Abstract
Indicators of a poorer food environment such as low food environment index scores, food deserts, and food swamps have been linked to increased mortality, and decreased survival among colorectal cancer (CRC) cases.
This retrospective study of 47,807 primary CRC cases from the Kentucky Cancer Registry (KCR) database diagnosed between January 1995 and December 2021 used Cox-proportional-hazard (Cox-PH) regression models to estimate adjusted hazard ratios (aHRs) with 95% confidence intervals (CIs) for CRC-specific mortality and all-cause mortality associated with residential fast-food exposures.
Residential fast-food exposure was not associated with CRC-specific mortality. However, residing within 0.5-mile of a fast-food outlet was associated with a 7% higher hazard of all-cause mortality (aHR: 1.07; 95%CI: 1.02–1.11; P-for trend: 0.001) compared with those living more than 3 miles away. Similar associations for all-cause mortality were observed for higher fast-food coverage (aHR: 1.05; 95%CI: 1.02–1.07; P-value: 0.001) and fast-food density (aHR: 1.05; 95%CI: 1.02–1.07; P-value: 0.001) within a 1-mile radius.
Residential fast-food exposures such as closer proximity, higher fast-food coverage and greater fast-food density may contribute to a modest increase in all-cause mortality among individuals diagnosed with CRC. These findings underscore the potential role of neighborhood food environments in the survival of CRC patients.
Keywords: Fast-food proximity, fast-food density, fast-food coverage, colorectal cancer (CRC)-specific mortality, all-cause mortality, Kentucky Cancer Registry (KCR)
Introduction
Colorectal cancer (CRC) is the third most diagnosed cancer and the second leading cause of cancer-related death worldwide1. In the United States (US), CRC was the fourth most common cancer based on incident and mortality rates (excluding non-melanoma skin cancers) with an age-standardized incidence rate of 36.7 CRC cases per 100,000 population in 2022 and an age-standardized mortality rate of 12.7 CRC deaths per 100,000 population in 20232. Annual new CRC cases and CRC deaths are also expected to increase by 20403. Despite being preventable4, CRC still remains a major public health concern due to its high treatment costs5, negative impact on quality of life6, and increasing rate of premature mortality, particularly among young adults7.
Dietary behaviors are known to influence the risk of both overall mortality and CRC-specific mortality8–12. Food environment plays a critical role in shaping an individual’s overall health and dietary behaviors13 by influencing the food purchases and consumption decisions based on food accessibility14. Food environment such as food deserts (i.e., residential areas with limited access to healthy foods) and food swamps (i.e., residential areas with higher access to unhealthy foods) was associated with poor survival outcomes among CRC cases15–17. Studies have suggested a strong positive association of residential fast-food exposures such as fast-food proximity and fast-food density with increased consumption of fast-food meals18–21. Additionally, a positive association between more frequent consumption of fast-food items and increased odds of CRC has been reported22–26. Studies also observed an association of increased consumption of fast-food items or meals prepared away from home with an increased risk of all-cause mortality and cancer mortality27,28. However, most prior studies have examined outcomes such as body weight or body mass index (BMI)29,30, mortality from cardiovascular disease (CVD) or stroke31–33, and all-cause mortality34. To date, no epidemiological studies in the US have specifically investigated whether residential fast-food exposures influence CRC-specific mortality or all-cause mortality.
Investigating the role of food environment such as residential fast-food exposure is essential in identifying the modifiable risk factors that might contribute to CRC-specific or all-cause mortality, particularly in the regions with disproportionately high CRC outcomes and higher fast-food exposure. Food deserts and food swamps, an indicator of geographical food environment, are highly prominent in the South or Southeastern region of the US17. Kentucky (KY), a southeastern state, reported the highest age-standardized incidence rate of CRC with 46.1 CRC cases per 100,000 population in 2022 and the 2nd highest age-standardized mortality rate of CRC with 17.5 CRC deaths per 100,000 population in 202335. In this study, we examined the association between residential fast-food exposures and CRC-specific and all-cause mortality among individuals diagnosed with CRC in Kentucky.
Methods
Study population
Information on incident cases of CRC diagnosed in Kentucky between January 1, 1995 and December 31, 2021 was obtained from the Kentucky Cancer Registry (KCR). The KCR is a population-based central cancer registry. Mortality statistics by KCR are updated annually through synchronization with the Kentucky Department of Vital Statistics, the National Death Index, and other sources36. KCR has been part of the National Cancer Institute's (NCI) Surveillance Epidemiology and End Results (SEER) program since 2000 and is also a member of the Centers for Disease Control and Prevention's (CDC) National Program of Cancer Registries37.
This retrospective study included CRC cases aged 18 or older from the KCR database diagnosed between January 1, 1995 and December 31, 2021. A total of 66,139 primary CRC cases were identified from the KCR database. CRC cases with unknown or missing information on the CRC stage (n=3,369), cause of death (n=940), and date of CRC diagnosis (n=1) were excluded. CRC cases with geocoded addresses derived from street intersections, mid-point of streets, ZIP code, city or county centroids, or manual lookups were excluded (n=10,863) as these coordinates were approximate or imprecise that will lead to inaccurate distances when estimating residential fast-food exposures. Finally, CRC cases with ≤1 month of follow-up time between the date of diagnosis and date of last contact or death (n=3,066), whose proximity to fast-food restaurants could not be estimated (n=91), and with missing urban/rural information (n=2) were also excluded. The final analytic data set contained 47,807 eligible KCR CRC cases (Figure S1). This study was reviewed and determined to be exempt by the University of Louisville Institutional Review Board.
Endpoints: CRC-specific and all-cause mortality
The primary outcome of interest was CRC-specific mortality. Vital status (dead or alive) for each CRC case was obtained from the KCR, and the overall survival time (OS) was calculated in months by subtracting the date of diagnosis from the date of last contact or death. CRC-specific cause of death was determined as the underlying cause of death by the KCR. CRC-related deaths prior to January 1, 1999 were determined using the International Classification of Diseases, Ninth Revision (ICD-9) codes for CRC cases (i.e.,153–154). CRC-related deaths after January 1, 1999 were determined using ICD-10 codes for CRC cases (i.e., C18–C20)38. For CRC-specific mortality analyses, CRC cases were censored at the date of last contact or death from other causes. As a secondary analysis, we examined the associations with all-cause mortality which was defined as any death within the study period. For all-cause mortality analyses, CRC cases were censored at the date of last contact or death.
Fast-food restaurant data source and definition
Addresses of all the food establishments in Kentucky until 2021 were obtained from the Division of Public Health Protection and Safety, Kentucky Department for Public Health. Using the combinations of previously used criteria29,39,40, fast-food restaurants were defined as those food establishments that: (i) have outlets nationwide or in multiple states and have multiple outlets within the state of Kentucky, (ii) serve complete meal orders without the assistance of waiters or waitresses39 or counter served, (iii) provide facilities for customers to consume their meals immediately on site, taken out, or delivered to the customer’s location39,40, and (iv) make customers pay before eating29,40.
Using the fast-food restaurant definition and number of outlets, we identified the top 25 fast-food restaurant in KY with 1,989 outlets that were opened in or after 1982 and were still operating until 2020. We did not include fast-food restaurants that were closed between 1982 to 2021 as information on dates of closure was not available. Using ArcGIS Pro software version 3.3.0 (ESRI, Redlands, California), all the identified fast-food restaurants’ addresses were geocoded to obtain the latitude and longitude coordinates. Fast-food outlets with unmatched geocoded addresses (n=22) or with airport-based addresses (n=1) were excluded, resulting in a total of 1,966 fast-food outlets within KY for final exposure estimation. Subway and McDonald’s accounted for approximately 25% of total fast-food outlets included in the analysis (Table S1).
Proximity to a fast-food outlet, coverage and density of fast-food restaurant
The CRC cases’ geocoded residential addresses at the time of CRC diagnosis were linked to geocoded fast-food outlet locations for proximity estimation. Among the active fast-food restaurants, only outlets with opening dates prior to the date of diagnosis for KCR CRC cases were included in exposure estimation. For instance, if a CRC case was diagnosed in 2010, only fast-food outlets established before 2010 were included.
The primary exposure of interest was proximity to a fast-food outlet, defined as the road network distance (in miles) from each CRC case’s residential address at the time of diagnosis to the nearest fast-food outlet in ArcGIS Pro software version 3.3.0 (ESRI, Redlands, California). For analytical purposes, fast-food proximity was categorized into four groups: ≤0.5 mile, >0.5–≤1.0 mile, >1.0– ≤3.0 miles, and >3.0 miles, consistent with categories used in prior literatures41.
