State Vital Statistics Research Articles

Clinical Outcomes Following Regionalization of Gastric Cancer Care in a US Integrated Health Care System.

In 2016, Kaiser Permanente Northern California regionalized gastric cancer care, introducing a regional comprehensive multidisciplinary care team, standardizing staging and chemotherapy, and implementing laparoscopic gastrectomy and D2 lymphadenectomy for patients eligible for curative-intent surgery. This study evaluated the effect of regionalization on outcomes. The retrospective cohort study included gastric cancer cases diagnosed from January 2010 to May 2018. Information was obtained from the electronic medical record, cancer registry, state vital statistics, and chart review. Overall survival was compared in patients with all stages of disease, stage I-III disease, and curative-intent gastrectomy patients using annual inception cohorts. For the latter, the surgical approach and surgical outcomes were also compared. Among 1,429 eligible patients with gastric cancer with all stages of disease, one third were treated after regionalization, 650 had stage I-III disease, and 394 underwent curative-intent surgery. Among surgical patients, neoadjuvant chemotherapy utilization increased from 35% to 66% (P < .0001), laparoscopic gastrectomy increased from 18% to 92% (P < .0001), and D2 lymphadenectomy increased from 2% to 80% (P < .0001). Dissection of ≥ 15 lymph nodes increased from 61% to 95% (P < .0001). Surgical complication rates did not appear to increase after regionalization. Length of hospitalization decreased from 7 to 3 days (P < .001). Overall survival at 2 years was as follows: all stages, 32.8% pre and 37.3% post (P = .20); stage I-III cases with or without surgery, 55.6% and 61.1%, respectively (P = .25); and among surgery patients, 72.7% and 85.5%, respectively (P < .03). Regionalization of gastric cancer care within an integrated system allowed comprehensive multidisciplinary care, conversion to laparoscopic gastrectomy and D2 lymphadenectomy, increased overall survival among surgery patients, and no increase in surgical complications.

CDC's Efforts to Quantify Prescription Opioid Overdose Deaths Fall Short.

In a 2018 report titled, Quantifying the Epidemic of Prescription Opioid Overdose Deaths, four senior analysts of the Centers for Disease Control and Prevention (CDC), including the head of the Epidemiology and Surveillance Branch, acknowledged for the first time that the number of prescription opioid overdose deaths reported by the CDC in 2016 was erroneous. The error, they said, was caused by miscoding deaths involving illicitly manufactured fentanyl (IMF) as deaths involving prescribed fentanyl. To understand what caused this error, the authors examined the CDC’s methodology for compiling drug-related mortality data, beginning with the source data obtained from approximately 2.8 million death certificates received each year from state vital statistics registrars. Systemic problems often begin outside the CDC, with a surprisingly high rate of errors and omissions in the source data. Using the CDC’s explanation for what caused the error, the authors show why an international program used by the CDC for reporting mortality is ill-suited for compiling and reporting drug overdose deaths. Except for heroin, methadone, and opium, each of which has an individual program code, all other opioids are separated in just two program codes according to whether they are synthetic or semisynthetic/opiates. Methadone-involved deaths pose a special problem for the CDC because methadone has dual indications for treating pain and for treating opioid use disorder (OUD). In 2019, more than seven times more methadone was administered or dispensed for OUD treatment than was prescribed for pain, yet all methadone-involved deaths are coded by the CDC as involving the prescribed form of the drug. The authors conclude that the CDC was at fault for failing to recognize and correct this problem before 2016. Public policy consequences of this failure are briefly mentioned.

Validating the Matching of Patients in the Linkage of a Large Hospital System's EHR with State and National Death Databases.

