Systematic review of observational coding systems to assess patient-healthcare professional interaction

Objectives: Systematic coding is used to study interactions between patients and healthcare professionals from an independent observer perspective. Many coding systems are available, but an up-to-date overview is lacking. We aimed to: (1) provide a comprehensive overview of systems for systematic coding of patient-healthcare professional interactions; and describe their 2) general characteristics and development, and 3) validation and adaptation.
Methods: We systematically searched peer-reviewed empirical articles in five databases (Cochrane, Embase, PubMed/Medline, CINAHL, PsychINFO) using variations of the following keywords: (i) patient and/or other stakeholder, (ii) healthcare professional (iii), interactions in healthcare, (iv) coding tool, and (v) development and/or validation. All titles/abstracts and full-texts were screened independently and in duplicate. Additionally, coding systems were identified through an earlier review, an open-access research database, and a forward-reference search of all included coding systems up to that point. For all eligible systems, we extracted characteristics and psychometric properties.
Results: From a total of 6950 identified articles from literature databases, 188 full-text articles were screened. Thirty-five articles were included from additional sources. In total, we included 124 articles describing 98 coding systems. Systems were highly variable in terms of topic (e.g., patient-centered communication, shared decision making, behavior change counseling), clinical context (e.g., oncology, pediatrics, generic), rigor of development and reporting, coding complexity, and extent of psychometric testing. Inter-rater reliability was reported for most coding systems; only few were tested for other types of reliability or for validity.
Conclusions: A plethora of coding systems are available, but more systematic reporting and psychometric testing are urgently needed to enhance evidence of validity. Testing may initially focus on the most relevant and broadly applicable coding systems.
Practice implications: These results can aid researchers in selecting the most suitable coding system for their purposes. Researchers may consider using or adapting existing systems, before developing new coding systems.