Introduction

Pharmacy and therapeutics committees (PTCs) are multidisciplinary hospital teams of healthcare professionals involved in the medication use process. One of the main tools of a PTC is a continually updated list of medications, known as a formulary, to be preferentially prescribed, along with related information to ensure safe and rational medication use.1–3 The PTC should also provide guidance on subprocesses of the medication use process; for example, generic substitution and therapeutic interchange guidelines, or strategies integrating computerised physician order entry systems combined with clinical decision support systems.4–7

Assessment tools allow hospitals to determine the implementation level and quality of activities. For example, the Swiss Society of Public Health Administration and Hospital Pharmacists (GSASA) has developed the quality management standards for hospital pharmacies (QRHP).4–8 Numerous surveys related to PTCs and formulary systems exist and some have developed tools to improve PTC decision-making. However, no validated tool is available to comprehensively assess PTCs’ operating qualities.

Thus, the objective of this work was to systematically develop a PTC assessment tool that could thoroughly assess the operating quality and functionality of a specific hospital’s PTC, and therefore enable the identification of areas for improvement.

MethodsDevelopment of the assessment toolLiterature review

A scoping review in PubMed and Embase was conducted to identify relevant publications for the following research question: What subjects are addressed in the current literature that are important to developing a structured assessment of a hospital’s PTC and formulary system? The search strategy was elaborated in collaboration, and the selection and data extraction was performed by one reviewer. Index terms for PTCs and formulary systems were combined with index terms for guidelines, assessments and PTC responsibilities (ie, decision-making). Additionally, titles and abstracts were searched for these index terms and synonyms. Publications (published from January 2000 to March 2020) in English and German were retrieved and, after deduplication, screened for inclusion (description, assessment or practice guidelines for PTCs or formulary systems) and exclusion (focus on a specific substance or pharmacological class, settings other than hospitals, clinical studies or no full text available) criteria. The references of included publications were searched for additional relevant publications. (For details, see online supplemental appendix B.)

Consolidation of the assessment tool

The identified subjects were converted into potential assessment items. The items were categorised into focus topics and tabulated. To consolidate the assessment tool, we developed a weighting system, consisting of item weights and publications weights. The publication weight was estimated by assigning the following five score-types for each included publication and adding up the points.

Publication type: practice guideline (7 points), literature review (6), survey (4), research article (3), decision-making tool (2) and overview or summary (1).

2017 health expenditure by financing schemes of the country of origin, current US$ per capita: >9000 (4 points), ≤9000 and >4000 (3), ≤4000 and >2000 (2) and ≤2000 (1).

Publication period: 2010 or after (2 points), before 2010 (1).

SCImago Journal Rank: >1.000 (2 points), ≤1.000 or not applicable (1).

Setting: university hospital (2 points), other (1).

The lowest possible publication weight for an individual publication was five and the highest was 17. The relevance of every item (item weight) was calculated by adding up publication weights of all included publications addressing the particular item in their full text. Assessment items with a weight <30, calculated based on the weights of the publications mentioning the item, were removed. It was important to exclude items to ensure the applicability of the assessment tool and the relevance of the assessment item. The reasoning behind a cut-off of 30 was that an assessment item was to be included if it was covered by at least two highly relevant publications, and excluded if it was only covered by one publication, to minimise the influence of selection bias.

Applicability of the assessment tool

The tool was pilot tested in an 800-bed Swiss university hospital. The individual items were answered in a descriptive manner by reviewing available documentation and conducting a semi-structured interview with a pharmacist PTC member. The answers were verified by another pharmacist PTC member. As the information gathered was purely descriptive, the tool on its own could not analyse a PTC’s operating quality. Therefore, the collected information had to be additionally evaluated, which was done with a SWOT (strengths, weaknesses, opportunities and threats) analysis. This method is widely used for the structured identification of future organisational strategies. The internal factors (strengths and weaknesses that can be influenced by the institution) were based on the information obtained by the applied assessment tool. For the external factors (opportunities and threats) an environmental analysis was necessary. To structurally conduct it, the PEST framework (political, economic, social and technological) was used. The strategic consequences result from all possible combinations of the factors. To minimise subjectivity, three experts reached consensus about the SWOT and PEST categories. When identifying improvement opportunities, the focus of the future strategies should be the minimisation of weaknesses that may otherwise jeopardise patient safety. For this reason, the ‘weaknesses combined with opportunities’ and ‘weaknesses combined with threats’ strategies were the basis for the formulation of our improvement strategies.

ResultsAssessment tool development

A total of 1704 publications were identified. After removal of duplicates, title/abstract searching and full-text screening, 93 publications met our inclusion criteria. Additional eight publications were identified through other sources.