Background Though electronic health record (EHR) data have been linked to national and state death registries, such linkages have rarely been validated for an entire hospital system's EHR. Objectives The aim of the study is to validate West Virginia University Medicine's (WVU Medicine) linkage of its EHR to three external death registries: the Social Security Death Masterfile (SSDMF), the national death index (NDI), the West Virginia Department of Health and Human Resources (DHHR). Methods Probabilistic matching was used to link patients to NDI and deterministic matching for the SSDMF and DHHR vital statistics records (WVDMF). In subanalysis, we used deaths recorded in Epic ( n = 30,217) to further validate a subset of deaths captured by the SSDMF, NDI, and WVDMF. Results Of the deaths captured by the SSDMF, 59.8 and 68.5% were captured by NDI and WVDMF, respectively; for deaths captured by NDI this co-capture rate was 80 and 78%, respectively, for the SSDMF and WVDMF. Kappa statistics were strongest for NDI and WVDMF (61.2%) and NDI and SSDMF (60.6%) and weakest for SSDMF and WVDMF (27.9%). Of deaths recorded in Epic, 84.3, 85.5, and 84.4% were captured by SSDMF, NDI, and WVDMF, respectively. Less than 2% of patients' deaths recorded in Epic were not found in any of the death registries. Finally, approximately 0.2% of “decedents” in any death registry re-emerged in Epic at least 6 months after their death date, a very small percentage and thus further validating the linkages. Conclusion NDI had greatest validity in capturing deaths in our EHR. As a similar, though slightly less capture and agreement rate in identifying deaths is observed for SSDMF and state vital statistics records, these registries may be reasonable alternatives to NDI for research and quality assurance studies utilizing entire EHRs from large hospital systems. Investigators should also be aware that there will be a very tiny fraction of “dead” patients re-emerging in the EHR.

Distance to an Abortion Provider and Its Association with the Abortion Rate: A Multistate Longitudinal Analysis.

Although one in four U.S. women has an abortion in her lifetime, barriers to abortion persist, including distance to care. This study evaluates the association between distance to care and the abortion rate, adjusting for abortion demand. Two analyses were conducted using a data set linking provider locations and 2000-2014 county-level abortion data for 18 states; data sources included the Census Bureau, state vital statistics offices and the Guttmacher Institute. First, a series of linear regression models were run, with and without adjustment for demographic covariates, modeling distance as both a continuous and a categorical variable. Then, an instrumental variable analysis was conducted in which being 30 or more miles from a large college-enrolled female population younger than age 25 was used as an instrument for distance to a provider. The outcome variable for all models was abortions per 1,000 women aged 25 or older. All models were adjusted for state, year and state-year interaction fixed effects. Increased distance to a provider was associated with a decreased abortion rate. Each additional mile to a provider was associated with a decrease of 0.011 in the abortion rate. Compared with being within 30 miles of a provider, being between 30 and 90 miles from a provider was associated with 0.80-1.46 fewer abortions per 1,000 women. In the instrumental variable analysis, being 30 or more miles from a provider was associated with 5.26 fewer abortions per 1,000 women. Distance to a provider may present a barrier to abortion by preventing access to care. Therefore, policies that increase travel distances have potential for harm.

Evaluation of an active population-based surveillance system for major birth defects in South Carolina.

Birth defects are common, costly, and contribute substantially to infant mortality. The South Carolina Birth Defects Program (SCBDP) employs active population-based surveillance to monitor major birth defects statewide. We evaluated SCBDP's system attributes using published CDC guidelines. To determine timeliness, completeness, and accuracy of birth defects information, we examined SCBDP's reports, program and education materials, advisory group meeting minutes, and strategic plan. We also met with program staff and stakeholders (n = 10) to discuss program goals and data utilization. We calculated the percentage of birth defects cases found 6 months after a birth cohort year for 2016-2018. SCBDP identifies 900-1,200 birth defects cases for a birth population of approximately 55,000 live births annually through active case reviews. SCBDP uses trained medical staff to abstract detailed information from maternal and infant medical records; SCBDP also has established auto-linkage with state vital statistics to capture demographic and birth data. SCBDP is timely and captures 97.1% (range 96.7-97.6%) of birth defects cases within 6 months after the birth cohort year closes. Active case identification using medical records as the primary data source improves quality assurance and completeness, while prepopulating demographic information improves timeliness. Given that birth defects significantly contribute to infant morbidity and mortality, monitoring these conditions is important to understand their impact on communities and to drive public health actions. SCBDP active surveillance and rigorous data quality practices provide the program with timely, complete, and accurate birth defects data.

