Kirk A. Bol

SURVIVAL OF INFANTS WITH NEURAL TUBE DEFECTS IN THE PRESENCE OF FOLIC ACID FORTIFICATION

OBJECTIVE. Neural tube defects (NTDs) are preventable through preconceptional and periconceptional folic acid intake. Although decreases in the prevalence of NTDs have been reported since folic acid fortification of United States grain products began, it is not known whether folic acid plays a role in reducing the severity of occurring NTDs. Our aim was to determine whether survival among infants born with spina bifida and encephalocele has improved since folic acid fortification and to measure the effects of selected maternal, pregnancy, and birth characteristics on first-year (infant) survival rates. METHODS. A retrospective cohort study was conducted and included 2841 infants with spina bifida and 638 infants with encephalocele who were born between 1995 and 2001 and were registered in any of 16 participating birth defects monitoring programs in the United States. First-year survival rates for both spina bifida and encephalocele cohorts were measured with Kaplan-Meier estimation; factors associated with improved chances of first-year survival, including birth before or during folic acid fortification, were measured with Cox proportional-hazards regression analysis. RESULTS. Infants with spina bifida experienced a significantly improved first-year survival rate of 92.1% (adjusted hazard ratio: 0.68; 95% confidence interval: 0.50–0.91) during the period of mandatory folic acid fortification, compared with a 90.3% survival rate for those born before fortification. Infants with encephalocele had a statistically nonsignificant increase in survival rates, ie, 79.1% (adjusted hazard ratio: 0.76; 95% confidence interval: 0.51–1.13) with folic acid fortification, compared with 75.7% for earlier births. CONCLUSIONS. Folic acid may play a role in reducing the severity of NTDs in addition to preventing the occurrence of NTDs. This phenomenon contributes to our understanding of the efficacy of folic acid. Additionally, as survival of NTD-affected infants improves, health care, education, and family support must expand to meet their needs.

Fatality and Severity of Firearm Injuries in a Denver Trauma Center, 2000-2013

This study uses trauma center registry data to describe the severity and mortality of gunshot wounds seen at a Colorado trauma center and its catchment area between 2000 and 2013.

Previous Suicide Attempt and Its Association With Method Used in a Suicide Death

INTRODUCTION Non-fatal suicide attempts are a risk factor for suicide, but less is known about its effect on the method of injury used in an eventual suicide death. This study examines the association between history of non-fatal suicide attempt and the risk of a poisoning suicide versus firearm or hanging suicide in Colorado. METHODS Nine years (2004-2012, N=7,020) of suicide deaths in Colorado were compiled through the National Violent Death Reporting System. With these data, a retrospective cohort study was conducted in 2015, examining the risk associated with a history of suicide attempt and an eventual suicide death by poisoning versus a firearm or hanging suicide death. Multivariable log-binomial regression modeling was used to analyze the possible confounders of age and county type and gender as an effect modifier. RESULTS Suicide decedents with evidence of prior suicide attempt were twice as likely to eventually die by suicide via self-poisoning rather than by firearm injury (relative risk=1.94, 95% CI=1.8, 2.1, p<0.001). A significant interaction (p<0.001) between prior attempt and gender was identified and the relationship between prior attempt and method (poisoning versus firearm) was stronger among male than female suicides (female attempt history, 1.16, 95% CI=1.07, 1.27, p<0.001; male attempt history, 1.91, 95% CI=1.69, 2.16, p<0.001). CONCLUSIONS Individuals with a history of non-fatal suicide attempt are more likely to die by suicide via poisoning rather than firearm. Gender has a significant effect on this association. This information can be useful when discussing means restriction and suicide prevention efforts.

Sudden Infant Death Syndrome and Residential Altitude

BACKGROUND: Theories of sudden infant death syndrome (SIDS) suggest hypoxia is a common pathway. Infants living at altitude have evidence of hypoxia; however, the association between SIDS incidence and infant residential altitude has not been well studied. METHODS: We performed a retrospective cohort study by using data from the Colorado birth and death registries from 2007 to 2012. Infant residential altitude was determined by geocoding maternal residential address. Logistic regression was used to determine adjusted association between residential altitude and SIDS. We evaluated the impact of the Back to Sleep campaign across various altitudes in an extended cohort from 1990 to 2012 to assess for interaction between sleep position and altitude. RESULTS: A total of 393 216 infants born between 2007 and 2012 were included in the primary cohort (51.4% boys; mean birth weight 3194 ± 558 g). Overall, 79.6% infants resided at altitude <6000 feet, 18.5% at 6000 to 8000 feet, and 1.9% at >8000 feet. There were no meaningful differences in maternal characteristics across altitude groups. Compared with residence <6000 feet, residence at high altitude (>8000 feet), was associated with an adjusted increased risk of SIDS (odds ratio 2.30; 95% confidence interval 1.01–5.24). Before the Back to Sleep campaign, the incidence of SIDS in Colorado was 1.99/1000 live births and dropped to 0.57/1000 live births after its implementation. The Back to Sleep campaign had similar effect across different altitudes (P = .45). CONCLUSIONS: Residence at high altitude was significantly associated with an increased adjusted risk for SIDS. Impact of the Back to Sleep campaign was similar across various altitudes.

PS2-20: Obesity Mapping in Colorado: A Novel System for Monitoring and Tracking BMI.