The included publications resulted in 101 potential assessment items. The subsequent weighting process determined 60 items with a weight of ≥30, categorised into eight PTC-related focus topics (for the focus topics and an item example see figure 1; for the final tool, see online supplemental appendix A)

Figure 1

Top: Process carried out to systematically develop the assessment tool with subsequent pilot testing and identification of opportunities for improvement by using the developed tool in combination with a SWOT analysis. Bottom: Example of the tool application. CPOE, computerised physician order entry; PTC, pharmacy and therapeutics committee; SWOT, strengths, weaknesses, opportunities and threats.

Applicability of the assessment tool

In the pilot, all 60 items could be answered. The assessment tool allowed us to obtain a comprehensive picture of the pilot PTC’s current situation and to identify 15 strengths and 18 weaknesses. In addition, we identified seven opportunities and nine threats by applying the PEST framework. These findings combined were used as a basis to identify future improvement strategies, which were grouped into four areas for improvement (see figure 1).

Discussion

To our knowledge, this is the first comprehensive tool to assess PTCs. As we aimed to extensively assess PTCs, we used a broad research question and eligibility criteria for the scoping review. This yielded many records. Nevertheless, most included publications proved relevant for the formulation of the assessment items, as only four did not mention any included assessment item. Relevance was assigned with the self-developed weighting system instead of categorising levels of evidence. This approach was chosen because the information extracted from the publications were not primarily focused on outcomes. For instance, a PTC practice guideline is more relevant for the formulation of assessment items than a systematic literature review focused on one specific PTC task. In addition to the publication type, publications originating from a country with similar healthcare expenditure to Switzerland were graded higher than countries with lower expenditures, which potentially face different challenges in their PTCs.9 10 Articles published after 2010 were graded higher to acknowledge more contemporary findings. Studies conducted in a university hospital were graded higher as we aimed to assess the highest possible complexity of healthcare systems. With a cut-off value of 30, we included 60 out of 101 assessment items. In our opinion, the applied weighting system did not exclude any relevant item and was especially successful in minimising redundant information.

We found two frameworks that assess PTCs and formulary systems.10 11 Unlike our tool, which is more comprehensive, Lima-Dellamora’s framework11 is limited to structures and processes needed for medication selection and has a strong focus on Brazilian hospitals. The second framework outlines similar topics within a different structure; however, these relevant topics are rather descriptive and no assessment items are extracted.12 Our assessment tool includes key PTC aspects, formulated as performance indicators elsewhere.13 14 The Swiss QRHP includes a formulary section and our tool covers all topics addressed there as well.8 Our tool has a higher level of detail than existing instruments and includes contemporary topics, like the use of technology to guide formulary medication use and considerations on biosimilars or drug shortages.

The assessment tool was pilot tested in one hospital and allowed us to obtain a comprehensive picture of this PTC. The tool helped to identify opportunities for improvement when combined with a SWOT analysis. However, further validation in other hospitals—and most importantly through researchers not involved in the development—would be advisable.

Limitations

The main limitations of the assessment tool are that the development based solely on a literature review and the many items may limit its usability. To improve this, the tool’s scope could be narrowed to the most relevant items by applying a consensus method (eg, the Delphi technique). We addressed this problem by weighting the items, which proved to be beneficial in our case, as we showed that the most important items are still included. One limitation of this system, however, is that it is not validated but developed and piloted by our research team only. Another limitation is that the selection and data extraction processes of the literature review were not conducted independently by more than one reviewer. This could have been especially critical, as some extracted data were qualitative, and consequently, the development of the assessment tool relied on the interpretation of the reviewer. Since the assessment tool aims to evaluate contemporary PTCs, this was the main rationale for excluding articles published before the year 2000. However, relevant work was published before 2000; for example, a set of PTC indicators.14 Nevertheless, we do not believe that additional publications would have caused extensive changes to the assessment tool, as we showed that the relevant topics are covered even when comparing our indicator set with the aforementioned indicators published earlier. As the tool was only tested in one hospital, its generalisability needs further research.

Conclusion

We systematically developed a comprehensive PTC assessment tool based on a scoping literature review. The tool was pilot tested and used in combination with a SWOT analysis to identify improvement opportunities for the pilot hospitals’ PTC. Further consolidation and validation of the tool to facilitate its application in different hospitals would be advantageous.

Data availability statement

Data are available upon reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information. Data most relevant to the study are included in the article or uploaded as supplementary information. Additional data are available upon reasonable request.

Ethics statementsPatient consent for publication

Not applicable.

Ethics approval

Not applicable.

Acknowledgments

We thank Professor Andrea Burden, Assistant Professor of Pharmacoepidemiology, ETH Zurich, Switzerland for the opportunity to conduct this research project in her group.

References↵↵↵↵↵↵↵↵

Schweizerischer Verein Der Amts- und Spitalapotheker (GSASA), Agentur für die promotion und die evaluation Der Qualität Im Gesundheitswesen (APEG). Referenzsystem Qualität für Spitalapotheken, RQS .

↵↵↵↵↵↵