Neonatal Seizures Among Low-Risk Pregnancies at Term: Risk Factors and Adverse Outcomes.

To examine risk factors and adverse outcomes for neonatal-maternal dyads among low-risk pregnancies at term with subsequent neonatal seizures. United States vital statistics data sets were used for this retrospective study. Inclusion criteria were low-risk women (without hypertensive disease or diabetes) with nonanomalous singleton pregnancies, who delivered after labor at 37-41 weeks of gestation. The primary composite neonatal adverse outcome included 5-minute Apgar score less than 5, assisted ventilation longer than 6 hours, and neonatal death. A secondary outcome was composite maternal adverse outcome. Multivariable Poisson regression models with robust error variance were used, with adjusted relative risk (aRR) and 95% CI reported. Of 19.76 million live births during the study interval, 11.7 million (59.4%) met inclusion criteria. The rate of neonatal seizures after low-risk pregnancies delivered at term was 0.2 per 1,000 live births. The maternal risks factors associated with neonatal seizures included no prenatal care, smoking during pregnancy, being overweight or obese, and gestational age of 41 weeks. The strongest risk factors for neonatal seizures were chorioamnionitis (relative risk [RR] 5.04, 95% CI 4.40-5.77; aRR 3.27, 95% CI 2.84-3.76) and route of delivery, with operative vaginal (RR 3.62, 95% CI 3.20-4.09; aRR 3.02, 95% CI 2.66-3.43) and cesarean (RR 4.13, 95% CI 3.81-4.48; aRR 3.14, 95% CI 2.86-3.45) higher than spontaneous vaginal. Compared with neonates without seizures, those with seizures had higher risk of composite neonatal adverse outcome (RR 64.55, 95% CI 61.83-67.39; aRR 37.09, 95% CI 35.20-39.08). Compared with women who delivered neonates without seizures, those who delivered neonates with seizures had higher risk of composite maternal adverse outcome (RR 16.27, 95% CI 13.66-19.37; aRR 9.70, 95% CI 8.15-11.53). We identified modifiable maternal risk factors associated with neonatal seizures among low-risk pregnancies at term. Though infrequent, neonatal seizures are associated with higher risk of adverse outcomes in neonatal-maternal dyads.

Provider–patient Language Discordance and Cancer Operations: Outcomes from a Single Center Linked to a State Vital Statistics Registry

Patterns of worldwide immigration have resulted in high rates of discordance between medical providers and the patients they treat. For example, in the USA, 25 million individuals in the USA self-identified that they speak English less than "very well." Previous studies have generated mixed results regarding differences in postoperative outcomes between English proficient (EP) and limited English proficient (LEP) patients. Our objective was to determine whether a difference in outcomes exists for non-English-speaking patients compared to English-speaking patients after operations commonly performed to treat cancer. A retrospective cohort study was performed in an urban, safety net and tertiary referral medical center over a five-year period. Adult patients undergoing cancer operations were stratified as EP and LEP. We evaluated 30-day revisit to the ED, length of stay (LOS), long-term all-cause mortality, and any major complication on index admission. Regression was used to adjust for baseline comorbidities, case risk, and socioeconomic factors. A total of 2467 patients were included. There was no difference in case risk between language groups, but EP had a larger proportion of high comorbidity scores. Patients in the non-English group were more likely to be uninsured/self-pay andlive in neighborhoods with lower median income. After adjustment, we found no difference in long-term mortality [hazard ratio: 0.87 (95% CI 0.52-1.45)]. LEP patients had the same LOS compared to primary EP patients with an IRR of 0.99 (95% CI 0.88-1.10). There was no difference in the odds of revisit to hospital for LEP versus EP, with an OR of 1.08, 95% CI [0.75-1.53] and no difference in major complication (OR 0.76 (95% CI 0.39-1.45). We found no association between language and outcomes after cancer operations. This lack of difference may reflect local efficacy at treating non-English-speaking patients, and health systems with fewer services for LEP patients might show different results.

