PRAMS Information for Researchers

PRAMS was developed by the Centers for Disease Control and Prevention (CDC) in 1987. Washington PRAMS began to collect data in June of 1993. The PRAMS survey asks about topics related to risk factors before and during pregnancy and infancy, access to prenatal and children's health care, and content of prenatal care.

The PRAMS survey answers give us information about what behaviors and factors may affect the health of the mother and her new baby. The answers help us to learn more about ways to improve the health of mothers and babies in Washington.

PRAMS Data Collection

In Washington, about 2,200 surveys are sent to participants each year. Approximately 206 mothers are drawn each month from birth certificate data using a random sample based on race and ethnicity. The sample includes the following racial/ethnic groups: White not Hispanic, African American not Hispanic, Asian/Pacific Islander not Hispanic, Native American not Hispanic, and Hispanic.

Washington PRAMS collects data through a statewide mailing of the survey. The PRAMS survey is sent to new mothers two to six months after they deliver their babies. Telephone follow-up is done for those who do not respond to the mailing. The mail and telephone surveys are available in English and Spanish. Washington PRAMS' unweighted response rate for 2021 is 54%.

Data Collection Cycle

  Multiple Mailings

• Pre-letter introducing survey
• 1st mailing with survey
• Reminder to complete survey
• 2nd mailing

  Telephone follow-up for non-respondents

• Up to 15 attempts per good phone number

  Online platform version available starting with 2023 cohort

PRAMS Limitations

Self-report information collected may not always be accurate

• Not able to accurately remember (recall bias)
• Unwilling to report behaviors accurately (social desirability bias)

Low response rates among some demographics

• Harder to reach populations less likely to respond
• Possible results in the non-response population differ from those who responded (non-response bias)

Small sample size

• ∼1000 respondents per year
• Smaller sub-group analysis is usually not possible without combining groups of years of data

Our Team