Trends in Postpartum Contraceptive Use in 20 U.S. States and Jurisdictions: The Pregnancy Risk Assessment Monitoring System, 2015-2018.

INTRODUCTION: In the last decade, state and national programs and policies aimed to increase access to postpartum contraception; however, recent data on population-based estimates of postpartum contraception is limited. METHODS: Using Pregnancy Risk Assessment Monitoring System data from 20 sites, we conducted multivariable-adjusted weighted multinomial regression to assess variation in method use by insurance status and geographic setting (urban/rural) among people with a recent live birth in 2018. We analyzed trends in contraceptive method use from 2015 to 2018 overall and within subgroups using weighted multinomial logistic regression. RESULTS: In 2018, those without insurance had lower odds of using permanent methods (adjusted odds ratio [AOR], 0.72; 95% confidence interval [CI], 0.53-0.98), long-acting reversible contraception (LARC) (AOR, 0.67; 95% CI, 0.51-0.89), and short-acting reversible contraception (SARC) (AOR, 0.61; 95% CI, 0.47-0.81) than those with private insurance. There were no significant differences in these method categories between public and private insurance. Rural respondents had greater odds than urban respondents of using all method categories: permanent (AOR, 2.15; 95% CI, 1.67-2.77), LARC (AOR, 1.31; 95% CI, 1.04-1.65), SARC (AOR, 1.42; 95% CI, 1.15-1.76), and less effective methods (AOR, 1.38; 95% CI, 1.11-1.72). From 2015 to 2018, there was an increase in LARC use (odds ratio [OR], 1.03; 95% CI, 1.01-1.05) and use of no method (OR, 1.05; 95% CI, 1.02-1.07) and a decrease in SARC use (OR, 0.97; 95% CI, 0.95-0.99). LARC use increased among those with private insurance (OR, 1.05; 95% CI, 1.02-1.08) and in urban settings (OR, 1.04; 95% CI, 1.02-1.07), but not in other subgroups. CONCLUSIONS: We found that those without insurance had lower odds of using effective contraception and that LARC use increased among those who had private insurance and lived in urban areas. Strategies to increase access to contraception, including increasing insurance coverage and investigating whether effectiveness of existing initiatives varies by geographic setting, may increase postpartum contraceptive use and address these differences.

Variation in Postpartum Use of Most and Moderately Effective Contraceptive Methods Among Louisiana Women.

BACKGROUND: Women experiencing unintended and short-interval pregnancies are at increased risk for adverse outcomes. Nationally, researchers report disparities in women's use of effective contraceptive methods based on demographic, cultural, financial and system-level factors. Despite 58% of Louisiana births being unplanned, researchers have not reported on these relationships in Louisiana. METHODS: We used Louisiana Pregnancy Risk Assessment Monitoring System data from 2015 to 2018. Among postpartum women who were not abstinent, pregnant, or trying to become pregnant, we estimated use of five categories of effective contraception versus no effective method. We used multivariable multinomial logistic regression to investigate the association between effective contraceptive use and race/ethnicity, postpartum insurance and education. RESULTS: Among Louisiana postpartum women who were not abstinent, pregnant, or trying to become pregnant, 35.4% were not using effective contraception. Women with public insurance had greater odds of using long-acting reversible contraception than women with private insurance (adjusted odds ratio [AOR] 1.55; 95% confidence interval [CI] 1.11-2.16). Compared to women with a bachelor's or higher, women with less than high school (AOR 2.09; CI 1.22-3.56), high school (AOR 3.11; CI 2.01-4.82) or some college education (AOR 2.48; CI 1.64-3.75) had greater odds of using permanent contraception. Black (AOR 3.83; CI 2.66-5.54) and Hispanic (AOR 3.85; CI 2.09-7.11) women, women with less than high school (AOR 6.79; CI 2.72-16.94), high school (AOR 7.26; CI 3.06-17.21) and some college (AOR 7.22; CI 3.14-16.60), and women with public insurance (AOR 1.91; CI 1.28-2.87) had greater odds of using injectable contraception. DISCUSSION: Results showed variation in effective contraceptive method use by race/ethnicity, insurance and education. These findings highlight the need for state-level research into the individual, provider, and policy-level factors that influence women's contraceptive choices.

Racial Inequities in Preventable Pregnancy-Related Deaths in Louisiana, 2011-2016.