Accuracy in Reporting of Kentucky Certified Nurse‐Midwives as Attendants in Birth Registration Data

Birth certificate data are used nationally to determine health care policy and health care reimbursement and to demonstrate the legitimacy and value of certified nurse-midwives (CNMs) and certified midwives (CMs) in perinatal and neonatal outcomes. However, the validity of birth certificate data is questionable, in part because of the data collection process. These data are particularly crucial for midwife-attended births because the correct birth attendant is not always accurately identified on the birth certificate. The purpose of this project was to examine the actual number of CNM-reported births compared with those recorded by the Kentucky Office of Vital Statistics and to examine the process used by birth registrars to complete the birth certificate. CNMs attending births in hospitals in Kentucky in 2017 logged their birth statistics. These numbers were compared with the 2017 Kentucky Vital Statistics Birth Certificate data of CNM-attended births. Kentucky birth registrars (50%) who work in facilities where CNMs attend births completed a 32-question survey to describe their process of collecting birth certificate data. The comparison data revealed that CNM-attended births in Kentucky are underrepresented in the state vital statistics by as much as 19.2%. Birth registrars identified barriers to collecting accurate data including lack of training, multiple sources of data, incomplete prenatal records, and absence of systems to help ensure accuracy. CNMs/CMs should keep personal and practice birth logs and routinely compare these with hospital data kept by the birth registrar. The state office of vital statistics and hospitals should target training to specific facilities that have the most inaccurate data. The Improving Midwifery Birth Numbers Initiative through the American College of Nurse-Midwives Division of Research Data Management Section should continue to encourage midwifery students to complete this research in all 50 states.

Brief report: Assessing the risk of reporting bias in a RCT for adolescent mothers

IntroductionMost adolescent sexual and reproductive health studies rely solely upon self-report surveys to assess key variables such as pregnancy and sexually transmitted infections (STIs). The current study investigated the risk of reporting bias that may result from using such surveys to measure outcomes in randomized controlled trials (RCTs). MethodsAs part of an RCT of a multicomponent intervention to delay repeat pregnancy among adolescent mothers, we compared survey data on repeat pregnancy with birth records from a state's vital statistics system. The survey assessed contraceptive use, pregnancy status, and future pregnancy intentions. The sample consisted of American adolescents ages 18 and 19 years who were at least 28 weeks pregnant or less than nine weeks postpartum. ResultsFor 14 of 331 study participants (4 percent), we found a birth record in the vital statistics system for a mother who reported not having gotten pregnant on the survey. We found no evidence of underreporting for the other survey respondents. The rate of underreporting was similar for the intervention and control groups. ConclusionsA low rate of underreporting for repeat pregnancy in adolescent mothers was found in the present sample.

Prenatal exposure to nitrate in drinking water and the risk of congenital anomalies

