Abstract
BACKGROUND:
Compared with urban, rural patients experience more postoperative complications, including surgical site infections and anastomotic leaks after colon cancer resection. As these complications often require long-term care, we investigated rural–urban differences in postoperative outcomes and discharge disposition.
STUDY DESIGN:
We identified patients who underwent colon cancer resection from the Surveillance, Epidemiology, and End Results–Medicare database (2014 to 2019). We used multivariable logistic regression models to evaluate rural–urban differences in complications, ICU admission, discharge disposition, readmissions, and postdischarge reoperations. County-level availability of postacute care services (skilled nursing, rehabilitation, home healthcare) was examined using the Area Health Resource File.
RESULTS:
Among 31,635 patients (urban = 82%; rural = 18%), more rural patients were White (88% vs 77%; p < 0.001), underwent open surgery (69% vs 60%; p < 0.001), and had surgery at nonacademic centers (44% vs 64%; p < 0.001). Stage was similar between the groups. Rural patients had lower Charlson Comorbidity Indices (p < 0.001) but more complications (24% vs 22%; p = 0.0039) and ICU admissions (34% vs 30%; p < 0.001). Rural patients had lower odds of being discharged with postacute care services (odds ratio 0.76, 95% CI 0.71 to 0.82); readmissions did not differ, but rural patients underwent more postdischarge reoperations (0.6% vs 0.3%, p = 0.0006). Rural counties had more postacute care facilities (8 [interquartile range [IQR] 5 to 13] vs 4 [IQR 3 to 5], p < 0.001) and home healthcare agencies per 100,000 (3 [IQR 0 to 7] vs 2 [IQR 1 to 5], p < 0.001) compared with urban.
CONCLUSIONS:
Rural patients experienced more complications and ICU admissions but had lower odds of being discharged with postacute care services, despite higher per-capita services and facilities available in rural areas. Improving postacute care delivery and usage in rural areas may reduce geographic inequities after colon cancer resection.
Rural individuals represent 20% of the US population, but rural patients with colon cancer (CC) have worse outcomes including higher mortality after surgical resection compared with urban patients.1–3 Some disparities in outcomes have been attributed to rural patients’ lower likelihood of undergoing minimally invasive resection, receiving adequate lymphadenectomy, and initiating chemotherapy in a timely manner.4,5 However, in addition to appropriate clinical and oncologic treatment, access to postacute care services impacts long-term outcomes and is critical among rural patients who tend to live further from hospitals.6,7
Surgical complications after CC resection, including surgical site infections (SSIs) and anastomotic leaks, may lead to prolonged hospitalizations, deconditioning, and the need for postacute care services.8–11 After nonsurgical hospitalizations, rural patients are less commonly discharged with postacute care services (inpatient rehabilitation, skilled nursing facilities [SNFs], home healthcare), and those who are discharged home without services have higher mortality compared with urban individuals discharged home.12,13 Although rurality may be associated with an increased risk of postoperative complications after CC resection, the interplay between rurality, postoperative complications, postacute care needs, and discharge disposition after CC resection is not clear.
We aimed to examine associations between rurality and discharge disposition in a nationwide sample of patients with CC undergoing surgical resection by developing a novel needs-based model to identify whether patients with postacute care needs received postacute care services at discharge. To further understand the rural–urban disparity in access to postacute care services, we investigated county-level densities of SNFs and beds, as well as home healthcare agencies. We hypothesized that rural patients would have higher postacute care needs but receive services less often due to worse access to services in rural regions.
METHODS
Data source
The Surveillance, Epidemiology, and End Results (SEER)–Medicare linked database (2014 to 2019) containing claims data from 2013 to 2020 for Medicare beneficiaries was used.14,15 The SEER–Medicare linked database allows for the examination of demographic and clinical data along with claims data for individuals with cancer diagnoses and Medicare insurance. This study was deemed exempt from review by the Boston University IRB.
Study population
Individuals aged 66 years and older with continuous Medicare Parts A and B enrollment and without HMO insurance from 1 year prediagnosis through 6 months postdiagnosis or death who were diagnosed with CC were included. We included patients who underwent surgical resection between 1 month before and 6 months after diagnosis to capture both emergent (where resection preceded histopathologic diagnosis) and elective resections.16,17 Individuals were excluded if they were managed nonoperatively or had surgery outside the study window, were diagnosed before January 2014 or were missing a diagnosis date, if diagnoses were obtained from death certificates or autopsies, or had a date of birth or death differing by 3 months or longer between SEER and Medicare data. Patients for whom it was not known whether they had been admitted to the ICU, those with unknown stage, and those who died before discharge or were discharged against medical advice were excluded.
