1. Introduction
People living in rural regions and those experiencing socioeconomic deprivation are known to have poorer cancer outcomes.1–4 In prior work, we found that a later stage at diagnosis for colorectal cancer (CRC) was associated with living in rural areas or having Medicaid insurance, but there were no differences according to local Area Deprivation Index (ADI).5 These findings suggested poor access to screening or symptom evaluation. Cancer treatment is demanding, and people in rural and low-income areas may have difficulty with transportation and other support needed to engage in cancer care.6 Our goal was to characterize relationships between rurality and area socioeconomic deprivation and treatment for CRC in Maine.7 A better understanding of the drivers of cancer care variation will help inform health policy and health system changes to improve access to care.
2. Methods
2.1. Cancer registry data
We obtained deidentified data from the Maine Cancer Registry for all incident CRC cases in adults (21 years and older) from 2017 to 2021. The stage at diagnosis was defined by the SEER Summary Stage Manual (https://seer.cancer.gov/tools/staging/index.html). We collapsed the stages into groups—local, regional, and distant spread—and excluded cases with in-situ stage for this analysis (n = 89). For analyses of treatment differences, we separated colon cancers (n = 1988) from rectal (n = 883) and anal (n = 214) cancers.
2.2. Rurality and area deprivation index
Rurality was determined by mapping 5-digit ZIP codes to the 2010 Rural-Urban Commuting Area (RUCA) Codes (https://depts.washington.edu/uwruca/ruca-uses.php).8 We grouped RUCA codes into tiers: urban, large rural, small rural, and isolated rural. To examine the impact of poverty in addition to rurality, we used the state-level ADI that was ascertained using a crosswalk from the University of Wisconsin Neighborhood Atlas (https://www.neighborhoodatlas.medicine.wisc.edu/). ADI is a composite measure of neighborhood socioeconomic disadvantage based on factors such as income, education, employment, and housing quality ranked from 1 (least disadvantaged) to 10 (most disadvantaged) at the state level. We collapsed groups into quintiles to ensure adequate statistical power without losing the ability to discriminate between groups. We excluded people for whom we could not assign RUCA (n = 4) and ADI (n = 45) because this factor was our primary interest for this study. After the exclusions, 3085 cases were included in the analysis.
2.3. Statistical analysis
We used descriptive statistics (chi-square and t tests) to describe the overall population of CRC cases in the registry. Using R version 4.2.1, separate logistic regression models were constructed with treatment (surgery, chemotherapy, or radiation) as the dependent variable, with separate models for colon and rectal/anal cancers. Independent variables included age, sex, race, insurance type, rurality, ADI, stage, and year of diagnosis. The MaineHealth Institutional Review Board approved this study.
3. Results
Table 1 shows the demographics of the 3085 CRC cases. Most cases occurred in older adults with a median (SD) age of 67 (14) years. Cases were split evenly by sex; 98% (3016 of 3085) were White and the remaining 2% (69 of 3085) were another or unknown race. Medicare (43% [1335 of 3085]) or commercial insurance (31% [969 of 3085]) were most common. Also, 34% (1063 of 3085) of people resided in small or isolated rural areas, and 43% (1323 of 3085) resided in areas in the highest 2 quintiles of ADI. Regional or distant spread at diagnosis was present among 60% (1902 of 3085) of cases. Cases were diagnosed evenly across calendar years.
Table 1.
Characteristics of Patients with Incident Colorectal Cancer, Maine Cancer Registry, 2017–2021 (N = 3085).
