Data Talks: Obesity-Related Influences on US Mortality Rates

Background
In the US, obesity is an epidemiologic challenge and the population fails to comprehend this complex public health issue. To evaluate underlying obesity-impact patterns on mortality rates, we data-mined the 1999-2016 Center for Disease Control WONDER database's vital records.


Methods
Adopting SAS programming, we scrutinized the mortality and population counts. Using ICD-10 diagnosis codes connected to overweight and obesity, we obtained the obesity-related crude and age-adjusted causes of death. To understand divergent and prevalence trends we compared and contrasted the tabulated obesity-influenced mortality rates with demographic information, gender, and age-related data.


Key Results
From 1999 to 2016, the obesity-related age-adjusted mortality rates increased by 142%. The ICD-10 overweight and obesity-related death-certificate coding showed clear evidence that obesity factored in the male age-adjusted mortality rate increment to 173% and the corresponding female rate to 117%. It also disproportionately affected the nation-wide minority population death rates. Furthermore, excess weight distributions are coded as contributing features in the crude death rates for all decennial age-groups.


Conclusions
The 1999-2016 data from ICD-10 death certificate coding for obesity-related conditions indicate that it is affecting all segments of the US population.


Introduction
The span of the United States (US) adult obesity rate has arrived at a point where its colossal consequential risks are associated with burdensome public health concerns (Abarca-Gómez et al., 2017;www.scholink.org/ojs/index.php/rhs Research in Health Science Vol. 3, No. 3, 2018 66 Published by SCHOLINK INC. Wyatt et al., 2006). Major peer-reviewed articles have documented that the US population has basic understanding and awareness of obesity's incidence rate and its proportional detrimental health prevalence (Rosenthal et al., 2017). What's more, the cost of managing obesity within the overall economy is well cited and understood (Colditz, 1999;Guarneri, 2017).
Several longitudinal studies have directly correlated the adverse obesity impacts on human health Fitzmaurice et al., 2017;Probst et al., 2004). Body-mass index (BMI) assessments are shown to be both, consequential (Klatsky et al., 2017;Preston et al., 2018), and as being less important Kennedy et al., 2018). However, several extensive investigations have made clear that obesity does lead to persistent, long-standing, non-communicable health conditions that require sustained medical interventions (Despré s et al., 2001;Katz et al., 2000;Poulain et al., 2008;Wahba et al., 2007).
For official morbidity and mortality statistics, the Center for Disease Control (CDC) Wide-ranging Online Data for Epidemiologic Research (WONDER) database (Friede et al., 1993), presents descriptive tabulations of the US populations vital information incorporating the International Classification of Diseases 10 th Revision (ICD-10) coding (Bowman, 2008;Weiner, 2018). Due to healthcare's cumbersome workflow nature, inaccuracies in ICD-10 coding with misrepresentation in patient data (Lebovitz et al., 2018) are known to occur. Nonetheless, sensitivities of ICD-10 coding for overweight and obesity have proven to be reasonably accurate and helpful (George et al., 2018;Flegal et al., 2007;Keller et al., 2018;Yang et al., 2017). In addition, studies that assessed the accuracy of the ICD-diagnostic codes determined their conspicuous usefulness in empirical research (O'Malley et al., 2005), and in chronic disease management programs (Krueger et al., 2001;Vandyk et al., 2015).
To seek viable solutions that tackle the various aspects of adult obesity, big-data approaches are being employed (Barrett et al., 2013;Cutler et al., 2003;Jin et al., 2016;Wang et al., 2007). This real-time data-abundance revolution has positively impacted public health (Khoury et al., 2014;Murdoch et al., 2013) and public health policy (Athey, 2017;Beam et al., 2018). As a result, the university community has escalated its graduate and undergraduate data-science curriculum to include big-data analysis tools to advance the understanding of the trends in social lifestyle diseases (Elgin et al., 2017;De Veaux et al., 2017).

