Registry Data and the National Immunization Survey

1
Registry Data and the National Immunization Survey

• Lawrence Barker
• National Immunization Program
• Centers for Disease Control and Prevention
• Department of Health and Human Services
• Linda Piccinino, Mike Battaglia
• Abt Associates

2
Overview

• Description of National Immunization Survey
  (N.I.S.)
• Comparison of N.I.S. and Registry Data
• Findings
• N.I.S. data can be used to track progress toward
  the Healthy People 2010 registry goals (N.I.S.
  shows that registry data are good but can stand
  improvement)
• Registry data can be used to benefit N.I.S.
  (registries contain some vaccines not reported to
  N.I.S.)

3
National Immunization Survey

• The N.I.S. is the nations primary tool for
  assessing immunization coverage among 19 to 35
  month old children
• The N.I.S. yields estimates of national, state,
  and selected urban area coverage for vaccines and
  vaccine series (for example, 43133, 4 doses
  of DTaP,3 doses of polio vaccine, 1 dose of
  MCV, 3 doses of hib, 3 doses of hep B)

4
N.I.S. is More Than One Survey...

• Nashville
• Milwaukee
• NYC
• Boston
• Cleveland
• Chicago
• Indianapolis
• Memphis
• Dallas
• San Jose
• Los Angeles
• San Diego
• Phoenix
• El Paso
• San Antonio
• Houston

• Seattle
• Detroit
• Boston
• Newark
• Philadelphia
• Baltimore
• Atlanta
• Jacksonville
• Miami
• New Orleans
• Birmingham
• Columbus
• D.C.

5
N.I.S. Methods

• Random-digit dialing
• Collect information from household
• probe to identify all immunization providers
• ask for permission to contact providers
• registries are not routinely contacted (might
  change in the future)
• N.I.S. estimates are based on those households
  for which providers give us adequate data
• Responses are weighted to represent U.S. children
  aged 19 to 35 months

6
N.I.S. / Registry Comparison

• In quarter 3 2002 (one site) and quarter 4 2002
  (three sites), four mature registries
  participated in a comparison of N.I.S. and
  registry data
• Sites are not identified for reasons of
  confidentiality, but are three states and one
  city

7
N.I.S. / Registry Comparison, continued

• Immunization records of children in the registry
  and with provider-validated N.I.S. immunizations
  were compared, with care-givers consent to
  contact registry (almost 90 percent consented, in
  all sites)
• Each site had between 65 and 140 children with
  provider-verified immunizations during quarter of
  interest

8
N.I.S. / Registry Comparison goals of study

• To show feasibility of using N.I.S. data to track
  progress toward the Healthy People 2010 goal of
  95 percent participation in registries
• To determine if consulting registries (with
  caregivers consent) can benefit N.I.S.

9
Limitations

• Cant necessarily extrapolate to the population
• Children with both registry data and data in
  N.I.S. are not a true random sample
• results presented here are not weighted
• Can
• reach qualitative conclusions
• state that the issues discussed are real
• get a (very) rough sense of the seriousness of
  the issues

10
Children with Provider-Verified Data Who Had Data
in the Registry

• Percent of children with provider-verified data
  who had two or more vaccines in the registry
  (unweighted)
• Site A 88 percent
• Site B 73 percent
• Site C 80 percent
• Site D 60 percent
• These sites (and probably others) are doing a
  good job in capturing at least some immunizations

11
Children with Both Provider-Verified Data and
Registry Data with Vaccinations in NIS but Not in
Registry

• Percent of children with both provider-verified
  and registry data with vaccinations in NIS but
  not in registry (unweighted)
• Site A 64 percent
• Site B 43 percent
• Site C 34 percent
• Site D 31 percent
• Registry data are of variable completeness

12
Children with Both Provider-Verified Data and
Registry Data with Vaccinations in Registry But
Not in N.I.S.

• Percent of children with both provider-verified
  and registry data with vaccinations in registry
  but not in N.I.S. (unweighted)
• Site A 15 percent
• Site B 28 percent
• Site C 24 percent
• Site D 12 percent
• Registries indicate some vaccinations that N.I.S.
  does not

13
Children with Both Provider-Verified and Registry
Data that Registry Said Were 43133 Complete
but N.I.S. Did Not

• Percent of children with both provider-verified
  and registry data that registry said were
  43133 complete but N.I.S. did not
  (unweighted)
• Site A 3 percent
• Site B 10 percent
• Site C 6 percent
• Site D 2 percent
• Series completion differences are of concern

14
Children from Last Slide that had 4th dose of
D.T.a.P. in Registry but Not in N.I.S.

• 4th dose means greater or equal to 4 doses as
  opposed to less than 4 doses not necessarily
  the chronologically last dose
• Percent of children from last slide that had 4th
  dose of D.T.a.P. in registry but not in N.I.S.
  (unweighted)
• Site A 100 percent
• Site B 60 percent
• Site C 86 percent
• Site D 100 percent
• Detected series completion differences are mostly
  in the 4th dose of D.T.a.P.

15
Conclusions

• N.I.S. data can be used to track progress toward
  Healthy People 2010 goals (registry data are
  good, but need to improve)
• Registry data can benefit N.I.S.
• Some vaccinations (primarily 4th D.T.a.P. dose)
  in registry but not in N.I.S.
• Hypothetical explanations
• Duplicative (or otherwise erroneous) entries in
  registries
• Some providers report some immunizations to
  registries but not to N.I.S.
• N.I.S. does not identify all immunization
  providers for some children
• More research needed to learn reasons for
  discrepancies