National Center For Health Statistics Research Articles

Prescribing Narcotics for Pain: Reconsider the Fifth Vital Sign.

Prescribing Narcotics for Pain: Reconsider the Fifth Vital Sign.

Abstract P022: The Geographic Variations In Hypertension Medication Adherence

Objective: Hypertension (HTN) affects 47% of adults in the United States (US). Medication adherence is difficult to achieve due to its benign course. This study aims to characterize the difference in HTN medication adherence between major regions of the US and the incongruity between urban and rural areas. Methods: This is a cross-sectional study consisting of 11,494 persons, representing over 71 million adults in the US, with a diagnosis of HTN. Data was obtained from the National Health Interview Survey in 2020. Variables include HTN medication adherence, access to regular care, household region, and urban-rural classification as defined by the National Center for Health Statistics. Chi-square test and logistic regression was used to compare the study groups. Results: HTN medication adherence was 75.6% in large urban (LU), 81.5% in large fringe urban (FU), 81.1% in medium urban (MU), and 83.3% in nonmetro (NM) areas [p < 0.001]. Using urban as the baseline comparison for a logistic regression model, a statistically significant difference was observed between LU and each individual group [p < 0.001] . This association remained significant when adjusting for covariates of BMI, marital status, insurance status, race, and age [p < 0.001]. Also, patients in LU were least likely to have a regular healthcare provider at 94.7% followed by 95.9% in FU, and 97.0% in MU and NM [p < 0.01]. Adherence across U.S. regions, defined as Northeast, South, Midwest, and West was statistically significant [p <0.01]. On logistic regression, West had a significantly lower adherence rate at 76.6% when compared to other regions [OR 0.69; 95% CI 0.56, 0.85; p < 0.05]. The odds increased further when adjusting for covariates of BMI, marital status, insurance status, race, and age [AOR 0.74; 95% CI 0.58, 0.95; p < 0.05]. Conclusion: Urban areas are associated with lower rates of HTN medication adherence with increasing adherence in smaller communities. Additionally, Western US is correlated with a lower adherence rate compared to other regions. Further investigation into sources for this discrepancy is warranted, including potential differences in public health campaigns, prescribing practices, and follow up care. This knowledge could assist in the development of targeted solutions.

Lesion-Specific Congenital Heart Disease Mortality Trends in Children: 1999 to 2017.

Congenital heart disease (CHD) is a leading cause of premature death in infants and children. Currently limited data are available regarding lesion specific mortality over time. Our study aimed to describe pediatric mortality trends by CHD lesion in the United States. We conducted a 19 year analysis (1999 to 2017) of publicly available, deidentified multiple cause of death data compiled and produced by the National Center for Health Statistics. Analysis was stratified by CHD diagnosis and age using 3 age categories (infants, 1 to 4 years, and 5 to 17 years). Temporal trends of CHD mortality and the effect of contributing risk factors were analyzed by using joinpoint regression. Mortality was highest for in infants for all CHD lesions, in particular for total anomalous pulmonary venous return. Significant declines in infant CHD mortality occurred for most other lesions. Contributing risk factors, including prematurity, extracardiac birth defects, and genetic conditions, occurred in 19% of infant CHD deaths and demonstrated worse mortality trends in the majority of lesions. Mortality rates remained highest for single ventricle lesions in all ages, with an infant mortality rate plateau in the later half of the study and progressive increasing mortality rates for children 5 to 17 years. CHD mortality is decreasing for most lesions. Because of the heterogenicity of CHD lesions, there is expected variability in mortality trends by lesion and age group. Single ventricle lesions continue to contribute most heavily to premature death because of CHD demonstrated by significant increases in mortality rate for children aged 5 to 17 years.

Relationship between Infant Mortality, Household Income and Race in Maryland Counties: A Short Communication

We used the data collected by the Centers for Disease Control and Prevention, National Center for Health Statistics to analyze infant crude death rate data in Maryland counties between 1999 and 2020 and compared this data to household income data collected by the US Census in the same Maryland counties. We found a significant inverse relationship and correlation between infant deaths and household income in Maryland counties. It is likely that low income plays a role in low birth weight in these infants and therefore may be related to the major cause of infant mortality.

Lost on the frontline, and lost in the data: COVID-19 deaths among Filipinx healthcare workers in the United States.

