ABSTRACT The purpose of the study is to investigate if a monetary incentive can increase the response rate for vulnerable children and youth in a postal survey. A secondary purpose is to test if monetary incentive has an effect on the content of answers in a postal survey. The study is a randomized controlled trial where participants will be randomly allocated to receive a questionnaire together with a €15 voucher for a supermarket or to only receive a questionnaire. Data are taken from an intervention study aiming at creating network for different groups of vulnerable children and youth aged 8-23 years. The questionnaire survey in the present study is a follow up survey of the original study which are performed 1-2½ years after the participant have left the study. Recruitment to the original study is still ongoing and the investigators expect the total participation to be around 250. In the primary analysis the investigators will estimate the relative risk (RR) to assess whether there is a difference between group allocation and questionnaire response rate. The investigators will adjust for age in the analysis. In the secondary analyses the investigators will test whether scores on the five SDQ scales are different between group allocations. The investigators will adjust for the covariates age and gender, since the scores may depend on age and gender (4).

TITELPAGE Title: The effect of monetary incentive on survey response for vulnerable children and youth

Name and contact details for head of the project:

Seniorresearcher, Jan Hyld Pejtersen SFI - The Danish National Centre for Social Research Herluf Trolles Gade 11 DK-1052 Copenhagen K +45 3348 0898 jhp@sfi.dk

Name and affiliation of people in the project group:

Seniorresearcher, Turf Böcker Jakobsen SFI - The Danish National Centre for Social Research

Responsible institution:

SFI - The Danish National Centre for Social Research Herluf Trolles Gade 11 DK-1052 Copenhagen K

+45 3348 0800 sfi@sfi.dk

ABSTRACT The purpose of the study is to investigate if a monetary incentive can increase the response rate for vulnerable children and youth in a postal survey. A secondary purpose is to test if monetary incentive has an effect on the content of answers in a postal survey. The study is a randomized controlled trial where participants will be randomly allocated to receive a questionnaire together with a €15 voucher for a supermarket or to only receive a questionnaire. Data are taken from an intervention study aiming at creating network for different groups of vulnerable children and youth aged 8-23 years. The questionnaire survey in the present study is a follow up survey of the original study which are performed 1-2½ years after the participant have left the study. Recruitment to the original study is still ongoing and The inestigators expect the total participation to be around 250. In the primary analysis the investigators will estimate the relative risk (RR) to assess whether there is a difference between group allocation and questionnaire response rate. The investigators will adjust for age in the analysis. In the secondary analyses the investigators will test whether scores on the five SDQ scales are different between group allocations. The investigators will adjust for the covariates age and gender, since the scores may depend on age and gender (4).

BACKGROUND Non-responders in questionnaire surveys may introduce bias and lower the validity of the studies. A Cochrane review has identified trials evaluating different ways of increasing response rates (1). The odds of responding on postal questionnaires were almost doubled using monetary incentive, and the odds were higher when the incentive was given with the questionnaires compare to only given after participants had responded. However, to our knowledge none of the studies concerned vulnerable children and youth. The purpose of the study is to investigate if a monetary incentive can increase the response rate for vulnerable children and youth in a postal survey. A secondary purpose is to test if monetary incentive has an effect on the content of answers in a postal survey.

METHODS/DESIGN Study design The study is a randomized controlled trial. To study the effect of incentive on survey response participants will be randomly allocated to receive a questionnaire together with a €15 voucher for a supermarket or to only receive a questionnaire.

Participants Inclusion criteria: Participants are taken from an established intervention study aiming at creating network for different groups of vulnerable children and youth aged 10-23 years. The groups comprise: children in foster care, former foster youth, children of mental ill parents, children of parents with alcohol abuse and children raised in violent families. Participants who have given informed consent to take part in the evaluation of the established intervention study and thereby agree to fill out questionnaires and report their unique personal identification number are eligible to take part in the study.

Exclusion criteria: Based on the participants unique personal identification number (CPR-number in the Danish Civil Registration System) the Central Office of Civil Registration will provide us with addresses of the participants. Participants, who are not available in the Civil Register either because participants are dead, immigrated or have requested survey exemption, are excluded from the study. Participants who have reported incomplete identification number will also be excluded.

Participant recruitment At the moment 215 children and youth have given informed consent that they will participate in the evaluation of the original study. Recruitment to the original study is still ongoing and the investigators expect the total participation to be around 250. The questionnaire survey in the present study is a follow up survey of the original study which will be performed 1-2½ years after the participant have left the original study. The present study will be performed in 4 waves. The first wave will be performed in November 2012 and will include participants who have left the original study 1-2½ years ago. The second, third and fourth waves will be performed May 2013, November 2013 and May 2014, respectively, 1-1½ years after the participants have left the original study. The outcome will be measured over a perod of 1½ years.

Randomisation The randomization will be performed within each wave of the survey. The participants are given a unique identification number in the original study and this number will also be used for the present study. The first 3 digit in the identification number identifies the municipality of the participants and the last digit are given in the order the participants enters the original study. Within each wave of the study participants will be sorted according to their identification number and will be assigned a random number between 0 and 1. The investigators will use the Rand ('uniform') function in SAS to create the randomization sequence. If the numbers are less than 0.5 the participants will be allocated to control otherwise participants will be allocated to treatment.

Blinding Neither participant nor researchers are blinded to the intervention.

INTERVENTION The intervention group will receive a postal questionnaire together with a voucher worth €15 for the largest supermarket chain in Denmark. Non-responders will after three weeks receive a reminder together with another copy of the questionnaire. In both cases the questionnaire will be sent with a cover letter and a reply paid return envelope.

The control group will only receive a postal questionnaire. Non-responders will after three weeks receive a reminder together with another copy of the questionnaire. In both cases the questionnaire will be sent with a cover letter and a reply paid return envelope.

After 10 weeks duration the study period will be over and the participants in the control group will receive a similar voucher as the intervention group. This is done for ethical reason, so both groups are treated equal. Any responses after 10 weeks will be reported, but will not be included in the primary and secondary analyses.

OUTCOME MEASURES The postal questionnaire consists of 91 items. The core of the questionnaire is the strengths and difficulties questionnaire SDQ (3) covering children and young people's behaviours, emotions, and relationships. Further topics in the questionnaire are: Family and housing; education and training; leisure time; alcohol consumption and evaluation of the intervention in the original study.

The primary outcome measure will be questionnaire response rate which is defined as the proportion of questionnaires returned by participants. The secondary outcomes will be scores on the five multiitems scales in the strengths and difficulties questionnaire (3).

POWERCALCULATION The sample size calculation is determined by the sample size recruited to the original study. Consequently, the investigators did not undertake a formal power calculation to determine the sample size.

DATA ANALYSIS In the primary analysis the investigators will estimate the relative risk (RR) to assess whether there is a difference between group allocation and questionnaire response rate. The investigators will adjust for age in the analysis. The analysis will be performed using PROC GENMOD in SAS. Level of significant will be at 5% significance level.

In the secondary analyses the investigators will test whether scores on the five SDQ scales are different between group allocations. The investigators will adjust for the covariates age and gender, since the scores may depend on age and gender (4). If scores are normally distributed the investigators will use GLM in SAS, otherwise the investigators will use ordinal logistic regression (PROC GENMOD).

ETHICS The study is registered with the Danish Data Protection Agency. In Denmark studies of this kind do not need approval by the local ethical committee. This has been verified by the Ethical committee, protocol: H-1-2012-FSP.

The trial will be registered at www.clinicaltrials.gov

PUBLICATION The study will be published in a peer reviewed international journal as for instance Journal of Child Psychology and Psychiatry or similar.

The first participant will be included in November 2012 and the last participant will be included in May 2014. The overall time frame of the study will be 1½ years.

ORGANISATIONAL ISSUES/AUTHORS' CONTRIBUTIONS All authors contributed to the study design. JHP drafted the protocol, and TBJ reviewed and revised the manuscript. All authors read and approved the final version.

ACKNNOWLEDMENTS AND FUNDING The study is partly funded by the National Board of Social Services as part of the evaluation of the intervention project 'Lige Muligheder [in Danish], and partly by The Danish National Centre for Social Research.

COMPETING INTERESTS The authors declare that they have no competing interests.

REFERENCES

1. Edwards PJ, Roberts I, Clarke MJ, DiGuiseppi C, Wentz R, Kwan I, et al. Methods to increase response to postal and electronic questionnaires. Cochrane Database of Systematic Reviews 2009;(3).
2. Edwards P, Roberts I, Clarke M, DiGuiseppi C, Pratap S, Wentz R, et al. Increasing response rates to postal questionnaires: systematic review. BMJ 2002 May 18;324(7347):1183.
3. Obel C, Heiervang E, Rodriguez A, Heyerdahl S, Smedje H, Sourander A, et al. The Strengths and Difficulties Questionnaire in the Nordic countries. European Child & Adolescent Psychiatry 2004;13:32-9.
4. Niclasen J, Teasdale TW, Andersen AMN, Skovgaard AM, Elberling H, Obel C. Psychometric Properties of the Danish Strength and Difficulties Questionnaire: The SDQ Assessed for More than 70,000 Raters in Four Different Cohorts. Plos One 2012 Feb 27;7(2).

Inclusion Criteria:

• Participants are taken from an established intervention study aiming at creating network for different groups of vulnerable children and youth aged 10-23 years. The groups comprise: children in foster care, former foster youth, children of mental ill parents, children of parents with alcohol abuse and children raised in violent families. Participants who have given informed consent to take part in the evaluation of the established intervention study and thereby agree to fill out questionnaires and report their unique personal identification number are eligible to take part in the study.

Exclusion Criteria:

• Based on the participants unique personal identification number (CPR-number in the Danish Civil Registration System) the Central Office of Civil Registration will provide us with addresses of the participants. Participants, who are not available in the Civil Register either because participants are dead, immigrated or have requested survey exemption, are excluded from the study. Participants who have reported incomplete identification number will also be excluded.