Demographic data included age, sex, race (non-Hispanic White, non-Hispanic Black, Hispanic, Asian, other), and comorbidities were captured using the Charlson Comorbidity Index (CCI, based on diagnoses 1 year before CC diagnosis). Patients without CCI diagnoses within 1 year before cancer diagnosis were classified as having a CCI of 0. A validated claims-based preoperative frailty index was generated using ICD-9/−10 diagnosis codes and CPT codes from the 12 months before diagnosis (less than 0.1 = nonfrail, 0.1 to 0.2 = prefrail, greater than or equal to 0.2 = frail).18,19
Clinical data included surgical approach (open, minimally invasive), stage (see Supplemental Digital Content 1 at https://links.lww.com/JACS/A557 for definitions), location of cancer (right, transverse, left, overlapping, or unspecified sites combined), surgery at academic center, length of stay (LOS), ICU admissions, physical therapy (PT) evaluations (postoperative PT charge more than $0 for Medicare Provider Analysis and Review admission associated with surgery), discharge disposition, and 30-day hospital readmissions. Discharge dispositions included home without services, home with home healthcare, inpatient care facility (inpatient rehabilitation facilities, SNF), and other (hospice and transfer to a federal hospital, psychiatric facility, or critical access hospital).
Postoperative complications within 30 days were defined as respiratory failure, pneumonia, MI, deep vein thrombosis or pulmonary embolus, renal failure, hemorrhage, SSI, anastomotic leak, and other. We then identified patients who underwent a postdischarge reoperation related to a complication within 30 days of discharge, including abdominal washouts, diverting ostomies, and hemorrhage control procedures (see Supplemental Digital Content 1 at https://links.lww.com/JACS/A557 for further definitions of complications and reoperations).
Surgical resection
The ICD-9 and ICD-10 procedure codes obtained from the Medicare Provider Analysis and Review inpatient files were used to define surgical resection.20 Procedure codes for open or minimally invasive (laparoscopic + robotic) colectomy within 1 month before diagnosis until 6 months postdiagnosis were included (Supplemental Digital Content 1, https://links.lww.com/JACS/A557). The most recent surgical resection was selected as the index operation if patients had multiple resection claims within the study timeframe (fewer than 1% of patients).17
Exposure
The 2013 US Department of Agriculture’s Rural–Urban Continuum Codes (RUCC), which classifies counties on a scale of increasing rurality from 1 to 9, was used to define rurality.21 RUCC classification is based on population size and adjacency to urban areas. We categorized RUCC 1 to 3 as urban and RUCC 4 to 9 as rural, based on established rural–urban categorizations.22,23
Outcomes
The primary outcome was discharge with postacute care services (home healthcare, inpatient rehabilitation, SNF, other [short- or long-term inpatient care facilities, hospice]) vs discharge home without services. We also examined the subset of patients discharged to an inpatient facility (inpatient rehabilitation, skilled nursing, or other short- or long-term care facility) vs discharged home (with or without services). Secondary outcomes included LOS, ICU admissions, PT evaluations, complications, readmissions, and postdischarge reoperations.
Regional facilities measures
As a proxy for the regional availability of postacute care services, we examined county-level densities of SNF facilities, SNF beds, and home healthcare agencies by dividing the number of facilities or beds in the county by the county population. Densities are reported as SNF beds per 100,000 and home healthcare agencies per 100,000. Numbers of facilities, beds, and agencies were obtained from the 2017 Area Health Resources File, and county populations were drawn from the 2010 US Census Bureau data.24,25 We first calculated tertiles for both SNF density and home health agency density and categorized them into “low” (first tertile), “middle” (second tertile), and “high” (third tertile). Next, we created combined variables by urban–rural status with the following levels, (1) rural: low, (2) rural: middle, (3) rural: high, (4) urban: low, (5) urban: middle, and (6) urban: High. All maps were created using the “usmap” package26 version 0.7.1 in R (version 4.4.0).
Statistical analysis
Data were analyzed from February 2025 to April 2025. Urban–rural differences in demographic, clinical, and hospital characteristics were assessed using chi-square tests (for categorical variables) and Wilcoxon rank-sum tests (for continuous). A 2-tailed p value of 0.05 was considered statistically significant. Multivariable logistic regression models were used to evaluate factors associated with discharge with services and discharge to a facility. Covariates included age (older than vs younger than 75 years), sex, race (White, non-White), stage (0 to III vs IV), rurality (urban, rural), CCI (continuous), acuity of presentation (emergent + urgent combined, elective), complications (yes, no), and frailty (not frail, prefrail, and frail).
We then created needs-based models for discharge disposition to determine whether patients who predictably required postoperative support services, including those who were prefrail or frail and those who experienced complications, received them. For this needs-based model, we evaluated receipt of (1) any postoperative services and (2) discharge to a postacute care facility. Covaries included age, sex, race, stage, rurality, CCI, acuity of presentation, complications, and frailty as categorized previously. Sensitivity analyses, including area-level poverty as a covariate, did not change our findings. All analyses were conducted using SAS software version 9.4 (Cary, NC).