Secondary exposures of interest included fast-food coverage (i.e., restaurant counts) and fast-food density (i.e., restaurant counts per square mile) within geographical buffers of 1 mile, 3 miles, and 5 miles radius along the road network distance of each CRC case’s residence. The designated buffers along the road network distance of each CRC case’s residence were created to determine fast-food coverage within a buffer by linking the location of fast-food outlets to the road network area. For analysis, the fast-food coverage variable was categorized as a binary variable into two levels: ≥1 outlet within a road network buffer or 0 outlets within a road network buffer for each of the designated geographic buffers around the CRC case’s residence. Similarly, the area along the road network distance of each CRC case’s residence within a designated buffers were obtained to calculate fast-food density. Fast-food density was calculated by dividing the number of fast-food outlets within a designated buffer with the corresponding service area of the buffer29,42,43. For analysis, we dichotomized the fast-food density variable based on the median value for each level of designated geographical buffers around the CRC case’s residence. All the analyses for exposure estimation were performed using ArcGIS Pro.
Covariates
Information on individual-level covariates was obtained from the KCR database. Sociodemographic variables included age at the time of CRC diagnosis, sex (male and female), race (White, Black, and Asian/others/unknown), and marital status (married, not married, and unknown). Health related behaviors included smoking status (ever smoker, never smoker, and unknown/missing) and health insurance status (not insured, private insurance, public insurance, and unknown insurance type/status). Tumor characteristics at the time of diagnosis included tumor stage (in-situ, localized, regional, and distant), and tumor site (proximal, distal, rectal, and unspecified); and treatment modalities such as surgery (yes and no), chemotherapy (yes and no), and radiation (yes and no).
Because individual-level data was not available on socioeconomic status (SES) or other CRC risk factors (e.g., obesity, physical inactivity, and alcohol consumption), we linked KCR data with the CDC's Population Level Analysis and Community Estimates (PLACES) (2020 release)44 and American Community Survey (ACS) five-year estimates (2006–2010)45 in order to obtain census-tract level estimates. The census-tract level estimates were based on 2010 decennial census tract boundaries. Information on census-tract level SES factors was obtained from ACS (2006–2010) that included the percentage of adults who graduated high school or higher45. Similarly, information on census-tract level lifestyle variables were obtained from the 2018 CDC PLACES, which included the percentage of physically inactive adults (i.e., no leisure-time physical activity among adults), the percentage of binge drinking among adults, and the percentage of obesity among adults44. Likewise, the rural and urban status was obtained from the Rural-Urban Commuting Area (RUCA) codes, a census-tract level composite measure of urbanicity developed by the U.S. Department of Agriculture. RUCA codes classify areas based on population density, urbanization, and daily commuting patterns46.
Statistical analyses
Descriptive analyses of the CRC cases characteristics were conducted according to categories of proximity to a fast-food outlet. Normally distributed continuous variables were summarized as mean ± SD (standard deviation); non-normally distributed continuous variables as median ± with interquartile range (IQR); and categorical variables as frequencies. Crude survival rates by categories of proximity to a fast-food outlet were estimated using the Kaplan–Meier (KM) method, and the differences between the survival curves were assessed using the log-rank test (Figure S2 and S3)47,48.
Crude and multivariable Cox-proportional hazards (Cox-PH) regression models were used to estimate hazard ratios (HRs) with 95% confidence intervals (CIs) of CRC-specific and all-cause mortality associated with proximity to a fast-food outlet. Data that were considered as missing values or unknown status in the KCR database were included as a separate category (i.e., missing/unknown) in all analyses. The P-value for trend was calculated by treating proximity to a fast-food outlet as ordinal value. The proportional hazard (PH) assumption was examined using Schoenfeld residuals49. For covariates that violated the PH assumption, either stratification was applied, or a product term of log(time) and the covariate was included into the model to account for non-proportionality50. Covariates such as census tract level estimates of obesity and diabetes were considered as mediators and therefore not included into the model. Based on the descriptive table findings (Table 1), the DAG helped to identify the strongest individual level confounders such as age, sex, race, marital status, urbanicity and smoking. These confounders were included in Model 1 because they were associated with both the outcome and the exposure and were not considered mediators. Similarly, strong CRC mortality predictors such as tumor stage, tumor site, surgery, chemotherapy, radiation, and health insurance status which did not vary across the exposure categories were included in a separate Model 2 to explain outcome variability. Finally, census-tract level estimates of educational attainment, binge alcohol drinking, and no leisure-time physical activity were considered confounders, using DAG approach, because they varied across the exposure categories and were included in Model 3. For consistency, the covariates used in the model for primary exposure were the same for the model with secondary exposures (i.e., fast-food coverage and fast-food density).
Table 1a:
Age-adjusted characteristics of colorectal cancer (CRC) cases from Kentucky Cancer Registry (KCR) database diagnosed between 1995–2021 by residential proximity to a fast-food outlet at diagnosis (n = 47,807).
| Proximity to a fast-food outlet | ||||
|---|---|---|---|---|
| ≤0.5 mile (n=4,924) | >0.5 – ≤1 mile (n=9,524) | >1 – ≤3 miles (n=16,946) | >3 miles (n=16,413) | |
| Mortality and Follow-up * | ||||
| Follow-up time (years), median (IQR) | 4.33 (1.58,9.46) | 4.67 (1.75,9.71) | 5.08 (1.92,10.92) | 4.92 (1.92,10.42) |
| Deaths, %(n) | 66 (3259) | 64 (6140) | 63 (10719) | 61 (9937) |
| CRC-specific deaths, %(n) | 28 (1394) | 28 (2668) | 28 (4678) | 29 (4721) |
| Individual-Level Characteristics | ||||
| Age at diagnosis (in years), mean (SD)* | 67.81 (13.70) | 67.38 (13.74) | 66.19 (13.39) | 64.94 (12.93) |
| Male, %(n) | 47 (2332) | 48 (4576) | 51 (8582) | 54 (8875) |
| Race, %(n) | ||||
| White | 89 (4398) | 88 (8412) | 89 (15156) | 98 (16110) |
| Black | 10 (469) | 11 (1007) | 10 (1656) | 2 (251) |
| Asian/Others/Unknown | 1 (57) | 1 (105) | 1 (134) | 0 (52) |
| Marital Status, %(n) | ||||
| Married | 37 (1815) | 41 (3925) | 46 (7727) | 53 (8664) |
| Not Marrieda | 42 (2076) | 38 (3612) | 31 (5334) | 28 (4676) |
| Unknown | 21 (1033) | 21 (1987) | 23 (3885) | 19 (3073) |
| Urbanicity, %(n) | ||||
| Urban Areas | 64 (3175) | 69 (6559) | 69 (11613) | 34 (5595) |
| Rural Areas | 36 (1749) | 31 (2965) | 31 (5333) | 66 (10818) |
Values are means (SD) or medians (IQR) for continuous variables; percentages or frequencies or both for categorical variables and are standardized to the age distribution of the study population. Values of categorical variables may not sum to 100% due to rounding.
Values are not age adjusted.
Single, unmarried/domestic partner, separated, divorced, or widowed.
Effect modification for potential effect modifiers such as sex and urbanicity were explored using likelihood ratio test. A two-sided p-value of <0.05 was considered statistically significant. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Inc., Cary, NC).
Results
KCR CRC case’s characteristics
A total of 47,807 individuals diagnosed with primary CRC were included in the current analysis with a median (IQR) follow-up time of 4.83 years (1.83–10.33). The mean (SD) age at diagnosis was 66.17 years (13.38) with 51% male, and 92% white individuals. During the study period, there were 30,055 (63%) overall deaths including 13,461 (28%) deaths attributed to CRC (Table S2). A total of 4,924 (10%) individuals resided within 0.5-mile of a fast-food outlet with a mean (SD) age of 67.81 years (13.70), while 16,413 (34%) individuals resided at greater than 3 miles from a fast-food outlet with a mean (SD) age of 64.94 years (12.93) (Table 1). Compared to those who resided at greater than 3 miles from a fast-food outlet, individuals who resided within 0.5-mile of a fast-food outlet were more likely to be female (53% versus 46%), Black (10% versus 2%), ever smoker (38% versus 35%), have public insurance (67% versus 66%), and to have lived in urban areas (64% versus 34%) and were less likely to be married (37% versus 53%) after age-adjustments (Table 1). Furthermore, those who resided within 0.5-mile of a fast-food outlet had lower median (IQR) survival time of 6.08 years (5.67–6.58) for all-cause mortality compared to those who resided at greater than 3 miles from a fast-food outlet (7.67 years; 7.33–7.92) (Table S3).