Talent, and Success (UR-CATS).
Using open-source data coupled with ArcGIS and SAS techniques, Wesley undergraduates have shown  that US obesity rates were directly associated with the social inequalities of income and education. This same study also showed that age and education were key obesity risk factors within the State of Delaware. Furthermore, a senior honors capstone project (D'Souza et al., 2017) analyzed the 1999-2014 CDC WONDER Delaware death records where an ICD-10 code for overweight and obesity was listed as a contributing factor and found that the 2014 age-adjusted Delaware obesity-related mortality rates increased by 28.7%.
Acknowledging the ICD-10 coding limitations (Lebovitz et al., 2018), but spurred on by its merits in obesity reports and studies (D'Souza et al., 2017;George et al., 2018, Flegal et al., 2007Keller et al., 2018;O'Malley et al., 2005;Yang et al., 2017), this observational retrospective undertaking, uses the CDC WONDER database (Friede et al., 1993), to evaluate the 1999-2016 multiple causes of death information, to pursue important insights, associations, and shifting trends in the US death rates where there was any mention of overweight and obesity on the death certificate. Information from such a study is helpful as it can enhance public perceptions on body fat for inducing positive, individual, behavioral change.

Methods
For all US residents, the CDC WONDER compressed mortality files (Friede et al., 1993) lists population counts, numbers of deaths, age-adjusted and crude death rates (at 95% confidence intervals), manner-of-death (4-digit ICD-10 code or group of codes), age groups, gender, and demographic data.
To gain access to this query-based database, requires a signed consent form. The entry to this rich source of statistical research data is described (in detail) elsewhere (D'Souza et al., 2017), and is through search/browse variables and the structured query language (SQL). For all 50 US States, the entire mortality dataset, pre-constructed subsets, or customized subsets of the data were downloaded into MS Excel and SAS files. The age-adjusted mortality rates per 100,000 are available in CDC WONDER and are calculated using Equation 1 (D'Souza et al., 2017).
Mortality rate for year i = (number of deaths in year i ÷ population in year i) × 100,000 (1) The age-adjusted mortality rates are statistical processes that are used to compare death rates in populations with differing age groups, while the annual crude death rates are ratio of deaths for any given age group. In this extensive 1999-2016 investigation of ICD-10 coded overweight and obesity-related counting of deaths, a total of 459,528 mortality records were compiled, of which