Background/Aims There has been a marked increase in obesity prevalence in the US and around the world over the last twenty years. The causes of the rise in obesity involve many interrelated environmental and socioeconomic factors. Because interventions to address the obesity epidemic often occur at the community level, better surveillance data are needed to monitor obesity within communities. The only available community-level BMI data are from self-reports, provided by the Behavioral Risk Factor Surveillance System (BRFSS). Methods We have piloted a system for the collection of BMI data from multiple healthcare providers in Colorado (Kaiser Permanente Colorado, Denver Health, Children’s Hospital Colorado and High Plains Community Health Center). BMI information, objectively measured during routine care and collected in Electronic Medical Records (EMR), is combined with geocoded residence address and other demographic variables. These data, combined in a manner to protect confidentiality, is then linked with built and social environment data from the Colorado Department of Public Health and Environment (CDPHE) and public data sources collected by the University of Colorado Denver (UCD). The surveillance system utilizes the HMORN Virtual Data Warehouse (VDW) data framework, including the VDW vitals and enrollment tables. Results The BMI database will be available through a regional data sharing network, moving data from participating sites to a central data coordinating center (CDPHE) and to UCD for mapping and analysis. The network will enable users to both share data and perform queries within a single software environment. We intend to use the data model to track patients’ BMI over time and by county, census tract and block group geographies, and link BMI data with built and social environment data. We also intend to generate maps of BMI by census tract and block group and overlay built and social environment data to explore correlations between BMI and environmental factors. Conclusions Using the underlying framework of the VDW, we have created a multi-site regional data sharing network in Colorado for tracking individual patient-level BMI data, overlaid with built and social environment information.

The Spatial Distribution of Adult Obesity Prevalence in Denver County, Colorado: An Empirical Bayes Approach to Adjust EHR-Derived Small Area Estimates

Objectives: Measuring obesity prevalence across geographic areas should account for environmental and socioeconomic factors that contribute to spatial autocorrelation, the dependency of values in estimates across neighboring areas, to mitigate the bias in measures and risk of type I errors in hypothesis testing. Dependency among observations across geographic areas violates statistical independence assumptions and may result in biased estimates. Empirical Bayes (EB) estimators reduce the variability of estimates with spatial autocorrelation, which limits the overall mean square-error and controls for sample bias. Methods: Using the Colorado Body Mass Index (BMI) Monitoring System, we modeled the spatial autocorrelation of adult (≥ 18 years old) obesity (BMI ≥ 30 kg m2) measurements using patient-level electronic health record data from encounters between January 1, 2009, and December 31, 2011. Obesity prevalence was estimated among census tracts with >=10 observations in Denver County census tracts during the study period. We calculated the Moran’s I statistic to test for spatial autocorrelation across census tracts, and mapped crude and EB obesity prevalence across geographic areas. Results: In Denver County, there were 143 census tracts with 10 or more observations, representing a total of 97,710 adults with a valid BMI. The crude obesity prevalence for adults in Denver County was 29.8 percent (95% CI 28.4–31.1%) and ranged from 12.8 to 45.2 percent across individual census tracts. EB obesity prevalence was 30.2 percent (95% CI 28.9–31.5%) and ranged from 15.3 to 44.3 percent across census tracts. Statistical tests using the Moran’s I statistic suggest adult obesity prevalence in Denver County was distributed in a non-random pattern. Clusters of EB obesity estimates were highly significant (alpha=0.05) in neighboring census tracts. Concentrations of obesity estimates were primarily in the west and north in Denver County. Conclusions: Statistical tests reveal adult obesity prevalence exhibit spatial autocorrelation in Denver County at the census tract level. EB estimates for obesity prevalence can be used to control for spatial autocorrelation between neighboring census tracts and may produce less biased estimates of obesity prevalence.

Response to Letter to the Editor: “Questions Regarding Study Design and Analysis in ‘Previous Suicide Attempt and its Association With Method Used in a Suicide Death’”

We are in receipt of the letter “Questions Regarding Study Design and Analysis in ‘Previous Suicide Attempt and its Association With Method Used in a Suicide Death,’” by Lefebvre et al., which highlights some concerns with our previously published article cited in the title. We acknowledge the concerns they raised, and accept the provided input as intended to improve the quality of the research and applicability of results. Our responses to the points raised in this letter are as follows. The concern about the study design is understandable, given the unique design of the study and data source used. However, we believe the designation of the study as a retrospective cohort, and associated analytic approach, is appropriate. From among all suicide decedents eligible for the study (the cohort), we defined exposure as those having a documented history of non-fatal suicide attempt, and the unexposed as those without such a history. The cohort was then monitored to determine the outcome and method used in eventual suicide death. In response to the concern about our analysis on gender as an effect modifier, we felt that, based on the known patterns around gender and method of suicide, gender would have a differential effect on both the exposure and outcomes of interest. The results suggest that reducing access to all lethal means is important for all at-risk people. Particularly, it speaks to the cultural phenomena that men more often die by suicide via firearm injury, so healthcare practitioners and family members of suicidal men may be prone to focus on reducing access to a firearm. Our analysis reveals that it is important to reduce access to potential poisons as well, specifically when an individual has attempted suicide in the past. The results do not suggest that prioritizing reduction in access to lethal means for one gender over the other may be more efficacious. Rather, as elucidated in our Discussion section, the results promote the inclusion of counseling around potentially fatal poisons for all those caring for suicidal individuals. The layout of Table 1 was a design choice made early in manuscript development, and does make more sense when looking at it through the lens of the intended retrospective cohort study design. Specifically, with regard to the exposure of interest—history of suicide attempt or no history—and other decedent characteristics, understanding the distribution of means of eventual suicide death is key to informing which crude associations to assess and in justifying the research and multivariate analyses central to the research question.