BackgroundNitrate is a common water contaminant that has been associated with birth defects, although the evidence is limited. The purpose of this study was to examine whether maternal consumption of nitrate through drinking water is associated with an increased risk of congenital anomalies. MethodsThe study included a total of 348,250 singletons births from the state of Missouri between January 1, 2004 and December 31, 2008. Individual-level birth defect data and maternal and child characteristics were obtained from the Missouri birth defects registry and state vital statistics records. Outcomes were linked with county-specific monthly estimates of the nitrate concentration in finished water, based on data collected for compliance with the Safe Drinking Water Standard. Poisson models were fit to examine the association between nitrate exposure and birth defects. Average nitrate exposure during the first trimester and over 12 months prior to birth were modeled as continuous variables. Sensitivity analyses included restriction of the sample to counties with <20% and <10% private well usage to reduce exposure misclassification as well as limiting the analyses to residents of rural counties only to account for potential confounding by urbanicity. ResultsEstimated water concentrations of nitrate were generally low and below the Environmental Protection Agency’s maximum contaminant level of 10 mg/L. Nitrate exposure was associated with a significantly increased risk of limb deficiencies (RR for 1 mg/L (RR1) = 1.26, 95% CI = 1.05, 1.51) in models without well restriction. Nitrate was also weakly associated with an increased risk of congenital heart defects (RR1 = 1.13, 95%CI = 0.93, 1.51) and neural tube defects (RR1 = 1.18, 95%CI = 0.93, 1.51) in models with well restriction (<10%). ConclusionThe positive associations found between nitrate exposure via drinking water and congenital abnormalities are largely consistent with some previous epidemiologic studies. The results of this study should be interpreted with caution given limitations in our ability to estimate exposures and the lack information on some risk factors for congenital abnormalities. Our findings may have serious policy implications given that exposure levels in our study were well below current EPA standards for nitrate in drinking water.

Association of changes in red meat consumption with total and cause specific mortality among US women and men: two prospective cohort studies

ObjectiveTo evaluate the association of changes in red meat consumption with total and cause specific mortality in women and men.DesignTwo prospective cohort studies with repeated measures of diet and lifestyle...

A longitudinal intervention to improve young children\u2019s liking and consumption of new foods: findings from the Colorado LEAP study

BackgroundMany interventions have been conducted to improve young children’s liking and consumption of new foods however their impacts on children’s consumption have been limited. Consistent evidence supports the use of repeated exposure to improve liking for new foods however longitudinal effects lasting greater than 6 months often have not been demonstrated. Here we report the eating-related findings of the Colorado Longitudinal Eating And Physical Activity (LEAP) Study, a multi-component intervention, delivered primarily in the school setting, which aimed to improve children’s liking and consumption of a target food via repeated exposure and positive experiential learning.MethodsFour sites in rural Colorado, each housing Head Start preschool programs, matched on state vital statistics for childhood obesity rates, (2 intervention and 2 control sites) took part in a quasi-experimental study design which included 4 time points (baseline, post-intervention, one-year [Y1] and two- year [Y2] follow ups). A total of 250 children and families were enrolled (n = 143 intervention and n = 107 control; 41% Hispanic and 69% low-income). A 12-week intervention, Food Friends – Fun With New Foods®, delivered by trained preschool teachers and which focuses on positive and repeated experiences with new foods, and a 5-month (1 unit/month) social marketing “booster program” was delivered in kindergarten (one-year follow up) and 1st grade (two-year follow up). Main outcome measures included change in children’s liking for new foods, analyzed by ordinal regression using generalized estimating equations, and change in weighed consumption of new foods over time, analyzed using a hierarchical mixed effects model.ResultsThe intervention was delivered with good fidelity (87%). Both intervention and control groups demonstrated an increase in liking for the target food over time (p = 0.0001). The pattern of consumption of the target food was different, over time, for intervention and control groups (p < 0.005). In particular the change in intake between baseline and post-intervention was significantly greater in the intervention compared to the control group (p < 0.0001) though this pattern of change did not hold between baseline and Y2 follow up (p = 0.1144). Children in the intervention group who liked the target food consumed nearly double their baseline consumption at post-intervention (p < 0.0001;) and maintained this increase at Y2 follow up (p < 0.0001).ConclusionsThe Food Friends intervention, which utilized positive, repeated experiences with new foods, and was delivered with good fidelity by trained preschool teachers, found that larger improvements were observed in children’s eating behaviors than would be expected with developmentally-based changes in eating behaviors.Trial registration numberThis trial is registered at ClinicalTrials.gov: NCT01937481.Date registered: 09/09/2013; Retrospectively registered.Date first participant registered: 09/15/2010.