Residential fast-food exposures and CRC-specific mortality
We did not observe statistically significant associations between residential proximity to fast food restaurants and CRC-specific mortality. Individuals who resided within 0.5-mile of a fast-food outlet showed a non-significant 3% higher hazard for CRC-specific mortality (aHR: 1.03; 95%CI: 0.96–1.10; P-for trend: 0.51; Table 2) compared to individuals who resided greater than 3 miles from a fast-food outlet in a fully adjusted model (Model 3). Effect modification by sex (LRT p-value for interaction = 0.03) was observed, but not by urbanicity (LRT p-value for interaction = 0.66) for the association between fast-food proximity and CRC-specific mortality. Sex-stratified analysis showed that males who resided within 0.5-mile of a fast-food outlet (aHR: 1.08; 95%CI: 0.99–1,18; P-for trend: 0.71; Table S4) had a non-significant 8% higher hazard of CRC-specific mortality compared to males who resided at greater than 3 miles from a fast-food outlet in a fully adjusted model. Among females, no significant association between proximity to a fast-food outlet and CRC-specific mortality (aHR: 1.00; 95%CI: 0.92–1.09; P-for trend: 0.44; Table S4) was observed.
Table 2:
Crude and adjusted hazard ratios (95% confidence intervals) of colorectal cancer (CRC)-specific and all-cause mortality among CRC cases from Kentucky Cancer Registry (KCR) database diagnosed between 1995–2021 by residential proximity to a fast-food outlet at diagnosis (n = 47,807).
| Residential proximity to a fast-food outlet at diagnosis | n/Events | Crude (HR with 95%CI) |
Model 1a (HR with 95%CI) |
Model 2b (HR with 95%CI) |
Model 3c (HR with 95%CI) |
|---|---|---|---|---|---|
| CRC-Specific Mortality | 47,807/13,461 | ||||
| ≤0.5 mile | 4,924/1,394 | 1.03 (0.97–1.10) |
0.96 (0.91–1.02) |
1.03 (0.96–1.09) |
1.03 (0.96–1.10) |
| >0.5 – ≤1 mile | 9,524/2,668 | 0.99 (0.95–1.04) |
0.95 (0.90–0.99) |
0.99 (0.94–1.04) |
1.00 (0.95–1.05) |
| >1 – ≤3 miles | 16,946/4,678 | 0.95 (0.91–0.99) |
0.93 (0.89–0.97) |
0.99 (0.95–1.03) |
1.00 (0.95–1.04) |
| >3 miles | 16,413/4,721 | REF | REF | REF | REF |
| P-for trendd | 0.50 | 0.1 | 0.66 | 0.51 | |
| All-Cause Mortality | 47,807/30,055 | ||||
| ≤0.5 mile | 4,924/3,259 | 1.18 (1.14–1.23) |
1.03 (0.99–1.08) |
1.05 (1.00–1.09) |
1.07 (1.02–1.11) |
| >0.5 – ≤1 mile | 9,524/6,140 | 1.11 (1.08–1.15) |
1.00 (0.97–1.03) |
1.02 (0.98–1.06) |
1.04 (1.00–1.08) |
| >1 – ≤3 miles | 16,946/10,719 | 1.02 (0.99–1.04) |
0.96 (0.93–0.99) |
0.98 (0.95–1.01) |
1.01 (0.97–1.04) |
| >3 miles | 16,413/9,937 | REF | REF | REF | REF |
| P-for trendd | <.0001 | 0.15 | 0.02 | 0.001 |
Adjusted for age (years) at CRC diagnosis, sex (male and female), race (White, Black, and Asian/others/unknown), urbanicity (urban and rural areas), marital status (married, not married, and unknown), and smoking status (ever smoker, never smoker, and unknown/missing)
Adjusted for Model 1 covariates plus tumor stage (in-situ, localized, regional, and distant), tumor site (proximal, distal, rectal, and unspecified), surgery (yes and no), chemotherapy (yes and no), radiation (yes and no), and health insurance status (not insured, private insurance, public insurance, and unknown insurance type/status).
Adjusted for Model 2 covariates plus the percentage of the census tract level characteristics such as high school education or more, binge alcohol drinking, and no leisure-time physical activity.
Proximity to a fast-food outlet treated as ordinal value.
Similarly, fast-food coverage at ≤1 mile, ≤3 miles, and ≤5 miles geographical buffers were not significantly associated with CRC-specific mortality (Table 3). For the fast-food density analyses, there were no associations between fast-food density and CRC-specific mortality at the ≤1 mile and ≤3 miles buffers (Table 4); however, at the geographical buffer of ≤5 miles, fast-food density was associated with a modestly lower hazard of CRC-specific mortality (aHR: 0.95; 95%CI: 0.92–0.99; P-value: 0.007; Table 4).
Table 3a:
Crude and adjusted hazard ratios (95% Confidence Intervals) of colorectal cancer (CRC)-specific mortality among CRC cases from Kentucky Cancer Registry (KCR) database diagnosed between 1995–2021 by fast-food coverage (i.e., restaurant counts) within 1-, 3-, and 5- miles buffer area (n =47,807).
| Number of fast-food restaurants within a geographically defined buffer (i.e., restaurant counts) | n/Events | Crude (HR with 95%CI) |
Model 1a (HR with 95%CI) |
Model2b (HR with 95%CI) |
Model3c (HR with 95%CI) |
|---|---|---|---|---|---|
| CRC-Specific Mortality | |||||
| ≤1 mile buffer | |||||
| 0 outlet | 32,309/9,101 | REF | REF | REF | REF |
| ≥1 outlet | 15,498/4,360 | 1.03 (1.00–1.07) | 1.00 (0.96–1.03) | 1.02 (0.98–1.06) | 1.02 (0.98–1.05) |
| P-value | 0.07 | 0.79 | 0.39 | 0.44 | |
| ≤3 miles buffer | |||||
| 0 outlet | 16,214/4,675 | REF | REF | REF | REF |
| ≥1 outlet | 31,593/8,786 | 0.97 (0.94–1.00) | 0.93 (0.90–0.97) | 1.00 (0.96–1.04) | 1.00 (0.96–1.04) |
| P-value | 0.08 | 0.0004 | 0.87 | 0.98 | |
| ≤5 miles buffer | |||||
| 0 outlet | 11,551/3,386 | REF | REF | REF | REF |
| ≥1 outlet | 36,256/10,075 | 0.95 (0.91–0.99) | 0.92 (0.88–0.96) | 0.99 (0.95–1.03) | 0.99 (0.95–1.04) |
| P-value | 0.006 | <.0001 | 0.56 | 0.73 | |
Adjusted for age (years) at CRC diagnosis, sex (male and female), race (White, Black, and Asian/others/unknown), urbanicity (urban and rural areas), marital status (married, not married, and unknown), and smoking status (ever smoker, never smoker, and unknown/missing).
Adjusted for Model 1 covariates plus tumor stage (in-situ, localized, regional, and distant), tumor site (proximal, distal, rectal, and unspecified), surgery (yes and no), chemotherapy (yes and no), radiation (yes and no), and health insurance status (not insured, private insurance, public insurance, and unknown insurance type/status).
Adjusted for Model 2 covariates plus the percentage of the census tract level characteristics such as high school education or more, binge alcohol drinking, and no leisure-time physical activity.
Table 4a:
Crude and adjusted hazard ratios (95% Confidence Intervals) of colorectal cancer (CRC)-specific mortality among CRC cases from Kentucky Cancer Registry (KCR) database diagnosed between 1995–2021 by fast-food density (i.e., restaurant counts/miles2) within 1-, 3-, and 5- mile buffer area (n = 47,807).
| Fast-food density within a geographically defined buffer (i.e., restaurant counts/miles2) | n/Events | Crude (HR with 95%CI) |
Model 1a (HR with 95%CI) |
Model2b (HR with 95%CI) |
Model3c (HR with 95%CI) |
|---|---|---|---|---|---|
| CRC-Specific Mortality | |||||
| ≤1 mile buffer | |||||
| 0 outlet/mile2 | 32,311/9,101 | REF | REF | REF | REF |
| >0 outlet/mile2 | 15,496/4,360 | 1.03 (1.00–1.07) | 1.00 (0.96–1.03) | 1.02 (0.98–1.06) | 1.02 (0.98–1.05) |
| P-value | 0.07 | 0.80 | 0.39 | 0.43 | |
| ≤3 miles buffer | |||||
| ≤0.41 outlet/mile2 | 23,832/6,968 | REF | REF | REF | REF |
| >0.41 outlet/mile2 | 23,975/6,493 | 0.95 (0.92–0.99) | 0.94 (0.91–0.97) | 0.97 (0.94–1.01) | 0.97 (0.94–1.01) |
| P-value | 0.005 | 0.0003 | 0.15 | 0.13 | |
| ≤5 miles buffer | |||||
| ≤0.33 outlet/mile2 | 24,008/7,122 | REF | REF | REF | REF |
| >0.33 outlet/mile2 | 23,799/6,399 | 0.93 (0.90–0.96) | 0.94 (0.90–0.97) | 0.95 (0.91–0.98) | 0.95 (0.92–0.99) |
| P-value | <.0001 | 0.0003 | 0.005 | 0.007 | |
Adjusted for age (years) at CRC diagnosis, sex (male and female), race (White, Black, and Asian/others/unknown), urbanicity (urban and rural areas), marital status (married, not married, and unknown), and smoking status (ever smoker, never smoker, and unknown/missing).