BackgroundFilipinx Americans working in healthcare are at risk for COVID-19 death but lack consistent mortality data on healthcare worker deaths. The lack of disaggregated data for Asian subgroups proliferates anti-Asian structural racism as the needs of high-risk groups are systematically undetected to merit a proper public health response. We work around this aggregated data problem by examining how the overrepresentation of Filipinxs in healthcare contributes to COVID-19 mortality among Asian American populations.MethodsTo overcome the lack of COVID-19 mortality data among Filipinx American healthcare workers, we merged data from several sources: Kanlungan website (the only known public-facing source of systematically reported mortality data on Filipinx healthcare workers nationally and globally), National Center for Health Statistics, and 2014–2018 American Community Survey. We examined county-level associations using t-tests, scatterplots, and linear regression.FindingsA higher percentage of Filipinxs among Asian Americans was correlated with a higher percentage of COVID-19 decedents who are Asian Americans (r = 0.24, p = 0.01). The percentage of Filipinx in healthcare remained a strong predictor of COVID-19 deaths among Asian Americans even after adjusting for age, poverty, and population density (coef = 1.0, p < 0.001). For every 1% increase in Filipinx among the healthcare workforce, the percentage of Asian American COVID-19 decedents increased by 1%.InterpretationOur study shows that the overrepresentation of Filipinxs in healthcare contributes to COVID-19 mortality disparities among Asian Americans. Our findings advocate for systems change by practicing anti-racist data agendas that collect and report on Asian subgroups for effective real-time targeted approaches against health inequities.

Income Inequality and Pregnancy-Associated Homicide in the US: A Longitudinal, State-Level Analysis.

Pregnancy-associated homicide remains an understudied yet critical issue. Using restricted use mortality files provided by the National Center for Health Statistics and the National Violent Death Reporting System, annual state-level pregnancy-associated homicide ratios were estimated as the count of deaths divided by the number of live births. The exposure, the state Gini index, was categorized into tertiles to compare states by levels of income inequality. In the final adjusted longitudinal linear model, those who experienced the greatest amount of income inequality had a significant 1.28 per 100,000 homicide rate when compared to the lowest income inequality tertile.

Saving light, losing lives: How daylight saving time impacts deaths from suicide and substance abuse.

This paper estimates the impact of Daylight Saving Time (DST) on deaths from suicide and substance abuse in the United States. Using Multiple Cause-of-Death Mortality Data from the National Vital Statistics System of the National Center for Health Statistics from 1979 to 1988, the effect is identified in two ways: a regression discontinuity design that exploits discrete time changes in the Spring and Fall; and a fixed effects model that uses a policy change and a switching mechanism that introduces random variation to DST's start and end dates. This is one of the first attempts to estimate the impact of DST on deaths due to suicide and substance abuse and the first to use either identification strategy. The results from both methods suggest that the sleep disruptions during the Spring transition cause the suicide rate to rise by 6.25 percent and the death rate from suicide and substance abuse combined to increase by 6.59 percent directly after the time change. There is no evidence for any change in these outcomes during the Fall transition. The contrasting results from Spring to Fall suggest the entire effect can be attributed to disruptions in sleep patterns rather than changes in ambient light exposure.

A Female Career in Research

After a long career at the National Center for Health Statistics, I retired and joined the Stanford Prevention Research Center as an unpaid associate. I was once described by a former US Food and Drug Administration commissioner as "one of the great epidemiologists." The chair of the Harvard nutrition department, speaking on National Public Radio, once described my research as "rubbish." Both may be exaggerations. Here I address some of the events that led to these contrasting descriptions. I also address the extent to which the so-called Matilda effect may have influenced my career. Are women in science on an equal footing with men? The Matilda effect suggests not. Unlike the Matthew effect for scientists, whereby those of higher prestige accrue a disproportionate share of recognition and rewards, the Matilda effect proposes that women scientists are systematically undervalued and underrecognized. I could never get a faculty job and was often treated like an underling. Nonetheless I persevered to publish highly cited research on several high-profile and sometimes controversial topics. Though overt sexism in science and workplaces has diminished over the course of my career, progress toward eliminating unconscious bias has been slower. The Matthew and Matilda effects are still powerful forces that distort incentives and rewards in science.

