Maximizing Readmission Reduction in Colon Cancer Patients

Abstract

Background:

Multiple studies have identified risk factors for readmission in colon cancer patients. We need to determine which risk factors, when modified, produce the greatest decrease in readmission for patients so that limited resources can be used most effectively by implementing targeted evidence-based performance improvements. We determined the potential impact of various modifiable risk factors on reducing 30-day readmission in colon cancer patients.

Study Design:

We used a cohort design with the 2012–2020 American College of Surgeons’ National Surgical Quality Improvement Program (ACS-NSQIP) data to track colon cancer patients for 30 days following surgery. Colon cancer patients who received colectomies and were discharged alive were included. Readmission (to the same or another hospital) for any reason within 30 days of the resection was the outcome measure. Modifiable risk factors were the use of minimally invasive surgery (MIS) versus open colectomy, mechanical bowel preparation, preoperative antibiotic use, functional status, smoking, complications (deep vein thrombosis, pulmonary embolism, myocardial infarction, stroke, infections, anastomotic leakage, prolonged postoperative ileus, extensive blood loss, sepsis), serum albumin, and hematocrit.

Results:

111,691 patients with colon cancer were included in the analysis. About half of the patients were male, most were aged 75 or older, and were discharged home. Overall, 11,138 patients (10.0%) were readmitted within 30 days of surgery. In adjusted analysis, the reduction in readmission would be largest by preventing both prolonged ileus and by switching open colectomies to MIS (28.0% relative reduction) followed by preventing anastomotic leaks (6.2% relative reduction). Improving other modifiable risk factors would have a more limited impact.

Conclusions:

The focus of readmission reduction should be on preventing prolonged ileus, increasing the use of MIS, and preventing anastomotic leaks.

Introduction

Nearly 150,000 patients were newly diagnosed with colorectal cancer in the U.S. in 2020.(1) Hospital readmissions among CRC patients are common and costly with an estimated $300 million in readmission cost annually.(2) More than ten percent of CRC patients are readmitted within 30 days following hospital discharge.(3) These readmitted patients have worse outcomes, including worse prognosis.(4)

While multiple studies have identified risk factors for readmission in colon cancer patients such as older age, comorbid conditions, preoperative immunosuppressive therapy, postoperative complications, and nonhome discharge,(5) this tells us nothing about how the readmission burden might change if a risk factor were modified. Once risk factors are identified, we then need to determine which risk factors, when modified, produce the greatest decrease in readmission for patients so that policymakers, clinicians, and quality improvement staff can use their limited resources most effectively by implementing targeted evidence-based performance improvements. A recent example of this approach showed that improving pre-operative hematocrit through blood transfusion could reduce predicted readmission by 12% in general surgery patients.(6) However, many hospitals and local policymakers do not know what would make the biggest impact on readmission considering available resources.(7) It remains to be determined whether intensified quality improvement efforts can further decrease readmission rates.(8) In this study, we calculated the potential effect of changes in modifiable risk factors on reducing 30-day readmission in colon cancer patients.

Materials and Methods

Data source and patient selection

We used the 2012–2020 American College of Surgeons’ National Surgical Quality Improvement Program (ACS-NSQIP) Participant Use File and Targeted Colectomy File. The ACS-NSQIP measures the quality of surgical care by collecting patient-level data at participating hospitals. The University of Arkansas for Medical Sciences Institutional Review Board considered this study nonhuman subjects research. The IRB approved a waiver of informed consent. Preoperative patient characteristics, preoperative laboratory results, intraoperative procedure characteristics, surgical complications, and 30-day postoperative readmissions were abstracted from participating hospitals’ medical records by trained reviewers as detailed elsewhere. (https://www.facs.org/quality-programs/acs-nsqip) Patients who underwent resection for colon cancer (with or without obstruction) were eligible for inclusion. Patients who died during their hospitalization and whose American Society of Anesthesiologists (ASA) physical status was class V (not expected to survive without surgery) or missing were excluded.(9)

Because NSQIP captures readmission within 30 days of resection, rather than 30-day post-discharge, patients with a longer length of stay have less opportunity to be readmitted.(10) For example, a patient discharged on post-op day 25 will have only 5 days to be readmitted in the 30 days post-resection period. Therefore, we excluded patients with a length of stay longer than 20 days. We conducted a sensitivity analysis to determine the robustness of our findings.

