Norma I. Gavin

Perinatal depression: a systematic review of prevalence and incidence.

OBJECTIVE: We systematically review evidence on the prevalence and incidence of perinatal depression and compare these rates with those of depression in women at nonchildbearing times. DATA SOURCES: We searched MEDLINE, CINAHL, PsycINFO, and Sociofile for English-language articles published from 1980 through March 2004, conducted hand searches of bibliographies, and consulted with experts. METHODS OF STUDY SELECTION: We included cross-sectional, cohort, and case-control studies from developed countries that assessed women for depression during pregnancy or the first year postpartum with a structured clinical interview. TABULATION, INTEGRATION, AND RESULTS: Of the 109 articles reviewed, 28 met our inclusion criteria. For major and minor depression (major depression alone), the combined point prevalence estimates from meta-analyses ranged from 6.5% to 12.9% (1.0-5.6%) at different trimesters of pregnancy and months in the first postpartum year. The combined period prevalence shows that as many as 19.2% (7.1%) of women have a depressive episode (major depressive episode) during the first 3 months postpartum; most of these episodes have onset following delivery. All estimates have wide 95% confidence intervals, showing significant uncertainty in their true levels. No conclusions could be made regarding the relative incidence of depression among pregnant and postpartum women compared with women at nonchildbearing times. CONCLUSION: To better delineate periods of peak prevalence and incidence for perinatal depression and identify high risk subpopulations, we need studies with larger and more representative samples.

Racial and ethnic disparities in the use of pregnancy-related health care among Medicaid pregnant women.

Objective: To assess the extent to which racial and ethnic disparities exist in the use of prenatal services among Medicaid pregnant women. Methods: Medicaid claims data for Florida, Georgia, New Jersey, and Texas, with linked birth certificate data for Georgia and Texas, were used to investigate the use of selected prenatal services, including the initiation and adequacy of prenatal care visits; prescriptions for multiple vitamins and iron supplements; and claims for complete blood cell counts, blood type and RH status, hepatitis B surface antigen, ultrasound, maternal serum alphafetoprotein, drug screening, and HIV tests. We computed raw and adjusted odds ratios of having the health service of interest during pregnancy for women in three minority groups: black non-Hispanics, Hispanics, and Asian/Pacific Islanders. Results: We found racial and ethnic disparities in the use of every health service investigated. Compared with white non-Hispanics, minority women were less likely to receive services that the woman initiates, discretionary services, and services potentially requiring specialized follow-up care, whereas they were more likely to receive screening tests for diseases related to high-risk behaviors. Disparities were generally larger, more consistent across states, and less likely to be explained by other factors among black non-Hispanics than among either Hispanics or Asian/Pacific Islanders. Conclusions: Even among women who are provided equal financial access to health care services, unexplained racial and ethnic disparities persist in the initiation and use of both routine and specialized prenatal care services.

Welfare reform, insurance coverage pre-pregnancy, and timely enrollment: An eight-state study

Implementation of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) broke the automatic linkage between Medicaid eligibility/enrollment and welfare cash assistance for women eligible at welfare income levels. This study used data from the Pregnancy Risk Assessment Monitoring System (PRAMS) for the period 1996–1999 to examine insurance coverage of these and other pregnant women pre- and post-PRWORA. Controlling for individual characteristics and economic growth, the relative odds of having private insurance did not change while the odds of being Medicaid enrolled versus uninsured pre-pregnancy declined for welfare-eligible women post-PRWORA. The absolute effect was a decline of 7.9 percentage points in the probability of welfare-eligible women being insured. While these results apply to the early years of welfare reform, it is still likely that states can improve Medicaid outreach and enrollment of women eligible prior to pregnancy.

Determinants of hormone replacement therapy duration among postmenopausal women with intact uteri

ObjectiveTo investigate factors associated with hormone replacement therapy (HRT) duration among postmenopausal women with intact uteri. DesignA Cox proportional hazard model on time to HRT discontinuation is estimated for 2,632 postmenopausal HRT users with intact uteri who began a new episode of treatment between January 1990 and December 1994 in Saskatchewan, Canada. ResultsMajor contraindicating medical events were highly associated with HRT discontinuation among postmenopausal women. Women who were diagnosed with uterine cancer while taking HRT were almost four times as likely to discontinue HRT, and women who were diagnosed with breast cancer while taking HRT were nearly five times as likely to discontinue HRT. Other statistically significant factors associated with the duration of HRT episodes include administration mode and the ability to try different types and strengths of HRT. Women initiating HRT with a transdermal patch were 50% more likely to discontinue it. Women who were willing and able to experiment with different HRT reduced their likelihood of discontinuing by one-half to three-fourths. ConclusionsAlthough some of the factors associated with the hazard of HRT discontinuation among postmenopausal women who are taking the treatment for preventive benefits are immutable, clinicians may influence HRT continuation rates through initial drug choice or modifications in drug type or regimen over the course of therapy.

