ESRA 2017 Programme

Tuesday 18th July      Wednesday 19th July      Thursday 20th July      Friday 21th July     

     ESRA Conference App

Tuesday 18th July, 09:00 - 10:30 Room: F2 103

Power of Survey Research in Evidence-Based Policymaking 1

Chair Dr Young Chun (PSI Institute for Data Science and Interdisciplinary Research )
Coordinator 1Ms Giang Nguyen (ISR Foundation Center for Interdisciplinary Research)
Coordinator 2Ms Clara Kyung (McGill University)

Session Details

Co-Organizers: Cindy Won, Brown University; Hanyuying Wong, Cambridge University; Leying Guan, Stanford University

Evidence-Based Policymaking (EBP), a term coined from the idea of “evidence-based medicine” and spread to all spheres of public policy, involves the use of evidence collected by scientifically rigorous methods, such as randomized controlled trials. The EBP has garnered support from academia and governmental sectors across the Atlantic. For instance, Adrian Smith (1996) introduced the term in his presidential address to the Royal Statistical Society, questioning the current process of ideology-based policymaking and urging for an “evidence-based approach” in public policy. On the other side of the Atlantic, Davies (2004) echoed EBP as the integration of experience, judgement and expertise with “the best available external evidence from systematic research.” Listening to the call from the scientific research community and gleaning from the steps taken by the UK government, the Obama administration kicked off the Evidence-Based Policy Commission in summer of 2016 to institutionalize evidence-based policy development across all federal agencies.

Survey research-based data and survey data linked to other sources of data, such as administrative records and big data, are at the heart of evidence-based policy making. This session invites papers that demonstrate how survey research and administrative records are instrumental to evidence-based policymaking. We encourage submission of interdisciplinary papers that address the following topics:

1) Contributions of survey research applications to evidence-based policy making in public health, economy and education, just to name a few. For example, how survey research has informed the policymaking process, which intends to bring about social, economic, and fiscal benefits?

2) Papers demonstrating the use of administrative records linked to survey data in developing public policy, for example, in healthcare intervention, special education for the disabled, and economic welfare programs.

3) Papers involving experimental design, such as randomized controlled trials, in a broad context of survey research, to advance evidence-based policymaking with case studies (e.g., small business support, education initiatives for low-income children, healthcare programs to prevent or control epidemic disease like TB and Hepatitis).

4) Papers utilizing data visualization, a tool of showing agile comprehension of massive data, to communicate the public messages about policy issues particularly in developing countries. EBP is less implemented in developing countries due to lack of resources and limited civil liberty; the potential for evidence-based policy impact is greater.

Paper Details

1. Advancing Evidence-based Policymaking through Survey Integration Efforts
Dr Steven Cohen (RTI International)

Recent changes in the provision and organization of health care in the U.S. have served to enhance the importance of evidence-based policymaking to address issues focused upon access, use, and expenditures for care. The quality and content of national population-based economic and health care surveys to inform policy formulation are enhanced through integrated designs that include the conduct of inter-connected surveys to medical providers, businesses, and medical facilities. Analytical capacity is dramatically enhanced through their connectivity to existing secondary data sources at higher levels of aggregation and via direct matches to additional health and socioeconomic measures acquired for the same sample units from other sources of survey or administrative data. These administrative databases also may serve as sampling frames to facilitate a cost-efficient sample selection. These designs improve data collection strategies to meet target response rates, achieve reductions in nonresponse bias, and enhance data quality and analytical capacity. They permit extensions in longitudinal analyses necessary for policy formulation and permit methodological studies to assess the accuracy of household reported data. Advances in data science also serve to facilitate the effective and efficient utilization of statistical models and procedures in concert with big data applications. These design features and analytic enhancements are illustrated with examples drawn from national health-related surveys with integrated survey designs. They include the Medical Expenditure Panel Survey (MEPS), the MEPS Medical Organization Survey (MOS), the Health Interview Survey (NHIS) and the Medicare Current Beneficiary Survey (MCBS). Design limitations also are discussed.
This presentation will also include a summary of enhancements to the analytic capacity and utility of cancer clinical trial Project Data Sphere® content achieved through the integration of de-identified and demographically limited patient-level data with nationally representative health related data in the Medical Expenditure Panel Survey (MEPS). The inclusion of these measures has further stimulated hypothesis generation and the initiation of new studies that explore newly identified relationships. Attention is given to research efforts based on the integrated data that serve to identify health care disparities, advancing evidence-based policymaking focused on the issues of health care access and practice.

