Decentralization of youth care: the
neighborhood team approach of
Dutch municipalities
Lindy HouwingSupervised by dr. Jochen Mierau
19 January 2018
2
Table of content
1. Introduction ... 3
2. Literature review ... 5
2.1. Trends in the international health care market organization ... 5
2.2. Neighborhood teams in the Netherlands ... 7
2.3. Results from the Dutch neighborhood team approach ... 8
2.4. Crisis situations ... 9
3. Data collection and analysis ... 11
3.1. Data collection ... 11
3.2. Data analysis ... 12
4. Results ... 15
4.1. Organizational form of the youth care ... 15
4.2. Reasoning of the municipalities ... 16
4.3. Results experienced by municipalities ... 17
4.4. Results from the data analysis ... 18
4.5. Care suppliers ... 21
5. Conclusion ... 23
6. References ... 25
7. Appendices ... 27
Appendix A. Average youth care ... 27
Appendix B. The average amount of situations with as goal ‘stabilization of a crisis situation’ (half-yearly, starting at 2015) ... 35
3
1. Introduction
There have been trends of a more decentralized organization of international health care markets. In 2015, the Netherlands followed this trend and decentralized the care system, among which the youth care system. Before 2015, the responsibilities for youth care were highly fragmented. Municipalities fostered universal and preventive services. Health insurance companies were responsible for youth mental health care. Provinces provided more specialized care and support in case of serious development and/or parenting issues, as well as youth protection, youth probation, and foster care. The central government was responsible for the most specialized youth care, long-term care for physically or mentally disabled youths, and juvenile justice policy and related institutions. These responsibilities have shifted since 2015, when the new regulation came into effect. Since then, municipalities are responsible for different facets of youth care, from preventive- to highly specialized care.
There were several reasons for the Dutch government to support this change. According to the Rijksoverheid1 , the first is to make use of the social network of youth and their parents, so they can
work together with their social environment and professional health care operators to find solutions. Secondly, the government hopes to decrease the amount of medicine subscribed and the demand for healthcare. Moreover, it wants to provide individually suiting care earlier for vulnerable youth. Creating coherent care for families by having one responsible organization with one plan for their care is another important reason. Finally, they want to create more room for the youth care professionals and decrease their work pressure concerning the arrangement of care. Summarizing, the most important reason for the change in the youth care system was the expectation that the municipalities would be able to fit the individual needs of the youth in their area better. This could create an efficiency gain through the possibility of earlier intervention, preventing the need for more specialized care.
This paper focusses on the provision and use of youth care, trying to find out whether the use of youth care has decreased after the regulation change. Is the goal of the government, to decrease youth care by making use of the efficiency gain, met? Moreover, how do municipalities handle youth care differently now compared to before 2015 and what are their reasons for this? How does this compare to the international systems? Furthermore, this paper looks at the escalation of youth care
4 since the regulation change, by comparing the amount of crisis situations from 2015 and 2016. Does the new regulation create more or less crisis situations?
5
2. Literature review
2.1. Trends in the international health care market organization
Differing trends have occurred in the organization of health care markets. On the one hand, countries have had a highly fragmented and dispersed provision of health services, whereas other countries have had mostly publicly funded and provided health services. Both types have tried to move more towards the other end, where the highly fragmented countries have introduced some form or regional or national controls. Reasons behind this according to Mills (1994) are to avoid duplication of services and the proliferation of high technology, leading to cost control. The countries where health services are arranged more publicly have tried to decentralize responsibilities, mainly on planning and management. A reasoning behind this is making services more efficient, more able to response to local needs and demands, and more accountability to local communities. As we have seen in the previous section, the Dutch government uses similar reasoning to support the recent trend of decentralization in the Netherlands.
In 1994, Mills (1994) discusses the decentralization of health care markets. She identifies four forms of decentralization: de-concentration, devolution, delegation, and privatization. Devolution is the form where subnational levels of government are created or strengthened, where these levels are substantially independent from the national government on a certain defined set of functions. Therefore, the Dutch decentralization of 2015 can be seen as a devolution of health care services. In 1994, it was quite rare to completely devolve the responsibility for health. Precursors on this matter were Scandinavian countries like Denmark and Sweden.
