Medication errors are one of the main causes of morbidity in hospitalized patients. Between 10% and 70% of medication histories contain at least an error, up to 1/3 of these errors have the potential to cause harm to the patient and more than 50% of medication errors occur in the transition between care.1 Medication reconciliation (MR) is recognized as an important tool in avoiding discrepancies. MR is defined as the process of analyzing a patient's best possible medication history (BPMH), whenever there are changes and should be carried out at vulnerable/critical points of care transition, namely at hospital admission and discharge and at unit transfers.1,2 It has previously been demonstrated that MR interventions carried out by hospital pharmacy are cost-effective.3 Healthcare institutions must promote the implementation of the MR process, namely, through the adoption of a systematic approach to MR, involving a multidisciplinary team that identifies and establishes appropriate strategies for implementing the process. Ideally, the possibility of simultaneously implementing this process at all critical points should be evaluated. If this is not possible, an operational plan must be defined, considering the first critical point as the admission that results in hospitalization.1 Pharmacist involvement at all stages of the reconciliation process for every patient may not be feasible because of reasons such as time commitment, number of pharmacists on staff, and other professional responsibilities. For this reason, Ordem dos Farmacêuticos, which is a professional public association that represents Portuguese pharmacists and regulates their activity in the national territory, published a guideline that lists some of the criteria for selecting patients. As this list is quite extensive, they recommended that the pharmaceutical intervention priorities considered most relevant in relation to the type of patients being monitored, before starting the service and in accordance to available human resources.4

Although the term “high-risk patient” is referred to in the literature as a criterion for selecting patients for whom therapeutic reconciliation should be performed, the criteria were undefined or defined in an empirical way.5 Patients undergoing major orthopedic surgeries such as hip arthroplasties/hip and knee prostheses, represent a particularly high-risk group.6,7 These patients are often elderly, with multiple comorbidities and subject to polymedication, which increases the complexity of therapeutic management and the risk of drug interactions.8 For this reason, during the development of the MR process by the hospital pharmacy team, there was a need to identify and establish a consensus over which patients are at high risk of medication errors and error-related adverse events.

This original research aims to validate the criteria that identify patients at high risk of experience medication errors and error-related adverse events undergoing major orthopedic surgery, in terms of home drug reconciliation.

The multidisciplinary research group drafted the form with the high-risk criteria in MR previously defined, based on the probability and risk of the adverse event to happen. This form included 29 inclusion criteria (Fig. 1) and 2 new inclusion criteria were suggested by experts: opioids and systemic antifungals. 10 experts were invited, and the invitation was accepted by all. The number of experts who responded to the form ranged from 9 in the first round to 8 in the second round and finally 6 in the third round. (Fig. 1).

Fig. 1.

Delphi flowchart. Created by the authors.

(0.31MB).

Several studies show the relevance of therapeutic reconciliation in the transition of care, as well as the pharmacist's role. This process is, for example, part of the Medication Management and Use Standard Expectation by Joint Commission International.12 This activity is time-consuming, so it is necessary to define high-risk criteria. The Ordem dos Farmacêuticos guideline mentioned before lists some of the criteria for selected patients, but this list is quite extensive, so they recommended that the pharmaceutical intervention prioritize certain ones with previously defined criteria.4 Patients undergoing major orthopedic surgeries represent a particularly high-risk group, not only due to patients characteristics (age, comorbidities, and polymedication), but also because orthopedist surgeon physicians and anesthesiologists have a greater focus on therapies directly related to the surgical procedure.13–16 For this reason, the development of a consensus regarding high-risk criteria in therapeutic reconciliation makes it possible to optimize resources and direct available resources in these patients who truly benefit from pharmaceutical care.

