Mobile Integrated Social Services Increasing employment Outcomes for people in Need ( MISSION) RCT in Kortrijk Belgium

Last registered on March 20, 2018

Pre-Trial

Trial Information

General Information

Title
Mobile Integrated Social Services Increasing employment Outcomes for people in Need ( MISSION) RCT in Kortrijk Belgium
RCT ID
AEARCTR-0002786
Initial registration date
March 19, 2018

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
March 19, 2018, 6:49 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
March 20, 2018, 7:18 AM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Region

Primary Investigator

Affiliation
University of Leuven

Other Primary Investigator(s)

PI Affiliation
University of Antwerp

Additional Trial Information

Status
On going
Start date
2017-04-06
End date
2020-02-28
Secondary IDs
Abstract
The research seeks to assess the impact of a program targeted to increase the take-up of local social services with a particular focus on employment amongst disadvantaged families in the Belgian city of Kortrijk. The study is designed as a RCT which will enable us to test whether close counseling by multi-disciplinary case handlers leads to increased take-up rates of social services (and to some extent benefits). The effectiveness is tested by randomization of poor households into two groups: an intervention group and a control group. Approximately 235 participants will be equally divided between control and treatment. In addition to testing a strategy to reduce non-take-up of social services and benefits, the research aims to understand the underlying mechanism of non-take-up (NTU) and examine other possible outcomes of the intervention. Therefore the RCT is complemented by a mixed method approach where participants in the experiment are surveyed to shed light on determinants of NTU and examine as well outcomes including well-being, social network, trust in institutions, satisfaction with social services as well as living conditions. The hypothesis is that the integrated delivery of social services through multidisciplinary case handlers leads to better employment outcomes for service users. More precisely, it is assumed that pro-active action will improve the take-up of employment support, income support and other social services provided directly to the person and that the outcome and sustainability of employment support will improve when effective action is taken in several domains of life, delivered in an integrated manner through single points of contact. The study and program are funded under the umbrella of the EASI (Employment and Social Innovation) programme (PROGRESS axis) of the European Commission.
External Link(s)

Registration Citation

Citation
Decancq, Koen and Wim Van Lancker. 2018. "Mobile Integrated Social Services Increasing employment Outcomes for people in Need ( MISSION) RCT in Kortrijk Belgium." AEA RCT Registry. March 20. https://doi.org/10.1257/rct.2786-3.0
Former Citation
Decancq, Koen and Wim Van Lancker. 2018. "Mobile Integrated Social Services Increasing employment Outcomes for people in Need ( MISSION) RCT in Kortrijk Belgium." AEA RCT Registry. March 20. https://www.socialscienceregistry.org/trials/2786/history/26925
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
A significant challenge in setting up the pilot programme is how to reliably identify the targeted population, in casu disadvantaged families that are eligible for local employment services. In order to identify reliably this rather hard to reach group, the study makes use of home visits of district nurses by the Flemish agency for Child and Family Welfare (‘Kind en Gezin’, K&G hereafter). The nurses visit around 96,5 % (2014) of all new- borns in Flanders and identify disadvantaged household making use of a multidimensional deprivation indicator based on 6 dimensions: monthly household income, parental educational level, child development, parental employment, housing situation and health status. Families with new-born children that are identified as living in disadvantaged circumstances by Kind en Gezin (K&G) are the main target group of the study.
The intervention group gets the opportunity to have integrated delivery of social services through multi-disciplinary case handlers who will be reaching out to the disadvantaged families making innovative use of home-visits. The intervention will focus on the integrated delivery of a wide array of local social services, synthesized in six broader domains: income support services; employment support services; housing support services; health care support services; child development support services and education & training support services. The integrated service delivery is implemented no only by case handlers but as well through two other instruments: an integrated IT-system and a multi-disciplinary multi-actor network.
Multi-disciplinary case handlers will closely council disadvantaged families in order to guide them towards the appropriate services that are on offer at the local level. The case handlers will act as single point of contact for the families involved, aiming at successful integration into the labour market or training programmes. In addition, the case handler will also be mandated to guide the families towards other services and benefits they might be entitled to depending on their needs, including access to adequate income and cash benefits, housing services, health services, and childcare services. In short, the families in the control group are entitled to exactly the same services and benefits as the families in the intervention group, but for the latter group the case handler acts as an in-between. The period of the intervention and thus the availability of a case handler, is set at 12 months in total with a intensive period of 6 to 8 months: depending on how quickly the case handler can build a relationship and start the intensive support this can vary between 6 and 8 months. After the intensive period the support continues up to 12 months but gradually diminishes and the case handler’s support is less intensive with the goal to hand over to existing local services. During the intensive support period case handlers will visit the families at least twice a month, and will be in contact with the family at least once a week. The case handlers will be trained beforehand to ensure that all families in the intervention group get a similar minimum treatment. Depending on the situation, the case handler may choose to pay more visits to a particular family or to invest more time in counselling that family. What is constant across all families for the intervention is the availability of a case handler during 6 to 8 months intensively and for 12 months in total be able to provide integrated support customised to family’s needs.
The needs of the family, hence the selection of services most appropriate for these families, will be determined after a thorough ‘needs assessment’ conducted by the case handler together with the beneficiary using a newly developed software application suitable for smartphones and tablets (‘app’). This app will generate individual profiles of the beneficiaries which will be automatically linked to all the wide array of benefits, right & services that are available at the local level. The software will be accessible for the case handler, the beneficiary and the service providers involved.
Finally a multi-disciplinary multi-actor network is established to provide for ad hoc need based consultation and has a policy formulation function. Depending on the need, the network involves representatives of the local Public Authority, the local Centres for Social Work, Local Job Centre’s and service providers & NGO’s in the domains of social housing, health care, child care & family matters, social partners such as unions & health insurance companies, local social enterprises, end users of services (represented through the case handlers) and civil society organisations.
Intervention Start Date
2017-04-06
Intervention End Date
2019-10-06

