MEC Interview:

1


MEC Interview Measuring Sensitive Behaviors
Debra Brody, MPH, Geraldine McQuillan, PhD,
Benny Kottiri, PhD, Elizabeth Hintz, MS
Centers for Disease Control and
Prevention National Center for Health Statistics
2
Impact of complete reporting of sensitive
behaviors

• Understanding relationships between behavior and
  disease outcomes
• Modeling estimates of risk
• Evaluating prevention programs

OB142S002
3
Overview

• Two bilingual interviewers per team
• Two computer-based interview modes
• Interview time 10 30 minutes

• Personal interview (CAPI)
• Self-administered interview (A-CASI)

OB142S003
4
MEC Interview Three sections

• Current health status, health behaviors, and
  reproductive health history
• Mental health assessments
• Risk behaviors and other sensitive topics

OB142S004
5
Current health status and behavior

• Recent illness, quality of life (12)
• Incontinence (20-59)
• Tobacco use past 5 days (20)
• Alcohol use (20)
• Reproductive health (females 12)
• Physical Activity (12-15)

OB142S005
6
Mental health assessmentsComposite
International Diagnostic Interview (CIDI)

• 20-39 years (1/2 sample)
• Panic disorder
• Generalized anxiety disorder
• Major depression

OB142S006
7
Mental health assessmentsDiagnostic Interview
Schedule forChildren (DISC IV)

• Based on DSM IV/ ICD-10
• Ages 8 19 years
• Youth informant (in-person)
• Parent informant (telephone)

OB142S007
8
Diagnostic Interview Schedule for Children
(DISC) Modules

• Youth modules

• Parent modules

• Depression
• Eating
• Conduct
• Panic
• GAD

• Depression
• Eating
• Conduct
• Elimination
• ADHD

OB142S008
9
Audio Computer Assisted Self-Interview (A-CASI)

• Questions simultaneously seen and heard
• Greater privacy than face-to-face interview
• Multilingual administration possible
• Literacy issues minimized

OB142S009
10
NHANES A-CASI topics

• Youth conduct disorder (12-19 years)
• Alcohol use (12-19 years)
• Tobacco use (12-19 years)
• Drug abuse (12-59 years)
• Sexual behavior (14-59 years)
• Mens health (18 years)

OB142S010
11
A-CASI screen
12
A-CASI non-response screen
You did not answer the previous question. Was it
because

• You really meant to answer?
• You prefer not to answer?
• You dont know the answer?

OB142S011
13
Non-response Male-to-male sex CAPI vs. A-CASI
NHANES 1988-94
NHANES 1999-00

• Refused 5
• Dont know 3

• Refused lt1
• Dont know lt1

OB142S012
14
Summary

• Expansion of sensitive risk behavior questions in
  NHANES
• Improved accuracy of reporting with A-CASI
  approach

• Minimizes item non-response
• Reduces underreporting

OB142S013