RESULTS
Patient characteristics
Among 31,635 individuals (rural = 5637; 17.8%), a higher proportion of rural individuals were men (46.0% vs 43.8%, p = 0.0021), White (87.9% vs 76.8%, p < 0.0001), and underwent open surgery (69.4% vs 60.3%, p < 0.0001); more urban patients presented emergently (37.3% vs 35.8%, p = 0.0269) and were treated more often at academic centers (63.6% vs 44.0%, p < 0.0001). Stages of disease were similar among the groups (p = 0.7125), and proportions of non-frail, prefrail, and frail individuals were distributed similarly between rural and urban populations (p = 0.0914; Table 1).
Table 1.
Demographic, Clinical, and Hospital Characteristics of Study Population
| Patient characteristic | Overall | Urban | Rural | p Value |
|---|---|---|---|---|
| Total patients, n (%) | 31,635 (100) | 25,998 (82.2) | 5,637 (17.8) | — |
| Age, y, median (IQR) | 77 (71–83) | 77 (71–83) | 76 (71–82) | <0.0001 |
| Sex, m, n (%) | 13,969 (44.2) | 11,376 (43.8) | 2,593 (46.0) | 0.0021 |
| Race, n (%) | <0.0001 | |||
| White | 24,910 (78.7) | 19,957 (76.8) | 4,953 (87.9) | |
| Non-White | 6,725 (21.3) | 6,041 (23.2) | 684 (12.1) | |
| Charlson Comorbidity Index, median, (range) | 1 (0–13) | 1 (0–13) | 1 (0–11) | <0.0001 |
| Frailty, n (%) | 0.0914 | |||
| Not frail | 1,787 (5.7) | 1,502 (5.8) | 285 (5.1) | |
| Prefrail | 20,201 (63.9) | 16,563 (63.7) | 3,638 (64.5) | |
| Frail | 9,647 (30.5) | 7,933 (30.5) | 1,714 (30.4) | |
| Emergent operation, n (%) | 11,722 (37.1) | 9,706 (37.3) | 2,016 (35.8) | 0.0269 |
| Open resection, n (%) | 19,581 (61.9) | 15,672 (60.3) | 3,909 (69.4) | <0.0001 |
| Academic center, n (%) | 19,003 (60.1) | 16,522 (63.6) | 2,481 (44.0) | <0.0001 |
| Stage, n (%) | 0.7125 | |||
| Stage 0 | 780 (2.5) | 634 (2.4) | 146 (2.6) | |
| Stage I | 8,844 (28.0) | 7,265 (27.9) | 1,579 (28.0) | |
| Stage II | 12,220 (38.6) | 10,058 (38.7) | 2,162 (38.4) | |
| Stage III | 6,369 (20.1) | 5,253 (20.1) | 1,116 (19.8) | |
| Stage IV | 3,422 (10.8) | 2,788 (10.7) | 634 (11.3) | |
| Site, n (%) | 0.1018 | |||
| Right colon | 19,179 (60.6) | 15,819 (60.9) | 3,360 (59.6) | |
| Transverse colon | 3,331 (10.5) | 2,756 (10.6) | 575 (10.2) | |
| Left colon | 8,361 (26.4) | 6,801 (26.2) | 1,560 (27.7) | |
| Overlapping sites/unspecified | 764 (2.4) | 622 (2.4) | 142 (2.5) |
IQR, interquartile range
Univariate analysis: postoperative care requirements
A larger proportion of rural individuals experienced a postoperative complication (24.1% vs 22.2%, p = 0.0013) and more rural individuals were admitted to the ICU (34.0% vs 30.4%, p < 0.0001), whereas fewer were admitted to step-down units (10.4% vs 14.8%, p < 0.0001). Compared with urban, rural individuals were evaluated by physical therapists less often (53.5% vs 63.8%%, p < 0.0001), and fewer rural patients who were prefrail and frail were evaluated by PT (prefrail: 48.5% vs 58.7%, p < 0.0001; frail: 68.0% vs 78.9%, p < 0.0001). However, rural patients had slightly shorter LOS compared with urban patients for both minimally invasive surgery (median 4 days, interquartile range [3 to 7 days] vs 4 [3 to 8]; p = 0.0039) and open colectomies (7 [5 to 11] vs 7 [5 to 12]; p < 0.0001; Table 2).
Table 2.
