Tuesday 16th July
Wednesday 17th July
Thursday 18th July
Friday 19th July
Download the conference book
Collecting and Analysing Physical Measure and Biomarker Data in Surveys |
|
| Chair | Dr Stephanie Mcfall (University of Essex) |
Physical measurements are taken on a number of cross-sectional and longitudinal studies in the UK as well as internationally. This paper outlines our approach to developing an interviewer accreditation process for conducting physical measurements in the Age 11 survey of the Millennium Cohort Study (MCS).
The MCS is a large-scale birth cohort study following over 19,000 children born in the UK in 2000/1. So far there have been five waves of the study at 9 months, 3 years, 5 years, 7 years and most recently at age 11. Since age 3, interviewers have taken the cohort child's physical measurements in the home.
It is generally well-established that, with appropriate training, field interviewers are able to carry out physical measurements in a home setting. At age 11 we developed a formal accreditation process to ensure interviewers were confident and accurate in measuring children's height. Our approach drew on formal training and accreditation schemes developed by a number of other large-scale and longitudinal studies internationally, while considering the specific context and constraints of this particular study.
The process entailed developing a measurement protocol for handling the equipment, positioning the child and taking an accurate reading, based on techniques developed by experts at the Child Growth Foundation. Training, practice and accreditation sessions were built into the interviewer briefing programme and supplemented by an additional check in the Field.
This paper reports on the protocols developed and concludes with lessons learnt and implications for future practice.
The UK Household Longitudinal Study (UKHLS), Understanding Society, is a large longitudinal study with annual interviews of members of sampled households. In spring 2011, we assessed the feasibility of collection of biomeasures by trained non-clinical interviewers. Building on established measurement protocols, we implemented anthropometrics, blood pressure, grip strength and collection of saliva and dried blood spots (DBS). We evaluated processes for recruitment and training of interviewers, and collected data from 92 adult participants. The study also provided information about time, participation and quality of biological samples. There were multiple positive outcomes supporting feasibility. Ten interviewers were selected for training and nine achieved certification through a performance-based evaluation. Interviewers met objectives for number of participants and biological samples, though the rate of refusal was high. Among participants, completion of individual measures ranged from 86-100%. The percentage providing DBS was higher than for nurse-collected whole blood (in a different study component); satisfaction was also high among respondents providing DBS. Quality and number of DBS were sufficient for three analytes. Quality and quantity of DNA extracted from saliva was adequate. Other dimensions were less convincing of feasibility, including the amount of time for data collection and adequacy of the DBS as a long-term research resource. There is more to feasibility than "getting data in the door." We conclude by describing how the pilot study will influence the design of the UKHLS.
Abstract. Dengue Hemorrhagic Fever (DHF) is an acute febrile disease caused by the dengue virus which is transmitted by the Aedes aegypti mosquito. DHF is a serious public health problem in Indonesia causing death within a short time. Bogor is a city in West Java with a level as high as 1504 cases in 2009 (Incidence rate (IR): 1.564 per 1000 ; Case Fatality Rate (CFR): 0.26%) . To suppress the spread of the disease, early detection is crucial. In this context identification of areas that have great potential to become endemic is of great importance.
Two different methods have frequently been applied to identify possible high incidence regions: choropleth mapping of the non-smoothed SIRs and Bayesian methods to smooth the SIRs. We compare both methods in terms of Mean Square Error and conclude that the Empirical Bayes method based on the Poison - Gamma model performs best