OBJECTIVE: To examine preventable pregnancy-related deaths in Louisiana by race and ethnicity and maternal level of care to inform quality improvement efforts. METHODS: We conducted a retrospective observational descriptive analysis of Louisiana Pregnancy-Associated Mortality Review data of 47 confirmed pregnancy-related deaths occurring from 2011 to 2016. The review team determined cause of death, preventability, and contributing factors. We compared preventability by race-ethnicity and maternal level of care of the facility where death occurred (from level I: basic care to level IV: regional perinatal health center) using odds ratios (ORs) and 95% CIs. RESULTS: The rate of pregnancy-related death among non-Hispanic black women (22.7/100,000 births, 95% CI 15.5-32.1, n=32/140,785) was 4.1 times the rate among non-Hispanic white women (5.6/100,000, 95% CI 2.8-10.0, n=11/197,630). Hemorrhage (n=8/47, 17%) and cardiomyopathy (n=8/47, 17%) were the most common causes of pregnancy-related death. Among non-Hispanic black women who experienced pregnancy-related death, 59% [n=19] of deaths were deemed potentially preventable, compared with 9% (n=1) among non-Hispanic white women (OR 14.6, 95% CI 1.7-128.4). Of 47 confirmed pregnancy-related deaths, 58% (n=27) occurred at level III or IV birth facilities. Compared with those at level I or II birth facilities (n=2/4, 50%), pregnancy-related deaths occurring at level III or IV birth facilities (n=14/27, 52%) were not less likely to be categorized as preventable (OR 2.0, 95% CI 0.5-8.0). CONCLUSION: Compared with non-Hispanic white women, pregnancy-related deaths that occurred among non-Hispanic black women in Louisiana from 2011 to 2016 were more likely to be preventable. The proportion of deaths that were preventable was similar between lower and higher level birth facilities. Hospital-based quality improvement efforts focused on addressing hemorrhage, hypertension, and associated racial inequities may prevent pregnancy-related deaths in Louisiana.

Pregnant? Validity of the pregnancy checkbox on death certificates in four states, and characteristics associated with pregnancy checkbox errors.

BACKGROUND: Maternal mortality rates in the United States appear to be increasing. One potential reason may be increased identification of maternal deaths after the addition of a pregnancy checkbox to the death certificate. In 2016, 4 state health departments (Georgia, Louisiana, Michigan, and Ohio) implemented a pregnancy checkbox quality assurance pilot, with technical assistance provided by the Centers for Disease Control and Prevention. The pilot aimed to improve accuracy of the pregnancy checkbox on death certificates and resultant state maternal mortality estimates. OBJECTIVE: To estimate the validity of the pregnancy checkbox on the death certificate, and to describe characteristics associated with errors using 2016 data from a 4-state quality assurance pilot. MATERIALS AND METHODS: Potential pregnancy-associated deaths were identified by linking death certificates with birth or fetal death certificates from within 1 year preceding death or by pregnancy checkbox status. Death certificates that indicated that the decedent was pregnant within 1 year of death via the pregnancy checkbox, but that did not link to a birth or fetal death certificate, were referred for active follow-up to confirm pregnancy status by either death certifier confirmation or medical record review. Descriptive statistics and 95% confidence intervals were used to examine the distributions of demographic characteristics by pregnancy confirmation category (confirmed pregnant, confirmed not pregnant, and unable to confirm). We compared the proportion confirmed pregnant and confirmed not pregnant within age, race/ethnicity, pregnancy checkbox category, and certifier type categories using a Wald test of proportions. Binomial and Poisson regression models were used to estimate prevalence ratios for having an incorrect pregnancy checkbox (false positive, false negative) by age group, race/ethnicity, pregnancy checkbox category, and certifier type. RESULTS: Among 467 potential pregnancy-associated deaths, 335 (72%) were confirmed pregnant via linkage to a birth or fetal death certificate, certifier confirmation, or review of medical records. A total of 97 women (21%) were confirmed not pregnant (false positives) and 35 (7%) were unable to be confirmed. Women confirmed pregnant were significantly younger than women confirmed not pregnant (P < .001). Deaths certified by coroners and medical examiners were more likely to be confirmed pregnant than confirmed not pregnant (P = .04). The association between decedent age category and false-positive status followed a dose-response relationship (P < .001), with increasing prevalence ratios for each increase in age category. Death certificates of non-Hispanic black women were more likely to be false positive, compared with non-Hispanic white women (prevalence ratio, 1.41; 95% confidence interval, 1.01, 1.96). The sensitivity of the pregnancy checkbox among these 4 states in 2016 was 62% and the positive predictive value was 68%. CONCLUSION: We provide a multi-state analysis of the validity of the pregnancy checkbox and highlight a need for more accurate reporting of pregnancy status on death certificates. States and other jurisdictions may increase the accuracy of their data used to calculate maternal mortality rates by implementing quality assurance processes.

Policy change is not enough: engaging provider champions on immediate postpartum contraception.