The association between Toxoplasma infection and mortality: the NHANES epidemiologic follow-up study

BackgroundToxoplasma gondii has been reported to be associated with higher mortality in patients with schizophrenia. This study aimed to explore the relationship between T. gondii infection and 25-year mortality based on data from the Third National Health and Nutrition Examination Survey (NHANES III) database.MethodsCases with serum T. gondii antibody test results were included in this study and the corresponding mortality dataset was obtained from the US National Center for Health Statistics (NCHS). Propensity score matching (PSM) was used to match age and sex between groups. The Cox proportional hazards model was used to evaluate the effect of T. gondii infection on mortality.ResultsA total of 14,181 cases were included in the analysis, of which 3831 (27.0%) were seropositive for T. gondii antibody. The median follow-up time of the whole cohort was 22.5 (interquartile range 16.3, 24.5) years. A total of 5082 deaths were observed in this cohort, a mortality rate of 35.8%. All-cause mortality was significantly higher in the seropositive group than in the seronegative group (50.0% vs 30.6%, P < 0.001). Kaplan–Meier analysis showed a significant difference in the survival time between two groups before and after PSM. Multivariate analysis showed that T. gondii infection was independently associated with higher all-cause mortality after adjusting for potential confounders.ConclusionsToxoplasma gondii infection is associated with higher mortality in general population.Graphical

Racial Differences in Firearms Used and Autopsies Following Gun Deaths Exist

IntroductionDisparities in gun violence across race are well documented. Studying these disparities is essential to reduce preventable gun deaths. This study evaluates the relationship between sociodemographic factors and firearms used in gun deaths. Materials and methodsThis retrospective cohort study of firearm mortalities from 2009 to 2018 used the Mortality Multiple Cause-of-Death Public Use Record from the National Center for Health Statistics. The primary outcome was the type of firearm used and the secondary outcome was autopsy status. Factors of interest include race, ethnicity, gender, marital status, age, education, and place of death. Factors significantly associated with outcomes in univariate analyses were included in separate multivariate logistic regression models for assaults, intentional self-harm, and accidents. ResultsA total of 276,127 firearm deaths from 2009 to 2018 were analyzed. Compared to White victims, Black victims were less likely to die from handguns (accident: odds ratio [OR] = 0.70, P < 0.05; self-harm: OR = 0.84, P < 0.001; assault: OR = 0.58, P < 0.001) and rifles, shotguns, or large firearms (accident: OR = 0.30, P < 0.001; self-harm: OR = 0.37, P < 0.001; assault: OR = 0.28, P < 0.001). Black decedents were more likely to undergo autopsy than White decedents (accident: OR = 2.14, P < 0.001; intentional self-harm: OR = 2.02, P < 0.001; assault: OR = 2.02, P < 0.001). Ethnicity, gender, marital status, age, education, and other racial identities were also associated with firearms used and autopsy rates (P < 0.05). ConclusionsDifferences in firearms used and autopsy rates following gun deaths exist by race, ethnicity, gender, marital status, age, and education. Future studies should investigate the relationship between sociodemographic factors and firearms used and autopsy status following gun deaths.

Deprescribing: What About Vitamins, Minerals, and Other Nutritional Supplements?

Deprescribing: What About Vitamins, Minerals, and Other Nutritional Supplements?

Why did preterm birth rates fall during the COVID-19 pandemic?

Why did preterm birth rates fall during the COVID-19 pandemic?

Association Between Rental Assistance Programs and Hemoglobin A1c Levels Among US Adults