6 this decision were also the ability to provide services close to home, adjusted to fit the priorities and needs of the individual, and mobile instead of static.
In their 2004 paper, Thornicroft and Tansella (2004) also discuss community mental health teams, which are non-specialized teams providing the full range of interventions for a certain defined geographical area. According to them, there are some clear benefits to this method: being able to improve engagement with services, increase user satisfaction, increase met needs, improve adherence to treatment, and having more developed continuity of care and service flexibility. However, previous literature has not found that symptoms or social function have improved by using this method (Thornicroft and Tansella (2004) and the references therein). There are more specialized community health teams used in addition to general teams. These can be divided into two sub-categories: assertive community treatment teams and early intervention teams.
Assertive community treatment teams provide specialized mobile outreach treatment, focusing on people with disorders that are more disabling than others. According to Thornicroft and Tansella (2004), there are three advantages specifically in areas where a high level of resources are available. Firstly, there will be reduced admissions to hospitals, and less use of acute beds. Secondly, accommodation status and occupation can improve, and finally, the service user satisfaction increases. Moreover, it can possibly reduce the cost of in-patient services, but not total costs of care. However, assertive community treatment teams, as the general community treatment teams cannot be shown to improve mental state or social behavior. In addition, when the usual services are already at high levels of continuity of care, the assertive community treatment teams may be less effective. Early intervention teams are teams where the focus is on immediate identification and treatment of illness at an early stage. Research has shown that a longer duration of untreated psychosis predicts worse outcomes for the disorder. Therefore, this method could be effective in improving an individuals’ health in the long run. However, not much controlled trials have been published, making it hard to draw conclusions about the effectiveness of this method.
7 The method arisen in a great part of the Dutch municipalities after the decentralization can be compared to above-mentioned methods. In this research, we look at the neighborhood teams in the Netherlands. Can they be compared to the more general community mental health teams, or to the case management method? Alternatively, do they look more like the specialized community mental health teams such as the assertive community treatment teams and early intervention teams? In other words, what organizational form do the Dutch neighborhood teams adopt? At the start of the decentralization, Arum and Schoorl (2016) have already looked into this matter, which will be discussed in the next sub-section.
2.2. Neighborhood teams in the Netherlands
8 municipalities have also mentioned some issues that still need attention in the future. The integral approach could be improved by more cooperation and alignment between the different neighborhood teams and care suppliers. Furthermore, municipalities find the new regulations have resulted in a high work- and caseload and a loss of time due to administrative tasks and registration.
2.3. Results from the Dutch neighborhood team approach
9 A possible result of the neighborhood approach, especially when using the form of a generalist team, could also be the escalation of situations. When a generalist in the neighborhood team does not recognize the severity of a situation, for instance, and as a result does not refer an individual that in fact should be referred to more specialized care. Moreover, when the generalist retains a situation for too long, this can also lead to a need for longer or more specialized care than would have been necessary if the generalist would have promptly referred the youth. There can be several reasons for this, for example the generalist not being specialized enough and therefore not recognizing the severity of the situation. Furthermore, it could occur that a generalist does not want to come across as being unable to handle the situation, and therefore keeps trying to solve it while indeed not being sufficiently able to do so. When a situation escalates to a so-called ‘crisis situation’, the costs will be much higher than they potentially could have been when the individual had gotten the right, specialized care at an earlier stage. This is considering that, in that case, they might have needed the specialized care for a shorter period of time. Moreover, they might have needed less specialized care than what they need in the escalated situation. These crisis situations have been looked at before by Hospers and Vernhout (2016 and 2017), and their findings are discussed in the next sub-section.