| Characteristic | Overall n = 3085 | Colon n = 1988 | Rectal/Anal n = 1097 |
|---|---|---|---|
| Age at diagnosis, median (SD), y | 67 (14) | 69 (14) | 63 (13) |
| Age group, No. (%), y | |||
| 21-49 | 338 (11) | 177 (8.9) | 161 (15) |
| 50-64 | 993 (32) | 548 (28) | 445 (41) |
| 65-74 | 873 (28) | 576 (29) | 297 (27) |
| 75 and older | 881 (29) | 687 (35) | 194 (18) |
| Sex, No. (%) | |||
| Female | 1552 (50) | 1036 (52) | 516 (47) |
| Male | 1533 (50) | 952 (48) | 581 (53) |
| Race, No. (%) | |||
| White | 3016 (98) | 1940 (98) | 1076 (98) |
| Other/Unknown | 69 (2.2) | 48 (2.4) | 21 (1.9) |
| Insurance, No. (%) | |||
| Commercial | 969 (31) | 563 (28) | 406 (37) |
| Medicare | 1335 (43) | 938 (47) | 397 (36) |
| Medicaid | 233 (7.6) | 127 (6.4) | 106 (9.7) |
| Medicare/Medicaid | 332 (11) | 218 (11) | 114 (10) |
| Veterans/public health service | 110 (3.6) | 69 (3.5) | 41 (3.7) |
| Not insured | 61 (2.0) | 40 (2.0) | 21 (1.9) |
| Unknown | 45 (1.5) | 33 (1.7) | 12 (1.1) |
| RUCA category, No. (%) | |||
| Urban | 1469 (48) | 946 (48) | 523 (48) |
| Large rural | 553 (18) | 351 (18) | 202 (18) |
| Small rural | 442 (14) | 298 (15) | 144 (13) |
| Isolated rural | 621 (20) | 393 (20) | 228 (21) |
| ADI quintile, No. (%) | |||
| ADI Q1 | 466 (15) | 313 (16) | 153 (14) |
| ADI Q2 | 613 (20) | 385 (19) | 228 (21) |
| ADI Q3 | 683 (22) | 436 (22) | 247 (23) |
| ADI Q4 | 766 (25) | 483 (24) | 283 (26) |
| ADI Q5 | 557 (18) | 371 (19) | 186 (17) |
| Stage at diagnosis, No. (%) | |||
| Localized | 1183 (38) | 755 (38) | 428 (39) |
| Regional | 1313 (43) | 845 (43) | 468 (43) |
| Distant | 589 (19) | 388 (20) | 201 (18) |
| Year of diagnosis, No. (%) | |||
| 2017 | 612 (20) | 400 (20) | 212 (19) |
| 2018 | 589 (19) | 393 (20) | 196 (18) |
| 2019 | 658 (21) | 443 (22) | 215 (20) |
| 2020 | 591 (19) | 346 (17) | 245 (22) |
| 2021 | 635 (21) | 406 (20) | 229 (21) |
| Surgery, No. (%) | 2484 (84) | 1762 (90) | 722 (71) |
| NA | 120 | 40 | 80 |
| Chemotherapy, No. (%) | 1370 (50) | 678 (40) | 692 (68) |
| Missing | 355 | 275 | 80 |
| Radiation, No. (%) | 644 (22) | 43 (2.2) | 601 (58) |
| Missing | 100 | 44 | 56 |
Abbreviations: ADI, Area Deprivation Index; RUCA, Rural-Urban Commuting Area; NA, Patients for whom treatment was too risky, died, or refused treatment were excluded in each treatment specific category.
No differences in treatment according to rurality and ADI were observed in bivariate analyses (data not shown). In multivariable logistic regression models (Fig. 1) with treatment (surgery, chemotherapy, or radiation) as the dependent variable, we observed no differences in the odds of receiving treatment according to rurality or ADI (see Appendixes for odds ratios and confidence intervals). Because so few people with colon cancer received radiation (n = 43), we do not show the radiation findings for colon cancer (Fig. 1, Appendix A). We observed lower odds of receiving chemotherapy among older patients with colon cancer, lower odds of all treatment types among older patients with rectal/anal cancers, and lower odds of radiation among men with rectal/anal cancer (Appendix B). We also observed lower odds of receiving surgery for people with rectal/anal cancer diagnosed in 2019 through 2021.
Fig. 1.
Forest Plots of Adjusted Odds Ratios of Treatment with Surgery and Chemotherapy according to Rurality and ADI from Multivariable Models by Cancer. ADI, Area Deprivation Index, and Insurance Status.
4. Discussion
We found no difference in the treatment for colon, rectal, and anal cancers in rural dwellers or people living in areas with high area poverty. Older people (75 years and older) had lower odds of receiving treatment compared with people ages 50 years and younger, which may be clinically appropriate. The COVID-19 pandemic resulted in decreased rates of cancer diagnoses nationally in 2020 and 2021,7 although we did not observe this trend. Changes in treatment guidelines for rectal and anal cancers may explain the lower odds of receiving surgery and higher radiation use for these cancers during this period. Our findings contradict several studies summarized in a recent review that described CRC treatment disparities in rural populations9 and other studies showing rural-urban differences.11,12 Because the population in Maine is predominantly White, we are unable to adequately examine racial differences in treatment or the intersectionality of race and small area deprivation together, as others have done.10
Our study is limited to a single state cancer registry. We do not have information on comorbidities to determine whether treatment was medically indicated, particularly to understand the treatment variation we observed among older patients. We also did not have access to survival data. The study may not be generalizable across the United States or between states, in part because the demographics of Maine are relatively homogeneous with regard to race. Specifically, it would be difficult to generalize our findings to more urban areas of the United States, areas with higher Medicaid populations, or other rural areas with higher racial, ethnic, and age diversity. In addition, we had a limited numbers of cases, particularly rectal and anal cancers, to explore differences in treatment.
In summary, we found no consistent evidence of variation in care for colon, rectal, and anal cancers according to rurality or area level poverty in Maine. Future efforts to understand colorectal care quality for people living in rural and low-income areas will require a more nuanced analysis of barriers to care and clinical outcomes data, particularly in older adults.
Acknowledgments
We appreciate the Maine Cancer Registry’s support in preparing the dataset.
Funding/Support
Research reported in this publication was supported by the National Institute of General Medical Sciences of the National Institutes of Health under the Northern New England Clinical and Translational Research grant (NNE-CTR, U54GM115516). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Before issuing a press release about the outcome of this research, please notify the National Institutes of Health awarding institute/center to allow for coordination.
A.
Logistic Regression Results for Treatment Outcomes: Colon Cancer
| Surgery |
Chemotherapy |
|||||
|---|---|---|---|---|---|---|
| Variable | Odds Ratio | 95% CI | P-value | Odds Ratio | 95% CI | P-value |
| Age Group | 0.3 | <0.001 | ||||
| Age 21–49 | — | — | — | — | ||
| Age 50–64 | 0.91 | 0.43, 1.86 | 0.68 | 0.41, 1.12 | ||
| Age 65–74 | 0.78 | 0.33, 1.82 | 0.44 | 0.24, 0.79 | ||
| Age 75 and older | 0.55 | 0.23, 1.29 | 0.15 | 0.08, 0.27 | ||
| Sex | 0.062 | 0.5 | ||||
| Female | — | — | — | — | ||
| Male | 0.70 | 0.47, 1.02 | 1.09 | 0.83, 1.41 | ||
| Race | 0.087 | 0.2 | ||||
| White | — | — | — | — | ||
| Other/Unknown | 0.39 | 0.14, 1.15 | 0.53 | 0.21, 1.32 | ||
| Insurance | 0.7 | 0.059 | ||||
| Commercial | — | — | — | — | ||
| Medicare | 1.02 | 0.53, 1.92 | 1.16 | 0.76, 1.79 | ||
| Medicaid | 0.61 | 0.29, 1.32 | 0.82 | 0.47, 1.45 | ||
| Medicare/Medicaid | 0.73 | 0.35, 1.54 | 0.62 | 0.37, 1.04 | ||
| Veterans/public health service | 1.34 | 0.48, 4.03 | 1.86 | 0.84, 4.20 | ||
| Not insured | 1.37 | 0.35, 6.99 | 0.66 | 0.26, 1.69 | ||
| Unknown | 0.60 | 0.14, 3.20 | 0.72 | 0.23, 2.17 | ||
| RUCA Category | 0.5 | 0.3 | ||||
| Urban | — | — | — | — | ||
| Large rural | 1.32 | 0.75, 2.39 | 1.07 | 0.73, 1.55 | ||
| Small rural | 1.20 | 0.64, 2.29 | 1.50 | 0.97, 2.32 | ||
| Isolated rural | 0.87 | 0.52, 1.45 | 1.25 | 0.86, 1.82 | ||
| ADI Quintile | 0.7 | 0.4 | ||||
| ADI_Q1 | — | — | — | — | ||
| ADI_Q2 | 0.67 | 0.34, 1.28 | 1.28 | 0.83, 1.98 | ||
| ADI_Q3 | 0.73 | 0.37, 1.40 | 1.41 | 0.91, 2.18 | ||
| ADI_Q4 | 0.74 | 0.39, 1.38 | 1.06 | 0.69, 1.61 | ||
| ADI_Q5 | 0.60 | 0.29, 1.24 | 1.00 | 0.61, 1.64 | ||
| Year of diagnosis | 0.4 | 0.5 | ||||
| 2017 | — | — | — | — | ||
| 2018 | 0.94 | 0.52, 1.68 | 0.95 | 0.63, 1.45 | ||
| 2019 | 1.56 | 0.86, 2.84 | 1.08 | 0.72, 1.61 | ||
| 2020 | 1.08 | 0.60, 1.96 | 0.86 | 0.56, 1.31 | ||
| 2021 | 0.93 | 0.52, 1.65 | 0.77 | 0.51, 1.16 | ||
| Stage at diagnosis | <0.001 | <0.001 | ||||
| Localized | — | — | — | — | ||
| Regional | 1.26 | 0.59, 2.71 | 33.3 | 22.9, 50.0 | ||
| Distant | 0.02 | 0.01, 0.04 | 83.2 | 53.6, 133 | ||
Abbreviations: CI = Confidence Interval, OR = Odds Ratio.
B.
Logistic Regression Results for Treatment Outcomes: Rectal/Anal Cancer
| Variable | Surgery |
Chemotherapy |
Radiation |
||||||
|---|---|---|---|---|---|---|---|---|---|
| Odds Ratio | 95% CI | P-value | Odds Ratio | 95% CI | P-value | Odds Ratio | 95% CI | P-value | |
| Age Group | 0.002 | 0.008 | 0.022 | ||||||
| Age 21–49 | — | — | — | — | — | — | |||
| Age 50–64 | 0.41 | 0.24, 0.71 | 0.52 | 0.29, 0.90 | 0.53 | 0.34, 0.83 | |||
| Age 65–74 | 0.39 | 0.19, 0.76 | 0.44 | 0.21, 0.90 | 0.52 | 0.29, 0.93 | |||
| Age 75 and older | 0.27 | 0.13, 0.56 | 0.26 | 0.12, 0.57 | 0.42 | 0.22, 0.79 | |||
| Sex | <0.001 | 0.5 | 0.012 | ||||||
| Female | — | — | — | — | — | — | |||
| Male | 2.03 | 1.46, 2.84 | 0.88 | 0.62, 1.25 | 0.69 | 0.51, 0.92 | |||
| Race | 0.5 | 0.13 | 0.3 | ||||||
| White | — | — | — | — | — | — | |||
| Other/Unknown | 0.63 | 0.21, 2.29 | 0.39 | 0.12, 1.30 | 0.56 | 0.19, 1.57 | |||
| Insurance | 0.4 | 0.3 | 0.071 | ||||||
| Commercial | — | — | — | — | — | — | |||
| Medicare | 0.68 | 0.40, 1.15 | 0.94 | 0.52, 1.68 | 1.44 | 0.89, 2.34 | |||
| Medicaid | 1.13 | 0.62, 2.11 | 1.10 | 0.59, 2.10 | 0.80 | 0.48, 1.34 | |||
| Medicare/Medicaid | 0.89 | 0.49, 1.66 | 1.50 | 0.75, 3.01 | 1.55 | 0.90, 2.70 | |||
| Veterans/public health service | 1.05 | 0.42, 2.99 | 1.84 | 0.70, 4.91 | 1.87 | 0.82, 4.41 | |||
| Not insured | 0.71 | 0.21, 2.54 | 1.19 | 0.34, 4.54 | 0.49 | 0.16, 1.37 | |||
| Unknown | 5.57 | 0.61, 149 | 0.22 | 0.04, 1.12 | 0.28 | 0.05, 1.24 | |||
| RUCA Category | 0.4 | >0.9 | >0.9 | ||||||
| Urban | — | — | — | — | — | — | |||
| Large rural | 1.32 | 0.83, 2.13 | 1.08 | 0.65, 1.80 | 1.00 | 0.66, 1.52 | |||
| Small rural | 0.80 | 0.46, 1.42 | 1.05 | 0.56, 1.96 | 0.95 | 0.57, 1.60 | |||
| Isolated rural | 0.92 | 0.58, 1.46 | 0.97 | 0.59, 1.60 | 1.06 | 0.70, 1.61 | |||
| ADI Quintile | 0.6 | 0.8 | 0.13 | ||||||
| ADI_Q1 | — | — | — | — | — | — | |||
| ADI_Q2 | 0.98 | 0.58, 1.67 | 1.23 | 0.68, 2.21 | 0.98 | 0.60, 1.60 | |||
| ADI_Q3 | 1.25 | 0.72, 2.17 | 1.43 | 0.77, 2.64 | 1.39 | 0.84, 2.32 | |||
| ADI_Q4 | 1.38 | 0.81, 2.34 | 1.07 | 0.59, 1.91 | 0.86 | 0.53, 1.41 | |||
| ADI_Q5 | 1.37 | 0.72, 2.61 | 1.20 | 0.58, 2.47 | 0.78 | 0.43, 1.42 | |||
| Year of diagnosis | 0.002 | 0.11 | 0.4 | ||||||
| 2017 | — | — | — | — | — | — | |||
| 2018 | 1.11 | 0.63, 1.97 | 1.19 | 0.68, 2.11 | 1.39 | 0.86, 2.23 | |||
| 2019 | 0.64 | 0.38, 1.08 | 1.56 | 0.90, 2.72 | 1.62 | 1.02, 2.58 | |||
| 2020 | 0.55 | 0.33, 0.90 | 1.53 | 0.89, 2.65 | 1.25 | 0.80, 1.97 | |||
| 2021 | 0.45 | 0.27, 0.74 | 0.84 | 0.49, 1.45 | 1.27 | 0.81, 2.00 | |||
| Stage at diagnosis | <0.001 | <0.001 | <0.001 | ||||||
| Localized | — | — | — | — | — | — | |||
| Regional | 0.64 | 0.44, 0.91 | 31.2 | 20.2, 50.1 | 10.1 | 7.20, 14.3 | |||
| Distant | 0.05 | 0.03, 0.08 | 11.6 | 7.19, 19.3 | 1.37 | 0.94, 1.99 | |||
Abbreviations: CI = Confidence Interval, OR = Odds Ratio.
Footnotes
Ethics approval
This work was reviewed and found exempt by the MaineHealth Maine Medical Center Institutional Review Board.
Conflict of interest disclosures
None reported.
Other disclosures (eg, a Trials number from clinicaltrials.gov)
None reported.
Previous presentations
An abstract from this work was presented as a poster at the Maine Public Health Association statewide meeting on October 27, 2025.
References
- 1.Kenzik KM, Davis ES, Franks JA, Bhatia S. Estimating the impact of rurality in disparities in cancer mortality. JCO Oncol Pract. 2024;20(7):993–1002. doi: 10.1200/OP.23.00626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Althans AR, Meshkin D, Holder-Murray J, et al. Deprivation and rurality mediate income inequality’s association with colorectal cancer outcomes. Am J Prev Med. 2024;67(4):540–547. doi: 10.1016/j.amepre.2024.06.002. [DOI] [PubMed] [Google Scholar]
- 3.Fairfield KM, Black AW, Ziller EC, et al. Area Deprivation Index and rurality in relation to lung cancer prevalence and mortality in a rural state. JNCI Cancer Spectr. 2020;4(4):pkaa011. doi: 10.1093/jncics/pkaa011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Monroe AC, Ricketts TC, Savitz LA. Cancer in rural versus urban populations: a review. J Rural Health. 1992;8(3):212–220. doi: 10.1111/j.1748-0361.1992.tb00354.x. [DOI] [PubMed] [Google Scholar]
- 5.Fairfield KM, Murray K, Cloutier LM, et al. Stage at diagnosis for common cancers according to rurality, area deprivation, and insurance, 2017 to 2021. Cancer Prev Res (Phila). 2025;18(8):465–474. doi: 10.1158/1940-6207.CAPR-24-0587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lee KMN, Hunleth J, Rolf L, et al. Distance and transportation barriers to colorectal cancer screening in a rural community. J Prim Care Community Health. 2023;14:21501319221147126. doi: 10.1177/21501319221147126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Burus T, Lei F, Huang B, et al. COVID-19 and rates of cancer diagnosis in the US. JAMA Netw Open. 2024;7(9):e2432288. doi: 10.1001/jamanetworkopen.2024.32288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hart LG, Larson EH, Lishner DM. Rural definitions for health policy and research. Am J Public Health. 2005;95(7):1149–1155. doi: 10.2105/AJPH.2004.042432. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sepassi A, Li M, Zell JA, Chan A, Saunders IM, Mukamel DB. Rural-urban disparities in colorectal cancer screening, diagnosis, treatment, and survivorship care: a systematic review and meta-analysis. Oncologist. 2024;29(4):e431–4e46. doi: 10.1093/oncolo/oyad347. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Yousef M, Yousef A, Chowdhury S, et al. Molecular, socioeconomic, and clinical factors affecting racial and ethnic disparities in colorectal cancer survival. JAMA Oncol. 2024;10(11):1519–1529. doi: 10.1001/jamaoncol.2024.3666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Singh GK, Williams SD, Siahpush M, Mulhollen A. Socioeconomic, rural-urban, and racial inequalities in us cancer mortality: part i-all cancers and lung cancer and part ii-colorectal, prostate, breast, and cervical cancers. J Cancer Epidemiol. 2011;2011:107497. doi: 10.1155/2011/107497. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ladage VP, Ramkumar N, Wang Q, et al. Rurality and use of metastasis-directed therapies for Medicare beneficiaries with metastatic colon cancer: a retrospective cohort study. Ann Surg Oncol. 2026;33(1):365–373. doi: 10.1245/s10434-025-18412-4. [DOI] [PMC free article] [PubMed] [Google Scholar]