Asthma-related medication use among children in the United States

BACKGROUND Asthma is one of the most common chronic conditions in children and has a major impact on health care use and quality of life. The Best Pharmaceuticals for Children Act mandates the federal government to sponsor pediatric studies of drugs approved for use in the United States but lacking evaluation in the pediatric population and lacking interest of commercial sponsors. As input into the drug selection and prioritization process, information is needed on the percentage of children who receive asthma-related medications. OBJECTIVE To estimate the percentage of children who receive asthma-related medications. METHODS Retrospective analysis of outpatient medical and drug claims from members of commercial health care insurance plans enrolled any time from January 1, 2004, through December 31, 2005. The study population included 4,259,103 children throughout the United States aged birth through 17 years. RESULTS Fifteen percent of all children were dispensed an asthma-related medication. Among 218,943 children with an asthma diagnosis, 188,286 (86%) had a dispensed asthma-related medication at any time during the 2-year study period. Among children without any asthma diagnoses, 398,880 (10%) had a dispensed medication. Fifty-nine percent of children with an asthma diagnosis were dispensed an anti-inflammatory medication within 90 days after a claim with a diagnosis of asthma. CONCLUSIONS Asthma-related medications are dispensed to a large percentage of the pediatric population, including many who do not have claims with asthma diagnoses listed. Data on the pharmacokinetics and safety of these drugs in children are largely unknown and difficult to obtain. Clinical studies that use new tools and approaches are needed to resolve this information gap.

Racial Disparities in Medicaid Enrollment and Prenatal Care Initiation Among Pregnant Teens in Florida : Comparisons Between 1995 and 2001

Background:Teens and racial and ethnic minority women are less likely to initiate prenatal care (PNC) in the first trimester of pregnancy than their counterparts. Objective:This study examines the impact of Medicaid program changes in the late 1990s on the timing of Medicaid enrollment and PNC initiation among pregnant teens by race and ethnicity. Research Design:Using Medicaid enrollment and claims data and a difference-in-differences method, we examine how the patterns of prepregnancy Medicaid enrollment, PNC initiation, and racial and ethnic disparities in PNC changed over time after controlling for person- and county-level characteristics. Subjects:We included 14,089 teens in Florida with a Medicaid-covered delivery in fiscal years 1995 and 2001. Measures:Prepregnancy enrollment was defined as enrollment 9 or more months before delivery; late or no PNC was defined as initiation of PNC within 3 months of delivery or not at all. Results:For teens enrolled in traditional welfare-related categories, the proportion with prepregnancy Medicaid enrollment increased and the proportion with late or no PNC declined from 1995 to 2001. Teens enrolled under the Omnibus Budget Reconciliation Act (OBRA) expansion category in 2001 were less likely than welfare-related teen enrollees to have prepregnancy coverage but were more likely to initiate PNC early. Racial disparities were found in PNC initiation among the 1995 welfare-related group and the 2001 expansion group but were eliminated or greatly reduced among the 2001 welfare-related group. Conclusions:Providing public insurance coverage improves access to care but is not sufficient to meet Healthy People 2010 goals or eliminate racial and ethnic disparities in PNC initiation.

The State Children's Health Insurance Program (SCHIP) and prepregnancy coverage of teenage mothers.