2.4. Crisis situations
11
3. Data collection and analysis
3.1. Data collection
Data is collected of the biggest municipalities in the Netherlands, supplemented with some municipalities we were referred to by other interviewees. In total, this amounts to 36 municipalities. The qualitative data is gathered by conducting interviews with some of these municipalities and a care supplier. There have been interviews with municipalities which differ in approach of the new regulations. Some of them do provide youth care within the neighborhood teams, and some of them do not. Questions asked mainly focused on the choice of approach and the reasoning behind this. Furthermore, they were asked what results they notice. In the interview with the care supplier, the focus was on their view on the new approaches and, more specifically, the role of care suppliers. A list of participating municipalities and questions can be found in Appendix A.
Quantitative data was gathered through Statline, the database of the Statistics Netherlands (CBS). Data on the total amount of youth care, the amount of youth, care performed within the neighborhood teams, and care performed outside of the neighborhood team was gathered. Definitions and descriptive statistics of the quantitative data can be found in Appendix B. A limitation of the data on care performed within or outside the neighborhood teams is that this data is only available for 2015 and 2016. Furthermore, only for 7 municipalities this data is trustworthy, as they are the only ones supplying data to the CBS on this matter. Concerning the rest of the municipalities, it is therefore not possible to make the distinction between municipalities who do not provide youth care within the neighborhood teams, and municipalities who do not offer data to the CBS. Therefore, we take this into consideration while performing the interviews, asking the municipalities whether they provide youth care within the neighborhood teams. Whilst this does make it more clear which municipalities do or do not provide youth care within the neighborhood teams, we are still not able to get the numbers. Therefore, we perform the quantitative tests with the 7 municipalities for which the data is available.
12 are taken relative to the total amount of youth care trajectories. Once again, this data only covers the years 2015 and 2016.
Summarizing, although we do have some quantitative data to give us a first grasp of the situation, the most important implications of this paper are nested in the qualitative data. In the next sub-section, the analysis of the quantitative data is explained. In section 4, the results of the qualitative as well as the quantitative data is discussed.
3.2. Data analysis
To get an insight in the trends of youth care of the past years, we perform some estimation with previously mentioned data variables, set up as panel data. An advantage of using panel data is that time-invariant unobservable effects, correlated with the observable variables are corrected for. We cannot assume that the different observations are independent, as the same variables are observed over time. Conditions required to achieve unbiasedness, consistency and efficiency using an OLS estimation are that the expected value of the error term ε is equal to 0, as well as the expected correlation between the explanatory variable and the error term ε.
Firstly, we look at the total amount of youth care relative to the amount of youth. To see whether the use of youth care has had an increasing trend over the past years, we regress the youth care variable over time:
𝐴𝑉𝐸_𝑌𝐶 = 𝛽0+ 𝛽1𝑌𝐸𝐴𝑅 + 𝜀 ( 1 ) Unfortunately, there is not enough data to perform a structural break test, so we look at the graphs to visually interpret the trends. Furthermore, we rank the municipalities by average youth care performed and then look at the correlation of the rank and the lagged rank. Similarly, we generate a lag variable of the average youth care provided and look at the correlation of this with the average youth care. These correlations can also give us an interpretation on the youth care trends and its changeability. The results, graphs and interpretation of the graphs are discussed in section 4.