Adjusted for Model 1 covariates plus tumor stage (in-situ, localized, regional, and distant), tumor site (proximal, distal, rectal, and unspecified), surgery (yes and no), chemotherapy (yes and no), radiation (yes and no), and health insurance status (not insured, private insurance, public insurance, and unknown insurance type/status).
Adjusted for Model 2 covariates plus the percentage of the census tract level characteristics such as high school education or more, binge alcohol drinking, and no leisure-time physical activity.
Residential fast-food exposures and all-cause mortality
Closer proximity to a fast-food outlet was significantly associated with a modest increase in rates of all-cause mortality. In a fully adjusted model (Model 3), individuals who resided within 0.5-mile and >0.5-≤1.0 mile of a fast-food outlet showed a higher hazard by 7% (aHR: 1.07; 95%CI: 1.02–1.11; P-for trend: 0.001) and 4% (aHR: 1.04; 95%CI: 1.00–1.08; P-for trend: 0.001), respectively, for all-cause mortality compared to individuals who resided at greater than 3 miles from a fast-food outlet (Table 2). We did not find evidence that the association between fast-food proximity and all-cause mortality differed by sex (LRT p-value for interaction = 0.25) or by urbanicity (LRT p-value for interaction = 0.34). We also observed that both fast-food coverage and fast-food density at geographical buffer of ≤1 mile was significantly associated with all-cause mortality. In a fully adjusted model, the hazard of all-cause mortality was 5% higher (aHR: 1.05; 95%CI: 1.02–1.07; P-value: 0.001; Table 3) for individuals with ≥1 fast-food outlet within 1 mile of their residence compared to individuals with no fast-food outlet within 1 mile. We also observed an increase in hazard of all-cause mortality by 5% for individuals with fast-food density of >0 outlet/mile2 within 1 mile of their residence compared to individuals with fast-food density of 0 outlet/mile2 within 1 mile after adjusting for Model 3 covariates (aHR: 1.05; 95%CI: 1.02–1.07; P-value: 0.001; Table 4).
Discussion
In this retrospective study of primary CRC cases in Kentucky, we observed that residing closer to a fast-food outlet was significantly associated with a modestly higher rate of all-cause mortality among CRC cases. Despite generally null association with CRC-specific mortality, sex-stratified analysis showed that males, but not females, living closer to a fast-food outlet showed a modest increase in CRC-specific mortality. Similarly, fast-food density and fast-food coverage at geographical buffers of ≤1 mile around CRC case’s residence showed a significant modest increase in hazard of all-cause mortality, but not with CRC-specific mortality. While fast-food coverage and fast-food density were not associated with CRC-specific mortality at buffers closer to one’s residence, we observed that higher fast-food density at the ≤5-miles buffer around CRC case’s residence was associated with a small decrease in hazard of CRC-specific mortality, which is likely due to chance or confounding.
We observed a modest increase in hazard of all-cause mortality associated with residing closer to a fast-food outlet. While no other epidemiological studies have been conducted on this topic to allow corroboration, studies have reported that increased consumption of red meat51,52, diets high in saturated fat53, and sodium54,55 (which are consistent with nutritional profile of foods from fast-food restaurants) were found to be associated with increased risk of overall mortality and cancer-specific mortality. Similarly, sugar-sweetened drinks (SSD), which are widely served in fast-food restaurants, have been suggested to increase all-cause mortality56 and worsen survival among CRC patients after diagnosis11,12. Some US-based studies have suggested an association with consumption of fast-food, or meals prepared away from home and all-cause mortality and death from cancer27,28, while other studies reported a null association between increased consumption of meals prepared away from home and western dietary pattern with all-cause mortality57,58.
Poorer food environment index (i.e., environment with high food insecurity and lower access to grocery stores) has been associated with higher colorectal cancer incidence and mortality59. Similarly, indicators of food environment such as food desert and food swamp which can be considered a proxy for fast-food coverage and fast-food density variables used in our study have been associated with worse survival among CRC patients15–17. A high food swamp score was associated with a higher age-adjusted odds of obesity-related cancer (including CRC) mortality17 and a higher odds of 90-day mortality16. Similarly, food desert residents showed a higher mortality risk among CRC patients15 and a higher odds of 90-day mortality16. Consistent with these studies, we observed that higher fast-food density and fast-food coverage within 1 mile buffer around CRC case’s residence showed a modest but significant increase in hazard of all-cause mortality. Similar to our study, a Canadian study of ecological cross-sectional design observed a significant association between higher odds of all-cause mortality (OR: 2.52; 95%CI: 1.54–4.13; P-value: <0.001) and fast-food density34. Likewise, other cross-sectional or ecological studies have also reported higher CVD mortality and mortality from other causes per 10% increase in fast-food density32. While literature on fast-food density and mortality acknowledge a role, the literature on fast-food coverage and mortality is sparse with most studies focusing on fast-food coverage and its relationships with body weight29,30 or risk of stroke33.
More people may die prematurely from consuming processed foods and fast foods than from smoking cigarettes60. There are several mechanisms that can possibly explain the association between residential fast-food exposures and mortality. Residential proximity to fast-food outlets, along with higher fast-food density and broader fast-food coverage, increases accessibility to these food sources, potentially leading to a higher likelihood of purchasing and consuming fast-food meals18–21. Given the nutritional quality of fast-food meals such as low nutritional value, high calories, red and processed meats, higher amount of unhealthy saturated fatty acids (SFAs), trans-fatty acids (TFAs), cholesterol, salt, and added refined sugar61–64, increased fast-food consumption can have adverse effects on the survival of individuals with a history of CRC65. Increased fast-food consumption due to increased accessibility to fast-food outlets contributes to mortality by promoting chronic conditions such as obesity and metabolic syndrome, which can significantly increase the risk of cardiovascular diseases, diabetes, and its associated complications29,30,66. In addition, fast-food outlets serve food items such as burgers that contains artificial food additives such as sweeteners, emulsifiers, thickeners, preservatives, and colours67. Epidemiological evidence suggests that higher intake of artificially sweetened beverages increases the risk of all-cause mortality68. Evidence also suggests a link between poor prognosis and pro-inflammatory diets among individuals after cancer diagnosis69. Carboxymethyl cellulose, a widely used dietary emulsifier, is suggested to promote inflammation by reducing gut microbial diversity70 while higher circulating concentrations of proinflammatory cytokines suggests poor prognosis in individual diagnosed with CRC71. Finally, increased fast-food consumption may also increase the risk of mortality from cancer through exposure to potential carcinogens such as N-nitroso compounds, heterocyclic amines (HCAs), and polycyclic aromatic hydrocarbons (PAHs)72–75.
This study should be interpreted considering several limitations. First, due to the nature of the population-based registries, post-diagnosis residential histories and duration of stay at the residence at the time of diagnosis are not captured and prohibited estimation of cumulative post-diagnosis. Residential moves after diagnosis may introduce misclassification of residential fast-food exposure after diagnosis; however, it has been estimated that approximately 85% of Kentuckians stay in the same residence for at least a year76. Second, information on the dietary patterns of individuals who reside closer to fast-food restaurants or in areas with greater density of fast-food restaurants was not known. Finally, given the observational nature of the study, residual confounding due to poorly measured or unknown confounders such as food habits cannot be ruled out. The present study design has several notable strengths. First, this study contributes to the limited body of literature on residential fast-food exposures and mortality risk at the individual-level and is a novel approach within population-based cancer registries that captures a large population of Kentucky. The residential fast-food exposures estimation was based on road network distance that is considered to reflect the accessible space in which the fast-food restaurants are located (i.e., along the road network)29 and aligns with common practices used in fast-food accessibility research42,43,77.