Readmission

Readmission was defined as any readmission (to the same or another hospital), for any reason, within 30 days of the resection. Patients who were discharged from their index hospital stay after their primary procedure, and within 30 days of the primary procedure, are subsequently formally readmitted by a qualified practitioner as an inpatient to an acute care bed or otherwise have a subsequent hospital (or facility-based) encounter (receiving outpatient, emergency department or observation services) that crossed at least two midnights.

Modifiable and non-modifiable risk factors for readmission

Potential risk factors were selected based on previous studies of risk factors of 30-day readmission among persons with colon cancer. (5, 11, 12) These factors included patient demographic characteristics (sex, age group, Hispanic ethnicity, race, and year of surgery), treatment-related variables for colon cancer, comorbidities, behavior-related variables, surgical complications, and biomarkers of underlying disease. Detailed descriptions were provided below. Some of the following risk factors are modifiable during the course of treatment while others are not.

Treatment-related risk factors included type of surgery, mechanical bowel preparation, preoperative antibiotic use, chemotherapy, AJCC TNM stage, length of hospitalization in days, tumor location, emergency surgery, steroid use 30 days prior resection, receipt of transfusion from the start of surgery to 72 hours post-surgery, and return to the OR. Colostomy and ileostomy were identified by CPT codes.(13) Based on clinical guidelines,(14–16) the use of minimally invasive surgery (MIS) versus open colectomy, mechanical bowel preparation, and preoperative antibiotic use are modifiable opportunities for potential improvement to reduce readmission.

Comorbidities included patient functional status (partially dependent or totally dependent versus independent), hypertension, congestive heart failure, renal failure, preoperative loss of blood necessitating a transfusion, weight loss>10% in the last six months, dyspnea, chronic obstructive pulmonary disease (COPD),(17) and American Society of Anesthesiologists (ASA) physical status classification system. All were measured within 30 days prior to surgery. The NSQIP data captures functional status by collecting data about the best physical functional status/level of self-care as demonstrated by the patient within the 30 days prior to the primary procedure or at the time the patient is being considered a candidate for surgery. Patient diabetes status was classified in the NSQIP data as “no diabetes” (no diagnosis of diabetes or diabetes controlled by diet alone), “diabetes not requiring insulin” (diabetes requiring therapy with a non-insulin anti-diabetic agent, such as oral agents or other non-insulin agents), or “diabetes requiring insulin” (diabetes requiring daily insulin therapy). The latter two groups of patients were contrasted with the “no diabetes” group in the analysis. A potentially modifiable risk factor is improvement in functional status since prehabilitation and geriatric assessments may improve this.(18)

Behavior-related risk factors included smoking status and BMI. Patients who had smoked cigarettes in the year prior to admission for surgery were considered smokers. The patient’s most recent height and weight, documented in the medical record within the 30 days prior to the colectomy or at the time the patient was being considered a candidate for surgery, was used to calculate BMI in kg/m². Patients were classified as underweight (BMI<18.5), normal weight (18.5≤BMI<25.0), overweight (25.0≤BMI<30.0), class 1–2 obesity (30.0≤BMI<40.0), or class 3 obesity (BMI≥40.0). Smoking is considered modifiable based on a meta-analysis that showed that smoking cessation lowers postoperative wound healing problems and surgical site wound infections following surgery,(19) which are risk factors for readmission. BMI may be improved through physical activity as part of prehabilitation although evidence for this is weak, especially when focusing on readmission.(20) Recent guidelines state a lack of evidence regarding diet and weight management interventions during cancer treatment.(21)

Complications included deep vein thrombosis, pulmonary embolism, myocardial infarction, stroke, infections (surgical site, wound, deep), anastomotic leakage, and prolonged postoperative ileus, extensive blood loss, and having at least one occurrence of sepsis up to 30 days after surgery. A wound infection included superficial or deep incisional surgical site infection or any other wound infection. Sepsis was defined by Systemic Inflammatory Response Syndrome criteria and either a positive blood culture/clinical documentation or a suspected preoperative clinical condition of infection. All complications were potentially modifiable.

Biomarkers included the following preoperative laboratory values with abnormal value cutoffs noted in parentheses: white blood cell count (≤4.5 or ≥11.0 × 10³/μL); glomerular filtration rate; serum albumin (<3.0 mg/dL); serum alkaline phosphatase (>125 mg/dL); serum bilirubin (≥1.0 mg/dL); blood urea nitrogen (BUN, ≥40 mg/dL) and BUN-to-creatinine (BUN/creatinine, >20) ratio;(22) platelet count (150.0–450.0 × 10³/μL); partial thromboplastin time (>35); and hematocrit. Hematocrit values were treated as a categorical variable with the following cutoffs for anemia: severe (<25%), moderate (25% to <29%), mild (29% to <37%), and no anemia (≥37%).(23) Glomerular filtration rate (GFR) was calculated based on race, sex, age, and creatinine levels with a cutoff GFR>60 mL/min per 1.73 m².(24) Anemia based on hematocrit was a potentially modifiable factor.

