performance can be achieved was insufficient. Professionals did not recognise the positive impact that the bundle would have on their everyday care. It was not clear how quality indicators to evaluate the implementation of the bundle should be measured, registered or given as feedback to professionals.
Secondly the necessity of reallocation of tasks was not clear. Currently, there are no clear agreements about tasks and responsibilities, despite the fact that the bundle was released as early as January 2007. The bundle did not explicitly state who, where and when to perform the different parts of medication reconciliation. There were several opinions among professionals on how best performance should be reached within the process of medication reconciliation. Various studies conclude that medication history taking by pharmacists or pharmacy technicians results in fewer errors compared to history taking by clinicians [1
]. Despite this, clinicians, in particular clinicians from non-surgery specialities, were opposed to reallocating their tasks. Their unwillingness originates from their autonomous way of working. Clinicians feel they should undertake this task, not least because ultimately as a clinician they are legally responsible for the complete treatment of the patient. Professionals indicate that when tasks are performed by someone other than the person responsible, it will result in uncertainties. Often clinicians do not trust the medication list if a nurse, pharmacy technician or pharmacists has done the history taking. All this resulted in inefficiencies and in different ways of working in the various departments. This complicated medication reconciliation at hospital admission, at transfers within the hospital and on discharge.
The third barrier is the impact market competition has on communication, understanding and collaboration. The relationship between community and hospital caregivers has become worse since the introduction of market-based competition in the Dutch health care system. Community pharmacies are more reserved in communicating medication information to other pharmacies. This is because, in their opinion, this information could also be used to lure patients with multiple medications to those competing pharmacies. Pharmacies gain most of their income from those patients. Since many hospitals currently also include hospital pharmacies, community pharmacies are equally reserved in sharing medication information with those hospitals. The probability exists that in the future performing medication reconciliation will be reimbursed by insurance. But this too would not encourage the cooperation between community and hospital pharmacists. They both want to do the job, because it is profitable. A lack of communication, understanding and collaboration between hospital caregivers and community caregivers is an important barrier to the medication reconciliation process [34
An important driver found in our study was obligation by the government. It is obligatory to perform medication reconciliation in every Dutch hospital. The attitude of professionals changed when they had no choice but to implement it into their work. The hospital management reinforced the obligation of the government by assigning responsibility for the implementation to departmental heads and installing a professional who facilitates the process. There were indicators formulated, as described in the method section, to monitor the implementation of the bundle. Up to now, however, departments who do not co-operate have not been sanctioned.
Secondly, several interviewees mentioned the importance of a planned phase of implementation. A multidisciplinary team should be involved from the start comprising all stakeholders in the implementation of medication reconciliation. In particular this should include community care professionals such as community pharmacists. This team should standardise the process of medication reconciliation through the development of protocols and forms, which include all the wishes and needs of the professionals involved. If the implementation phase is carefully planned, the process of medication reconciliation standardised and the environment in which the intervention is implemented taken into account, then it is more likely to succeed.
Thirdly, patient awareness should be improved. Professionals indicate that medication reconciliation is limited by poor health and medication literacy. That is that patients are not aware of the medication reconciliation process and do not realise that theirs is an important task in this process. They are not aware of the importance of having a clear and up-to-date insight into their own medication.
The barriers and drivers found in this study are consistent with results of previous similar studies, all carried out in the U.S. [24
]. These studies also found that: it is crucial that all parties involved have clearly defined roles and responsibilities; that there is a lack of uniformity across hospitals; that pharmacists do not play a significant enough role in the medication reconciliation process; that information was fed back infrequently; and that patients have little knowledge of their medication. The important drivers mentioned include: phased implementation; a multidisciplinary approach where hospital and community caregivers generate a common vision; and collaboration between the involved stakeholders. Several barriers found in these studies, such as ‘medication list not available’, ‘no access to outside records’ and ‘cumbersome hospital systems’ could be overcome with a regional or national electronic patient medication file. Other research also focuses on electronic tools as driving the implementation of medication reconciliation [36
]. The importance of patients in medication reconciliation is recognised by Varkey et al., who emphasise the importance of patient education [37
Strategies can be drawn up to improve the implementation of medication reconciliation based on the barriers and drivers identified. These have been summarised in Table . These are found to influence implementation on different levels, for example on patient, professional, and organisational level. Therefore, to improve implementation a multifacitated and multitargeted strategy which intervenes on different levels should be considered. Some of the suggestions mentioned in Table are discussed hereafter in more detail.
Suggestions for strategies based on barriers and drivers found
Professionals with more awareness of the importance of medication reconciliation are more likely to change their performance [38
]. An analysis of the process of medication reconciliation gives insight into the current process of care and its inefficiencies. Collecting feedback about the implementation, and about the reduction in medication errors keeps professionals informed and engaged. A lack of clarity about tasks and responsibilities can be resolved with a clear written policy. Research into the effectiveness of task reallocation of the medication history taking to pharmacy technicians should be emphasised. They are most specialised in relation to their lower salary, probably leading to higher cost-effectiveness.
A lack of collaboration between the many health care professionals involved in medication reconciliation can be addressed by a partnership between hospital and community pharmacy providers. This is important to ensure uninterrupted communication both in the inpatient and outpatient settings [24
]. Community pharmacies should be considered as a partner in medication reconciliation, especially with regard to high risk patients. Community pharmacies have frequent and direct contact with patients, resulting in a complete overview of patients’ medication history and offers opportunities to educate patients.
Finally, in every aspect of care patient empowerment will become more and more important. Therefore it is essential to create awareness among patient of the importance of carrying an accurate and up-to-date list of medications. Patients should be encouraged to take their own responsibility. They want to be in control of their own care, and thus in control of their medication [24
A methodological strength of this study is that we applied qualitative methods to explore, in-depth, all possible barriers and drivers. Interviews have proven to be a useful method of providing in-depth information on barriers and drivers with regard to implementation while at the same time exploring and understanding the motivations underlying behaviour [39
]. The interviews enabled us to identify the most relevant barriers and drivers perceived by the persons who were involved in undertaking the implementation of medication reconciliation. Our analysis of barriers provided detailed information for professionals or organisations, regionally or nationally, to develop multifaceted implementation strategies for improving the implementation process of medication reconciliation.
Even so, several limitations should be considered when interpreting our findings. The selection of interviewees from one hospital and the selection of a limited number of a different kind of health care professionals might raise questions about the generalisability of our findings. The results are, however, consistent with previous studies on this subject. A study including a larger sample in different types of hospitals could be performed to confirm our findings. Secondly, neither patients nor GP’s were involved in this study, while medication reconciliation is a multi disciplinary process. Including the patients’ and GP’s perspective would have strengthened the findings of this study. Finally, the effectiveness, the cost-effectiveness and feasibility of the strategies suggested are unknown and have yet to be tested in well-designed controlled evaluations.