Baseline Information for Evaluating the Implementation of the Health Insurance Portability and Accountability Act of 1996: Final Report. D. Policy Databases

10/01/1998

The policy databases we abstracted are:

  1. Community Rating: Issues and Experience
  2. Health Insurance Regulation: Variation in Recent State Small Employer Health Insurance Reforms
  3. State Experiences with Community Rating and Related Reforms
  4. Health Insurance Portability: Reform Could Ensure Continued Coverage for up to 25 Million Americans
  5. Understanding Individual Health Insurance Markets
  6. Summary of State Insurance Laws
  7. Major Health Care Policies: Fifty State Profiles, 1997
  8. New Federalism State Database
  9. HIAA Summary of Current Developments for HIPAA

 They were screened for the following content:

  • Guaranteed Issue
  • Guaranteed Renewal
  • Preexisting Condition/Portability
  • Individual Market Risk Pool
  • Rating Reform

 1. CPS

Organization Census Bureau
Contact General: (301) 457-3806 Health: Robert Bennefield (301) 763-8578 
Type public-use
Availability/Cost $150/CD, 

available about 6 months after data collection

can extract data or generate tables from /ferret.bls.census.gov/cgi-bin/ferret

or extract data from /www.census.gov/DES/www/welcome.html 

Time Period annual; conducted in March
Reference Period health insurance questions pertain to any time during previous calendar year.

Starting in 1995, also asks about current coverage.

Scope of coverage noninstitutionalized civilians and members of Armed Forces in civilian housing units
Analysis unit person, family, household
Sample size 48,000 households, 

136,000 persons

Response rate 92-93% provide basic labor force information

80-82% of households complete supplement

Geography 50 states and DC

Allows for state-level analyses, though might want to pool consecutive years to improve the estimates for smaller states. (The states with at least 1,400 people sampled are CA, NY, FL, TX, PA, IL, OH, MI, NJ, NC, MA) 

States can be identified.

Design sample selected using the 1990 Decennial Census

probability sample stratified by state to select households.

oversample hispanics

Data collection mode primarily telephone, some in-person
Content insurance coverage:
  • current or former employer/union
  • individual
  • public

current work status

recent work history

industry code of worker

occupation code of worker

health status

(April 1993 include offers of employer health insurance)

Note: Because of changes to the set of health insurance questions starting in March 1995, caution should be used when making comparisons to earlier years.


2. SIPP

Organization Census Bureau
Contact Michael McMahon (301) 457-3819 or Robert Bennefield (301) 763-8578
Type public-use
Availability/Cost $200/CD(wave), $1050 for longitudinal file.

can extract data or generate tables from /ferret.bls.census.gov/cgi-bin/ferret

or extract data from /www.census.gov/DES/www/welcome.html 

Time Period continuous series of panels from 2.5 to 4 years; most recent 2/93-2/96
Reference Period health insurance questions pertain to any time in the four months since the previous interview
Scope of coverage noninstitutionalized civilians and members of Armed Forces in civilian housing units
Analysis unit person (household and family composition is not constant throughout reference period)
Sample size 1993 panel: 20,000 households 
Response rate 91% of occupants in eligible living quarters were interviewed
Geography 50 states and DC

States can be identified, but state-level estimates are subject to high variance and are not recommended.

Allows for regional analyses.

Design multi-staged stratified sample 
Data collection mode decentralized telephone and in-person
Content insurance coverage:
  • current employer
  • cobra
  • individual
  • public

current work status

recent work history

industry code of worker

occupation code of worker

health status

Note: reinterviews at 4 month intervals allow for examination of changes in insurance status


3. NHIS

Organization National Center for Health Statistics (conducted by Westat)
Contact Owen Thornberry, Jr, (301) 436-8500
Type public-use
Availability/Cost $21 for CD purchase through NTIS (through 1994)

$60 for CD purchase through GPO (through 1994)

$645 for tape purchase through NTIS (through 1995)

1995 data is expected to be available on CD in July, 1998

order from www.cdc.gov/nchswww/products/catalogs/subject/nhis/tapeform.htm

Time Period annual; continuous sampling covers past 12 months; most recent available: 1995
Reference Period health insurance questions pertain to current status and changes in past 12 months
Scope of coverage noninstitutionalized civilians 
Analysis unit person
Sample size 36,000-49,000 households per year

92,000-125,000 persons

Response rate 90% of households completed interviews
Geography 50 states and DC

starting in 1995, design of sample modified to allow state-level estimates for the larger states, or can pool across years to obtain estimates for each state.