A statewide quality improvement (QI) initiative for better health outcomes and family stability among pregnant women with opioid use disorder (OUD) and their infants

A collaborative led by state health and human service agencies, academic leaders, and stakeholders tested interventions to expand use of medication assisted treatment (MAT) through a maternal medical home (MMH) model that coordinated behavioral health and prenatal care with social supports for pregnant women with opioid use disorder (OUD) enrolled in Medicaid. The program was anchored in four clinical organizations with distinct models of care: community behavioral health, residential behavioral health, hospital-based obstetrical practice, and co-located obstetrical and behavioral health. A modified version of the Institute for Healthcare Improvement Breakthrough Series Model for Improvement was implemented using monthly performance data feedback to conduct small tests of change and improve care. Administrative data from the state's Medicaid, vital statistics, and child welfare systems were linked to evaluate the impact of MOMS on 252 mother-infant dyads compared to a sample of 846 Medicaid beneficiaries with OUD in the third trimester of pregnancy. MOMS participation was associated with increased likelihood of MAT in trimesters one, two and three (AOR = 2.30, 4.40, 2.75, respectively), behavioral health counseling during trimesters two and three (AOR = 3.75 and 2.07, respectively), retention in MAT during postpartum months one through three and four through six (AOR = 2.86, 2.40, respectively), and marginally lower out-of-home placement of infants born to mothers with OUD (AOR = 0.66). Within the MOMS program, greater participation in behavioral health treatment and MAT (χ2(3) ≥ 12.09) was observed in the co-located behavioral health/obstetrical care practice site compared to behavioral health-led and obstetrical provider-led sites.

State Based Perinatal Quality Collaborative Pilot-Statewide Reduction of Colorado Primary Cesarean Delivery Rate [04OP

INTRODUCTION: The Colorado Perinatal Care Quality Collaborative performed a pilot quality improvement project to increase adherence to guidelines from the Consensus for the Prevention of the Primary Cesarean Delivery in a statewide cohort of hospitals to safely reduce their nulliparous, term, singleton, vertex (NTSV) cesarean delivery (CD) rate. METHODS: This quality improvement initiative was performed at 6 community hospitals across Colorado. Pre (January 2017 to May 2018) and post intervention (June to Oct 2018) NTSV CD and composite maternal and neonatal morbidity rates were determined for each hospital and the overall cohort using Colorado State Vital Statistics data. Interventions included distributing institution and provider specific NTSV CD rates, education for providers and nurses, implementation of a pre-cesarean labor dystocia checklist, and convening two forums for collaboration between hospitals. NTSV CD and composite maternal and neonatal morbidity rates were compared before and after the interventions for each hospital and the cohort. RESULTS: The 6 participating hospitals perform 19% of live births in Colorado. Pre-intervention hospital NTSV CD rates ranged from 19.3% to 33.8%. The cohort pre-intervention NTSV CD rate was 24.6%, and decreased to 21.9% post-intervention (P=0.04, 95% CI 0.77-0.99). There was no significant increase in composite maternal or neonatal morbidity post-intervention (8.6% vs 7.1% and 7.1% vs 6.8% respectively). CONCLUSION: A quality improvement effort led by a state-based Perinatal Quality Collaborative was effective at safely reducing NTSV CD rates across multiple hospitals. Future efforts will aim at expanding the number of participating hospitals and sustainability.

Wake Forest University long-term follow-up of type 2 myocardial infarction: The Wake-Up T2MI Registry.

BackgroundThe Wake‐Up T2MI Registry is a retrospective cohort study investigating patients with type 2 myocardial infarction (T2MI), acute myocardial injury, and chronic myocardial injury. We aim to explore risk stratification strategies and investigate clinical characteristics, management, and short‐ and long‐term outcomes in this high‐risk, understudied population.MethodsFrom 1 January 2009 to 31 December 2010, 2846 patients were identified with T2MI or myocardial injury defined as elevated cardiac troponin I with at least one value above the 99th percentile upper reference limit and coefficient of variation of 10% (>40 ng/L) and meeting our inclusion criteria. Data of at least two serial troponin values will be collected from the electronic health records to differentiate between acute and chronic myocardial injury. The Fourth Universal Definition will be used to classify patients as having (a) T2MI, (b) acute myocardial injury, or (c) chronic myocardial injury during the index hospitalization. Long‐term mortality data will be collected through data linkage with the National Death Index and North Carolina State Vital Statistics.ResultsWe have collected data for a total of 2205 patients as of November 2018. The mean age of the population was 65.6 ± 16.9 years, 48% were men, and 64% were white. Common comorbidities included hypertension (71%), hyperlipidemia (35%), and diabetes mellitus (30%). At presentation, 40% were on aspirin, 38% on β‐blockers, and 30% on statins.ConclusionImproved characterization and profiling of this cohort may further efforts to identify evidence‐based strategies to improve cardiovascular outcomes among patients with T2MI and myocardial injury.

Breast Cancer: Does Type of Hospital Where You Get Surgery Affect Survival?

Under the Affordable Care Act (ACA), more women are insured with Medicaid, which should improve healthcare access. We sought to determine whether there are survival differences among patients with breast cancer undergoing surgery at facilities with varying proportions of Medicaid patients. We used New York State (NYS) Vital Statistics death records data linked with NYS discharge inpatient and ambulatory surgery databases to examine 90-day survival after surgery from 2008 to 2013. We used all Medicaid discharges to calculate and create quintiles of facilities based on Medicaid volume. We calculated survival hazard ratios using a marginal Cox model controlling for clustering of patients within hospitals, age, race, insurance, year of surgery, and comorbidities. Women who received surgery in facilities with the highest quintile of Medicaid volume had higher 90-day mortality (2.1% vs. 0.07%, p < .001) compared with those treated in facilities with lowest Medicaid volume, even after adjusting for multiple confounders. Consequently, although the ACA may improve access, healthcare quality remains questionable because patients treated at facilities with high proportions of Medicaid volume appear to have worse 90-day survival, likely due to quality of surgical and postsurgical care. Policymakers must ensure that quality of care is not negatively impacted by programs to reduce costs.

Alive or dead: Validity of the Social Security Administration Death Master File after 2011.

To determine the reliability of the Social Security Death Master File (DMF) after the November 2011 changes limiting the inclusion of state records. Secondary data from the DMF, New York State (NYS) and New Jersey (NJ) Vital Statistics (VS), and institutional data warehouse. Retrospective study. Two cohorts: discharge date before November 1, 2011, (pre-2011) or after (post-2011). Death in-hospital used as gold standard. NYS VS used for out-of-hospital death. Sensitivity, specificity, Cohen's Kappa, and 1-year survival calculated. Patients matched to DMF using Social Security Number, or date of birth and Soundex algorithm. Patients matched to NY and NJ VS using probabilistic linking. 97069 patients January 2007-March 2016: 39075 pre-2011; 57994 post-2011. 3777 (3.9 percent) died in-hospital. DMF sensitivity for in-hospital death 88.9 percent (κ=0.93) pre-2011 vs 14.8 percent (κ=0.25) post-2011. DMF sensitivity for NY deaths 74.6 percent (κ=0.71) pre-2011 vs 26.6 percent (κ=0.33) post-2011. DMF sensitivity for NJ deaths 62.6 percent (κ=0.64) pre-2011 vs 10.8 percent (κ=0.15) post-2011. DMF sensitivity for out-of-hospital death 71.4 percent pre-2011 (κ=0.58) vs 28.9 percent post-2011 (κ=0.34). Post-2011, 1-year survival using DMF data was overestimated at 95.8 percent, vs 86.1 percent using NYS VS. The DMF is no longer a reliable source of death data. Researchers using the DMF may underestimate mortality.

Differences in Firearm Suicides by Residential Location in Texas, 2006–2015

This study used Texas state vital statistics records, 2006–2015, to examine firearm use rates among 28,010 suicide decedents by residential location (urbanized vs. all others). Firearms were responsible for 44% of all teenage suicides and 76–90% and 50–60% of suicides of men and women aged 60 + years, respectively, and firearm use rates remained steady for both genders during the study period. Logistic regression analysis showed a significantly higher firearm use rate (AOR = 1.35, 95% CI = 1.28–1.42) among decedents who resided in nonurbanized areas. Differences in firearm use rates by residential location likely reflect higher firearm ownership in smaller communities and rural than urban areas. The findings underscore the importance of community- and individual-level suicide prevention strategies.

The Washington State Merged Longitudinal Administrative Database

This paper describes a uniquely comprehensive database constructed from merged state administrative data. State Unemployment Insurance (UI) systems provide an important source of data for understanding employment effects of policy interventions but have also lack several key types of information: personal demographics, non-earnings income, and household associations. With UI data, researchers can show overall earnings or employment trends or policy impacts, but cannot distinguish whether these trends or impacts differ by race or gender, how they affect families and children, or whether total income or other measure of well-being change. This paper describes a uniquely comprehensive new administrative dataset, the Washington Merged Longitudinal Administrative Database (WMLAD), created by University of Washington researchers to examine distributional and household economic effects of the Seattle $15 minimum wage ordinance, an intervention that more than doubled the federal minimum wage. &#x0D; WMLAD augments UI data with state administrative voter, licensing, social service, income transfer, and vital statistics records. The union set of all individuals who appear in any of these agency datasets will provide a near-census of state residents and will augment UI records with information on age, sex, race/ethnicity, public assistance receipt, and household membership. In this paper, we describe 1.) our relationship with the Washington State Department of Social and Health Services that permits this data access and allows construction of this dataset using restricted personal identifiers; 2.) the merging and construction process, including imputing race and ethnicity and constructing quasi-households from address co-location; and 3.) planned benchmarking and analysis work.

Building A Longitudinal Cohort From 9-1-1 to 1-Year Using Existing Data Sources, Probabilistic Linkage, and Multiple Imputation: A Validation Study.

The objective was to describe and validate construction of a population-based, longitudinal cohort of injured older adults from 9-1-1 call to 1-year follow-up using existing data sources, probabilistic linkage, and multiple imputation. This was a descriptive cohort study conducted in seven counties in Oregon and Washington from January 1, 2011, through December 31, 2011, with follow-up through December 31, 2012. The primary cohort included all injured adults ≥ 65 years served by 44 emergency medical services (EMS) agencies. We used nine existing databases to assemble the cohort, including EMS data, two state trauma registries, two state discharge databases, two state vital statistics databases, the Oregon Physician Order for Life-Sustaining Treatment registry, and Medicare claims data. We matched data files using probabilistic linkage and handled missing values with multiple imputation. We independently validated data processes using 1,350 randomly sampled records for probabilistic linkage and 3,140 randomly sampled records for variables created from existing data sources. There were 15,649 injured older adults in the primary cohort, with 13,661 (87.3%) total matched records and 9,337 (59.7%) matches to the index ED/hospital visit. The sensitivity of linkage was 99.9% (95% confidence interval [CI]= 99.3%-100%) for any match and 98.3% (95% CI= 96.2%-99.4%) for index event matches. The specificity of linkage was 95.7% (95% CI= 93.7%-97.2%) for any match and 100% (95% CI= 99.2%-100%) for index event matches. Name, date of birth, home zip code, age, and hospital had the highest yield for linkage. Patients with matched records tended to be higher acuity than unmatched patients, suggesting selection bias if unmatched patients were excluded. Compared to hand-abstracted values, the sensitivity of electronically derived variables ranged from 18.2% (abdominal-pelvic Abbreviated Injury Scale score ≥ 3) to 97.4% (in-hospital mortality), with specificity of 88.0% to 99.8%. A population-based emergency care cohort with long-term outcomes can be constructed from existing data sources with high accuracy and reasonable validity of resulting variables.