Subsequently, we would like to identify the change in above-mentioned trend as a result from the decentralization. Therefore, we add a dummy variable which is equal to 1 for the years 2015 and 2016 (the years after the decentralization) and equal to 0 for the years 2011 until 2014. We then run the regression with an additional variable, namely the dummy variable multiplied with the year variable:
13 The coefficient of this additional variable gives an interpretation of the change in the yearly trend after the decentralization. We expect the yearly trend to positively change after the decentralization, considering the suction effect municipalities mention, which is discussed in more detail in section 4. Next, we want to investigate whether providing care within the neighborhood team reduces care provided outside of the neighborhood team. Moreover, is this relationship one-to-one, so that care provided within the neighborhood team perfectly substitutes care that would otherwise have been handled outside of this team? To find this relationship, we take the variables of youth care provided within the neighborhood teams and youth care provided outside of the neighborhood teams, and take both these variables relative to the total amount of youth are trajectories. We run a regression of care provided outside of the neighborhood team on the care provided inside the neighborhood team:
𝐿𝑂𝐺_𝐴𝑉𝐸_𝑂𝑈𝑇 = 𝛽0+ 𝛽1𝐿𝑂𝐺_𝐴𝑉𝐸_𝐼𝑁 + 𝜀 ( 3 ) We expect that there is a negative relationship between these two variables, thus that an increase in care provided inside the neighborhood team results in less care provided outside of the neighborhood team. However, we do not expect this to be a one-on-one shift from the care outside of the neighborhood teams to care provided inside of the teams. The reasoning behind this is the by municipalities indicated suction effect. We expect that more need for light youth care will be discovered by being closer to the society, and thus increasing the youth care inside of the neighborhood team. These lighter cases will not end up at care suppliers outside of the neighborhood teams, as these mostly handle more complex cases. Thus, the increase of care provided by the neighborhood team will be flowing in from two sources: the cases that in the previous situation were not discovered and thus not helped, as well as the lighter cases that now do not need to be handled by care suppliers. Therefore, the care provided within the neighborhood teams will expectantly grow more than the care outside of the teams will decrease. Whether this indeed is the case will be discussed in section 4.
14 𝐴𝑉𝐸_𝑆𝑇𝐵 = 𝛽0+ 𝛽1𝐻𝐴𝐿𝐹𝑌𝐸𝐴𝑅 + 𝜀 ( 4 ) Furthermore, we investigate the youth care trajectories with- and without stay, both relative to the total amount of trajectories. We generate a variable for the youth care trajectories with stay relative to the trajectories without stay, and regress this over time:
15
4. Results
4.1. Organizational form of the youth care
Almost all municipalities use either the neighborhood team or the CJG for at least the access to, and sometimes also the provision of (lighter) youth care. There is only one exception, which we will come back to later. The CJG can be seen as another form of a neighborhood team, as it has many similar characteristics as a neighborhood team. Both are closer to the society, more approachable and accessible. Therefore, we do not differentiate between using a neighborhood team or a CJG. We do, however, distinguish municipalities offering youth care within the neighborhood team or CJG from municipalities where the neighborhood teams or CJG do not provide youth care themselves.
Of the 23 municipalities for which it became clear whether they do or do not provide care within the neighborhood team or CJG, 15 indicated that they do. Most of the municipalities working with a neighborhood team, have created a foundation around the neighborhood teams or CJG. Subsequently, this foundation is funded by the municipality through subsidies. As a result, the employees in the neighborhood teams or the CJG are not employed by the municipality, but by the foundation.
Like said, 15 municipalities use either neighborhood teams or the CJG to provide youth care. We also looked into what kind of functional form these neighborhood teams have, and compare this to the different approaches mentioned by Thornicroft and Tansella (2004). Recall there were four different forms or methods when using neighborhood teams, or as they call them, community health teams: the generalist community health team, the assertive community health team, the early intervention teams and case management.
16 On the contrary, 8 municipalities indicated that they do not provide care within a neighborhood team. Most of these municipalities work with teams that organize the access to youth care, but do not provide care themselves. Sometimes, these teams also have a consultation role. Like said before, one of the municipalities arranges things differently: they have generalists working at different geographical locations in the city, which arrange the organization of youth care together with other parties such as care suppliers. Their aim is that all parties contact each other and work together on the youth care market. The generalists are employed by this municipality, enabling them to have a close involvement.
The organizational form of the fore-mentioned municipalities not providing care within the neighborhood team can be sub-divided under the case management approach. Instead of being actively involved in youth care trajectories, these teams manage how, and by who, the case is handled.
4.2. Reasoning of the municipalities
17 the youth care. A disadvantage of separating responsibilities, is that it becomes harder to work integrally. The final reason mentioned is that municipalities do not only have an insight in the care path of civilians, but also on other matters. To protect the privacy of the civilian, it is better to make it impossible to connect all ongoing things with each other.
