Research Proposal MSc placement Health Sciences 2015-2016
Name, student number, address, phone and email address, specialization
2.Name and address of institute/organization
National Coordination Centre for Communicable Disease Control (LCI)
Center for Infectious Disease Control (CIb)
National Institute for Public Health and the Environment
Antonie van Leeuwenhoeklaan 9
3721 MA Bilthoven
Project leader (main supervisor)
Jim van Steenbergen, MD, PhD.
Consultant Communicable Disease Control
Project coordinator (daily supervisor)
Mart Stein, MSc.
Researcher Infectious Disease Epidemiology
Short description of the research proposal
4.Title of internship
5.Scientific background of the research
People are interconnected and so their health is interconnected . Our friends, colleagues and relatives influence our opinion, beliefs, decisions and behaviour, which often referred to as social contagion . Similar to how infectious diseases are transmitted, studies demonstrated that social contagion follows the patterns of social contacts. Health behaviour is often ‘clustered’ (co-occurrence of the trait in connected individuals) in social networks [3-5]. Analysing social networks can help expand our understanding of whether and how certain social determinants influence individual decision-making with respect to prevention programmes. Although the determinants of well-informed decision-making for health behaviour are extensively studied in independent individuals in the general population, little research is done on the distribution of determinants within social networks. Risk communication regarding preventive (screening) programmes might benefit from using social networks.
The individual-level studies into the decision making process do not help explain how underlying determinants of informed decision making (IDM) vary between individuals in the same social network. A recent study that used data from Twitter to measure the evolution and distribution of sentiments towards a novel vaccine found that both negative and positive opinions were clustered in this on-line social network . While this analysis of data from on-line social media offers great potential to measure risk perception and health behaviour, extracting information from short on-line text messages for the purpose of assessing perceptions and behaviours also has its limitations: users may not be a representative sample of the target population, text messages may be wrongly interpreted and only limited information per individual is obtained. Therefore, development of other methods is needed in order to obtain more extensive knowledge on how certain communities are socially structured, how determinants for IDM are distributed within these communities, and the extent of clustering of these determinants and preventive behaviours.
In this innovative research we will use respondent-driven methods (RDM) to recruit individuals and their social contacts from target populations of screening and/or vaccination programmes. RDM is similar to snow-ball sampling and allows researchers to study real-world social networks .
Societal and policy need
Influenza vaccination is offered free of charge by general practitioners annually to all patients in their registry ages 60 and above. Uptake used to be over 77%, but is decreasing over the past few years. A pneumococcal vaccination is under study to be administered also to the same patients. The Health Council is waiting the results of this study before advising to administer pneumococcal vaccine to all people ages 60 and above.
To improve information material for target groups, the RIVM is interested is perceptions, intention and information needs of the target groups.
1.Smith KP, Christakis NA (2008) Social networks and health. Annual Review of Sociology
2.Berkman LF, Glass T (2000) Social integration, social networks, social support, and health. In: Berkman LF, Kawachi I, editors. Social Epidemiology. New York: Oxford University Press. pp. 137-173.
3.Christakis NA, Fowler JH (2007) The spread of obesity in a large social network over 32 years. N Engl J Med 357: 370-379.
4.Schuit AJ, van Loon AJ, Tijhuis M, Ocke M (2002) Clustering of lifestyle risk factors in a general adult population. Prev Med 35: 219-224.
5.Christakis NA, Fowler JH (2013) Social contagion theory: examining dynamic social networks and human behavior. Stat Med 32: 556-577.
6.Salathe M, Khandelwal S (2011) Assessing vaccination sentiments with online social media: implications for infectious disease dynamics and control. PLoS Comput Biol 7: e1002199.
7.Stein ML, van Steenbergen JE, Chanyasanha C, Tipayamongkholgul M, Buskens V, et al. (2014) Online respondent-driven sampling for studying contact patterns relevant for the spread of close-contact pathogens: a pilot study in Thailand. PLoS One 9: e85256
8.van Leeuwen AW, de Nooijer P, Hop WC (2005) Screening for cervical carcinoma. Cancer 105: 270-276.
9.Visser O, van Peppen AM, Ory FG, van Leeuwen FE (2005) Results of breast cancer screening in first generation migrants in Northwest Netherlands. Eur J Cancer Prev 14: 251-255.