Primary Outcomes

Primary Outcomes (end points)
- Take-up of local employment services (enrolment as well as extent and nature of use)
Depending on availability of administrative data:
- Take-up of other social services (enrolment as well as extent and nature of use)
- Take-up of certain benefits and subsidies (application and reception)
Primary Outcomes (explanation)
Randomizing the intervention across treatment and control groups, we will evaluate its impact on different categories of outcomes gathered from both administrative data and a survey carried out with all participants to the study. The main objective of the study is to identify the causal effect of the intervention by comparing differences between treatment and control group in terms of take up (enrolment and use) of local public employment services (PES) and to some extent take up of certain benefits and subsidies and other local services on offer. Therefore the primary outcome examined is whether the availibility of personal counseling of disadvantaged families by a well-trained case handler during a period of 10 to 12 months is more effective in increasing take-up of employment services at the local level compared with treatment as usual (TAU). The outcome will contribute in understanding strategies to reduce no-take up of (local employment) services by testing this pilot project. To measure the effectiveness of take-up of local employment services we consider administrative data of local employment services (inscription but also trajectories) and compare these data between control and intervention groups. As such it becomes possible to objectively test whether families in the intervention are more likely to be enrolled in local employment services compared with families in the control group. The primary outcome that will be measured by means of the experimental design will be a lower bound estimate of the actual effect (e.g. families may experience improved access to multiple local services as well benefits and subsidies instead of only the employment services). Depending on the availability of administrative data take up of these other services and certain benefits and subsidies will also will measured and additionally self-reported take-up through the survey, although less reliable, will triangulate or supplement the administrative data on take up.

Secondary Outcomes

Secondary Outcomes (end points)
- Wellbeing
- Trust in institutions
- Satisfaction and use of professional assistance
- Social support
- Living conditions including housing situation, activity (employment) status in the household, household financial and material situation and finally use of childcare services.
- Determinants of (non-)take-up of social services and benefits
Secondary Outcomes (explanation)
Broader secondary outcome measures are examined in order to qualify the primary outcomes obtained by means of a survey. This will provide deeper insight into how participants experience and use local service delivery, shed more light on the determinants of non- take-up (NTU) of local social services and benefits and subsidies and explain how the pilot programme affects the intervention group compared to TAU on several categories of outcomes: well-being, social support, trust in institutions and living conditions including housing situation, activity (employment) status in the household, household financial and material situation and finally use of childcare services.