States can NOT be identified on main file, but region and large MSAs are identified.

Design multistage probability sample (with oversampling of black and Hispanic households)
Data collection mode in-person
Content insurance coverage:
  • current or former employer/union
  • private, not through employer/union
  • public

change in insurance status (# of months not covered in past 12 months)

denied or restricted coverage because of health condition

premiums (individual/employee contribution)

current work status

recent work history 

industry code of worker

occupation code of worker

health status

Note: In the past, separate state files have been created to allow for state-level estimates of the larger states (in 1994, 39 states were included on the file). However, the researcher is cautioned about sample size and coverage issues for the smaller states. The 1995 NHIS was redesigned to improve the ability to perform state-level estimates but still has the same limitations for the smaller states. The 1995 state file will not be available before the end of 1998.


4. MEPS-HC

Organization AHCPR (conducted by Westat)
Contact MEPS project manager at AHCPR (301) 594-1406
Type public-use
Availability/Cost $60/CD purchase from NTIS (800) 533-6847 or www.ntis.gov/fcpc

Free from AHCPR Publications Clearinghouse (800) 358-9295 orwww.ahcpr.gov

Time Period continuous rotating panels starting in 1996; most recent round completed in 12/97.
Reference Period health insurance questions pertain to coverage in effect at any time during the round 
Scope of coverage noninstitutionalized civilians 
Analysis unit person, family
Sample size 24,676 persons in round 1 eligible households (23,612 with positive weights) 
Response rate 78%
Geography 50 states and DC

allows for regional, but NOT state-level analyses.

census region identified on public-use file (state NOT identified) 

Design drawn from NHIS sampling frame 
Data collection mode five times in-person, sixth time by phone
Content insurance coverage:
  • current employer
  • COBRA
  • previous employer (not COBRA)
  • individual
  • public

changes in insurance status (across rounds)

employer offers health insurance

premiums (individual/employee contribution can be obtained by linking to MEPS-IC)

current work status

recent work history (can be ascertained after multiple rounds)

health status

Note: Can be linked to detailed plan information from those employers identified as providing insurance to household respondents.

Also can be linked to data collected from the respondents’ medical providers and insurance providers.

Panel rounds allow for examination of changes in insurance status.

Citation: Cohen, J. W., Monheit, A. C., et al. "The Medical Expenditure Panel Survey: A National Health Information Resource," Inquiry 33: 373-389, 1996/1997.


5. CTS

Organization Center for Studying Health System Change (conducted by Mathematica)
Contact center@hschange.com, Peter Cunningham (202) 554-7549
Type public-use
Availability/Cost free; will be available through ICPSR later in 1998
Time Period in field 7/96-7/97
Reference Period health insurance questions pertain to current status and any change in past 12 months
Scope of coverage non-institutionalized civilians 
Analysis unit person, family (insurance unit), healthcare "community"
Sample size 36,200 families:

1,225 per high-intensity metropolitan site

375 per low-intensity metropolitan site

4,500 total from non-metropolitan areas

3,500 total from unclustered national sample

Response rate 65%
Geography 12 high-intensity metropolitan areas, 36 low-intensity metropolitan areas, 12 non-metro areas.

Supplemental sample allows for national estimates. 

Also allows for analysis of 12 high-intensity sites.

Design 60 communities: separate probabilistic samples of high-intensity communities, low-intensity metropolitan areas, and non-metro areas stratified geographically.

supplement: unclustered national sample

Data collection mode telephone (in-person for those without phone in high-intensity communities)
Content insurance coverage:
  • current or former employer/union
  • individual
  • public

change in insurance status (has coverage changed in past 12 months?)

employer offers insurance

premiums

current work status

industry code of worker (open-ended)

health status 

Note: There is also a linked Insurer Followback Survey that is ongoing (expected to be completed in fall 1998). They are attempting to contact all identifiable insurers and should have 1,400-1,500 health plan entities. The survey’s main purpose is to obtain managed care characteristics, and though they are collecting premium data, the quality is suspect.

Citation: Kemper, B., Blumenthal, D., Corrigan, J., et al, "The Design of the Community Tracking Study: A Longitudinal Study of Health System Change and Its Effects on People," Inquiry 33: 195-206. Summer 1996.


6. BRFSS

Organization Division of Adult and Community Health, National Center for Chronic Disease Prevention and Health Promotion, CDC
Contact Craig Lutzinger at (770) 488-5304 or ccdinfo@cdc.gov 
Type public-use
Availability/Cost free through GPO
Time Period annually; most recently available: 1996
Reference Period health insurance questions pertain to current status
Scope of coverage noninstitutionalized civilian adults (18 and above)
Analysis unit person
Sample size 1,100 to 4,000 interviews per state
Response rate 78% cooperation rate (median of rates by state)
Geography 50 states, DC, and 3 territories

allows for state-level analyses.

States can be identified.

Design random phone within state (design varies by state)
Data collection mode telephone
Content insurance coverage:
  • employer
  • individual
  • public

change in insurance status -- how long have you been without coverage?

current work status

recent work history (if not currently working, have you been out of work less than 1 year?)

health status

Note: identifies main reason you don’t currently have insurance coverage identifies main reason you didn’t have coverage at some time in past year


7. NSAF

Organization Urban Institute (conducted by Westat)
Contact (202) 828-1815 or email anfinfo@ui.urban.org
Type public-use
Availability/Cost (will be available in 1998)
Time Period data collection: February through November 1997 (follow-up survey in 1999 or 2000).
Reference Period health insurance questions pertain to current status.
Scope of coverage Families with children under age 18 and non-aged adults without children under age 18
Analysis unit person, family
Sample size sampled approximately 44,000 persons
Response rate still preliminary
Geography 13 states: AL, CA, CO, FL, MA, MI, MN, MS, NJ, NY, TX, WA, WI plus 5,000 respondents from rest of nation to allow for national estimates.

Allows for state-level estimates of the 13 states.

States can be identified.

Design random digit dial survey with oversampling of low-income households (<200% poverty)
Data collection mode telephone survey (3.5% of sample are non-telephone households that were supplied with cellular phones)
Content insurance coverage:
  • current or former employer/union
  • individual
  • public

change in insurance status over past 12 months (asked of most knowledgeable adult and up to 2 children)

current work status

industry code of worker (open-ended)

occupation code of worker (open-ended)

health status (asked of most knowledgeable adult and up to 2 children)

 

8. RWJ Family

Organization Robert Wood Johnson Foundation (conducted by Mathematica)
Contact ICPSR (313)-763-5010
Type public-use
Availability/Cost available through ICPSR at /www.icpsr.umich.edu/archive1.html (study #=6894)
Time Period data collection: last half of 1993 and early 1994
Reference Period health insurance questions pertain to current status and ever in past 12 months
Scope of coverage noninstitutionalized civilians 
Analysis unit person, family (insurance unit) 
Sample size 62,549 persons
Response rate range from 77-91% by state
Geography 10 states - CO, FL, MN, NM, NY, ND, OK, OR, VT, WA.

allows for state-level analyses.

States can be identified.

Design random digit dialing by state to screen eligible families. 

oversampled families containing one or more Medicaid or uninsured persons.

Also selected families without telephones in a few geographic areas per state.

Data collection mode telephone supplemented with in-person interviews for those without phones
Content insurance coverage:
  • current employer
  • cobra
  • retiree group
  • individual
  • public

change in insurance status (over past 12 months)

employer offers insurance

premium (individual policies only)

current work status

recent work history (if not currently working, were you looking for work during past 4 weeks?)

industry code of worker (open-ended)

health status

Note: if employer offers insurance but employee not covered, identifies why not.

Citation: Cantor, J. C., Long, S. H., Marquis, M. S., "Challenges of State Health Reform: Variations in Ten States," Health Affairs(January/February 1998): 191-200.