2. Improving TB Diagnosis in India: The Potential of the Private Sector and New Technologies
Miss Clara (Yehyun) Kyung (McGill University)
Mr Akira Camargo (Brown University)
Mr Elliott Chun (University of North Carolina at Chapel Hill)
Miss Hanyuying Wang (University of Cambridge)
Dr Asaph Young Chun (U.S. Census Bureau)
Miss Helene Cho (International Strategy and Reconciliation Foundation)
Ms Jiwon Sophia Lee (International Strategy and Reconciliation Foundation)

India carries the greatest burden of TB in the world, with over 2 million new cases every year. Despite efforts, there has hardly been any decline in TB incidence. This may be explained in part by the structure of India’s healthcare sector. The private healthcare sector treats over 50% of India’s TB cases, but it is highly unregulated and unorganized, and it often uses practices that are not evidence-based. As a result, TB is frequently misdiagnosed in the private sector. Misdiagnosis leads to inadequate or inaccurate treatment, and this can have serious consequences, ranging from transmission of TB to other people, death from untreated TB, or even the development of new strains of drug resistant TB.

The majority of TB cases can be cured with proper treatment, but in order to receive proper treatment, a patient must first receive a correct and timely diagnosis. India’s healthcare sector is struggling to complete this crucial first step, especially in the private sector. The purpose of this study is to identify and evaluate the ways in which the private sector and new technologies can advance the accuracy of TB diagnosis in India in order to create evidence-based policy recommendations. We use survey data and administrative records, including the Indian data and WHO administrative records, to address the questions of how the accuracy of TB diagnosis in India can be improved, especially in relation to diagnosis of active pulmonary TB and of drug resistant TB. In addition, we aim to address the question of how private and public healthcare sectors effectively address the management of TB in India by being sensitive to their respective strengths and drawbacks and leveraging new technologies that are effective for diagnosing TB. Findings may provide evidence to inform the process of public health policy in India, neighbouring countries and UN agencies.

3. Monitoring health and well-being with survey research in Finland – reporting results for national and regional level decision making
Ms Oona Pentala (National Institute for Health and Welfare Finland)
Mr Jukka Murto (National Institute for Health and Welfare Finland)
Mr Timo Koskela (National Institute for Health and Welfare Finland)
Dr Satu Helakorpi (National Institute for Health and Welfare Finland)

Finnish municipalities and cities have broad responsibilities e.g. in providing citizens with social and health care services and in arranging preventive work. Moreover, the Health Care Act (2010) obliges municipalities to monitor the health of their population and its subgroups. National Institute for Health and Welfare (THL) has obligatory responsibility to study and monitor the welfare and health of the population, the factors affecting and problems related to them, the prevalence of these problems and opportunities for preventing them, and to develop and promote welfare and health and reduce welfare and health problems.

In 2010, THL launched a new questionnaire survey called the Regional Health and Well-Being Study (ATH from its Finnish initials). During three-year period 2013−2015 in collaboration with municipalities, the sample size was increased up to 170 000. The aim was to collect follow up data from such health and well-being promotion actions that cannot be found from registers. Questions cover e.g. living and work conditions, well-being, health, functional and working capacity, risk behavior, service use and satisfaction.

Questionnaire forms are described in a machine readable XML format which provides an effective way to describe, save and check the data. Obtained data from paper and online forms is saved in a database and paired with metadata and register data (ex. education and income). Register data is also available for non-respondents, and that information is used to calculate adjustment weights to correct the non-response bias (Härkänen et al 2014). Overall response rate has been 54%. The gathered data is ready for analysis within 6 to 8 months from the beginning of the data collection period.

Survey data is processed into indicators together with substance experts. International indicator definitions such as WHO quality of life (WHOQOL-BREF) and Mental Health Index (MHI-5) are also used. For each indicator a metadata table is written, which includes the description of the indicator, what phenomena it measures, and how it is relevant for the public health and well-being. Metadata may also contain examples of cost burden of the phenomena, and some advice what authorities in the local authority could do to affect the phenomena.

These indicators are reported in THL’s interactive online service ( with tables, graphs and thematic maps (InstantAtlas). In the service it is possible to view the results divided into different population groups like age-, gender and education groups. Is also possible to compare different regions and compare municipality’s results with regional and national results. This online service has been developed together with regional and national decision-makers.

The indicator data is also available as Excel and csv sheets, in case the local authority needs to make their own reports. Indicator data is also available via open interface ( for database use and different applications. THL offers educational events where the use of these reporting services is presented and the data is also available for further analyses for free.