¹Universität Heidelberg, Medizinische Fakultät Heidelberg / Universitätsklinikum Heidelberg, Medizinische Klinik (Krehl-Klinik), Innere Medizin IX – Abteilung für Klinische Pharmakologie und Pharmakoepidemiologie, Heidelberg, Germany

Text

Introduction: Computerized Physician Order Entry (CPOE) systems for medication orders are evaluated in many hospitals. Heterogeneity of reported outcomes impairs the comparison of evaluation studies in systematic reviews or meta analyses. To increase comparability, core outcome sets (COS) define a standardized set of outcomes that should be measured as a minimum in all trials in a specific area [1]. In the first step of COS development, existing outcomes need to be identified and classified into categories. To this end, we (1) performed a rapid review to identify outcomes used in CPOE evaluation studies and (2) categorized the outcomes. The Core Outcome Measures in Effectiveness Trials (COMET) initiative recommends a general taxonomy for such classifications [2]. However, this taxonomy was developed for outcomes at an individual patient level. Therefore, the proposed domains are not necessarily suitable for CPOE evaluation studies. The specific aim of this work was to develop an adapted classification suitable for interventions targeting not only individual patients but also processes in health care.

Methods: A systematic search in two databases (Pubmed, Embase) was combined with a validated data saturation approach [3] to extract the reported outcomes of 125 CPOE evaluation studies. Those outcomes were classified into a generic level, consisting of core areas and domains and a subordinated specific level containing the explicit outcomes. COMET’s recommended outcome taxonomy, especially the core areas (a) death, physiological or clinical, (b) adverse events, (c) resource use and (d) life impact were used as a basis. During the classification, this taxonomy was tailored to the specific needs of CPOE evaluation outcomes by modifying and adding core areas and domains. All outcomes were classified by two researchers independently. Results were compared and differences were discussed with a third researcher until consensus was reached.

Results: 365 outcomes were classified into 8 core areas and 36 domains, containing 348 specific outcomes. Similar to (a), (b) and (c) of the COMET taxonomy, we included the core areas (1) clinical outcomes, including domains like mortality, (2) adverse effects, errors, safety (e.g. medication errors) and (3) resource use (e.g. cost). We changed (d) to (4) user related outcomes, which covers domains like user satisfaction or workload. We added (5) medication process, (e.g. duration of order entry) (6) system related/technical, (e.g. usability), (7) implementation related (e.g. implementation strategies) and (8) CPOE adoption, including user and hospital perspectives.

Discussion: Classifying identified outcomes is an important step in COS development. As recommended, we used COMET’s standard taxonomy as a basis. To cover all aspects of CPOE evaluation, especially outcomes that are not measured on the individual patient level, we provide an enhanced taxonomy for this topic. To further discuss the classification, we will conduct expert interviews.

Conclusion: The definition of 8 core areas and 36 domains allows a structured overview of the identified outcomes. This is an important preparation for the identification of the most important core areas and domains in a consensus process, which will be the following step in the development of a COS for CPOE evaluation studies.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.

References

[1] Williamson PR, Altman DG, Bagley H, Barnes KL, Blazeby JM, Brookes ST, Clarke M, Gargon E, Gorst S, Harman N, Kirkham JJ, McNair A, Prinsen ACA, Schmitt J, Terwee CB, Young B. The COMET Handbook: version 1.0. Trials. 2017;18(Suppl 3):280. DOI: 10.1186/s13063-017-1978-4
[2] Dodd S, Clarke M, Becker L, Mavergames C, Fish R, Williamson PR. A taxonomy has been developed for outcomes in medical research to help improve knowledge discovery. J Clin Epidemiol. 2018;96:84-92. DOI: 10.1016/j.clinepi.2017.12.020
[3] Veen KM, Joseph A, Sossi F, Blancarte Jaber P, Lansac E, Das-Gupta E, Aktaa S, Takkenberg J. Standardized approach to extract candidate outcomes from literature for a standard outcome set: a case- and simulation study. BMC Med Res Methodol. 2023;23(1):261. DOI: 10.1186/s12874-023-02052-x