Oregon PRAMS: Presentations
The Use of PRAMS to Assess Newborn Hearing Screening,
Discussion 1 |
Discussion 2 |
Discussion 3 |
The Use of PRAMS to Assess Newborn Hearing Screening,
Ken Rosenberg, MD, MPH
Alfredo P. Sandoval, MS, MBA
Department of Human Services
Office of Family Health
Seventh Annual MCH Epidemiology Conference
December 12, 2001
Clearwater Beach, Florida
NOTES: Universal Newborn Hearing Screening (UNHS) is a set of technologies that can screen newborns for deafness. It's the first step in getting infants with hearing loss into early intervention. A state law mandating that hospitals do NHS (for all babies born in hospitals with more than 200 live births per year) was passed by the Oregon legislature in 1999 and went into effect July 1, 2000. We developed a method to estimate the effect of the new law: the percent of newborns getting NHS.
- Universal Newborn Hearing Screening (UNHS)
- Oregon law mandating UNHS for all babies born in hospitals with more than 200 live births per year: July 1, 2000.
NOTES: The Pregnancy Risk Assessment Monitoring System (PRAMS) is a stratified random survey of postpartum women with oversampling of women from racial and ethnic minorities and whose babies had low birth weight. The survey is sent when the baby is 2-6 months old. Oregon PRAMS began in 1998. Starting in May 2000, a question was inserted into Oregon PRAMS asking whether the baby got a hearing screening test before newborn discharge. The question was "In the hospital or birthing center, after your new baby was born, did your baby get a hearing screening?" We excluded from the analysis answers that the baby got the screening after going home or if the mother did not know. We did not ask the question in telephone follow-up. This analysis was done using unweighted data. We excluded all Missings and Unknowns from this analysis.
- PRAMS: Pregnancy Risk Assessment Monitoring System
- Began asking about NHS: May 2000
- Question in mailed surveys: "In the hospital or birthing center, after your baby was born, did your baby get a hearing screening?"
- Unweighted data
NOTES: The question was answered by 564 women. The responses were categorized according to child's date of birth. The arrow indicates the time when the law went into effect, July 2000. Numbers went from less than 60% in March and April to 80% in June to over 90% in every month after the law went into effect.
NOTES: The MCHB Block Grant asks what percent of babies are getting NHS. Working with Karl White, we had developed a roster of which hospitals did UNHS in Oregon. We knew that 6 hospitals did NHS on all babies and another 6 did NHS on all NICU babies. That added up to about 30% of newborns up through the autumn of 1999, a few months after the new law was passed. In contrast, using PRAMS, we found that over 60% of newborns were getting NHS in 3/00. There are 2 likely reasons for the difference: 1. The law, passed in the summer of 1999, caused some hospitals to start or expand NHS programs before we started PRAMS surveillance. 2. The Roster Method may have missed hospitals that had been doing some NHS all along. The PRAMS method may be a better way to know how many newborns are getting NHS.
- Roster Method derived from Karl White of Utah State University
- Oregon roster, autumn 1999:
- 12 hospitals with universal NHS
- 2 more hospitals did NHS on NICU babies
- Sum = 37% of newborns getting NHS
- 3/00 PRAMS (baseline): over 60%
NOTES: There are many potential biases in this data. PRAMS selects mothers through versampling (of low birthweight babies and those with racial and ethnic minorities). But we did not ask about NHS in telephone surveys. We made no attempt to handle the results as weighted data. So oversampled minority mothers and those with low birthweight babies are more likely to have been surveyed.I n some ways, this sample we are analyzing is more like a convenience sample than the usual PRAMS sample. We do not know whether babies whose mothers did not know whether their babies were tested are more or less likely to have been tested. Recall bias is a potential problemBut the trend over time is more robust than the actual numbers for any individual month, since the biases are not likely to change over a short time.
- 1. Weighted sampling; unweighted analysis
- 2. Mothers who did not know whether their babies were tested
- 3. Recall bias
- Trend over time is robust
NOTES: We found that a law mandating UNHS by hospitals can dramatically increase the proportion of newborns receiving NHS. The proportion of newborns who received NHS began to dramatically increase in the month before the law took effect. Almost all newborns were getting NHS one month after the law took effect. And this trend continued for as long as we surveyed. We believe that the law helped make UNHS part of the standard of practice in Oregon.
- Law mandating UNHS by hospitals increases NHS.
- NHS increased in the month before the law took effect.
- Almost all newborns were getting NHS after the law took effect.
- UNHS: standard of practice in Oregon.
NOTES: 1. Improved infrastructure like PRAMS creates flexibility and opportunities for creativity in addressing a wide range of public health needs, some unanticipated. PRAMS was an elegant way to estimate the proportion of newborns receiving hearing screening during the period when the new law went into effect. 2. The Maternal and Child Health Bureau asks all states to report on the percent of newborns getting NHS for the Title V Information System. The use of PRAMS for routine annual assessment of this Performance Measure would create a uniform national methodology to increase the reliability of comparisons between states.
- PRAMS: infrastructure permits flexibility
- MCHB Title V Performance Measure on Newborn Hearing Screening
Return to the top of the page.