Programs that provide affordable and stable housing, such as federal rental assistance, may be associated with improved mean blood glucose levels and related diabetes outcomes. To assess whether 2 different types of federal rental assistance programs are associated with glycated hemoglobin A1c (HbA1c) levels among middle-aged and older US adults. This cohort study used data from the National Health and Nutrition Examination Survey (NHANES) linked with US Department of Housing and Urban Development records of rental assistance participation. Adults aged 45 years or older who were receiving 2 types of rental assistance (project-based housing or housing vouchers) at the time of the NHANES interview and those who would receive rental assistance within the subsequent 2 years (waitlist group) were included. Data were collected from January 1999 to December 2016 and analyzed in October 2021. Rental assistance participation, including project-based housing (subsidized housing developments including public housing) and housing vouchers (tenant-based subsidies for private market housing). The primary outcome was continuous HbA1c level, a common measure of blood glucose reflecting diabetes control. Linear regression was used to estimate the association between the 2 rental assistance programs and HbA1c level. Logistic regression was used to assess the association between rental assistance programs and HbA1c cut points (prediabetes: 5.7% to ≤6.5%; diabetes: >6.5%; uncontrolled diabetes: ≥9% [to convert to proportion of total Hb, multiply by 0.01]). Analyses used weights created by the National Center for Health Statistics that adjust for linkage eligibility. Among 1050 adults in the study (41.6% aged ≥65 years; 70.1% female), 795 were receiving rental assistance at time of the NHANES interview (450 lived in project-based housing, and 345 had housing vouchers), and 255 received rental assistance within 2 years after the interview. Participants in project-based housing had lower HbA1c levels compared with individuals in the waitlist group (β, -0.290; 95% CI, -0.599 to 0.020), but the difference was not significant. No significant differences in HbA1c levels were found between those receiving housing vouchers and those in the waitlist group (β, 0.051; 95% CI, -0.182 to 0.284). Receiving project-based housing was associated with a reduced likelihood of uncontrolled diabetes (-3.7 percentage points; 95% CI, -7.0 to -0.0 percentage points) compared with being in the waitlist group. In this cohort study of a nationally representative sample of US adults, living in project-based, federally subsidized housing was associated with a reduced likelihood of uncontrolled diabetes. The findings suggest that affordable housing programs may be associated with improved diabetes outcomes.

Joint effect of cognitive function and C-reactive protein on all-cause mortality risk: 1999–2002 NHANES

Examine the joint effect of cognitive function and C-reactive protein (CRP) on all-cause mortality risk in older U.S. adults. Sample included 1335 older adult (≥60 years of age) participants in the 1999-2002 National Health and Nutrition Examination Survey. A four-level variable was created using cognitive function and CRP concentration. Mortality was assessed using National Center for Health Statistics linked death records from the National Death Index. Increased risk of all-cause mortality was revealed in adults with high CRP and low cognitive function and in those with low to average CRP and low cognitive function (P &lt; .0001 for both). Sex-stratified analyses revealed increased all-cause mortality risk in males with low cognitive function, independent of CRP concentration. However, in females, a significant increase in all-cause mortality risk was only observed in those with low to average CRP and low cognitive function. Low cognitive function was associated with increased all-cause mortality risk independent of CRP concentration. However, the joint effect of cognitive function and CRP on all-cause mortality risk differed according to sex.

Association of Optimal Gestational Weight Gain Ranges With Perinatal Outcomes Across Body Mass Index Categories in Twin Pregnancies

The existing gestational weight gain (GWG) recommendations for twin pregnancies are lacking for underweight individuals and are not stratified by obesity class. To identify optimal GWG ranges associated with reduced adverse perinatal outcomes stratified by prepregnancy body mass index (BMI) categories in twin pregnancies. This population-based cohort study of twin pregnancies using data from the National Center for Health Statistics was conducted between January 1, 2014, and December 31, 2018. Statistical analysis was performed from October 24, 2021, to May 7, 2022. The study population comprised 262 604 individuals between 18 and 45 years of age with live-born twins without congenital malformation between 24 and 42 weeks of gestation. Two approaches were used to determine the optimal GWG ranges: a statistics-based approach calculating IQRs of GWG in a low-risk population, and an outcome-based approach identifying GWG thresholds below or above which an adverse perinatal outcome increased. Gestational weight gain. Preterm birth less than 36 weeks, gestational hypertensive disorders, small for gestational age status, large for gestational age status, and a composite outcome defined as any occurrence of the individual outcomes. The main sample comprised 200 810 individuals with twin pregnancies (mean [SD] maternal age, 30.4 [5.5] years; 1624 [0.8%] American Indian or Alaska Native, 13 031 [6.5%] Asian or Pacific Islander, 36 423 [18.1%] Black, and 149 732 [74.6%] White; and 137 409 [68.4%] multiparous). In the low-risk subgroup (n = 61 794), the IQRs of the total GWG after 36 weeks of gestation as assessed using a statistics-based approach and based on BMI group were 15.9 to 22.7 kg for underweight, 15.4 to 22.7 kg for normal weight, 12.7 to 22.2 kg for overweight, 10.0 to 20.0 kg for class 1 obesity, 7.7 to 18.1 kg for class 2 obesity, and 5.9 to 16.3 kg for class 3 obesity. The absolute risk of the composite outcome showed U-shaped associations with GWG across BMI categories. The optimal GWG ranges by 36 weeks identified using an outcome-based approach and BMI group were 17.5 to 24.9 kg for underweight, 15.0 to 24.9 kg for normal weight, 15.0 to 24.9 kg for overweight, 10.0 to 19.9 kg for class 1 obesity, 7.5 to 17.4 kg for class 2 obesity, and 5.0 to 9.9 kg for class 3 obesity. The multivariable logistic models assessed using the validation sample (n = 49 275) showed that GWG defined outside those optimal ranges was associated with preterm birth at less than 36 weeks, gestational hypertensive disorders, and small or large for gestational age. This population-based cohort study found that optimal GWG ranges were similar for individuals with underweight and normal weight but decreased with increasing severity of obesity. The current US Institute of Medicine GWG recommendations may be too high for individuals with moderate or severe obesity.