In conclusion, this study provides an important step toward understanding the potential impact of residential fast-food exposures on all-cause and CRC-specific mortality among individuals diagnosed with CRC. This study showed that living closer to a fast-food outlet, higher fast-food availability, and greater fast-food density within a mile buffer around CRC case’s residence was associated with a worse prognosis among CRC cases. Similarly, males, but not females, living closer to fast-food restaurants showed a non-significant modest increase in CRC-specific mortality. These findings support the need for further research into the contribution of environmental risk factors such as residential fast-food exposures impacting all-cause and CRC-specific mortality. Furthermore, identifying environmental risk factors such residential fast-food exposures that contribute to mortality can help inform policy decision making and the development of effective intervention strategies to reduce mortality.
Supplementary Material
Table 1b:
Age-adjusted characteristics of colorectal cancer (CRC) cases from Kentucky Cancer Registry (KCR) database diagnosed between 1995–2021 by residential proximity to a fast-food outlet at diagnosis (n = 47,807).
| Proximity to a fast-food outlet | ||||
|---|---|---|---|---|
| ≤0.5 mile (n=4,924) | >0.5 – ≤1 mile (n=9,524) | >1 – ≤3 miles (n=16,946) | >3 miles (n=16,413) | |
| Tumor Characteristics | ||||
| CRC Stage, %(n) | ||||
| In-situ | 7 (363) | 7 (708) | 8 (1359) | 7 (1108) |
| Localized | 52 (2562) | 52 (4928) | 51 (8690) | 51 (8335) |
| Regional | 24 (1172) | 24 (2284) | 24 (4132) | 25 (4165) |
| Distant | 17 (827) | 17 (1604) | 16 (2765) | 17 (2805) |
| CRC Sub-type, %(n) | ||||
| Proximal Colon Cancer | 38 (1892) | 41 (3899) | 39 (6685) | 38 (6252) |
| Distal Colon Cancer | 34 (1679) | 33 (3121) | 34 (5802) | 35 (5716) |
| Rectal Cancer | 23 (1110) | 22 (2052) | 22 (3647) | 23 (3698) |
| Unspecified | 5 (244) | 5 (452) | 5 (811) | 5 (747) |
| Treatment Characteristics | ||||
| Surgery, %(n) | 88 (4333) | 90 (8553) | 90 (15275) | 89 (14617) |
| Chemotherapy, %(n) | 36 (1787) | 36 (3467) | 37 (6208) | 39 (6335) |
| Radiation, %(n) | 13 (640) | 13 (1197) | 13 (2145) | 14 (2294) |
Values are means (SD) or medians (IQR) for continuous variables; percentages or frequencies or both for categorical variables and are standardized to the age distribution of the study population. Values of categorical variables may not sum to 100% due to rounding.
Table 1c:
Age-adjusted characteristics of colorectal cancer (CRC) cases from Kentucky Cancer Registry (KCR) database diagnosed between 1995–2021 by residential proximity to a fast-food outlet at diagnosis (n = 47,807).
| Proximity to a fast-food outlet | ||||
|---|---|---|---|---|
| ≤0.5 mile (n=4,924) | >0.5 – ≤1 mile (n=9,524) | >1 – ≤3 miles (n=16,946) | >3 miles (n=16,413) | |
| Health Behaviors and Risk Factors | ||||
| Insurance Status, %(n) | ||||
| Not Insured | 3 (148) | 2 (227) | 2 (367) | 3 (424) |
| Private Insurance | 20 (993) | 22 (2123) | 24 (4053) | 20 (3226) |
| Public Insurance | 67 (3280) | 65 (6172) | 63 (10603) | 66 (10790) |
| Unknown Insurance Type/Status | 10 (503) | 11 (1002) | 11 (1923) | 12 (1973) |
| Smoking History, %(n) | ||||
| Never User | 37 (1834) | 37 (3526) | 38 (6355) | 38 (6185) |
| Ever User | 38 (1862) | 37 (3518) | 36 (6128) | 35 (5743) |
| Unknown/Missing | 25 (1228) | 26 (2481) | 26 (4463) | 27 (4484) |
| Census-Tract level Characteristics, median (IQR) * | ||||
| % High School or More | 83.00 (76.75,88.90) | 83.80 (76.90,91.20) | 84.80 (77.30,91.60) | 78.00 (71.70,84.30) |
| % Adult Binge Alcohol Drinking | 15.00 (13.20,17.20) | 15.30 (13.50,17.60) | 15.30 (13.50,17.80) | 14.00 (13.10,15.00) |
| % Adult No Physical Activity | 33.40 (27.90,37.50) | 32.90 (27.20,37.70) | 31.60 (26.20,37.40) | 35.30 (31.70,39.00) |
| % Adult Obesity | 37.20 (32.80,40.40) | 36.80 (31.50,40.50) | 36.70 (31.90,40.50) | 39.30 (37.00,41.40) |
| % Adult Diabetes | 13.10 (11.10,15.20) | 13.00 (10.90,15.20) | 12.50 (10.60,14.90) | 14.10 (12.60,15.60) |
Values are means (SD) or medians (IQR) for continuous variables; percentages or frequencies or both for categorical variables and are standardized to the age distribution of the study population. Values of categorical variables may not sum to 100% due to rounding.
Values are not age adjusted.
Table 3b:
Crude and adjusted hazard ratios (95% Confidence Intervals) of all-cause mortality among CRC cases from Kentucky Cancer Registry (KCR) database diagnosed between 1995–2021 by fast-food coverage (i.e., restaurant counts) within 1-, 3-, and 5- miles buffer area (n = 47,807).
| Number of fast-food restaurants within a geographically defined buffer (i.e., restaurant counts) | n/Events | Crude (HR with 95%CI) |
Model 1a (HR with 95%CI) |
Model2b (HR with 95%CI) |
Model3c (HR with 95%CI) |
|---|---|---|---|---|---|
| All-Cause Mortality | |||||
| ≤1 mile buffer | |||||
| 0 outlet | 32,309/19,985 | REF | REF | REF | REF |
| ≥1 outlet | 15,498/10,070 | 1.12 (1.10–1.15) | 1.03 (1.01–1.06) | 1.04 (1.01–1.07) | 1.05 (1.02–1.07) |
| P-value | <.0001 | 0.01 | 0.004 | 0.001 | |
| ≤3 miles buffer | |||||
| 0 outlet | 16,214/9,826 | REF | REF | REF | |
| ≥1 outlet | 31,593/20,229 | 1.06 (1.04–1.09) | 0.98 (0.96–1.01) | 1.00 (0.98–1.03) | 1.03 (1.00–1.06) |
| P-value | <.0001 | 0.12 | 0.86 | 0.08 | |
| ≤5 miles buffer | |||||
| 0 outlet | 11,551/7,091 | REF | REF | REF | REF |
| ≥1 outlet | 36,256/22,964 | 1.03 (1.00–1.06) | 0.96 (0.94–0.99) | 0.98 (0.95–1.01) | 1.01 (0.98–1.04) |
| P-value | 0.05 | 0.007 | 0.23 | 0.58 | |
Adjusted for age (years) at CRC diagnosis, sex (male and female), race (White, Black, and Asian/others/unknown), urbanicity (urban and rural areas), marital status (married, not married, and unknown), and smoking status (ever smoker, never smoker, and unknown/missing).
Adjusted for Model 1 covariates plus tumor stage (in-situ, localized, regional, and distant), tumor site (proximal, distal, rectal, and unspecified), surgery (yes and no), chemotherapy (yes and no), radiation (yes and no), and health insurance status (not insured, private insurance, public insurance, and unknown insurance type/status).
Adjusted for Model 2 covariates plus the percentage of the census tract level characteristics such as high school education or more, binge alcohol drinking, and no leisure-time physical activity.
