Study-design
For this study a multiple case design is used, in which several cases and qualitative
methods are taken into account 21], 22]. The study sample consisted of seven cases, each case being a neighbourhood in the
province (region) of Noord-Brabant. In this context, a neighbourhood is defined as
an operationally viable geographical or organisational area (e.g., part of municipality,
village) 23]. The cases were selected by managers of three different regional public health services
(GGDs) and one regional primary care support structure (ROS Robuust) which falls under
the control of the provincial authorities 24], 25]. They selected the neighbourhoods mainly on the basis of existing contacts, open
attitude for collaboration between public health and primary care (e.g., GP and municipality),
location (at least two municipalities of each participating regional public health
service) and size (two medium-sized and five small municipalities). The province contains
67 municipalities 26]. In each neighbourhood a stepwise approach to develop integrated district plans or
activities was implemented in the period July 2013 to July 2014 (roughly a year).
To supervise the process, a core team was assembled for each neighbourhood, consisting
of an epidemiologist and a policy advisor from the regional public health services
(GGD) and a policy advisor from the regional primary care support structure (ROS).
To help them with their work, core team members attended three ‘inspiration days’
(devoted to collaborating and networking in the neighbourhood, preparing and holding
policy dialogues, and embedding plans or activities) and materials were made available
(e.g., specimen district health profiles and policy dialogue work forms). A project
group concerned in the province and a steering committee consisting of experts in
the fields of public health, health care, the primary care association, health insurance
companies, advocacy citizens, municipalities, health centres and knowledge institutes
advised on this multiple-case study.
A stepwise process approach
In each of seven neighbourhoods, a stepwise approach was used to develop an integrated
district plan or activities, focusing on collaboration between public health and primary
care with a view to improving the health of the local population. District health
profiles and policy dialogues were the central tools used for the approach. The stepwise
approach is summarised in Fig. 1.
Fig. 1. Stepwise approach
The stepwise approach for this study was developed on the basis of existing tools
for district profiles and intersectoral action for health 15], 16], and nine exploring interviews (4 regional public health service, 2 regional primary
care support structure, 2 municipalities, 1 province). These exploring interviews
focused on enabling and disabling factors in working with district profiles or policy
dialogues, but also on neighbourhood-focused working and collaborating between public
health and primary care in practice.
Data collection
Use of the tools (district health profiles and policy dialogues) used in the stepwise
integrated district (activity) plan development process was studied by means of documentary
analysis, digital questionnaires, interviews and observations. In order to answer
the two research questions, data were gathered regarding both the tool content and
the (collaboration) process:
Documentary analysis of district health profile presentations, dialogue reports and
district plan or activity papers (content data).
(Group) interviews with each of the seven core teams (see Table 1). Two semi-structured interviews lasting about ninety minutes were held with each
core team, one partway through the process (after step 3) and one at the end (during
step 7). Hence, two sets of seven interviews took place. In addition, partway through
the process, an interview was held with each of the managers of the seven core teams,
because of the role that they played in starting up the process in the neighbourhoods
(e.g., contacting municipal councillors and municipal officials). As some managers
were responsible for two teams in total four interviews were held. A total of eighteen
interviews was carried out, focusing on the tools, the process steps and collaboration
between public health and primary care. The interviews were all recorded using a digital
voice recorder and transcribed (content and process data).
Table 1. Overview number of (group) interviews and questionnaires
Online questionnaires completed both by the core teams and by relevant local actors
(see Table 1). Relevant partners are partners in the field or municipality, primary health care
professionals, social welfare professionals and local residents. A total of fifty-one
actors from six pilot neighbourhoods were approached and forty of them returned the
questionnaire (a response rate of 78Â %). The forty respondents included sixteen of
the eighteen core team members (89Â %) and twenty-four of the thirty-three relevant
actors (73Â %). In one of the seven neighbourhoods no digital questionnaires were sent
out, because this neighbourhood was significantly behind the others in terms of progression
of the process. The questionnaire was made up of questions with closed answer categories
(no, more no than yes, neutral, more yes than no, yes) about the district health profile,
the policy dialogue and the collaboration between public health and primary care.
Respondents were also asked to rate (by giving marks ranging from 1 to 10) both tools
(content and process data).
Observations at policy dialogue sessions. Two members of the research team attended
nine policy dialogue sessions to obtain an impression of how the district health profile
was used in the dialogue and to identify any links between the public health sector
and the primary care sector that might have developed. In a number of neighbourhoods,
several policy dialogues were held. The research team members recorded their observations
in a report (process data).
In the interviews and digital questionnaires, the questions regarding the tools were
based on McMaster University’s evaluation of policy dialogues 27]. The process-related questions were based on the Intersectoral Collaboration Checklist
(e.g., collaboration) and the evaluation questions of the Regional Public Health Reporting
(e.g., content, use and usability 28]–31]. The questions were developed specifically for this multiple-case study. Practical
implementation of the plans and collaboration was due to take place after the project
ended and was not therefore studied.
Participants in the interviews were informed that contributions included in the results
would be made anonymous. The statements were also to be presented as group results
and would not be reducible to individuals. On the basis of these conditions and prior
to the execution of the interviews recorded on tape, participants agreed to take part
and gave verbal informed consent to use the results in publications. This study was
not subject to the Dutch Medical Research Involving Human Subjects Act. That means
research activities including human participants is exempted from ethics approval
in case they do not meet the criterion that participants are subjected to (invasive
or bothersome) procedures or are required to follow rules of behaviour.
Data-analysis
In order to assess how the tools were used in the cases (research question 1), various
aspects of each district health profile (e.g., area level, reference field, public
health and primary care indicators, sources and presentation form) and policy dialogues
(e.g., group size, dialogue work form, actors involved in the neighbourhood and chosen
themes) were analysed. The concrete activities or plans and collaboration agreements
arising from use of the tools were also included in the analysis. Information on similarities
and differences between the seven neighbourhoods regarding the various aspects was
based on analysis of data from documents, interviews and digital questionnaires.
In the assessment of how local actors viewed the stepwise approach in the cases (research
question 2), various aspects of the two tools and collaboration were analysed. Where
the tools were concerned, those aspects were the mean marks for the tools, the value
of the profile in the policy dialogue (e.g., comprehensible presentation, relevant
data), the value of the dialogue in the discussion of problems, priorities or solutions
(e.g., defining themes, identifying solutions, use of results after dialogue). Based
on experiences of the cases, also the dos and don’ts associated with the use of the
tools were included in the analysis. Where the collaboration was concerned, the aspects
were the extent to which the stepwise approach had contributed to strengthening bridges
between sectors (e.g., interest in participation, appropriate actors involved, satisfied
with participants’ input, consensus about focus, ties between sectors) and the appetite
for continued collaboration. That was done for the seven neighbourhoods by analysis
of data from the interviews, questionnaires and observations. The data analyses were
performed by two researchers.