Wellbeing is measured by letting the main respondent (usually the mother) complete a wellbeing questionnaire containing questions of the Personal and Social Wellbeing module of European Social Survey (Round 6), measuring 6 different dimensions of wellbeing which are accounting for measuring both positive feelings and positive functionings of wellbeing: evaluative, emotional, functioning, vitality, supportive relations and community wellbeing are measured. Social support is assessed in the survey gauging for emotional, instrumental, financial and informational support the main respondent is receiving. Trust in institutions is evaluated in the survey by using a 10-point scale for trust assessing several institutions such as the police, politicians, the local employment offices, the Flemish agency for Child and Family Welfare (‘Kind en Gezin’, K&G hereafter), trade unions, etc. Outcomes related to the housing situation are assessed by having information on type of housing, satisfaction, costs and application to subsidies and benefits regarding housing as provided in the survey. Activity status contains the current state of activity by respondent (and partner if applicable) (employee, independent, student, unemployed, housewife/men, invalidity, etc.) as well as the work intensity of the household in the past 12 months as reported in the survey. Household financial and material situation constitutes survey information on monthly household net income, material deprivation, household debt and assistance with debt management/counselling. Outcomes regarding childcare services are assessed on amount paid per day and wether or not the household makes use of the services and reasons reported why household is making use or not of these services. When comparing outcomes regarding professional assistance, survey questions on quantity of professional assistance and it’s nature are used, as well as the reported satisfaction of the respondent measured through general judgment of professional assistance organization used as well as respondent’s judgment on the harmonization of assistance, and the quality/availability of support of professional assistance in general. Some of these secondary outcomes are expected on the mid-or long term, whereas measurement is done first of all in the shorter term (6 months after the baseline). Ideally after 1 year and if possible 2 years after the intervention employment and all outcomes should be measured again to evaluate the mid-term impact of the intervention.

Furthermore, as said above, the study wants to understand better the mechanism behind non-take up and therefore the survey contains questions on both self-reported reasons of non-take up of services and benefits/subsidies, as well as information on possible determinants. To grasp how the intervention possibly affects take-up of local services and benefits/subsidies determinants will be examined pre-and post intervention together with outcomes determinants. The following possible determinants are included in the survey: trigger events, knowledge and information related costs, perception of eligibility, need, attitudinal barriers against social benefits, perception of utility of benefit/services, process costs (difficulty application procedure, filling in forms, etc.,), psychological and social costs: stigma (personal, social, claims), personal values (regarding giving personal information, dependency on government, requirements unrealistic or unacceptable to fulfil eligibility requirements), (mis)perception of risks, number of people using benefits/subsidies in respondent network and information context (sources of information, information seeking behaviour, advice to take up, perception on availability and quality of benefit related information, etc.)

The information gathered in the survey will allows us to control for basic characteristics between control and intervention group such as household composition, age, migration background and socio-economic situation.

Experimental Design

Experimental Design
The effectiveness is tested by randomization of poor households into two groups: an intervention group and a control group. A sample of 235 participants, with 2 equal groups is expected. There is only one treatment: availability during a fixed period of integrated delivery of social services through a outreaching multidisciplinary case handler. Participants (both from the comparison and treatment groups) are interviewed before being assigned into treatment and control group and are surveyed again 6 months after the baseline measurement and most probably as well 1 year after baseline measurement (intervention ended). Administrative data containing information on take–up of employment services will be used at different points in time: at baseline level, after 6 months, 12 months.
Experimental Design Details

Randomization Method
Done in office manually by an independently researcher assigning on a coming in basis into treatment and control groups (holding both groups equal in size). This is done as our target group only is given on a monthly basis, depending on new births and following identification of families being deprived by the district nurses after visiting newborns.
Randomization Unit
Individual unit: Stratified randomization with one covariate which is nationality (native Belgians vs non-native Belgian mother of the household. Each household is first grouped into native or non native stratum based on the mother ‘s nationality. Within each stratum, individual household are then randomly assigned to treatment or control.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
0
Sample size: planned number of observations
235
Sample size (or number of clusters) by treatment arms
Equally divided between treatmet and control: 117 and 118
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Power calculations have been carried to estimate the sample size required based upon the average effect size of similar studies on a related topic. Although previous research on the specific issue of improving take-up of local employment services by means of home visits by case handlers is absent, experimental research into the effectiveness of job counselors allows to estimate a minimum detectable effect (MDE) of the intervention compared with the control group. Rosholm (2014) provides an overview of studies that have examined the effect of ‘job counselor’ meetings with the participants to get them into an employment trajectory, and found substantial effects of a 10% to 30% difference between control and intervention groups. Given this, we assume a MDE of about 20%. For the primary outcome, assuming a base rate take up of 18 % in the control group, a sample of 235 participants, with 2 equal groups, the minimal detectable effect size is 11% with power 0.8 at a significance level of 5%. Avoiding attrition in conducting RCTs is a challenge, and in many case attrition rates exceeding 20% lead to bias in outcomes (Wood et al, 2004). Recent experimental studies involving home visits by case managers or counsellors, however, usually report lower attrition rates. Moreover, a recent study showed that home visits are an effective approach to reduce attrition rates (Peterson et al, 2012). Assuming a worst case scenario, an additional power analysis shows that a lower bound estimate of 176 participants (25% attrition) would still be sufficient to detect an increase in take up of 13%.
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials