Health Topics

• Data Development
• Health Care Reform
• Health Information Technology
• Medicaid and CHIP
• Medicare
• Mental Health and Substance Abuse
• Payment Policies and Costs
• Public and Population Health
• Quality and Health System Functioning
• State Health Policy

•  

Health Tracking Household Survey and Community Tracking Study Household Survey

The Health Tracking Household Survey (HTHS), formerly the Community Tracking Study Household Survey (CTS), is funded by the Robert Wood Johnson Foundation (RWJF) and conducted under the direction of our research partner, the Center for Studying Health System Change (HSC). This large-scale, long-term survey is designed to provide a sound information base for decision making by health care leaders. It does so by collecting information on how the health system is evolving across the United States and how these changes affect people.

The HTHS-CTS, which Mathematica has conducted since 1996, is a longitudinal project that relies on periodic site visits and surveys of households. Our work includes coordinating survey design, instrument development, sample design and implementation, interviewing, and weighting.

The HTHS-CTS survey instrument was administered by telephone, using computer-assisted telephone interviewing (CATI) technology. Respondents in round 5 were selected through list-assisted random-digit dialing (RDD) sampling methodology, with no clustering by site, no sampling of prior round sample members, and no field component (these were design components in previous rounds).

In round 1 of the survey, completed in July 1997, over 32,000 families were interviewed from the RDD sample; 635 were interviewed from the field sample. The families included close to 50,000 adults and 11,000 children. The weighted family-level response rate was 64.8 percent; 95 percent of adults completed the questions in a self-response module.

In round 2, completed in November 1999, over 31,000 families were interviewed from the RDD sample; close to 800 were interviewed from the field sample.  The families included almost 48,000 adults and over 10,000 children. The weighted family-level response rate was 62.5 percent; 94 percent of adults completed the questions in the self-response module.

Round 3 was fielded between September 2000 and September 2001. The final total consisted of close to 32,000 families from the RDD sample and 925 from the field sample. The families included about 50,000 adults and 10,000 children. The weighted family-level response rate was 58.6 percent; 94 percent of adults completed the self-response module.

In round 4, which began in February 2003 and was completed in February 2004, the sample size was reduced. The final total consisted of almost 25,000 families from the RDD sample and 800 from the field sample.  The families included over 39,000 adults and 7,000 children. Sample targets were reduced in the fourth wave of data collection by removing the unclustered national (“supplemental”) sample and by reducing the RDD samples in the 12 high-intensity sites. The weighted response rate was 56.5 percent, and 94 percent of the self-response modules were completed.

Data collection for round 5 took place between April 2007 and January 2008. The final sample of completed interviews consisted of 9,400 families, including over 15,000 adults and 2,600 children. The unweighted and weighted response rates were both 43.5 percent, and 90.5 percent of the self-response modules were completed by people who had also finished core interviews.

The findings from round 5 of this survey have been widely disseminated and reported in publications such as the New York Times and Wall Street Journal. Mathematica staff have presented methodological papers about the study at the Joint Statistical Meetings and the Association for Public Opinion Research. In addition, many other researchers have used the data to conduct important studies of health care issues.