martes, 10 de diciembre de 2019

Using Geospatial Analyses of Linked Electronic Health Records and Tobacco Outlet Data to Address the Social Determinants of Smoking

Using Geospatial Analyses of Linked Electronic Health Records and Tobacco Outlet Data to Address the Social Determinants of Smoking

PCD logo



Using Geospatial Analyses of Linked Electronic Health Records and Tobacco Outlet Data to Address the Social Determinants of Smoking

GIS SNAPSHOTS — Volume 16 — November 14, 2019

Scott D. Siegel, PhD, MHCDS1,2; Madeline M. Brooks, MPH1; Bayo M. Gbadebo, MBA1; James T. Laughery1 (View author affiliations)

Suggested citation for this article: Siegel SD, Brooks MM, Gbadebo BM, Laughery JT. Using Geospatial Analyses of Linked Electronic Health Records and Tobacco Outlet Data to Address the Social Determinants of Smoking. Prev Chronic Dis 2019;16:190186. DOI: http://dx.doi.org/10.5888/pcd16.190186external icon.
PEER REVIEWED
On This Page

Two maps of New Castle County, Delaware depict, at the census tract level, tobacco outlet density as the number of tobacco outlets per 1,000 adults, alongside a ratio of current to former smokers identified from a hospital-based population. Comparison shows that census tracts with higher tobacco outlet density tend to have more current smokers relative to former smokers, especially in northeast New Castle County. This effect is pronounced in the city of Wilmington, which has a higher tobacco outlet density and ratio of current to former smokers compared to the county as a whole.
Two maps of New Castle County, Delaware depict, at the census tract level, tobacco outlet density as the number of tobacco outlets per 1,000 adults, alongside a ratio of current to former smokers identified from a hospital-based population. Comparison shows that census tracts with higher tobacco outlet density tend to have more current smokers relative to former smokers, especially in northeast New Castle County. This effect is pronounced in the city of Wilmington, which has a higher tobacco outlet density and ratio of current to former smokers compared to the county as a whole. [A text description of this figure is available.]
Top

Background

The medical community now widely recognizes that social determinants of health (SDOH) have a significant effect on population health (1). If health systems are to move beyond a focus on traditional clinical services to successfully addressing SDOH, new models of care delivery are needed to guide this transition. Toward that end, the Centers for Disease Control and Prevention (CDC) developed a conceptual framework to facilitate collaboration between health systems and public health practitioners. This framework prioritizes increasing the adoption of clinical care that has a strong public health evidence base, innovating new forms of care delivery outside of clinical settings, and implementing community-wide interventions (2). As a first step toward operationalizing this framework, health systems can look beyond the “walls of the hospital” to develop an understanding of SDOH within their communities. By linking patients’ electronic health record (EHR) clinical data to area-level measures (3), geospatial analyses can be employed to inform the development of novel interventions that address local SDOH.
To illustrate the potential of this approach, we linked patient-level smoking status and address data from a Mid-Atlantic health system EHR to local area–level SDOH measures. Despite notable successes in tobacco control over the last 50 years or more, smoking remains the leading preventable cause of morbidity and mortality in the US, with slower declines in smoking rates for people of low socioeconomic status (4). Hospital-based smoking cessation interventions are effective (5); however, people of low socioeconomic status are more likely to reside in neighborhoods with more tobacco outlets and other challenges (6), which can undermine smoking cessation efforts (7). Therefore, we included area-level measures of socioeconomic status and tobacco outlet exposure as indicators of the social determinants of smoking status.
Top

Data Sources and Map Logistics

We included all adult patients with a smoking history admitted to Christiana Care Health System (CCHS) in New Castle County, Delaware, from January 1, 2015, through June 30, 2018, who resided in New Castle County. Smoking status (current, former, never) was assessed through a standardized interview administered at admission by the patient’s inpatient nurse and documented in the EHR. Patient addresses were cleaned and geocoded using ArcGIS 10.6 (Esri) (match rate = 94%). To adjust for geographic variation in the number of patients with a smoking history who were hospitalized at CCHS, a ratio of current to former smokers was calculated for each census tract. Tobacco outlet addresses were obtained from a public state business license database (8) and geocoded to calculate tobacco outlet density (TOD) for each census tract (number of tobacco outlets/1,000 adults). Patient-level tobacco outlet exposure was calculated as each patient’s Euclidean distance (in miles) to the nearest tobacco outlet and the number of tobacco outlets within a half-mile radius of their address. American Community Survey data provided census tract poverty and race/ethnicity characteristics (9). Census tracts were classified as high poverty if the percentage of residents who lived below the poverty line was equal to or greater than the 75th percentile and predominant minority if a nonwhite racial or ethnic group constituted the highest proportion of the tract population. Choropleth maps were created for smoker ratios and TOD by using natural breaks classification. Independent Samples t tests and χ2 tests were used to compare TOD by census tract classification and current versus former smokers on demographic, neighborhood, and tobacco outlet exposure characteristics.
Top