4.3. Results experienced by municipalities
The most mentioned result of the new youth care approach by all municipalities, whether they do or do not provide health care within the neighborhood team, is the suction effect on youth care demand. Almost all municipalities notice an increasing youth care demand, however they state that this does not necessarily mean that the demand has actually increased. A possible explanation is that municipalities are better capable of identifying the care needs, as their neighborhood teams are physically closer to the civilians. If the suction effect is a sign of more approachable youth care institutions, this is a positive effect. However, it could be that the increased supply and approachability created by the neighborhood teams creates the increasing demand. In short, it is possible that the supply creates demand instead of the other way around. If this is the case, and people are put in a trajectory while not necessary, this is a waste of resources.
Opinions on whether the new approach is for the better or not differ. Most municipalities feel better able to provide youth care, as they are closer to society and care has become more approachable. Furthermore, municipalities feel more responsible for the care, as they do now see the cases and results, whereas they first had almost nothing to do with this. Some municipalities, however, do not feel better able to provide health care. Most mentioned problems are the workload and getting used to a new way of working. A positive aspect of the new responsibilities of municipalities is that they now have an insight in every aspect of youth care. Previously, employees of municipalities did not actually see the cases that were handled in their area, whereas now they are more closely involved. This brings youth care more personal for employees of the municipalities, which makes them more inclined to find a solution where all youth is given the help they need as fast as possible.
18
4.4. Results from the data analysis
When performing the regressions discussed in the previous section, we find that average youth care has a slightly negative trend over time. Furthermore, the trend increases with a very small number when including the dummy variable. These results can be found in Table 1, and none of these results are significant on a 10% confidence level (p > 0.1 in all cases). We can conclude from this that the trend of average youth care most likely is constant over time.
Result estimation (1) Result estimation (2)
YEAR -0.0007 (9.964) -0.0042 (0.0088) YEAR*D_YEAR -0.0042 (0.0088) Constant 1.5559 (0.0049) 8.4940 (17.7994) Standard errors in parentheses
Table 1. Estimation results of estimation (1) and (2).
This is supported by the correlation between the rank of municipalities based on average youth care, and the lagged rank. These two are strongly correlated: the correlation between them is 0.8798. Therefore, this gives us another reason to believe that the rank of municipalities on youth care provided is constant, irrespective of their approach after the new regulations. Moreover, the correlation between average youth care and it’s lag is 0.9677, providing even more evidence of a constant trend. These correlations can be found in Table 2 and 3.
RANK_AVE_YC LAG_RANK_AVE_YC
RANK_AVE_YC 1.0000
LAG_RANK_AVE_YC 0.8798 1.0000
Table 2. The correlation between the rank of municipalities and the lagged rank.
AVE_YC LAG_AVE_YC
AVE_YC 1.0000
LAG_AVE_YC 0.9677 1.0000
19 When looking at the graph of average youth care over time of the municipalities, we do visually see an increase in youth care after the decentralization. This is reflected in Graph 1 for 4 municipalities and is in line with our expectations. The rest of the graphs can be found in Appendix C.
Graph 1. The average youth care of 4 Dutch municipalities.
Result estimation (3)
LOG_AVE_IN -0.2357*
(0.0200)
Constant -0.6773*
(0.0338) Standard errors in parentheses
*p < 0.01 Table 4. Estimation results of estimation (3).
20 neighborhood team increases with 1%, the youth care outside the teams decreases by only 0.24%. This is in line with the expectation of a negative effect.
Result estimation (4) Result estimation (5)
HALFYEAR -0.0009 (0.0010) -0.0001 (0.0028) Constant 0.0204* (0.0028) 0.1305* (0.0078) Standard errors in parentheses
*p < 0.01 Table 5. Estimation results of estimation (4) and (5).
21 Graph 2. The average amount of situations with as goal ‘stabilization of a crisis situation’ for 4 municipalities (half-yearly, starting at 2015).
When looking at Graph 2, we see that the average amount of situations which had the goal ‘stabilization of a crisis situation’ has decreased for most municipalities. The results for the other municipalities can be found in Appendix D. In Graph 3, the amount of care with stay relative to car without stay is reflected for 4 municipalities. The rest of the graphs can be found in Appendix E. We can see that these results differ per municipality: some have had a decreasing trend, whereas some have seen an increasing trend. Therefore, there can be no clear conclusion drawn on the increase or decrease of crisis situations in 2015 and 2016.
Graph 3. The average amount of care with stay relative to care without stay of 4 municipalities (half-yearly, starting at 2015).
4.5. Care suppliers
23
5. Conclusion
Almost all municipalities in the Netherlands provide access to youth care through neighborhood teams or the CJG. However, municipalities differ in the provision of youth care, where some neighborhood teams provide care themselves, and others are merely an access point. A mix of organizational forms of neighborhood teams are used by Dutch municipalities, in which the generalist health teams and case management are the most commonly mentioned. Many municipalities with generalist teams report that they use T-shaped professionals with different backgrounds and expertise for this. These approaches are mostly similar to internationally adopted organizational forms.
Municipalities have different reasons to provide youth care within the neighborhood teams or not. Reasons for providing care within the teams are picking up signals of a care need earlier and being able to handle accordingly in an earlier stage, expecting to work more preventively, and working more integrally. More different reasons are given not to provide youth care within the neighborhood teams, starting with municipalities not having the necessary knowledge and expertise, and wanting to give more trust and responsibility to the care suppliers. Furthermore, splitting up responsibilities of indication and care provision makes that all parties are more aware of what their responsibilities are. Additionally, some municipalities find that the main task of the municipality is to direct the care market, and find that therefore, it is better to keep a distance from the actual provision of youth care.
From the youth care data available at this time, it is hard to draw strong conclusions, but it is possible to get a grasp of trends in youth care. The trend of average youth care seems to be mostly constant from the data and slightly positive after the decentralization from the visual representation. Furthermore, the trend does not differ much between municipalities: the ranking based on average youth care provided is almost completely constant. In a few years, much more could be substantiated and there will be stronger findings from analyzing the grown databases.
24 municipality is better able to work preventively. However, there could also be more care need created unnecessarily, because of increasing supply.
When it comes to the role of care suppliers, there are different opinions and approaches. There is a trade-off between the link and cooperation between neighborhood teams and care suppliers, and the agency problem. For care supplier employees to be included in the neighborhood teams, trust in the right incentives of these employees is essential, as well as maintaining the link between the neighborhood team and the care supplier.
25
6. References
Arum, S. van, Schoorl, R., 2015, Sociale (wijk)teams in beeld, February 2016, Movisie.
Batterink, M., Hoven, W., Lapajian, I., Tazelaar, P., 2017, Sturing op specialistische jeugdhulp: Exploratief onderzoek naar het zorglandschap specialistische jeugdhulp, Significant, 2017.
Eijk, L. van, Tiège, E. de, Mutsaers, P., Evenboer, E., Grietens, H., Reijneveld, M., 2016, Werken in een winkel die verbouwd wordt, 2016, C4Youth.
Hospers, S., Vernhout, T., 2016, Crisis in context: Verkennend onderzoek naar toename crisismeldingen en -plaatsingen voor jeugdhulp in 2015, October 2016, Argos advies.
Hospers, S., Vernhout, T., 2017, Crisis in context: Nader onderzoek spoedplaatsingen JeugdzorgPlus, April 2017, Sophie Hospers coaching en advies.
Hospers, S., Zijden, Q. van der, 2017, Rapportage Herhaald Nader Onderzoek Toegang Jeugdhulp: Lokale inkleuring landelijke beleidsinformatie jeugdhulp - eerste halfjaar 2016 -, 14 April 2017, Hoogmade: Partners in jeugdbeleid.
Interviews with municipalities and a care supplier, 2 November 2017 until 14 December 2017.
Mills, A., 1994, Decentralization and accountability in the health sector from an international perspective: what are the choices?, 1994, Public Administration and Development.
Mol, J., 2017, Jeugdzorgdata en de integrale blik, June 2017, Centraal Bureau voor de Statistiek.
Thornicroft, G., Tansella, M., 2002, Balancing community-based and hospital-based mental health care, June 2002, World Psychiatry.
Thornicroft, G., Tansella, M., 2004, Components of a modern mental health service: a pragmatic balance of community and hospital care, 2004, The British Journal of Psychiatry.
26 Verbeek, M., 2012, A guide to modern econometrics, 2012, John Wiley & Sons Ltd.
Vermeulen, W., 2015, Decentralization of social policy in the Netherlands, CPB Background Document, November 2015, The Hague: CPB Netherlands Bureau for Economic Policy Analysis.
Zijden, Q. van der, 2015, Nader onderzoek toegang jeugdhulp: Lokale inkleuring landelijke beleidsinformatie jeugdhulp - eerste kwartaal 2015 -, 21 October 2015, Hoogmade: Partners in jeugdbeleid.
27
7. Appendices
Appendix A. Interviews; participants and questions
Interviews were held with multiple municipalities, mostly by phone. Exceptions are the interviews with Den Haag, Groningen, and Parnassia, which were held face-to-face. The following parties participated in the interviews:
Amersfoort Apeldoorn Den Haag Dordrecht Eindhoven Emmen Groningen Haarlem Haren
Parnassia (care supplier) Stadskanaal
Tilburg Zoetermeer
During the interviews, follow-up questions were asked when necessary for a thorough
understanding. However, in all interviews a certain line of questioning was adopted, which can be found below. Sometimes, some questions were not asked, for example when a municipality did not have a neighborhood team, it was not necessary to go into the composition of neighborhood teams.
- Does your municipality have neighborhood teams? What is their role?
- What kind of care is provided inside the neighborhood teams, and what care by other parties? Which other parties are these?
- What is the composition of the neighborhood teams? What kind of professionals are involved in youth care?
- What is the role of the municipality concerning youth care? - What are reasons behind the current approach?
- What were the expectations of the new approach?
28 - What is the role of care suppliers in youth care, and what is their view on the new approach? In the interview with the care supplier, the focus was on this last question. Moreover, we discussed what the ideal situation in the healthcare market would be from the perspective of the care
29
Appendix B. Data description
Variable Definition Obs. Mean Standard
deviation
Min. Max.
YEAR Year 216 2011 2016
D_YEAR Dummy variable which is 0 for 2011 until 2014, and 1 for 2015 and 2016
216 0 1
HALFYEAR Half-year (1 = first half-year of 2015, 4 = last half-year of 2016)
144 1 4
Youth The number of inhabitants with age < 20
216 34,656.8 26,046.64 4,254 137,874
Youth care trajectories
The amount of youth care trajectories
216 3,859.86 3,292.38 345 20,300
AVE_YC Youth care trajectories divided by youth
216 0.1235 0.1239 0.0717 1.0075
Care inside neighborhood teams
The amount of youth care trajectories offered by the neighborhood team
14 2,393.21 2,627.22 270 8,295
Care outside neighborhood teams
The amount of youth care trajectories offered by other parties than the neighborhood team
14 5,378.93 3,716.57 2,595 11,260
AVE_IN Care inside neighborhood teams divided by the amount of youth care trajectories
14 0.2597 0.1293 0.0668 0.4242
AVE_OUT Care outside neighborhood teams divided by the amount of youth care trajectories
14 0.7400 0.1293 0.5758 0.9319
Stabilization The amount of situations with the goal to ‘stabilize a crisis situation’ relative to the amount of youth care trajectories
30 AVE_STB Stabilization divided by the
amount of youth care trajectories
134 0.0181 0.0131 0.0028 0.0853
Trajectories with stay
The amount of youth care trajectories with stay
144 479.93 419.08 25 2,065
Trajectories without stay
The amount of youth care trajectories without stay
144 3,773.30 3,487.74 325 20,170
AVE_WS The number of youth care trajectories with stay divided by the amount of youth care trajectories
144 0.1142 0.0293 0.0554 0.2066
AVE_WOS The number of youth care trajectories without stay divided by the amount of youth care trajectories
31
35
39