Univariate Differences in Postoperative Outcomes Among Rural and Urban Patients
| Outcomes | Overall | Urban | Rural | p Value |
|---|---|---|---|---|
| Total patients, n (%) | 31,635 (100) | 25,998 (82.2) | 5,637 (17.8) | — |
| LOS-open, d, median (IQR) | NA | 7 (5–12) | 7 (5–11) | <0.0001 |
| LOS-MIS, d, median (IQR) | NA | 4 (3–8) | 4 (3–7) | 0.0039 |
| Complication, n (%) | 7,119 (22.5) | 5,759 (22.2) | 1,360 (24.1) | 0.0013 |
| ICU admission, n (%) | 9,810 (31.0) | 7,895 (30.4) | 1,915 (34.0) | <0.0001 |
| Physical therapy evaluation, n (%) | 19,609 (62.0) | 16,593 (63.8) | 3,016 (53.5) | <0.0001 |
| 30-d readmission, n (%) | 4,589 (14.5) | 3,732 (14.4) | 857 (15.2) | 0.1012 |
| Postdischarge reoperation, n (%) | 113 (0.4) | 79 (0.3) | 34 (0.6) | 0.0006 |
| Discharge disposition, n (%) | <0.0001 | |||
| Home without services | 17,014 (53.8) | 13,718 (52.8) | 3,296 (58.5) | |
| Home with home healthcare | 6,078 (19.2) | 5,145 (19.8) | 933 (16.6) | |
| Inpatient rehabilitation | 1,071 (3.4) | 894 (3.4) | 177 (3.1) | |
| Skilled nursing facility or other long-term care facility | 6,873 (21.7) | 5,868 (22.6) | 1,005 (17.8) | |
| Other* | 599 (1.9) | 373 (1.4) | 226 (4.0) |
Other: transfer to a federal hospital, critical access hospital, psychiatric facility, or home hospice/hospice facility.
IQR, interquartile range; LOS, length of stay; MIS, minimally invasive surgery; NA, not applicable.
Rural patients were discharged home without postacute care services more often than urban patients (58.5% vs 52.8%), and fewer rural patients received home healthcare (16.6% vs 19.8%). Additionally, rural patients were less often discharged to SNFs or other short- or long-term care facilities (17.8% vs 22.6%, p < 0.0001). Rural and urban patients were similarly readmitted at 30 days postoperatively (p > 0.05). However, rural patients underwent more reoperations within 30 days of discharge (0.6% vs 0.3%, p = 0.0006; Table 2).
Multivariable logistic regression: discharge with services
Overall
In multivariable models controlling for age, sex, race, acuity of presentation, stage, CCI, complications, and frailty, rural patients had lower odds of being discharged with support services (odds ratio [OR] 0.76, 95% CI 0.71 to 0.82). Those who were older than 75 years (OR 2.65, 95% CI 2.51 to 2.80), women (OR 1.31, 95% CI 1.24 to 1.38), presented emergently (OR 4.28, 95% CI 4.06 to 4.52), had stage IV disease (OR 1.73, 95% CI 1.59 to 1.88), higher CCI (OR 1.12, 95% CI 1.10 to 1.14), complications (OR 2.86, 95% CI 2.68 to 3.04), and were prefrail (OR 1.64, 95% CI 1.44 to 1.87) and frail (OR 3.30, 95% CI 2.86 to 3.80) had higher odds of being discharged with services (Fig. 1).
Figure 1.
Odds of being discharged with postacute care services. Charlson, Charlson Comorbidity Index; LCL, lower confidence limit; OR, odds ratio; UCL, upper confidence limit.
Additionally, rural (OR 0.73, 95% CI 0.67 to 0.79) and non-White (OR 0.88, 95% CI 0.82 to 0.94) individuals had lower odds of being discharged to a facility, whereas older (OR 2.50, 95% CI 2.35 to 2.67), woman (OR 1.25, 95% CI 1.17 to 1.32), emergent (OR 4.28, 95% CI 4.03 to 4.54), late-stage (OR 1.10, 95% CI 1.01 to 1.20), higher CCI (OR 1.11, 95% CI 1.09 to 1.13), postcomplication (OR 3.27, 95% CI 3.07 to 3.48), prefrail (OR 1.84, 95% CI 1.53 to 2.21), and frail (OR 3.55, 95% CI 2.92 to 4.30) patients had higher odds of being discharged to a facility (Supplemental Digital Content 2, https://links.lww.com/JACS/A558).
Needs-based outcomes
When models were stratified by discharge needs, rural patients with needs (ie postoperative complications, prefrail or frail preoperatively), were less likely to be discharged with services compared with urban patients (OR 0.80, 95% CI 0.75 to 0.86). Individuals with needs who were older (OR 2.74, 95% CI 2.60 to 2.88), woman (OR 1.32, 95% CI 1.26 to 1.39), presented emergently (OR 4.21, 95% CI 3.99 to 4.45), had stage IV disease (OR 1.74, 95% CI 1.60 to 1.89), and higher CCI (OR 1.26, 95% CI 1.24 to 1.28) had higher odds of discharge with services (Fig. 2). Similarly, rural (OR 0.77, 95% CI 0.72 to 0.83) and non-White (OR 0.86, 95% CI 0.80 to 0.92) individuals with postacute care needs had lower odds of discharge to a facility (Supplemental Digital Content 3, https://links.lww.com/JACS/A558).
Figure 2.
