Investigation into coordinating dependencies between care pathways within mental healthcare: A qualitative case study and pilot testing of a new theoretical framework

Introduction

Mental healthcare organizations face pressures to reduce waiting times and delays within the care process while improving quality of care.^1,2 This need for change stimulated the development of care pathways, which are “a complex intervention of the mutual decision making and organization of care processes for a well-defined group of patients during a well-defined period of time”. ³ The implementation of care pathways brought about some benefits like reduction in length of stay for patients and waiting times, ⁴ increased throughput time, ⁵ efficiency gains and overall cost control. ⁶ Within the process of care, a patient can require several care pathways, comparable to a supply chain. ⁷ Each pathway describes the services provided during a specific period of time, ⁸ for example, the diagnosis and treatment phase. However, improvements in the care process (such as those stated above) are not a guaranteed result of successful single pathway implementation, due to dependencies between multiple pathways. For example, one pathway might achieve its goal, yet this may not have any positive effect unrelated pathways, resulting in limited gains for the entire process. ¹ As Malone et al. ⁹ explain, dependencies can be divided into: fit (when multiple activities collectively produce a single resource), flow (when one activity produces a resource that is used by another activity), and sharing (when multiple activities use the same resource). According to the Coordination Theory of Malone and Crowston, ¹⁰ in general organizational literature, coordination is required to manage these dependencies, regarding the care process as one single entity instead of a set of individual pathways. ¹¹

To manage the dependencies between the pathways, organizations can practice coordination in diverse ways. ¹⁰ To structure these coordination practices, one can consider the employees involved in coordination (the hierarchical level of the employees initiating coordination) and the (un-)intentional nature of coordination (i.e. if coordination is programmed beforehand or emerging spontaneously).^12–16 Integration of these theories in the current study resulted in the following matrix, analyzing the dependencies between individual pathways (see Figure 1). OPEN IN VIEWER

Boardroom instruction refers to a coordination practice where coordination is initiated vertically (by the strategic/tactical level employees) with a programmed character, as for example, could occur during a board meeting. During a ‘golf course chat' coordination practice, the employees of higher levels of an organization initiate coordination during spontaneous meetings, which may occur when directors meet on the golf course. The ‘water cooler talk' practice is initiated horizontally, by operational level employees, in a non-programmed manner, as might take place during interaction with colleagues while getting some water. The fourth and last practice is ‘workforce steering’, when coordination is initiated by operational level employees during planned meetings.

To recognize these dimensions in the workplace, the characteristics of coordination, based on the theory of Gittell ¹⁷ and Vissers and Beech, ¹⁸ are helpful (Figure 2). OPEN IN VIEWER

The coordination characteristics and matrix provided the basis for a new theoretical model that was used in the current study to research and compare coordination within organizations. This study can be seen as a first step in the search for the best practices for managing dependencies between pathways in mental healthcare. A case study was performed to explore current practices in coordination of dependencies between pathways in mental healthcare and to test the use of the theoretical model. The aim of this study was to investigate how (three departments within two) organizations coordinate dependencies between pathways, to discern what types of coordination are being used and apply these findings to the current knowledge base with a view to constructing a practical and theoretical framework.

Methods

Information about the two mental healthcare institutions and the departments studied is presented in Table 1.

OPEN IN VIEWER

Table 1.

Information of departments retrieved from the annual plans of the organizations or the departments themselves.

	Organization 1		Organization 2
Region	Eindhoven region		Tilburg, Breda, Etten-Leur region
FTE 31 December 2012	1.745		1.625
Clients in care in 2012	15.513		16.371
	Department 1	Department 2	Department 3
Employees	42 (25,19 FTE)	17 (13,28 FTE)	38 (28,16 FTE)
Clients	±500	125	±360
Target population	Adults with autism  spectrum disorder	Clients with  first psychosis	Adults with  autism spectrum disorder

From both mental health care institutions, one department serving the same client population was selected. In addition, a third department was included representing a different client population. This third sample was included so that the influence of client population and organization could be explored. Firstly, based on the document study and short face-to-face interviews with one key person per department (manager, quality coordinator, or team coordinator), the pathways that had been developed and implemented within the departments were researched in preparation of the in-depth interviews. Secondly, to investigate coordination in practice, eight to ten employees from each department representing all hierarchical levels and all care pathways were purposively selected and personally interviewed using a semi-structured protocol following the narrative approach of Riessman. ¹⁹ Anonymity was guaranteed. The interviews, around 40 minutes each and 27 in total, were conducted by one interviewer (first author) and covered the characteristics of coordination. To sort this data, the interviewer labeled all audio recordings and transcribed interviews as advised by Ritchie and Lewis, ²⁰ using the characteristics of coordination from the framework as a guide. Then, the labeled data per department were compared to the dimensions of coordination from the framework, leading to a description of coordination practice per department. Subsequently, the departments were compared.