10.Hartman E, van den Muijsenbergh ME, Haneveld RW (2009) Breast cancer screening participation among Turks and Moroccans in the Netherlands: exploring reasons for nonattendance. Eur J Cancer Prev 18: 349-353.
11.Statistics Netherlands (2014) Bevolking; generatie, geslacht, leeftijd en herkomstgroepering, 1 januari.
In this research project we will focus on measuring perceptions and behaviour in social networks. The project includes inter alia reaching and interviewing individuals and their contacts targeted for a prevention program, and social network analysis.
i. Under what circumstances is on-line or off-line RDM effective in reaching individuals targeted for (cancer) screening and/or influenza vaccination?
ii. How are perceptions and behaviour distributed within the sampled social networks? In other words: are individuals with similar perceptions and behaviour with respect to screening and/or vaccination programmes clustered in the selected target group?
And, if so, what is the extent of clustering (e.g. number of link steps that clustering extends to in an obtained network component) and what are the underlying determinants (e.g. related basic demographic characteristics)?
Note: The RIVM previously developed and validated a questionnaire to measure risk perception towards bowel cancer screening. The student will make a selection of the most relevant questions for inclusion in the RDM questionnaire. Previous developed RDM survey software will be used for the online survey.
7.Research design and methods
Our primary objective is to investigate the conditions for using on-line RDM to reach Dutch groups and their social contacts targeted for influenza vaccination, in order to conduct cross-sectional questionnaire surveys to measure determinants of IDM and preventive behaviour of these target populations.
Target sample size:
- web-based: 100 pairs (at least) of recruiter-recruited contact
- paper-based: 100 pairs (at least) of recruiter-recruited contact
- Descriptive analysis of total sample and per recruitment wave (in baseline tables/figures)
- Correlations between pairs for socio demographic characteristics; for one link step in network trees, and between any two individuals at different link steps in the same network tree.
- Regression analysis on drivers of peer-recruitment; including multi-level analysis to correct for non-randomness in sampling process.
8.Work to do
1. Literature review
What is known on:
- snowball sampling / RDM in a general population; paper-based versus web based?
- on risk perception and uptake of influenza and pneumococcal vaccination for specific subpopulations with non-western origin?
- which contact persons should be recruited?
2. Development of a short questionnaire
- Based on previous RIVM expertise, a short questionnaire will be developed for web based and paper RDM.
- “niet-WMO” statement has to be written and submitted to ethical committee of UMC Utrecht
3. Data collection
- Web based
- Paper based
4. Statistical analyses
5. Write article
Backup data for statistical analysis
If data collection is not successful within the intended timeframe and the student is unable to analyse the data, the student will be asked to analyse (already) collected data from a similar RDM project in Thailand. In this Thailand project, web-based RDM was used to study contact network patterns relevant for the spread of respiratory pathogens (such as influenza-like-illness) and to assess the clustering of influenza vaccination and vaccination beliefs in social networks in the Thai population. Using this dataset, the student will be able to perform a similar statistical analysis, as described above.
9.Expected result (scientific article and presentation)
1. An English written scientific article/manuscript of 3.000 to 4.0000 words (including Abstract, Introduction, Methods, Results, Discussion, Conclusion, References, Tables and Figures).
2. Oral presentation for the LCI department.
•Literature and conceptual framework March 2016
•Selection of, or development of questionnaire April 2016
•Paper-based and web-based survey May 2016
•Data analysis June 2016
•Report July 2016
-questionnaire surveys; web-based and paper-based
-data management in MS Excel
-quantitative data analyses in SPSS, and R (under supervision)
We are looking for both Dutch students and students with a non-western background. We note that one of the objectives of this RIVM project is to analyse risk perception and behaviour among non-western immigrants. Therefore, students with (grand-) parents coming from one of the large immigrant populations in the Netherlands are strongly recommended to apply. Mastery of the Dutch language is required.
The RIVM offers an internship fee of 554,- Euro (gross) per month (based on 40 hours per week).
To be filled in by the placement coordinator of the specialization
The university supervisor will be appointed by the placement coordinator of the specialization
Signature of on-site supervisor
Signature of university supervisor
Signature of student
BEWARE: Each research proposal has to be signed by the on-site supervisor, the university supervisor as well as the student before the start of the internship