Odds of being discharged with postacute care services among those with discharge care needs (experienced a complication, were prefrail or frail preoperatively). Charlson, Charlson Comorbidity Index; LCL, lower confidence limit; OR, odds ratio; UCL, upper confidence limit.
Regional postacute care facilities
Rural counties had higher densities of SNF facilities (median [interquartile range]; rural: 8.2 [4.8 to 13.1] vs urban: 3.6 [2.7 to 4.5], p < 0.0001), SNF beds (rural: 801 [520 to 1,225] vs urban: 471 [311 to 630], p < 0.0001), and home healthcare agencies (median [interquartile range]; rural: 3.2 [0 to 6.8] vs urban: 1.9 [0.6 to 5.2], p < 0.0001) per 100,000 compared with urban counties (Fig. 3).
Figure 3.
(A) Rural and urban county-level densities of skilled nursing facilities (SNFs) in Georgia, Iowa, Louisiana, and California. (B) Rural and urban county-level densities of home healthcare agencies in Georgia, Iowa, Louisiana, and California.
DISCUSSION
In this study of SEER–Medicare patients undergoing CC resection, we found that despite experiencing more postoperative complications and requiring more ICU admissions, rural patients had lower odds of being discharged with postacute care services compared with urban patients. Although rural and urban patients had similar readmission rates, rural patients underwent more reoperations within 30 days of discharge. We developed needs-based multivariable models to determine whether individuals received necessary postacute care services. Using this novel approach, we found that rural patients who experienced complications or were frail or prefrail had lower odds of being discharged with services compared with their urban counterparts. We also examined area-level availability of SNF facilities, beds, and home healthcare agencies and found that rural counties had higher densities of services compared with urban areas.
The SEER–Medicare database is ideal for studying receipt of postacute care services after CC surgery, given that older surgical patients are more likely to require services than younger patients, and all patients in the database have reliable insurance coverage.27 We found that rural patients experienced more postoperative complications, including SSIs and anastomotic leaks, which are often associated with postdischarge needs such as dressing changes, intravenous antibiotic administration, and ostomy care. However, rural patients experiencing complications still had lower odds of receiving services compared with urban patients with complications. Although there is a paucity of earlier data on rural–urban discharge disparities after CC resection, the orthopaedic literature presents similar findings. After elective joint arthroplasty, rural patients in a National Inpatient Sample study had 41% lower odds of receiving postacute care services compared with urban patients.13 Additionally, a Medicare claims study found that after total knee arthroplasty, patients in rural settings received 11% fewer home-based PT visits than their urban counterparts.28 A separate study of post-ICU Medicare beneficiaries found that rural individuals had 6% fewer home health visits than urban individuals.29 Notably, although we identified that rural patients were evaluated by PT less often than urban patients, we could not determine specific reasons rural patients were less likely to receive services, such as preexisting support at home or patient refusal.
In our needs-based model, we studied complications as well as frailty and found that rural prefrail and frail individuals had lower odds of being discharged with services or being discharged to a care facility. Additionally, PT evaluation is frequently a prerequisite to receiving services at discharge, but the rural prefrail and frail in our study were evaluated by PT while inpatient less often than their urban counterparts. Frail patients benefit from postacute care services, and 1 multicenter study demonstrated that receipt of postacute care services was associated with reduced readmissions, emergency room visits, and mortality among frail patients.30 Some authors have framed receiving services on discharge as a loss of independence and have included discharge with home healthcare in this definition, whereas others have defined loss of independence as discharge anywhere other than home.31,32 However, neither of these frameworks considers needs-based discharges, where discharge with services is an appropriate outcome. For example, patients with SSIs benefit from home healthcare to assist with dressing changes, and patients with new stomas may require help managing stoma appliances. We propose that evaluating whether individuals receive the services they need may be a more applicable and nuanced approach, which we accomplished by stratifying models by frailty and complications. These results have important implications for preoperative counseling and setting patients’ expectations about potential long-term postoperative courses.
Density or availability of services
To unpack the rural–urban disparity in receipt of postacute care services, we examined the densities of SNFs and home healthcare agencies in rural areas. We found that rural counties had significantly higher densities of SNF facilities, beds, and home healthcare agencies, likely due to rural counties’ smaller populations, although this varied among different states. However, as demonstrated in our maps, density is an imperfect measure, because services are meant to serve large geographic regions with dispersed populations, and some counties do not have any home health agencies. This issue is further compounded by SNF closures in rural regions.33 Additionally, rural home health agencies may struggle with staffing, infrastructure, and travel burdens, as staff face the challenge of traveling further distances between patients, reducing the number of visits providers can accomplish in a given day.12 In response, Centers for Medicare & Medicaid Services implemented a rural add-on policy in 2001, in which home health agencies were paid 10% more for visits in rural areas.34 However, the rural add-on adjustment slowly decreased and was phased out by 2023.35 Alternative incentives may be helpful for surpassing the previously mentioned logistical barriers and increasing availability of rural home healthcare services, including those targeted at increasing staffing and improving infrastructure.36 Because the rural–urban disparity in postacute care services was not explained by an unavailability of services in rural regions, provider bias and local discharge practices should be examined and addressed by hospital systems, as these practices result in rural patients receiving fewer services.
