In-home collection of saliva samples and physical measurements in the Millennium Cohort Study

Transcription

1 In-home collection of saliva samples and physical measurements in the Millennium Cohort Study Emla Fitzsimons CLOSER workshop on Combining social and bio-medical data collection, July 4 th 2017

2 About the Millennium Cohort Study Millennium Cohort Study (MCS) is a birth cohort study of around 19,000 individuals born in 398 areas of the UK in Highly multidisciplinary survey Surveys to date (planned): 9 months, 3, 5, 7, 11, 14, (17) 11,726 families at age 14 Joshi & Fitzsimons (2016). The UK Millennium Cohort Study: the making of a multipurpose resource for social science and policy. Longitudinal and Life Course Studies Vol. 7:

3 Outline Collection of physical measurements and saliva samples in the 6 th (age 14) sweep of MCS 1. Collection mode 2. Training and accreditation 3. Response rates and data quality

4 Collection modes Main models for bio-measure data collection in social surveys: 1. Medically trained personnel in specially equipped clinics 2. Home-visit setting with portable equipment by a mobile nurse field force 3. Home-visit setting with portable equipment by specially trained interviewers

5 Collection modes Main models for bio-measure data collection in social surveys: 1. Medically trained personnel in specially equipped clinics 2. Home-visit setting with portable equipment by a mobile nurse field force 3. Home-visit setting with portable equipment by specially trained interviewers

6 Field interviewers Advantages of model that uses field interviewers Evidence of higher response rates in a home setting collected alongside other study elements, rather than as part of follow-up clinic or nurse visit (Clemens et al. 2012) More cost-effective Fewer capacity issues than with nurse collection Caveat Appropriate training important to ensure high quality data collection: Formal accreditation increasingly common on surveys in the US and UK

7 Physical measurements Measuring child growth accurately and providing robust information on childhood obesity are important elements of the study Using field interviewers to collect physical measurements has been used for several waves of the MCS, since age 3 Age 9 mths Height and weight Waist measurement Body fat percentage Age 3 Age 5 Age 7 Age 11 Age 14

8 Saliva samples Collected at age 14 from cohort members and biological parents (where present in home) First time a triad of DNA samples collected from 2 biological parents and child in a large scale study Piloted at age 11 (Calderwood and Rose, 2013)

9 Saliva samples Collected at age 14 from cohort members and biological parents (where present in home) First time a triad of DNA samples collected from 2 biological parents and child in a large scale study Piloted at age 11 (Calderwood and Rose, 2013) Collaboration with University of Bristol store the samples, extract DNA, genotype Genotyping underway; process to take ~18 months (mid 2018); access likely via an independent Access Committee

10 Equipment physical measurements Height: Leicester height measure stadiometer: portable collapsible device with sliding head plate, base plate and 4 connecting rods marked with a measuring scale Weight and body fat: Tanita scales (BF-522W) Calibrated prior to being issued to interviewers

11 Equipment saliva samples Oragene DNA Self-Collection Kit OG-500

12 Training and Accreditation Protocols developed to ensure a standardised approach to collecting measures Training, practice and accreditation sessions built into (a packed!) 3-day interviewer briefing programme, to ensure protocols followed; supplemented by quality control procedures in field Briefing programme Attended by approx. 300 interviewers Held in 14 locations across the UK 7-45 interviewers per briefing

13 Training and Accreditation Briefing programme scheduled in order to facilitate training and accreditation: Days 1 and 2 of briefing on consecutive days, with a break before day 3 During break: interviewers required to conduct the physical measurements and saliva sample elements (among other things) in a home-setting with 2 specifically recruited 14 year olds On day 3, interviewers completed a competency-based assessment under test conditions on the physical measurements and saliva collection procedures, to ensure protocols being adhered to

14 Training and Accreditation Accreditation took place in groups of three: 1 accreditor and 2 interviewers per group Interviewers carried out the physical measurements and saliva collection procedures on each other, whilst the accreditor observed and completed an accreditation booklet Accreditation booklet set out marking criteria for each measurement, with major and minor errors listed for each measurement and a script for the accreditor to administer the accreditation. To pass each part of the accreditation interviewers needed to: Make no major errors Make no more than 3 minor errors All interviewers had to pass the accreditation prior to starting work. Re-accreditation arranged for those who did not pass first time MCS6 Technical Report (2016); Calderwood et al (2014)

15 Transporting saliva samples Samples packaged in accordance with the transportation of biological substances regulations and posted to the laboratory at the University of Bristol: 1. Interviewers placed the filled, sealed collection tube into a small plastic bag along with some absorbent material and then sealed the bag 2. Up to 15 samples were placed in a pre-addressed jiffy bag with a dispatch form listing the enclosed samples 3. Jiffy bags were dispatched to the laboratory on a weekly basis, or when they contained 15 tubes (whichever occurred first)

16 Response rates Physical measurements Cohort member Eligible Response (rate) 11,884 11,408 (96.0%)

17 Response rates Cohort member Eligible Physical measurements Response (rate) 11,884 11,408 (96.0%) 10,970 (96%) fully productive, i.e. provided all three measures 438 (4%) partially productive

18 Response rates Cohort member Natural mother Natural father Eligible Saliva samples Valid collection (rate) 11,806 9,758 (82.7%) 11,578 9,634 (83.2%) 7,215 5,202 (72.1%) Total 30,599 24,594

19 Response rates Cohort member Natural mother Natural father Eligible Saliva samples Valid collection (rate) 11,806 9,758 (82.7%) 11,578 9,634 (83.2%) 7,215 5,202 (72.1%) Total 30,599 24,594 DNA extracted from 23,499 samples 96% of collected samples Approx. 90% of samples gave yields of at least 20 µg, sufficient DNA for a range of genetic studies

20 Conclusion At-scale collection of biomeasures by field interviewers can yield high quality data, high response rates and be cost-effective Evidence of higher response rates and lower costs when embedded in routine survey, rather than as follow-up clinic or nurse visit; no evidence that quality compromised Training and accreditation important to ensure high quality data collection Interviewers could be trained to carry out other non-invasive bio-measures using a similar approach

21 Conclusion contd. Quality - Extensive research exists and validates saliva as an equivalent alternative to blood for genomic DNA Ease of collection - relative to blood: lower overall cost, lower infection risk, increased patient convenience, acceptability, compliance, and uptake; more appropriate for certain populations, e.g. children Processing - saliva samples compatible with high-throughput DNA processing, allowing integration into existing automated lab extraction procedures