Our study combined the development of inclusion criteria by the multidisciplinary group, and its validation according to the Delphi method, that is a flexible research technique well suited when there is incomplete knowledge about a phenomenon.10 It should be noted that Delphi method has some advantages when compared to a typical group interaction, namely because it allows anonymity of responses and from each other, reducing the pressure of group conformity. Another advantage is the opportunity to gather information from experts that can be geographically diverse. Disadvantages include loss of experts during rounds, weakening the consensus. Additionally, experts may be hesitant to share views that differ from other experts and may move towards consensus.17

As far as we know, this is one of the first studies to establish consensus regarding high-risk criteria for MR. Consensus was reached on 30 criteria: polypharmacy (defined as 5 or more medications daily),18 anticoagulants, oral chemotherapy (not hormone therapy), immunosuppressants, antiretrovirals, antimyasthenics (pyridostigmine and neostigmine), insulin, corticoids, neuroleptics, antiarrhythmics, digoxin, carbamazepine, phenytoin, valproate, thyroid drugs, antiglaucoma therapy, antiaggregants, urgent surgery, systemic antifungals, opioids, antiparkinsonics, beta-blockers, age ≥ 65 years, length of stay ≥5 days, lamotrigine, diuretics, antidepressants, ACE inhibitors, sulfonylureas, and anxiolytics. At the end, 1 criterion did not reach consensus: calcium channel blockers. Regarding calcium channel blockers, there is lack of controlled evidence, so no recommendations can be made either for or against the use of these drugs. Therapy should be continued or restarted based on the same criteria that would apply in the non-operative setting.19

Some included criteria deserve further discussion, like anxiolytics, namely benzodiazepines since they are the most commonly prescribed drugs in this class, we found in the published literature either indication to maintain or withdrawal in the perioperative period, due to the increased risk of serious withdrawal symptoms versus the risk of adverse effects.20

With the criteria that did not reach consensus, we proposed to hold a focus group to determine the reasons for not having consensus and decide whether to include it as a high-risk criterion. However, using modified Delphi method, it is important to know that this active participation of the steering group can cause bias through opinion of members.9 It should also be noted that the criterion of length of stay ≥5 days required clarification in the second round, at the request of an expert, as he did not understand the objective of including the length of stay in the definition of criteria, and it was explained that is related to an increased risk of developing complications.

With this work, in addition to having established a consensus regarding the high-risk criteria in therapeutic reconciliation, it was possible to develop an informatic alert system, based on these criteria, which will help us in the management of patients who require more attentive therapeutic reconciliation.

It is also important that, despite being an innovative study, there are methodological limitations, namely the small sample of experts who answered the questionnaire and, consequently, the relatively large attrition rate throughout the rounds, which may impact the final results. This limitation was acknowledged from the outset, and justifies the use of a flexible methodology for consensus, as explained in the methods.

In conclusion, we developed and validated a list of 30 criteria to identify patients at high risk of experience medication-related adverse events undergoing major orthopedic surgery. These criteria may help improve human resource management for clinical pharmacy activities by prioritizing patients who would benefit most from MR. This methodology could be replicated in other clinical areas.

Medication reconciliation is relevant in the transitional care. As hospital pharmacists, we would like to go further in our ability to optimize the quality of the pharmacotherapy our patients receive, but, given the limited resources, it is necessary to identify those who could benefit most from this activity. This publication is of special interest because it is the first, as far as we know, to provide a list of criteria to prioritize patients at high risk of experience medication-related adverse events.

These criteria may help to improve human resource management for clinical pharmacy activities by prioritizing patients who would benefit most from medication reconciliation.

No funding.

Mafalda Cavalheiro: Writing – review & editing, Writing – original draft, Visualization, Validation, Formal analysis. Jesús Cotrina-Luque: Validation, Supervision, Methodology, Investigation, Conceptualization. Gonçalo Duarte: Methodology, Investigation, Conceptualization. Patricia Silva: Writing – review & editing, Writing – original draft. Cátia Pereira: Supervision. Miriam Capoulas: Validation. Cláudia Santos: Validation.