Rates of short-interval pregnancies that result in unintended pregnancies remain high in the United States and contribute to adverse reproductive health outcomes. Long-acting reversible contraception methods have annual failure rates of <1%, compared with 9% for oral contraceptive pills, and are an effective strategy to reduce unintended pregnancies. To increase access to long-acting reversible contraception in the immediate postpartum period, several State Medicaid programs, which include those in Iowa and Louisiana, recently established reimbursement policies to remove the barriers to reimbursement of immediate postpartum long-acting reversible contraception insertion. We used a mixed-methods approach to analyze 2013-2015 linked Medicaid and vital records data from both Iowa and Louisiana and to describe trends in immediate postpartum long-acting reversible contraception provision 1 year before and after the Medicaid reimbursement policy change. We also used data from key informant interviews with state program staff to understand how provider champions affected policy uptake. We found that the monthly average for the number of insertions in Iowa increased from 4.6 per month before the policy to 6.6 per month after the policy; in Louisiana, the average number of insertions increased from 2.6 per month before the policy to 45.2 per month. In both states, the majority of insertions occurred at 1 academic/teaching hospital. In Louisiana, the additional increase may be due to the engagement of a provider champion who worked at both the state and facility level. Recruiting, training, engaging, and supporting provider champions, as facilitators, with influence at state and facility levels, is an important component of a multipart strategy for increasing successful implementation of state-level Medicaid payment reform policies that allow reimbursement for immediate postpartum long-acting reversible contraception insertions.

Louisiana implementation of the National Fetal and Infant Mortality Review (NFIMR) program model: successes and opportunities.

Common features of successful, local-level, Fetal Infant Mortality Review (FIMR) Programs are identified by the National Fetal and Infant Mortality Review (NFIMR) Program, including medical records abstraction and home interviews, case reviews by a case review team (CRT), and community systems action recommendations implemented by a community action team (CAT). This paper presents Louisiana's FIMR program, an adaptation of NFIMR recommendations. In 2001, the Louisiana Maternal and Child Health Program began a statewide FIMR Network (LaFIMR) based on the NFIMR model. Geographic areas of focus, case identification, staffing, data collection methods, and CRT and CAT membership and activities include modifications of the NFIMR recommendations unique to LaFIMR implementation. Adaptations made to the NFIMR model were advantageous to LaFIMR's success. Specifically, LaFIMR geographic areas of interest cover multiple natural communities. Compared with independent FIMR programs elsewhere, LaFIMR represents a Title V Program-based coordinated network of regional LaFIMR teams offering opportunities for expanded partnerships. Primary sources for LaFIMR case identification include obituaries and hospital logs, with secondary identification available through vital records. Improvements in vital records data systems are expected to enhance future LaFIMR case identification. LaFIMR-identified records that are linked with vital event certificates provide enhanced contextual findings for reviews and support continuous quality improvement processes. These differences in the LaFIMR implementation reinforce the NFIMR-supported uniqueness of FIMR programs across the United States, and may encourage other FIMR programs to consider how adaptations to NFIMR recommendations could benefit their programs.

Utility of childhood glucose homeostasis variables in predicting adult diabetes and related cardiometabolic risk factors: the Bogalusa Heart Study.

OBJECTIVE This study examines the usefulness of childhood glucose homeostasis variables (glucose, insulin, and insulin resistance index [homeostasis model assessment of insulin resistance {HOMA-IR}]) in predicting pre-diabetes and type 2 diabetes and related cardiometabolic risk factors in adulthood. RESEARCH DESIGN AND METHODS This retrospective cohort study consisted of normoglycemic (n = 1,058), pre-diabetic (n = 37), and type 2 diabetic (n = 25) adults aged 19-39 years who were followed on average for 17 years since childhood. RESULTS At least 50% of the individuals who ranked highest (top quintile) in childhood for glucose homeostasis variables maintained their high rank by being above the 60th percentile in adulthood. In a multivariate model, the best predictors of adulthood glucose homeostasis variables were the change in BMI Z score from childhood to adulthood and childhood BMI Z score, followed by the corresponding childhood levels of glucose, insulin, and HOMA-IR. Further, children in the top decile versus the rest for insulin and HOMA-IR were 2.85 and 2.55 times, respectively, more likely to develop pre-diabetes; children in the top decile versus the rest for glucose, insulin, and HOMA-IR were 3.28, 5.54, and 5.84 times, respectively, more likely to develop diabetes, independent of change in BMI Z score, baseline BMI Z score, and total-to-HDL cholesterol ratio. In addition, children with adverse levels (top quintile versus the rest) of glucose homeostasis variables displayed significantly higher prevalences of, among others, hyperglycemia, hypertriglyceridemia, and metabolic syndrome. CONCLUSIONS Adverse levels of glucose homeostasis variables in childhood not only persist into adulthood but also predict adult pre-diabetes and type 2 diabetes and relate to cardiometabolic risk factors.