Urbanization level and medical adverse event deaths among US hospital inpatients over the period 2010–2019

Urban-rural disparity constitutes a major source of health inequity also in high-income countries. This study aimed to compare the distribution of deaths due to medical adverse events across urbanization levels among US hospital inpatients. An open dataset from the National Center for Health Statistics (NCHS) comprised all certified deaths of US inpatients over the period 2010–2019. The urbanization level of each decedent was determined in accordance with the 2013 NCHS Urban-Rural Classification Scheme (large metropolitan, medium or small metropolitan, or nonmetropolitan). The outcome was death due to a medical adverse event (ICD-10 codes Y40—Y84) proportional to total inpatient deaths. The data were standardized for sex, ethnicity, and age, and analyzed with linear mixed models. Of the 8 071907 certified inpatient deaths during the study period, 21 444 (0.27%) were primarily attributed to medical adverse events. Decedents who resided in medium or small metropolitans and nonmetropolitans had approximately 0.5 units higher rate of adverse events per 1000 deaths (corresponding to a relative differece of 20%) when compared to decedents who resided in large metropolitans. Moreover, the urban–rural gradients showed an increasing trend towards the end of the study period, as the difference was found to increase at a rate of approximately 0.1 units per year (3%). There were no statistically significant differences between decedents from medium or small metropolitans and nonmetropolitans. The present findings highlight gradients in adverse event deaths between geographic areas, providing a basis for targeted preventive efforts. Future studies are invited to elucidate the underlying phenomena.

Trends in Assault Mortalities Among Asian Americans, 2009-2018.

As of yet, few studies have described the prevalence or rates of assault mortalities involving Asian victims on a national level. This study aimed to describe yearly trends and regional differences in assault mortalities among Asians. This repeat cross-sectional study from 2009 to 2018 used data from the Mortality Multiple Cause-of-Death Public Use Record from the National Center for Health Statistics and the Centers for Disease Control and Prevention Wide-Ranging Online Data for Epidemiologic Research. Total and proportion of assault mortalities involving Asians or Pacific Islanders were calculated by year. Total and rates (per 100,000 Asians or Pacific Islanders) of assault mortalities among Asians or Pacific Islanders were stratified by state and county classification where deaths occurred. In 2009, 344 assault mortalities among Asians or Pacific Islanders accounted for 2.07% of assault-related deaths. In 2018, 366 assault mortalities among Asians or Pacific Islanders accounted for 1.96% of assault-related deaths. Furthermore, there were more assault mortalities from 2009 to 2018 among Asians or Pacific Islanders in California (n=1116) and large central metropolitan counties (n=1707). However, the highest rates of assault mortalities were in Alaska and Mississippi (7.1 and 6.8 per 100,000, respectively) and noncore nonmetropolitan counties (2.9 per 100,000). These findings emphasize the importance of studying and addressing violence toward Asians in rural regions and Southern states. Future studies should use these results as a baseline to analyze mortality data from 2019 to 2021, when available, to examine the coronavirus disease 2019 pandemic's impact on assault mortalities among Asian.

Cancer disparities in the context of rurality: risk factors and screening across various U.S. rural classification codes.