Results and Discussion
Figure 1 provides a 1999-2016 national baseline comparison for race and ethnicity, while Figure 2 depicts the nation-wide combined effects of gender variation in the obesity-mortality relationship. Both Figures 1 and 2, show dramatic rises in mortality rates where an ICD-10 code for overweight and obesity mentioned excess weight as a contributory factor on the death certificate.
Prior published data has showed that American Indian or Alaska Native adults have less access to public health resources and generally do not take part in leisure-time activities (Probst et al., 2004).
Amongst the four major racial groups profiled in Figure 1,   On the other hand, for the Black or African-American population, there is a serious consistent above-average annual deviation from the Figure 1 median (national rate) curvature for the age-adjusted mortality rates where excess abdominal fat was a contributing factor. The mortality rate jumped from a in line with expectations as nationally, we have continually failed to stem this groups economic instability and healthcare disparities D'Souza et al., 2017;Flegal et al., 2007;Katz et al., 2000;Poulain et al., 2008;Preston et al., 2018;Probst et al., 2004;Rosenthal et al., 2017;Wang et al., 2007;Wyatt et al., 2006).
The exceedingly lower-than-average orientation from the national reference (Figure 1) in overweight and obesity-related age-adjusted mortality rates for Asian or Pacific Islanders is not unexpected and can be rationally explained by existing epidemiological obesity-specific data and the consequential cultural differences between Asian-Americans (Wang et al., 2007 (Despré s et al., 2001;Flegal et al., 2007;Kennedy et al., 2018;Klatsky et al., 2017;Poulain et al., 2008;Wahba et al., 2007;Wang et al., 2007;Wyatt et al., 2006).
To further tease apart any sex-difference facets within the two dominant US races, Figures 3 and 4, with 95% confidence intervals, portray the very distinct gender variations observed for the White and the Black or African American races. In Figure 3, throughout the 1999-2016 time-period, a complete separation is observed between the 95% confidence intervals for the White male and White female ICD-10 overweight and obesity-coded mortality rates. In addition, from 1999 to 2016, White males saw a significant increase (176.03%) in overweight and obesity influenced mortality rates while for White females the corresponding rate was lower (129.36%), but the gap is increasing. Also, in around 2005, the age-adjusted obesity-related mortality rates for White men began to trend above the national average. Such a movement could epitomize previously recorded White gender-differentiated cultural norms (Abarca-Gómez et al., 2017;Cutler et al., 2003;Keller et al., 2018;Probst et al., 2004;Wyatt et al., 2006) and their psychological well-being perceptions with regards to weight (Guarneri, 2017;Jin et al., 2016;Kennedy et al., 2018;Klatsky et al., 2017;Wang et al., 2007). In Figure 4, the 1999-2014 ICD-10 coded overweight and obesity-related age-adjusted mortality rate separations for Black or African American females and males are far-above the national average and between the two genders, the 95% confidence intervals begin to converge in 2015-2016. Additionally, the 1999-2016 mortality rates for Black or African-American females is undeniably higher and is possibly parallel to existing cultural and/or economic benchmarks (Abarca-Gómez et al., 2017;Cutler et al., 2003;Kennedy et al., 2018;Klatzky et al., 2017;Wang et al., 2007), and the current dysfunction with access to healthcare resources Jin & Yu, 2016;Keller et al., 2018;Preston et al., 2018;Rosenthal et al., 2017;Wyatt et al., 2006). Nonetheless, the 1999 to 2016 overweight and obesity-influenced mortality rate for Black or African-American females largely increased by 72.83%, which pales in comparison to the exponential 165.34% increase in the corresponding age-adjusted mortality rates containing ICD-10 overweight and obesity diagnostic codes, for Black or African-American males.  First in Table 1 and then in Figure 5, the CDC WONDER age-adjusted mortality rate national data set where an ICD-10 overweight and obesity code is marked off on the death certificate, is organized according to decennial age-groups. Surprisingly in 2016, for the youngest (15-24) and oldest (85+) age groups, there was a >95% (95.45% and 95.55%, respectively) increase in the crude death rates where being overweight and obese are listed as contributory factors. The 2016 data for the other six age groups, show surging crude death rates where the role of excess weight is noted on the death record.

Conclusions
Analyzing the 1999-2016 CDC WONDER mortality data when an ICD-10 overweight and obesity-code is flagged, identifies the divergence in the accrual of mortality rate trends for race and ethnicity, gender, and age-groups. The gender gap for age-adjusted mortality rates where excess weight is an indicated contributory factor, is widening. More pronounced racial differences from the national average age-adjusted mortality rate pattern, gives a compelling indication that disturbing obesity-related declines in overall health is complicated but prevalent in the underrepresented populations. The rates have also demonstrated that the cumulative economic and healthcare disadvantages for Black or African-American females and males has made obesity to be the underappreciated driver that systematically grows their longer-term age-adjusted mortality rate trend to be well-above the national mean. Big increases in the crude mortality rates were observed in the middle-age groups where broadening weight-related challenges are responsible for key chronic health conditions.
We believe that this examination of obesity at the time of death is likely yielding a marked underestimate of the impact of obesity on mortality. However, the statistics presented provide compelling evidence for the compounding unhealthy nature of excess abdominal body fat on chronic health conditions. Such outcomes provide useful information to improve public perceptions on weight control disparities and promotes constructive discussions to frame the importance of beginning proactive initiatives to reduce weight for optimum health.