Table 4b:
Crude and adjusted hazard ratios (95% Confidence Intervals) of all-cause mortality among CRC cases from Kentucky Cancer Registry (KCR) database diagnosed between 1995–2021 by fast-food density (i.e., restaurant counts/miles2) within 1-, 3-, and 5- mile buffer area (n = 47,807).
| Fast-food density within a geographically defined buffer (i.e., restaurant counts/miles2) | n/Events | Crude (HR with 95%CI) |
Model 1a (HR with 95%CI) |
Model2b (HR with 95%CI) |
Model3c (HR with 95%CI) |
|---|---|---|---|---|---|
| All-Cause Mortality | |||||
| ≤1 mile buffer | |||||
| 0 outlet/mile2 | 32,311/19,985 | REF | REF | REF | REF |
| >0 outlet/mile2 | 15,496/10,070 | 1.12 (1.10–1.15) | 1.03 (1.01–1.06) | 1.04 (1.01–1.07) | 1.05 (1.02–1.07) |
| P-value | <.0001 | 0.01 | 0.004 | 0.001 | |
| ≤3 miles buffer | |||||
| ≤0.41 outlet/mile2 | 23,832/15,354 | REF | REF | REF | REF |
| >0.41 outlet/mile2 | 23,975/14,701 | 1.03 (1.01–1.05) | 0.99 (0.97–1.01) | 1.00 (0.97–1.02) | 1.02 (0.99–1.04) |
| P-value | 0.02 | 0.41 | 0.93 | 0.24 | |
| ≤5 miles buffer | |||||
| ≤0.33 outlet/mile2 | 24,008/15,965 | REF | REF | REF | REF |
| >0.33 outlet/mile2 | 23,799/14,090 | 0.99 (0.97–1.01) | 0.99 (0.96–1.01) | 1.00 (0.97–1.03) | 1.02 (0.99–1.05) |
| P-value | 0.30 | 0.28 | 0.90 | 0.17 | |
Adjusted for age (years) at CRC diagnosis, sex (male and female), race (White, Black, and Asian/others/unknown), urbanicity (urban and rural areas), marital status (married, not married, and unknown), and smoking status (ever smoker, never smoker, and unknown/missing).
Adjusted for Model 1 covariates plus tumor stage (in-situ, localized, regional, and distant), tumor site (proximal, distal, rectal, and unspecified), surgery (yes and no), chemotherapy (yes and no), radiation (yes and no), and health insurance status (not insured, private insurance, public insurance, and unknown insurance type/status).
Adjusted for Model 2 covariates plus the percentage of the census tract level characteristics such as high school education or more, binge alcohol drinking, and no leisure-time physical activity.
Acknowledgements
Data used in this publication was provided by the Kentucky Cancer Registry (KCR), Lexington, KY and Division of Public Health Protection and Safety, Kentucky Department for Public Health. We would like to thank the KCR, who provided the CRC case data and the Division of Public Health Protection and Safety, Kentucky Department for Public Health, who provided the fast-food data utilized in this study.
Source of Funding
No funding source to be disclosed.
Footnotes
Declaration of interest
No potential conflicts of interest relevant to this article were reported. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
CRediT authorship contribution statement
Jeevan Adhikari: Writing – original draft, Writing- review & editing, Data curation, Conceptualization, Methodology, Formal analysis, Visualization. Sandra Kavalukas: Writing- review & editing, Methodology. Kira C. Taylor: Writing- review & editing, Methodology. Brian Guinn: Writing- review & editing, Methodology. Charlie H. Zhang: Writing – review and editing, Methodology. Natalie DuPré: Project administration, Supervision, Conceptualization, Resources, Methodology, Data curation, Writing – review & editing.
Data availability
The fast-food data used in this study are not publicly available and were provided by Division of Public Health Protection and Safety, Kentucky Department for Public Health upon request. The Kentucky Cancer Registry datasets analyzed in this study are not publicly available because it contains protected health information that requires appropriate IRB approvals. The US Census Bureau's American Community Survey (ACS) datasets, Centers for Disease Control and Prevention Population Level Analysis and Community Estimates (CDC PLACES) data, and 2010 Rural-Urban Commuting Area (RUCA) classification codes are publicly available online.
References
- 1.Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. May-Jun 2024;74(3):229–263. doi: 10.3322/caac.21834 [DOI] [PubMed] [Google Scholar]
- 2.United States Cancer Statistics: Data Visualizations. Compare by Cancer Site. U.S. Cancer Statistics Working Group. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute. Updated July 2025. Accessed August 13, 2025. https://gis.cdc.gov/Cancer/USCS/#/compare/cancer-site/incidence/mortality?indicator-1=count&timeperiod-1=1&sexes-1=1&races-1=1&ages-1=23&view-1=chart&indicator-2=count&timeperiod-2=26&sexes-2=1&races-2=1&ages-2=23&view-2=chart [Google Scholar]
- 3.Xi Y, Xu P. Global colorectal cancer burden in 2020 and projections to 2040. Transl Oncol. Oct 2021;14(10):101174. doi: 10.1016/j.tranon.2021.101174 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Boursi B, Arber N. Current and future clinical strategies in colon cancer prevention and the emerging role of chemoprevention. Curr Pharm Des. 2007;13(22):2274–82. doi: 10.2174/138161207781368783 [DOI] [PubMed] [Google Scholar]
- 5.Bhimani N, Wong GYM, Molloy C, et al. Cost of treating metastatic colorectal cancer: a systematic review. Public Health. Oct 2022;211:97–104. doi: 10.1016/j.puhe.2022.06.022 [DOI] [PubMed] [Google Scholar]
- 6.Gray NM, Hall SJ, Browne S, et al. Modifiable and fixed factors predicting quality of life in people with colorectal cancer. Br J Cancer. May 24 2011;104(11):1697–703. doi: 10.1038/bjc.2011.155 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jemal A, Siegel RL, Ma J, et al. Inequalities in premature death from colorectal cancer by state. J Clin Oncol. Mar 10 2015;33(8):829–35. doi: 10.1200/jco.2014.58.7519 [DOI] [PubMed] [Google Scholar]
- 8.Cheng E, Ou F-S, Gatten C, et al. Plant-based diet and survival among patients with metastatic colorectal cancer. JNCI: Journal of the National Cancer Institute. 2024;117(1):169–179. doi: 10.1093/jnci/djae213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Boggs DA, Ban Y, Palmer JR, Rosenberg L. Higher Diet Quality Is Inversely Associated with Mortality in African-American Women1, 2, 3, 4. The Journal of Nutrition. 2015/03/01/ 2015;145(3):547–554. doi: 10.3945/jn.114.195735 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Entwistle MR, Schweizer D, Cisneros R. Dietary patterns related to total mortality and cancer mortality in the United States. Cancer Causes Control. Nov 2021;32(11):1279–1288. doi: 10.1007/s10552-021-01478-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Fuchs MA, Sato K, Niedzwiecki D, et al. Sugar-sweetened beverage intake and cancer recurrence and survival in CALGB 89803 (Alliance). PLoS One. 2014;9(6):e99816. doi: 10.1371/journal.pone.0099816 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fung TT, Kashambwa R, Sato K, et al. Post diagnosis diet quality and colorectal cancer survival in women. PLoS One. 2014;9(12):e115377. doi: 10.1371/journal.pone.0115377 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Swinburn B, Sacks G, Vandevijvere S, et al. INFORMAS (International Network for Food and Obesity/non-communicable diseases Research, Monitoring and Action Support): overview and key principles. Obes Rev. Oct 2013;14 Suppl 1:1–12. doi: 10.1111/obr.12087 [DOI] [PubMed] [Google Scholar]
- 14.Singleton CR. Improving urban food systems requires emphasizing nutrition equity in interventions and policy action. Am J Clin Nutr. Apr 2022;115(4):981–982. doi: 10.1093/ajcn/nqac018 [DOI] [PubMed] [Google Scholar]
- 15.Fong AJ, Lafaro K, Ituarte PHG, Fong Y. Association of Living in Urban Food Deserts with Mortality from Breast and Colorectal Cancer. Ann Surg Oncol. Mar 2021;28(3):1311–1319. doi: 10.1245/s10434-020-09049-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Khalil M, Munir MM, Woldesenbet S, et al. Association of county-level food deserts and food swamps on postoperative outcomes among patients undergoing colorectal surgery. J Gastrointest Surg. Apr 2024;28(4):494–500. doi: 10.1016/j.gassur.2024.01.012 [DOI] [PubMed] [Google Scholar]
- 17.Bevel MS, Tsai MH, Parham A, Andrzejak SE, Jones S, Moore JX. Association of Food Deserts and Food Swamps With Obesity-Related Cancer Mortality in the US. JAMA Oncol. Jul 1 2023;9(7):909–916. doi: 10.1001/jamaoncol.2023.0634 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Athens JK, Duncan DT, Elbel B. Proximity to Fast-Food Outlets and Supermarkets as Predictors of Fast-Food Dining Frequency. J Acad Nutr Diet. Aug 2016;116(8):1266–75. doi: 10.1016/j.jand.2015.12.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.He M, Tucker P, Gilliland J, Irwin JD, Larsen K, Hess P. The influence of local food environments on adolescents' food purchasing behaviors. Int J Environ Res Public Health. Apr 2012;9(4):1458–71. doi: 10.3390/ijerph9041458 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Burgoine T, Forouhi NG, Griffin SJ, Wareham NJ, Monsivais P. Associations between exposure to takeaway food outlets, takeaway food consumption, and body weight in Cambridgeshire, UK: population based, cross sectional study. Bmj. Mar 13 2014;348:g1464. doi: 10.1136/bmj.g1464 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Moore LV, Diez Roux AV, Nettleton JA, Jacobs DR, Franco M. Fast-food consumption, diet quality, and neighborhood exposure to fast food: the multi-ethnic study of atherosclerosis. Am J Epidemiol. Jul 1 2009;170(1):29–36. doi: 10.1093/aje/kwp090 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Chang VC, Cotterchio M, De P, Tinmouth J. Risk factors for early-onset colorectal cancer: a population-based case-control study in Ontario, Canada. Cancer Causes Control. Oct 2021;32(10):1063–1083. doi: 10.1007/s10552-021-01456-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mahfouz EM, Sadek RR, Abdel-Latief WM, Mosallem FA, Hassan EE. The role of dietary and lifestyle factors in the development of colorectal cancer: case control study in Minia, Egypt. Cent Eur J Public Health. Dec 2014;22(4):215–22. doi: 10.21101/cejph.a3919 [DOI] [PubMed] [Google Scholar]
- 24.Peters RK, Garabrant DH, Yu MC, Mack TM. A case-control study of occupational and dietary factors in colorectal cancer in young men by subsite. Cancer Res. Oct 1 1989;49(19):5459–68. [PubMed] [Google Scholar]
- 25.Tayyem RF, Bawadi HA, Shehadah I, et al. Fast Foods, Sweets and Beverage Consumption and Risk of Colorectal Cancer: A Case-Control Study in Jordan. Asian Pac J Cancer Prev. Jan 27 2018;19(1):261–269. doi: 10.22034/apjcp.2018.19.1.261 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Tufail M, Wu C, Hussain MS. Dietary, addictive and habitual factors, and risk of colorectal cancer. Nutrition. Apr 2024;120:112334. doi: 10.1016/j.nut.2023.112334 [DOI] [PubMed] [Google Scholar]
- 27.Barrington WE, White E. Mortality outcomes associated with intake of fast-food items and sugar-sweetened drinks among older adults in the Vitamins and Lifestyle (VITAL) study. Public Health Nutr. Dec 2016;19(18):3319–3326. doi: 10.1017/s1368980016001518 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Du Y, Rong S, Sun Y, et al. Association Between Frequency of Eating Away-From-Home Meals and Risk of All-Cause and Cause-Specific Mortality. Journal of the Academy of Nutrition and Dietetics. 2021/09/01/ 2021;121(9):1741–1749.e1. doi: 10.1016/j.jand.2021.01.012 [DOI] [PubMed] [Google Scholar]
- 29.Reitzel LR, Regan SD, Nguyen N, et al. Density and proximity of fast food restaurants and body mass index among African Americans. Am J Public Health. Jan 2014;104(1):110–6. doi: 10.2105/ajph.2012.301140 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Jilcott SB, Wade S, McGuirt JT, Wu Q, Lazorick S, Moore JB. The association between the food environment and weight status among eastern North Carolina youth. Public Health Nutr. Sep 2011;14(9):1610–7. doi: 10.1017/s1368980011000668 [DOI] [PubMed] [Google Scholar]
- 31.Mazidi M, Speakman JR. Association of Fast-Food and Full-Service Restaurant Densities With Mortality From Cardiovascular Disease and Stroke, and the Prevalence of Diabetes Mellitus. J Am Heart Assoc. May 25 2018;7(11)doi: 10.1161/jaha.117.007651 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Daniel M, Paquet C, Auger N, Zang G, Kestens Y. Association of fast-food restaurant and fruit and vegetable store densities with cardiovascular mortality in a metropolitan population. Eur J Epidemiol. Oct 2010;25(10):711–9. doi: 10.1007/s10654-010-9499-4 [DOI] [PubMed] [Google Scholar]
- 33.Morgenstern LB, Escobar JD, Sánchez BN, et al. Fast food and neighborhood stroke risk. Ann Neurol. Aug 2009;66(2):165–70. doi: 10.1002/ana.21726 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Alter DA, Eny K. The relationship between the supply of fast-food chains and cardiovascular outcomes. Can J Public Health. May-Jun 2005;96(3):173–7. doi: 10.1007/bf03403684 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.United States Cancer Statistics: Data Visualizations. Compare by Geography. U.S. Cancer Statistics Working Group. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute. Updated June 2025. Accessed August 13, 2025. https://gis.cdc.gov/Cancer/USCS/#/compare/geography/incidence/mortality?cancer-1=76&indicator-1=value&timeperiod-1=1&sexes-1=1&races-1=1&ages-1=23&view-1=map&cancer-2=76&indicator-2=value&timeperiod-2=26&sexes-2=1&races-2=1&ages-2=23&view-2=map [Google Scholar]
- 36.Myint ZW, Shrestha R, Siddiqui S, et al. Ten-year survival outcomes for patients with early stage classical Hodgkin lymphoma: An analysis from Kentucky Cancer Registry. Hematol Oncol Stem Cell Ther. Mar 2020;13(1):17–22. doi: 10.1016/j.hemonc.2019.08.009 [DOI] [PubMed] [Google Scholar]
- 37.Lei F, Vanderpool RC, McLouth LE, et al. Influence of depression on breast cancer treatment and survival: A Kentucky population-based study. Cancer. Jun 15 2023;129(12):1821–1835. doi: 10.1002/cncr.34676 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Kentucky Cancer Registry (KCR). Kentucky Cancer Registry Online Manuals. Updated December 9, 2024. Accessed February 17, 2025. https://www.kcr.uky.edu/manuals/
- 39.Burdette HL, Whitaker RC. Neighborhood playgrounds, fast food restaurants, and crime: relationships to overweight in low-income preschool children. Prev Med. Jan 2004;38(1):57–63. doi: 10.1016/j.ypmed.2003.09.029 [DOI] [PubMed] [Google Scholar]
- 40.Creel JS, Sharkey JR, McIntosh A, Anding J, Huber JC. Availability of healthier options in traditional and nontraditional rural fast-food outlets. BMC Public Health. 2008/11/28 2008;8(1):395. doi: 10.1186/1471-2458-8-395 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Fiechtner L, Sharifi M, Sequist T, et al. Food environments and childhood weight status: effects of neighborhood median income. Child Obes. Jun 2015;11(3):260–8. doi: 10.1089/chi.2014.0139 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Sharkey JR, Johnson CM, Dean WR, Horel SA. Association between proximity to and coverage of traditional fast-food restaurants and non-traditional fast-food outlets and fast-food consumption among rural adults. Int J Health Geogr. May 20 2011;10:37. doi: 10.1186/1476-072x-10-37 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sharkey JR, Horel S, Dean WR. Neighborhood deprivation, vehicle ownership, and potential spatial access to a variety of fruits and vegetables in a large rural area in Texas. International Journal of Health Geographics. 2010/05/25 2010;9(1):26. doi: 10.1186/1476-072X-9-26 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Centers for Disease Control and Prevention (CDC). About PLACES. PLACES: Local Data for Better Health. Updated Updated October 18, 2021. Accessed February 17, 2025. https://www.cdc.gov/places/about/index.html
- 45.United States Census Bureau. Subjects Included in the Survey. American Community Survey. Updated August 19, 2024. Accessed February 17, 2025. https://www.census.gov/programs-surveys/acs/guidance/subjects.html [Google Scholar]
- 46.Rural-Urban Commuting Area Codes. Economic Research Service. U.S. Department of Agriculture. Updated May 08, 2025. Accessed June 24, 2025. https://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes [Google Scholar]
- 47.Peto R, Peto J. Asymptotically Efficient Rank Invariant Test Procedures. Journal of the Royal Statistical Society Series A (General). 1972;135(2):185–207. doi: 10.2307/2344317 [DOI] [Google Scholar]
- 48.Kaplan EL, Meier P. Nonparametric Estimation from Incomplete Observations. Journal of the American Statistical Association. 1958/06/01 1958;53(282):457–481. doi: 10.1080/01621459.1958.10501452 [DOI] [Google Scholar]
- 49.SCHOENFELD D Partial residuals for the proportional hazards regression model. Biometrika. 1982;69(1):239–241. doi: 10.1093/biomet/69.1.239 [DOI] [Google Scholar]
- 50.Kleinbaum DG, Klein M. Survival analysis a self-learning text. 3rd ed. Springer; New York, NY; 2012. [Google Scholar]
- 51.Pan A, Sun Q, Bernstein AM, et al. Red meat consumption and mortality: results from 2 prospective cohort studies. Arch Intern Med. Apr 9 2012;172(7):555–63. doi: 10.1001/archinternmed.2011.2287 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Sinha R, Cross AJ, Graubard BI, Leitzmann MF, Schatzkin A. Meat intake and mortality: a prospective study of over half a million people. Arch Intern Med. Mar 23 2009;169(6):562–71. doi: 10.1001/archinternmed.2009.6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Kim Y, Je Y, Giovannucci EL. Association between dietary fat intake and mortality from all-causes, cardiovascular disease, and cancer: A systematic review and meta-analysis of prospective cohort studies. Clin Nutr. Mar 2021;40(3):1060–1070. doi: 10.1016/j.clnu.2020.07.007 [DOI] [PubMed] [Google Scholar]
- 54.Messerli FH, Hofstetter L, Syrogiannouli L, et al. Sodium intake, life expectancy, and all-cause mortality. Eur Heart J. Jun 1 2021;42(21):2103–2112. doi: 10.1093/eurheartj/ehaa947 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Cook NR, Appel LJ, Whelton PK. Sodium Intake and All-Cause Mortality Over 20 Years in the Trials of Hypertension Prevention. J Am Coll Cardiol. Oct 11 2016;68(15):1609–1617. doi: 10.1016/j.jacc.2016.07.745 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Odegaard AO, Koh WP, Yuan JM, Pereira MA. Beverage habits and mortality in Chinese adults. J Nutr. Mar 2015;145(3):595–604. doi: 10.3945/jn.114.200253 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Kant AK, Graubard BI. A prospective study of frequency of eating restaurant prepared meals and subsequent 9-year risk of all-cause and cardiometabolic mortality in US adults. PLoS One. 2018;13(1):e0191584. doi: 10.1371/journal.pone.0191584 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Martínez-González MA, Zazpe I, Razquin C, et al. Empirically-derived food patterns and the risk of total mortality and cardiovascular events in the PREDIMED study. Clin Nutr. Oct 2015;34(5):859–67. doi: 10.1016/j.clnu.2014.09.006 [DOI] [PubMed] [Google Scholar]
- 59.Mo J, Juhua L, and Hendryx M. Food Environment and Colorectal Cancer Incidence and Mortality Rates. Journal of Hunger & Environmental Nutrition. 2022/05/04 2022;17(3):397–408. doi: 10.1080/19320248.2021.1893243 [DOI] [Google Scholar]
- 60.Murray CJ, Atkinson C, Bhalla K, et al. The state of US health, 1990–2010: burden of diseases, injuries, and risk factors. Jama. Aug 14 2013;310(6):591–608. doi: 10.1001/jama.2013.13805 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Bowman SA, Gortmaker SL, Ebbeling CB, Pereira MA, Ludwig DS. Effects of fast-food consumption on energy intake and diet quality among children in a national household survey. Pediatrics. Jan 2004;113(1 Pt 1):112–8. doi: 10.1542/peds.113.1.112 [DOI] [PubMed] [Google Scholar]
- 62.Bowman SA, Vinyard BT. Fast food consumption of U.S. adults: impact on energy and nutrient intakes and overweight status. J Am Coll Nutr. Apr 2004;23(2):163–8. doi: 10.1080/07315724.2004.10719357 [DOI] [PubMed] [Google Scholar]
- 63.Larson NI, Neumark-Sztainer DR, Story MT, Wall MM, Harnack LJ, Eisenberg ME. Fast food intake: longitudinal trends during the transition to young adulthood and correlates of intake. J Adolesc Health. Jul 2008;43(1):79–86. doi: 10.1016/j.jadohealth.2007.12.005 [DOI] [PubMed] [Google Scholar]
- 64.Paeratakul S, Ferdinand DP, Champagne CM, Ryan DH, Bray GA. Fast-food consumption among US adults and children: dietary and nutrient intake profile. J Am Diet Assoc. Oct 2003;103(10):1332–8. doi: 10.1016/s0002-8223(03)01086-1 [DOI] [PubMed] [Google Scholar]
- 65.Rock CL, Thomson CA, Sullivan KR, et al. American Cancer Society nutrition and physical activity guideline for cancer survivors. CA Cancer J Clin. May 2022;72(3):230–262. doi: 10.3322/caac.21719 [DOI] [PubMed] [Google Scholar]
- 66.Bahadoran Z, Mirmiran P, Azizi F. Fast Food Pattern and Cardiometabolic Disorders: A Review of Current Studies. Health Promot Perspect. 2015;5(4):231–40. doi: 10.15171/hpp.2015.028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Besedin D, Carter S, Shah R, Hao Van TT, Eri R. Beneath the Bun: Food additives in fast-food burgers and the path to healthier choices. Food and Humanity. 2025/12/01/ 2025;5:100646. doi: 10.1016/j.foohum.2025.100646 [DOI] [Google Scholar]
- 68.Qin P, Li Q, Zhao Y, et al. Sugar and artificially sweetened beverages and risk of obesity, type 2 diabetes mellitus, hypertension, and all-cause mortality: a dose-response meta-analysis of prospective cohort studies. Eur J Epidemiol. Jul 2020;35(7):655–671. doi: 10.1007/s10654-020-00655-y [DOI] [PubMed] [Google Scholar]
- 69.Zheng J, Tabung FK, Zhang J, et al. Association between dietary inflammatory potential and mortality after cancer diagnosis in the Women's Health Initiative. Br J Cancer. Feb 2023;128(4):606–617. doi: 10.1038/s41416-022-02079-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 70.Chassaing B, Compher C, Bonhomme B, et al. Randomized Controlled-Feeding Study of Dietary Emulsifier Carboxymethylcellulose Reveals Detrimental Impacts on the Gut Microbiota and Metabolome. Gastroenterology. Mar 2022;162(3):743–756. doi: 10.1053/j.gastro.2021.11.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71.West NR, McCuaig S, Franchini F, Powrie F. Emerging cytokine networks in colorectal cancer. Nat Rev Immunol. Oct 2015;15(10):615–29. doi: 10.1038/nri3896 [DOI] [PubMed] [Google Scholar]
- 72.Cross AJ, Sinha R. Meat-related mutagens/carcinogens in the etiology of colorectal cancer. Environ Mol Mutagen. 2004;44(1):44–55. doi: 10.1002/em.20030 [DOI] [PubMed] [Google Scholar]
- 73.Jägerstad M, Skog K. Genotoxicity of heat-processed foods. Mutat Res. Jul 1 2005;574(1–2):156–72. doi: 10.1016/j.mrfmmm.2005.01.030 [DOI] [PubMed] [Google Scholar]
- 74.Knize MG, Salmon CP, Pais P, Felton JS. Food heating and the formation of heterocyclic aromatic amine and polycyclic aromatic hydrocarbon mutagens/carcinogens. Adv Exp Med Biol. 1999;459:179–93. doi: 10.1007/978-1-4615-4853-9_12 [DOI] [PubMed] [Google Scholar]
- 75.Larsson BK. Formation of polycyclic aromatic hydrocarbons during the smoking and grilling of food. Prog Clin Biol Res. 1986;206:169–80. [PubMed] [Google Scholar]
- 76.Selected social characteristics in the United States. American Community Survey, ACS 5-Year Estimates Data Profiles, Table DP02. US Census Bureau; (2018). Accessed September 17, 2025. https://data.census.gov/table/ACSDP5Y2018.DP02?q=dp02&g=040XX00US21 [Google Scholar]
- 77.Dunn RA, Sharkey JR, Horel S. The effect of fast-food availability on fast-food consumption and obesity among rural residents: an analysis by race/ethnicity. Econ Hum Biol. Jan 2012;10(1):1–13. doi: 10.1016/j.ehb.2011.09.005 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The fast-food data used in this study are not publicly available and were provided by Division of Public Health Protection and Safety, Kentucky Department for Public Health upon request. The Kentucky Cancer Registry datasets analyzed in this study are not publicly available because it contains protected health information that requires appropriate IRB approvals. The US Census Bureau's American Community Survey (ACS) datasets, Centers for Disease Control and Prevention Population Level Analysis and Community Estimates (CDC PLACES) data, and 2010 Rural-Urban Commuting Area (RUCA) classification codes are publicly available online.