Prior cancer research is limited by inconsistencies in defining rurality. The purpose of this study was to describe the prevalence of cancer risk factors and cancer screening behaviors across various county-based rural classification codes, including measures reflecting a continuum, to inform our understanding of cancer disparities according to the extent of rurality. Using an ecological cross-sectional design, we examined differences in cancer risk factors and cancer screening behaviors from the Behavioral Risk Factor Surveillance System and National Health Interview Survey (2008-2013) across rural counties and between rural and urban counties using four rural-urban classification codes for counties and county-equivalents in 2013: U.S. Office of Management and Budget, National Center for Health Statistics, USDA Economic Research Service's rural-urban continuum codes, and Urban Influence Codes. Although a rural-to-urban gradient was not consistently evident across all classification codes, the prevalence of smoking, obesity, physical inactivity, and binge alcohol use increased (all ptrend < 0.03), while colorectal, cervical and breast cancer screening decreased (all ptrend < 0.001) with increasing rurality. Differences in the prevalence of risk factors and screening behaviors across rural areas were greater than differences between rural and urban counties for obesity (2.4% vs. 1.5%), physical activity (2.9% vs. 2.5%), binge alcohol use (3.4% vs. 0.4%), cervical cancer screening (6.8% vs. 4.0%), and colorectal cancer screening (4.4% vs. 3.8%). Rural cancer disparities persist across multiple rural-urban classification codes, with marked variation in cancer risk factors and screening evident within rural regions. Focusing only on a rural-urban dichotomy may not sufficiently capture subpopulations of rural residents at greater risk for cancer and cancer-related mortality.

A panel-based proxy for gun prevalence in US and Mexico

There is a consensus that the proportion of suicides committed with a firearm is the best proxy for gun ownership prevalence. Cerqueira et al. (2018) exploit socioeconomic characteristics of suicide victims in order to develop more refined prevalence indicators. They rest on location fixed effects of the victim's place of residence from a discrete-choice model for the likelihood of committing suicide with gun. We empirically assess this new indicator using gun ownership data from the Behavioral Risk Factor Surveillance System (BRFSS) and suicide registers of the US National Center for Health Statistics (NCHS) from 1995 to 2004. We show that the panel-based proxy variables correlate more with the percentage of households with firearms than the conventional proxy based on the proportion of suicides committed with a firearm does. We further estimate gun prevalence across Mexican states. This is a relevant application because there is no representative household survey in Mexico with information about gun ownership at the state level.

Cancer treatment and survivorship statistics, 2022.

The number of cancer survivors continues to increase in the United States due to the growth and aging of the population as well as advances in early detection and treatment. To assist the public health community in better serving these individuals, the American Cancer Society and the National Cancer Institute collaborate triennially to estimate cancer prevalence in the United States using incidence and survival data from the Surveillance, Epidemiology, and End Results cancer registries, vital statistics from the Centers for Disease Control and Prevention's National Center for Health Statistics, and population projections from the US Census Bureau. Current treatment patterns based on information in the National Cancer Database are presented for the most prevalent cancer types by race, and cancer-related and treatment-related side-effects are also briefly described. More than 18 million Americans (8.3 million males and 9.7 million females) with a history of cancer were alive on January 1, 2022. The 3 most prevalent cancers are prostate (3,523,230), melanoma of the skin (760,640), and colon and rectum (726,450) among males and breast (4,055,770), uterine corpus (891,560), and thyroid (823,800) among females. More than one-half (53%) of survivors were diagnosed within the past 10 years, and two-thirds (67%) were aged 65 years or older. One of the largest racial disparities in treatment is for rectal cancer, for which 41% of Black patients with stage I disease receive proctectomy or proctocolectomy compared to 66% of White patients. Surgical receipt is also substantially lower among Black patients with non-small cell lung cancer, 49% for stages I-II and 16% for stage III versus 55% and 22% for White patients, respectively. These treatment disparities are exacerbated by the fact that Black patients continue to be less likely to be diagnosed with stage I disease than White patients for most cancers, with some of the largest disparities for female breast (53% vs 68%) and endometrial (59% vs 73%). Although there are a growing number of tools that can assist patients, caregivers, and clinicians in navigating the various phases of cancer survivorship, further evidence-based strategies and equitable access to available resources are needed to mitigate disparities for communities of color and optimize care for people with a history of cancer. CA Cancer J Clin. 2022;72:409-436.