Additionally, existing literature highlights broad geographic variation in quality of SNFs with rural patients residing in lower-quality SNFs, including those with higher facility-acquired contracture rates and lower stars on the 5-star rating system developed by the Centers for Medicare and Medicaid Services.37–39 Including quality measures of postacute care facilities may provide greater insight into the complex issue of access to high-quality services among rural populations.
Limitations
Given our use of the SEER–Medicare database, we were not able to identify certain operative details, such as whether a hand-assisted resection was performed. We also could not determine whether patients resided in SNFs before surgery, although we expect this was a low percentage of our cohort. However, by using claims data, we were able to capture a wide range of discharge settings. We also used a claims-based frailty index which has been validated using Medicare claims data, but has not been validated in rural populations specifically; literature suggests some rural patients underuse healthcare, and therefore, frailty measures and CCI may underestimate comorbidities for rural individuals.40 Additionally, RUCC categories are based on population thresholds and adjacency to urban areas, but a classification system based on population density may have relevance for future studies investigating accessibility of postacute care services in rural areas.41,42 Finally, the RUCC codes in the SEER–Medicare database apply to patients’ home addresses rather than those of the treating hospitals; although some rural patients travel for cancer care, literature suggests rural inpatients at urban hospitals generally go to SNFs in rural regions near their homes.43
CONCLUSIONS
After CC resection, rural individuals had more ICU admissions and more complications, but fewer PT evaluations and shorter LOS. Rural patients, including the prefrail, frail, and those experiencing complications, had lower odds of receiving postacute care services. However, they also underwent more postdischarge reoperations. Taken together, these results suggest rural patients may benefit from more postacute care services. Such services could help fill the critical gap in care after discharge, particularly given that rural patients tend to live further from hospitals. Demand for hospital beds and pressure to discharge patients could be mitigated by increasing the availability of postacute care services. Finally, although density is an imperfect measure of access to high-quality services, the disparity we identified in discharge disposition is not explained by a dearth of services in rural counties. Underlying reasons for the rural–urban disparity in postacute care services may be further investigated with future qualitative studies.
Supplementary Material
Supplemental digital content is available for this article.
Support:
The collection of cancer incidence data used in this study was supported by the California Department of Public Health pursuant to California Health and Safety Code Section 103885; CDC National Program of Cancer Registries, under cooperative agreement 1NU58DP007156; the National Cancer Institute’s Surveillance, Epidemiology and End Results Program under contract HHSN261201800032I awarded to the University of California, San Francisco, contract HHSN261201800015I awarded to the University of Southern California, and contract HHSN261201800009I awarded to the Public Health Institute. Dr Kenzik’s institute is supported by NIH National Cancer Institute grant (R37CA266193). Drs Myers and Murillo are supported by National Research Service Award Institutional Postdoctoral Training Grant (T32HP10028).
Abbreviations and Acronyms
- CC
colon cancer
- CCI
Charlson comorbidity index
- LOS
length of stay
- OR
odds ratio
- PT
physical thera
- RUCC
Rural–urban continuum codes
- SEER
Surveillance, Epidemiology, and End Results
- SNF
skilled nursing facility
- SSI
surgical site infection
Footnotes
Disclosure Information: Nothing to disclose.
Disclosures outside the scope of this work: Dr Davids is a paid consultant to Olympus, Medtronic, and SafeHeal. Other authors have nothing to disclose.
Disclaimer: The ideas and opinions expressed here are those of the author(s) and do not necessarily reflect the opinions of the State of California, Department of Public Health, the National Cancer Institute, and the CDC or their Contractors and Subcontractors. The contents of this article are those of the authors and may not reflect the policies of the Health Resources and Services Administration, the US Department of Health and Human Services, or the US Government.
Presented at the American College of Surgeons 111th Annual Clinical Congress, Scientific Forum, Chicago, IL, October 2025.
Contributor Information
Sara Myers, Department of Surgery, Boston Medical Center, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA; Department of Surgery, Boston University Chobanian and Avedisian School of Medicine, Boston, MA.
Wang Pong Chan, Department of Surgery, Boston Medical Center, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA; Department of Surgery, Boston University Chobanian and Avedisian School of Medicine, Boston, MA.
Elizabeth S Davis, Department of Surgery, Brigham and Women’s Hospital, Boston, MA.
Anays Murillo, Department of Surgery, Boston Medical Center, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA.
Sing Chau Ng, Department of Surgery, Boston Medical Center, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA.
Olga Beresneva, Department of Surgery, Boston Medical Center, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA; Department of Surgery, Boston University Chobanian and Avedisian School of Medicine, Boston, MA.
Samantha Rivard, Department of Surgery, Boston Medical Center, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA; Department of Surgery, Boston University Chobanian and Avedisian School of Medicine, Boston, MA.
Teviah Sachs, Department of Surgery, Boston Medical Center, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA; Department of Surgery, Boston University Chobanian and Avedisian School of Medicine, Boston, MA.
Jennifer S Davids, Department of Surgery, Boston Medical Center, Chobanian and Avedisian School of Medicine, Boston University, Boston, MA; Department of Surgery, Boston University Chobanian and Avedisian School of Medicine, Boston, MA; Department of Surgery, Brigham and Women’s Hospital, Boston, MA..
Kelly M Kenzik, Department of Surgery, Brigham and Women’s Hospital, Boston, MA..
REFERENCES
- 1.US Census Bureau. Nation’s urban and rural populations shift following 2020 census. Available at: https://www.census.gov/newsroom/press-releases/2022/urban-rural-populations.html. Accessed April 23, 2025.
- 2.Raman V, Adam MA, Turner MC, et al. Disparity of colon cancer outcomes in rural America: making the case to travel the extra mile. J Gastrointest Surg 2019;23:2285–2293. [DOI] [PubMed] [Google Scholar]
- 3.Pandit V, Jehan F, Zeeshan M, et al. Failure to rescue in postoperative patients with colon cancer: time to rethink where you get surgery. J Surg Res 2019;234:1–6. [DOI] [PubMed] [Google Scholar]
- 4.Ramkumar N, Colla CH, Wang Q, et al. Association of rurality, race and ethnicity, and socioeconomic status with the surgical management of colon cancer and postoperative outcomes among Medicare beneficiaries. JAMA Netw Open 2022;5:e2229247. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chow CJ, Al-Refaie WB, Abraham A, et al. Does patient rurality predict quality colon cancer care? A population based study. Dis Colon Rectum 2015;58:415–422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ambroggi M, Biasini C, Del Giovane C, et al. Distance as a barrier to cancer diagnosis and treatment: review of the literature. Oncologist 2015;20:1378–1385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kelley KA, Young JI, Bassale S, et al. Travel distance influences readmissions in colorectal cancer patients—what the primary operative team needs to know. J Surg Res 2018;227:220–227. [DOI] [PubMed] [Google Scholar]
- 8.Khuri SF, Henderson WG, DePalma RG, et al. Determinants of long-term survival after major surgery and the adverse effect of postoperative complications. Ann Surg 2005;242:326–341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Silber JH, Rosenbaum PR, Trudeau ME, et al. Changes in prognosis after the first postoperative complication. Med Care 2005;43:122–131. [DOI] [PubMed] [Google Scholar]
- 10.Warps AK, Tollenaar RAEM, Tanis PJ, Dekker JWT. Postoperative complications after colorectal cancer surgery and the association with long-term survival. Eur J Surg Oncol 2022;48:873–882. [DOI] [PubMed] [Google Scholar]
- 11.Tevis SE, Kennedy GD. Postoperative complications: looking forward to a safer future. Clin Colon Rectal Surg 2016;29:246–252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kosar CM, Loomer L, Ferdows NB, et al. Assessment of rural-urban differences in postacute care utilization and outcomes among older US adults. JAMA Netw Open 2020;3:e1918738. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Burke RE, Jones CD, Coleman EA, et al. Use of post-acute care after hospital discharge in urban and rural hospitals. Am J Accountable Care 2017;5:16–22. [PMC free article] [PubMed] [Google Scholar]
- 14.SEER. Surveillance, Epidemiology, and End Results Program. Available at: https://seer.cancer.gov/index.html. Accessed January 29, 2025.
- 15.National Cancer Institute. Division of Cancer Control & Population Sciences. SEER-Medicare linked data resource. Available at: https://healthcaredelivery.cancer.gov/seermedicare/. Accessed January 29, 2025.
