Abstract
Introduction

Increasing demand for interprofessional collaboration in health care settings has increased focus on the influence of conditions, but we know little about the magnitude of interactions between conditions. Conditions are determinant factors that can affect interprofessional collaboration. This paper aims to examine the relationships of intervention conditions and context conditions at the professional and organizational levels and analyze how they influence the staff’s perceived success of interprofessional collaboration. Intervention conditions relate directly to integrated care models (e.g. strategy, clarity of roles and tasks). Context conditions do not directly relate to interventions, but they surround interprofessional collaboration (e.g. management support, trust amongst staff).

Theory/Methods

This study develops an analytical framework drawing on the RMIC-framework to operationalize intervention conditions and the iCoach-framework to operationalize context conditions. The study was conducted as a multilevel cross-sectional survey in March of 2019 in the second largest municipality in Denmark (Aarhus). The study population was all frontline-staff members and managers in nursing homes, home care units and health care units. The final sample consisted of 498 staff members and 27 managers. Confirmatory path analysis was used to analyze the data.

Results

The results indicate that context conditions greatly influence intervention conditions at the professional and organizational levels and that the professional and organizational levels moderately co-variate. Professional level context conditions have the biggest influence on staff’s perceived success, partly because some of its influence is confounded by intervention conditions.

Discussions

Our study stress the importance of opening the ‘black box’ surrounding context as a condition type. Based on our results we would expect that: conditions in general are a good predictor of the success of collaborations; context conditions are the most influential condition type with an indirect effect confounded by intervention conditions; conditions on different analytical levels co-variate. However, this raises questions about variation across country, sector, and case. For example, the most important individual context conditions may vary depending on setting. Nevertheless, the most important context conditions in the present study are well known in much of the literature, which might indicate low variance across settings.

Conclusions

Practice and research in health service delivery need to move from a broad understanding of context as unchangeable and inconsequential, to an understanding of context as an active condition type, which requires closer attention.

Lessons learned

Political and administrative decision-makers can purposely mould context conditions. In the present study, the four context conditions at the professional level are far from unchangeable. This means that leaders should analyze and work with context conditions as part of designing and implementing integrated care initiatives.

Limitations

Firstly, although the analytical framework is drawing on validated frameworks, it is not a thoroughly validated measurement tool itself. Secondly, the outcome measure is a proxy-variable. Lastly, there can be hypothesized additional paths and variables in the final path analysis model (e.g. patient and system level conditions).

Future research

Future research should open the ‘black box’ of context, by further exploring how different types of conditions on different analytical levels play out in different settings.