Results

The document study revealed that departments 1 and 2 started implementing care pathways in 2009, whereas department 3 started implementing care pathways in 2012 (Table 2). One clear difference between the organizations concerned the development of the pathways for the treatment phase. No such development could be discerned within organization 1 (department 1 and 2), whereas within organization 2 (department 3), this was considered a priority. From the interviews with key persons of the departments, it became clear that each department had tackled working with care pathways in its own way. Within all three departments, a flow dependency between the care pathways was found. Two out of the three departments also exhibited a second type of dependency.

OPEN IN VIEWER

Table 2.

Care pathways and dependencies in the departments.

	Information retrieved from documents		Information received during interviews
	Pathways in use	Year of implementation	Dependencies
Organization 1 Department 1	Assessment	2009	Flow dependency
	Diagnosis	2009
	Intake	2009
Department 2	Intake	2009	Flow and sharing  dependency
Organization 2 Department 3	Intake/diagnosis	2012	Flow and fit  dependency
	Treatment 5 x	2012

Based on the interviews with the employees of the departments, it was determined which coordination practices were used in the departments (Figure 3). Within department 1, the coordination practice of boardroom instruction was apparent, with the tactical/strategic employees initiating coordination of the pathways in a programmed manner. The operational level employees described their part in the care process as: “islands of disciplines”, yet one employee also stated: “it is preferable to discuss clients with other colleagues”. An explanation for the fact that employees seemed only to be focused on one particular pathway could be that: “it is caused by increased work pressure, uncertainty and lack of communication”. In department 2, the opposite practice was applied, namely the water-cooler talk coordination practice. Through direct and non-programmed interactions, coordination efforts were communicated, with a team approach defining relationships of shared knowledge and goals among employees. Employees stated that “there is a sense of safety within the team” and “all employees take responsibility for taking over each other’s tasks”, even though another employee disagreed and said: “the alignment within the team is less than I expected”. The coordination practice of department 3 lay somewhere between these departments with a workforce steering practice to manage the dependencies between pathways. Among the operational employees, there was little collaboration and interaction. An employee stated: “I dare not introduce topics at meetings, because I don’t know the team that well”. The lack of spontaneous communication to share knowledge and goals indicated a programmed coordination dimension. OPEN IN VIEWER

Findings did not point to any influence due to client population or organization.

Some additional information presented itself during the interviews. First of all, only within department 3 were all employees aware of the pathways as described in the documents. In the two departments of organization 1, employees stated: “I don’t see care pathways” or “the system of care pathways is not an issue”, frustrating employees when “the timeframes of care pathways are not achievable”. Secondly, employees elaborated on the subject of the monitoring of coordination practices. In department 1, one employee monitors coordination by controlling all incoming and outgoing information concerning pathways. In department 2, on the other hand, it seems that all employees monitor coordination practices. Within department 3, people did not monitor coordination. They did, however, have an integrated technical system that tracked the progress of employees within the pathway.

Discussion

This study led to new insights in how coordination takes place with a view to managing dependencies between care pathways in mental healthcare. The need to coordinate care pathways ¹ was recognized in all the departments that were studied. Contradictory to Malone and Crowston’s theory that similar dependencies are managed in similar ways, findings indicate that each department incorporated their own coordination practice even though a flow dependency is present at all cases. Although each department coordinated their care pathways in unique ways, it was found that two out of three departments seem to prefer coordination to be initiated by operational level employees (e.g. horizontal) and/or in a programmed manner. Whether this finding can be generalized to other departments needs to be verified in further research. The current study provides an exploratory look at managing care pathways and a first step towards a practical theoretical framework to this end. Practical advice for organizations working with pathways, which insights were gained from this study, include sharing knowledge on pathways and their dependencies with employees to achieve awareness and to implement a function or system to monitor coordination practices.

On the subject of further development of a practical framework, we suggest it might be worthwhile to look into other aspects that may have an effect on coordination, such as psychological safety and organizational climate.^21,22 Other aspects that can influence coordination that would be worthwhile studying are organizational aspects such as routines.^13,17

The current study has some strengths and weaknesses. A limitation was that analyses were carried out by one researcher, which may have impacted the objectivity. The verbatim transcriptions and comparisons of interviews, on the other hand, helped to validate the analyses. Another limitation was the small number of departments included in the study. The findings of the current study, therefore, need to be regarded as exploratory. Nevertheless, the current study represents a first in depth investigation into the overall coordination of dependencies between care pathways and a step towards a theoretical framework. Developing and implementing care pathways is a global phenomenon. To ensure implementation is a success, organizations need to manage the dependencies between care pathways. This study provides a first step towards that end.