Suggested Citation | Executive Summary | Introduction | Methodology | Appendices
Suggested Citation: Department of Human Services, Health Services. Oregon PRAMS First Year Report, 1998-99. Portland, Oregon: Office of Family Health, 2000.
Executive Summary: The Pregnancy Risk Assessment Monitoring System (PRAMS) is a new survey of women in Oregon interviewed approximately 3-4 months after giving birth. PRAMS is intended to help policymakers and program managers plan and evaluate programs for pregnant women and infants.
Oregon PRAMS collects data on maternal behaviors and experiences which, according to prior research, can affect the health of the infant. Topics covered in the interview include: unintended pregnancy; Medicaid coverage; prenatal care; participation in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC); maternal smoking and alcohol use; prenatal human immunodeficiency virus (HIV) prevention and test counseling; breastfeeding; infant sleep position; and birth control.
In its first year (November 1998 to October 1999 *), 1867 women were interviewed by mail or telephone. Every woman whose newborn had a birth certificate was eligible to be selected for the survey if the baby had been born in Oregon and the mother was a resident of Oregon. A random sample of women was selected each month. Overall, 64.0% of the women contacted statewide completed an interview.
Because one health concern is low birth weight, the PRAMS survey oversampled women whose babies had low birth weight (less than 2500 grams [5 pounds and 8 ounces]). Because women of color are more likely to have an infant with health problems, the PRAMS survey oversampled American Indian, Hispanic, African-American and Asian women. This oversampling was done to make sure that there would be enough responses for a meaningful analysis of issues related to race/ethnicity and low birth weight.
This report provides the statewide answers to the Year 1 questions. To ensure that the results are representative of the Oregon population of new mothers, the answers have been weighted to adjust for the disproportion of oversampling and for any bias caused by non-response.
The Appendices include the actual questions that were asked (Appendix A) (pdf) (50K) and some detail about the methods used in the survey (Appendix B).
Researchers who wish to use PRAMS data should contact Dr. Ken Rosenberg, PRAMS Project Director, 971-673-0237; E-mail: email@example.com.
Introduction The Pregnancy Risk Assessment Monitoring System (PRAMS) is part of the Office of Family Health's initiative to use data collection and analysis to support program development, program evaluation, and policy-making. PRAMS is an ongoing, population-based survey designed to identify and monitor selected maternal behaviors and experiences. The experiences and behaviors are those that occur before, during, and shortly after pregnancy. Oregon PRAMS began collecting data in November 1998. The women completed a mailed questionnaire or a telephone interview. They are a random sample of Oregon residents who delivered a live-born infant, as identified from an Oregon birth certificate. *
Oregon PRAMS was modeled after the PRAMS program run (since 1987) by the Centers for Disease Control and Prevention (CDC) and selected state health departments. Oregon PRAMS collects data on a variety of topics, including unintended pregnancy, prenatal care, Medicaid coverage, breastfeeding, smoking, drinking, infant sleep position, birth control, prenatal human immunodeficiency virus (HIV) prevention and HIV test counseling, and participation in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC).
This report presents an overview of the Oregon PRAMS project, a discussion of the Oregon PRAMS methodology, and a summary of the responses to the Oregon PRAMS survey for the first 12 months. The bulk of the report is the tabulation of the answers to the survey questions. Appendix A (pdf) (50K) presents the English-language mailed version of the survey. The Spanish-language mailed version and the phone interview scripts (in English and Spanish) are available upon request to the PRAMS Project Director.
This report does not contain a detailed analysis of each of the topics identified above. Our expectation is that programs and health researchers will analyze PRAMS data. Policymakers will then be able to use these analyses to monitor progress toward pregnancy-related health objectives. We are publishing the first PRAMS report in this form, with no analysis, to inform the broadest possible audience, as early as possible, of exactly what information we have -- so that programs and policymakers will think about how this and future data can inform their missions. We view dissemination of the data included in this report as a key step in the translation of PRAMS data into public health action, a primary goal for PRAMS. We hope that this report will serve as a valuable reference document for use in public health planning and policy development. Our goal is that analyses of these data will affect the work in local and state Maternal and Child Health (MCH) programs and perhaps generate questions for future versions of PRAMS.
Methodology Every month, a stratified random sample of 200-300 new mothers was selected from a frame of eligible birth certificates. The sampling frame included all Oregon resident women whose babies had been born in Oregon 60-180 days before the selection date and who had not previously been in the sampling frame. The PRAMS survey oversampled American Indian, Hispanic, African-American and Asian women and women whose babies had a birth weight less than 2500 grams (5 pounds and 8 ounces). This oversampling provided a sufficient number of responses to permit meaningful analysis of health risks related to race/ethnicity and low birth weight. Most women were contacted about 3 months after their baby was born. Each sampled mother was mailed an explanatory letter that introduced the survey, followed by a package containing the 20-page questionnaire one week later. A second package, containing the same questionnaire was mailed about 3 weeks later to those who did not respond. Telephone interviews were conducted by staff from Clearwater Research, Inc. of Boise, Idaho. Clearwater staff then telephoned those mothers who did not respond to the mailed survey and attempted to complete an interview.
Before analysis, the interview responses were weighted to make them representative of Oregon women with live births. Responses were first weighted to account for the sampling design (oversampling by race/ethnicity and birth weight): "over-sampling adjustment." The second layer of weighting ("unit non-response") accounted for non-response (e.g., young women were less likely to respond than older women). The third layer of weighting ("non-coverage adjustment") accounted for the very few birth certificates (about 0.03%) that were never in the sampling frame, most of whom are adopted or whose birth certificates were processed more than 180 days after birth (who were intentionally excluded). Details of the weighting methodology appear in Appendix D.
Additional information about the PRAMS methodology can be found in the appendices. Appendix B contains the questions that were asked in first 12 months of Oregon PRAMS. (As was noted above, copies of the later English version, as well as copies of the Spanish mailed questionnaire and the English and Spanish telephone interview scripts, are available upon request.) Appendix C describes the PRAMS data collection methodology in detail. Appendix D contains details on weighting, sample sizes, response rates, and stratification variables.
All tables in the report were produced using weighted PRAMS data. Percentages were calculated using SPSS. The number of respondents reported in each table is the number of women who answered that question. All missing (blank and "don't know") observations are excluded, unless "don't know" is an answer option.
Note that PRAMS data are representative of women whose pregnancies resulted in a live birth and are not generalizable to all pregnant women.
*Child's dates of birth were from 8/1/98 through 8/9/99.
Return to the top of the page.