Highlights

These maps depict a strong association between smoking status and tobacco outlet exposure, particularly when contrasting Wilmington to the county at large. Of the 22,112 patients included in this analysis, 9,303 (42%) were current and 12,809 (58%) were former smokers, yielding a current-to-former-smoker ratio of 0.73 across the county’s 130 census tracts. The smoker ratio across Wilmington’s 25 census tracts was 1.33, 82% higher than the county. That is, for every 100 former smokers, there were 133 current smokers in Wilmington and 73 in the county overall. Similarly, Wilmington’s TOD was more than double the county overall (3.51 vs 1.62). Notably, CCHS was located in the census tract with the greatest TOD in the county and in proximity to 42 tobacco outlets within a half-mile radius. More generally, TOD was significantly higher in census tracts classified as high poverty (3.37 vs 1.42, P < .001), predominant minority (3.21 vs 1.56, P < .001), and both high poverty and predominant minority (3.45 vs 1.61, P < .001). Compared with former smokers, current smokers were significantly younger and more likely to be male and racial/ethnic minorities and to reside in census tracts characterized as high poverty, predominant minority, or both (Table). Furthermore, compared with former smokers, current smokers lived, on average, 0.10 miles closer to the nearest tobacco outlet, approximately the equivalent of 2 Wilmington city blocks, and in proximity to >70% more tobacco outlets within a 1/2-mile radius of their address.
These findings extend prior community-based survey research on tobacco outlet exposure (6) to a health system population by linking area-level measures to patient-level EHR smoking history data. Greater tobacco outlet exposure can undermine smoking cessation efforts by increasing exposure to point-of-sale marketing and other smoking cues, easing access to cigarettes, and contributing to pro-smoking attitudes (7). Hospitalization represents an opportunity for smokers to make a quit attempt in a smoke-free environment with ready access to treatment. Unfortunately, returning to a neighborhood with high tobacco outlet exposure upon discharge can increase the likelihood of relapse.
The primary limitation of this snapshot is that it cannot support causal inferences. Without quit dates and contemporaneous addresses, it is unclear whether former smokers from this population were more likely to quit when living in neighborhoods with lower tobacco outlet exposure and higher socioeconomic status. Temporal considerations aside, greater TOD may contribute to higher rates of smoking, more demand for cigarettes may drive greater TOD, or “third variables” such as disinvestment may contribute to both the greater use of cigarettes to cope with living in low-socioeconomic status areas and more lenient zoning standards regarding tobacco outlets.
Top

Action

This snapshot can stimulate novel smoking cessation initiatives aligned with the CDC framework. At the clinical level, the results underscore the importance of ensuring access to evidence-based smoking cessation interventions given the SDOH many smokers face. In addition, behavioral interventions specifically designed to reduce reactivity to smoking cues (10) may prove uniquely beneficial for smokers with greater tobacco outlet exposure. Future research can evaluate whether such adjuvant treatments improve quit rates for these patients. At the community level, the fact that one of the CCHS hospitals was located in the census tract with the greatest TOD in the county would support extending smoking cessation programming to quite literally just outside the walls of the hospital (eg, partnering with local community organizations, deploying community health workers). At the population level, further research is needed to evaluate whether regulating TOD reduces smoking rates in low-socioeconomic status neighborhoods (7). Partnering with the local health department and advocacy organizations may facilitate such evaluation efforts. Taken together, this snapshot portrays how linking EHR and area-level data can guide more effective collaborations between health systems and public health practitioners to address the SDOH.
Top

Acknowledgments

This project was supported by the Delaware INBRE program, with a grant from the National Institute of General Medical Sciences – NIGMS (P20 GM103446) from the National Institutes of Health and the State of Delaware.
Top

Author Information

Corresponding Author: Dr. Scott D. Siegel, 4755 Ogletown-Stanton Road, 8E17, Newark, DE 19718. Telephone: 302-733-4380. Email: ssiegel@christianacare.org.
Author Affiliations: 1Value Institute, Christiana Care Health System, Newark, Delaware. 2Helen F. Graham Cancer Center and Research Institute, Christiana Care Health System, Newark, Delaware.
Top

References

  1. Adler NE, Glymour MM, Fielding J. Addressing social determinants of health and health inequalities. JAMA 2016;316(16):1641–2. CrossRefexternal icon PubMedexternal icon
  2. Auerbach J. The 3 buckets of prevention. J Public Health Manag Pract 2016;22(3):215–8. CrossRefexternal icon PubMedexternal icon
  3. Schinasi LH, Auchincloss AH, Forrest CB, Diez Roux AV. Using electronic health record data for environmental and place based population health research: a systematic review. Ann Epidemiol 2018;28(7):493–502. CrossRefexternal icon PubMedexternal icon
  4. National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health. The health consequences of smoking — 50 years of progress: A Report of the Surgeon General. Atlanta (GA): Centers for Disease Control and Prevention; 2014, p. 1–36.
  5. Rigotti NA, Clair C, Munafò MR, Stead LF. Interventions for smoking cessation in hospitalised patients. Cochrane Database Syst Rev 2012;16(5):CD001837. CrossRefexternal icon PubMedexternal icon
  6. Rodriguez D, Carlos HA, Adachi-Mejia AM, Berke EM, Sargent JD. Predictors of tobacco outlet density nationwide: a geographic analysis. Tob Control 2013;22(5):349–55. CrossRefexternal icon PubMedexternal icon
  7. Cantrell J, Anesetti-Rothermel A, Pearson JL, Xiao H, Vallone D, Kirchner TR. The impact of the tobacco retail outlet environment on adult cessation and differences by neighborhood poverty. Addiction 2015;110(1):152–61. CrossRefexternal icon PubMedexternal icon
  8. Delaware Department of Finance, Division of Revenue. Delaware business licenses. Delaware Open Data. https://data.delaware.gov/Licenses-and-Certifications/Delaware-Business-Licenses/5zy2-grhr. Published 2018. Accessed October 29, 2018.
  9. U.S. Census Bureau. 2012-2016 American Community Survey 5-Year Estimates. https://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml. Published 2016. Accessed August 31, 2018.
  10. Brandon TH, Unrod M, Drobes DJ, Sutton SK, Hawk LW, Simmons VN, et al. Facilitated extinction training to improve pharmacotherapy for smoking cessation: a pilot feasibility trial. Nicotine Tob Res 2018;20(10):1189–97. CrossRefexternal icon PubMedexternal icon

Publicado por en

No hay comentarios:

Publicar un comentario

Suscribirse a: Enviar comentarios (Atom)