Background:The 1997 State Childrens Health Insurance Program (SCHIP) program allowed states to expand Medicaid to uninsured children through age 18 in families under 200% of the federal poverty level. Prepregnancy insurance coverage of adolescents may help reduce unintended pregnancies, address other medical issues, and allow for early and adequate prenatal care for those carrying to term. Objectives:We tested the effects of SCHIP implementation on insurance coverage for teenage mothers and investigated whether these effects varied by type of state SCHIP program—Medicaid expansion, stand-alone program, or some combination of these. Research Design:We used Pregnancy Risk Assessment Monitoring System data from 1996 through 2000 and difference-in-differences analysis to analyze coverage changes for teenage mothers (age <20) relative to those for mothers aged 20–24 years old, a group whose Medicaid eligibility was not affected by SCHIP policies. Population Studied:Our raw sample of teenage and older mothers in Alaska, Oklahoma, South Carolina, Florida, Maine, New York, and West Virginia equaled 23,171 (811,638 weighted). Results:SCHIP implementation was associated with an almost 10 percentage point increase in prepregnancy coverage among teens under age 17. Although there were increases in both public and private coverage only the latter was statistically significant. The only statistically significant increase in Medicaid coverage, equal to almost 16 percentage points, was among 18-year-olds in states with Medicaid expansion programs. Conclusions:The temporary extension of SCHIP allows time to consider how to maintain the programs potentially positive effect on the reproductive health of adolescents.

Acute tocolysis for treatment of preterm labor: Review and meta-analysis

Abstract Objective: Preterm labor (PTL) is a prelude to preterm birth and to infant morbidity and mortality. We sought to review the literature on tocolysis for PTL and to do a meta-analysis of appropriate studies. Methods: We worked as part of the AHCPR Evidence Report on the Management of Preterm Labor. With ACOG and an advisory group, we developed search criteria. We did an exhaustive search, including the gray literature, for articles in English, French, or German from 1966 to 1998. We found 256 studies with the headings “PTL” and “tocolytic agents,” including individual tocolytics. We selected 55 studies with 40 or more subjects, tocolysis for first-line treatment, outcome measurement of gestational age at birth, and without membrane rupture at presentation. Reviewers and abstractors were masked to authorship and journal. Elements abstracted included definition of PTL, study design, population characteristics, masking, randomization, and analytic approach. Outcomes included delay in delivery, gestational age at birth, and proportion of births by prematurity (eg, less than 37 weeks of gestation and more than 30 weeks of gestation). Results: We will summarize the effectiveness of tocolytics for PTL by category of agent: β-mimetics, calcium channel blockers, magnesium, and nonsteroidal antiinflammatory drugs. For 20 randomized trials with sufficient detail, we will report meta-analysis results by category. Conclusions: We will present the conclusions of the team.

Antibiotics for treatment of preterm labor: review and meta-analysis

Abstract Objective: Inflammatory processes and infection contribute to the risk of preterm birth. We sought to review the literature on antibiotics for the treatment of preterm labor (PTL) and to do a meta-analysis of results from randomized trials. Methods: We worked as part of the AHCPR Evidence Report on the Management of Preterm Labor. With ACOG and an advisory group, we developed search criteria. We did an exhaustive search, including the gray literature, for articles in English, French, or German, from 1966 to 1998. We searched for randomized trials and observational studies on antibiotics for preterm birth among women with PTL. We found 107 articles; 15 had 40 or more subjects with outcome measurement of gestational age at birth and without rupture of membranes at enrollment. Abstractors were masked to authorship and journal. Elements abstracted included definition of PTL, study design, treatment regimen, population characteristics, masking, randomization, and analytic approach. Outcomes included delay in delivery, gestational age at birth, and proportion of births by prematurity (eg, less than 37 weeks of gestation and more than 30 weeks of gestation). Results: We will summarize the literature on antibiotics for treatment of PTL and reflect on the divergent regimens and findings in major studies. We will report the results of meta-analyses, including 13 randomized trials that examine antibiotic treatment and prolongation of pregnancy in days, gestational age at birth, and birth weight. Conclusion: We will present the conclusions of the team.

Explaining Racial Differences in Prenatal Care Initiation and Syphilis Screening Among Medicaid-covered Pregnant Women

Sexually transmitted diseases and their outcomes disproportionately affect non-Hispanic Blacks who also receive later prenatal care. We used a sample of low-income pregnant women insured by Medicaid to assess racial disparities in the receipt of first trimester prenatal care and any as well as early (by 2nd trimester) syphilis screening. We used an older but unique file of linked 1995 Georgia Medicaid claims and Pregnancy Risk Assessment Monitoring System (PRAMS) births (n = 1,096) to test the relative explanatory power of factors contained in administrative versus survey data. Using administrative data, we found non-Hispanic Blacks were less likely than non-Hispanic Whites to receive first trimester care but more likely to be screened. Adding in PRAMS survey data eliminated these differences. Having an outpatient department as usual source of care was a key factor. This may reflect unmeasured characteristics of minorities and their neighborhoods or differences in screening practices across provider settings.