Adherence to HAART among HIV-infected persons in rural Louisiana.

The purpose of this study was to examine factors associated with nonadherence to highly active antiretroviral therapy (HAART) in patients seen in HIV clinics throughout nonurban Louisiana. A convenience sample of 273 patients from 8 areas in nonurban Louisiana were interviewed to obtain demographic, clinical and adherence information. Associations between demographic, clinical, and behavioral factors and nonadherence were examined. Ideally, non-adherence should not exceed 5% in patients for whom HAART was prescribed. Mean age was 38.6 years (range, 19-66), 29.3% were female, 60.1% were African American, 34.4% reported nonadherence to their HAART medication (defined as the subject's self-report of missing any doses of HAART medication in the prior week). In the prior month, participants reported the following behaviors: binge drinking (12.8%), problem drinking (12.8%), and illicit drug use (16.5%). Depression was found in 49.8% of the respondents. In logistic regression analysis, problem drinking odds ratio [OR] (95% confidence interval [CI]): 3.92 (1.69,9.09) was found to be associated with nonadherence. Demographic and behavioral factors such as illicit drug use and depression were not associated with nonadherence on multivariable analysis. Problem drinking was associated with lack of adherence to HAART over the past week. Interventions to treat problem drinking are needed and may improve adherence to medication for HIV-infected persons living in rural, town, and small-city areas.

Framingham risk score is related to carotid artery intima-media thickness in both white and black young adults: the Bogalusa Heart Study.

The Framingham risk score (FRS), developed in a white cohort aged 30-74 years, is increasingly used in the early risk identification for coronary artery disease (CAD). This study examines the relationship between FRS and carotid artery intima-media thickness (IMT), a surrogate marker of coronary atherosclerosis, in black and white individuals aged 20-37 years. Five hundred seventeen young adults (aged 20-37 years; 71% white, 39% male) enrolled in the Bogalusa Heart Study had carotid artery ultrasonography. Age, gender, systolic blood pressure, total cholesterol to HDL cholesterol ratio, cigarette smoking habit, type 2 diabetes, and left ventricular hypertrophy (LVH) were used to calculate FRS. Results indicated a significant, positive linear relationship between tertiles of FRS and IMT of the common, bulb, and internal carotid segments in blacks and whites alike. In a multivariate analysis including FRS, race, BMI, parental history of CAD, stroke, type 2 diabetes, or hypertension, logtriglycerides, loginsulin, alcohol consumption (ml/week), and regular physical activity, the FRS was independently associated with all three carotid segments. Further, the FRS as a main predictor variable explained relatively more of the variance in the IMT of the carotid bulb (9%) than in the common (5%) or internal (3%) carotid segments. These results support the use of FRS in both white and black young adults and underscore the importance of prevention and control of multiple risk factors in youth.

Impact of multiple coronary risk factors on the intima-media thickness of different segments of carotid artery in healthy young adults (The Bogalusa Heart Study).

Although risk factors for coronary artery disease are also associated with increased carotid artery intima-media thickness (IMT) as measured by B-mode ultrasonography in middle-aged and older persons, information on the impact of multiple risk factors on the IMT of different segments of the carotid artery in young adults is limited. This relation was examined in a sample of 518 black and white subjects (mean age 32 years; 71% white, 39% male) enrolled in the Bogalusa Heart Study. IMT was thicker and more skewed in the bulb compared with other carotid segments. Race differences (blacks more than whites) were noted for the common carotid (p <0.001) and carotid bulb (bifurcation) IMT (women only, p <0.001). Men had a greater IMT in the common carotid (p <0.05), internal carotid (p <0.05), and carotid bulb (whites only, p <0.001). In a multivariate analysis, systolic blood pressure, race, age, low-density lipoprotein (LDL) cholesterol, and high-density lipoprotein (HDL) cholesterol weree entered into a model in that order and accounted for the 16.7% variance in the common carotid IMT; age, systolic blood pressure, HDL cholesterol, LDL cholesterol, race, and insulin levels explained the 19.4% variance in the carotid bulb IMT. Gender and body mass index (BMI) accounted for the 4.7% variance in the internal carotid IMT. Increases in IMT with increasing number of risk factors (cigarette smoking, higher total cholesterol to HDL cholesterol ratio, higher systolic blood pressure, greater waist circumference, and higher insulin level) were noted for the common carotid and carotid bulb segments (p for trend <0.001 for both). The observed deleterious trend of increasing IMT at different carotid segments with increasing number of risk factors in free-living, asymptomatic young subjects underscores the importance of profiling multiple risk factors early in life. Ultrasonography of carotid arteries, especially at the bifurcation, may be helpful along with measurements of risk factors for evaluation of asymptomatic atherosclerotic disease.