- 16.Romatoski KS, Chung SH, Sawhney V, et al. Factors associated with surgical management in gallbladder cancer—a surveillance, epidemiology, and end results Medicare–based study. J Surg Res 2024;304:9–18. [DOI] [PubMed] [Google Scholar]
- 17.Torres CM, Myers S, Janeway MG, et al. Exploring rural-urban differences in the receipt of nonelective cancer-specific gastrointestinal surgery using a multilevel mixed-effects approach. J Gastrointest Surg 2025;29:101858. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kim DH, Schneeweiss S, Glynn RJ, et al. Measuring frailty in Medicare data: development and validation of a claims-based frailty index. J Gerontol A Biol Sci Med Sci 2018;73:980–987. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kim DH, Patorno E, Pawar A, et al. Measuring frailty in administrative claims data: comparative performance of four claims-based frailty measures in the U.S. Medicare data. J Gerontol A Biol Sci Med Sci 2020;75:1120–1125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.National Cancer Institute. Division of Cancer Control & Population Sciences. BASE FFS Claims Files. October 9, 2024. Available at: https://healthcaredelivery.cancer.gov/seermedicare/medicare/claims.html. Accessed March 18, 2025. [Google Scholar]
- 21.USDA Economic Research Service. Rural-urban continuum codes. US Department of Agriculture. 2020. Available at: https://www.ers.usda.gov/data-products/rural-urban-continuum-codes.aspx. Accessed May 20, 2022. [Google Scholar]
- 22.Unger JM, Moseley A, Symington B, et al. Geographic distribution and survival outcomes for rural patients with cancer treated in clinical trials. JAMA Netw Open 2018;1:e181235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lewis-Thames MW, Langston ME, Khan S, et al. Racial and ethnic differences in rural-urban trends in 5-year survival of patients with lung, prostate, breast, and colorectal cancers: 1975–2011 Surveillance, Epidemiology, and End Results (SEER). JAMA Netw Open 2022;5:e2212246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Health Resources and Services Administration. Area Health Resources Files. Available at: https://data.hrsa.gov/topics/health-workforce/ahrf. Accessed April 2, 2025.
- 25.United States Census Bureau. Census Bureau Data. Available at: https://data.census.gov/. Accessed April 2, 2025.
- 26.Lorenzo PD. Usmap: US maps including Alaska and Hawaii. 2024. Available at: https://CRAN.R-project.org/package=usmap. Accessed March 2, 2025.
- 27.Wang YC, Chou MY, Liang CK, et al. Post-acute care as a key component in a healthcare system for older adults. Ann Geriatr Med Res 2019;23:54–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Falvey JR, Bade MJ, Forster JE, et al. Home-health-care physical therapy improves early functional recovery of Medicare beneficiaries after total knee arthroplasty. J Bone Joint Surg Am 2018;100:1728–1734. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Falvey JR, Murphy TE, Gill TM, et al. Home health rehabilitation utilization among Medicare beneficiaries following critical illness. J Am Geriatr Soc 2020;68:1512–1519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lee MC, Wu TY, Huang SJ, et al. Post-acute care for frail older people decreases 90-day emergency room visits, readmissions and mortality: an interventional study. PLoS One 2023;18:e0279654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Tanaka S, Iida H, Ueno M, et al. Preoperative risk assessment for loss of independence following hepatic resection in elderly patients: a prospective multicenter study. Ann Surg 2021;274:e253–e261. [DOI] [PubMed] [Google Scholar]
- 32.Owodunni OP, Mostales JC, Qin CX, et al. Preoperative frailty assessment, operative severity score, and early postoperative loss of independence in surgical patients age 65 years or older. J Am Coll Surg 2021;232:387–395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Sharma H, Bin Abdul Baten R, Ullrich F, et al. Nursing home closures and access to post-acute care and long-term care services in rural areas. J Rural Health 2024;40:557–564. [DOI] [PubMed] [Google Scholar]
- 34.Loomer L, Rahman M, Mroz TM, et al. Impact of higher payments for rural home health episodes on rehospitalizations. J Rural Health 2023;39:604–610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.CMS. Home health rural add-on policy. Available at: https://www.cms.gov/outreach-and-education/medicare-learning-network-mln/mlnproducts/fast-facts/home-health-rural-add-policy. Accessed April 3, 2025.
- 36.Landers S, Madigan E, Leff B, et al. The future of home health care: a strategic framework for optimizing value. Home Health Care Manag Pract 2016;28:262–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.CMS. Five-star quality rating system. Available at: https://www.cms.gov/medicare/health-safety-standards/certification-compliance/five-star-quality-rating-system. Accessed May 7, 2025.
- 38.Zuckerman RB, Wu S, Chen LM, et al. The five-star skilled nursing facility rating system and care of disadvantaged populations. J Am Geriatr Soc 2019;67:108–114. [DOI] [PubMed] [Google Scholar]
- 39.Bowblis JR, Meng H, Hyer K. The urban-rural disparity in nursing home quality indicators: the case of facility-acquired contractures. Health Serv Res 2013;48:47–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Nuako A, Liu J, Pham G, et al. Quantifying rural disparity in healthcare utilization in the United States: analysis of a large midwestern healthcare system. PLoS One 2022;17:e0263718. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Fischer A, Hutchinson JMS, Nelson KS. Where is rural? An analysis of the agreement between quantitative measures of rurality. J Rural Stud 2024;111:103424. [Google Scholar]
- 42.Davis ES, Franks JA, Bhatia S, Kenzik KM. Urban–rural differences in cancer mortality: operationalizing rurality. J Rural Health 2024;40:268–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Hoffman AF, Pink GH, Kirk DA, et al. What characteristics influence whether rural beneficiaries receiving care from urban hospitals return home for skilled nursing care? J Rural Health 2020;36:94–103. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.