Statistical analysis

Following descriptive statistics about the study population, we implemented four steps for estimating potential effects of changes in modifiable risk factors on readmission based on missing counterfactual observations.(25) The counterfactual model was used since it enhances causal inference in observational studies, standardizes risk factors allowing one to compare their values directly, and provides useful information for policymakers, clinicians, and quality improvement staff focused on interventions.

First, we estimated the association between all factors and readmission using multivariable logistic regression. We included all modifiable and nonmodifiable risk factors in the model because one of the assumptions of counterfactual modeling is that there are no unmeasured confounders. We included an interaction between type of surgery and prolonged ileus based on previously published studies.(26) We used the area under the receiver operating curve (AUC) and McFadden’s R square to describe the fit and discriminatory power of a model. Second, we used the model from step 1 to calculate the counterfactual probabilities of readmission for each patient while “setting” the values of a modifiable risk factor to different levels in the adjusted logistic model, holding everything else at its actual value for each patient. Third, we averaged the imputed probabilities for each patient across the whole population at each level of a modifiable risk factors. These averages of the individual probabilities estimated the “counter-factual” readmission rate for each level of the modifiable risk factor, if that “set” level of the risk factor had been present in all patients. This results in a population-wide predicted effect of changing the modifiable risk factor on the readmission rate in all patients. Fourth, we calculated confidence intervals around the effect estimate using bootstrapping. We then repeated steps 2–4 for all significant modifiable risk factors of interest and calculate their counterfactual readmission rates. This will allow for the comparison of the potential effect of changes in the modifiable risk factors on the readmission rate.

We calculated the absolute and relative reduction in the readmission rate to directly compare the potential impact of the modifiable risk factors. Absolute reduction is the difference between the predicted (counterfactual) probability of readmission at each level of a modifiable factor and the observed proportion of patients with readmission in this patient cohort (“observed admission rate”). We used the “number of patients needed to intervene to prevent one patient from readmission”, or the reciprocal of the absolute risk reduction, to assess the effectiveness of interventions on a modifiable risk factor.(27) A modified risk factor is considered a more favorable interventional option if its required number of patients to be intervened is smaller. Relative reduction is the absolute reduction expressed as a percentage from the observed readmission rate. Bootstrapping was used to calculate confidence intervals for absolute and relative reductions. We performed a sensitivity analysis to determine the robustness of our findings by excluding 1) the year 2020 because of the COVID-19 pandemic and 2) patients with a length of stay longer than the 75% (8 days). All analyses were conducted using STATA version 17 (Stata Corp LP, College Station, TX).

Results

Study sample

During the 2012–2020 study period, 119,131 patients with colon cancer were included in the NSQIP data. Patients were excluded because of missing readmission data (61), they died during their impatient stay (1562), ASA was 5 (189), missing ASA (153), length of stay>20 days (5,813), missing duration of surgery (191), or missing sex or age (4), resulting in 111,691 colon cancer patients.

Table 1 shows that about half of patients were male, most were aged 75 or older, and were discharged home. Overall, 11,138 patients (10.0%) were readmitted within 30 days of surgery. Mean length of stay was 5.9 days (median: 5). The 30-day readmission rate was 10.2% in 2012 and 9.6% in 2020. There was a slight yet statistically significant downward trend over time (p=0.001 from a logistic regression model).

Table 1.

Frequency, percentage readmitted, and adjusted association with 30-day readmission for nonmodifiable demographic and treatment risk factors (n=111,691).^*

Risk factor	Percentage of the study population (100.0%)	Unadjusted % readmitted	Adjusted odds ratio (95% confidence interval)
Demographics
Sex
Male	51.0	10.5	1.00 (0.96 – 1.05)
Female	49.0	9.4	Ref
Age group
<45	6.3	10.0	Ref
45–54	15.7	8.7	0.83 (0.75 – 0.92)†
55–64	23.2	9.6	0.88 (0.79 – 0.97)†
65–74	26.8	9.9	0.92 (0.83 – 1.02)
75+	28.0	11.1	1.08 (0.97 – 1.20)
Hispanic ethnicity*
Yes	5.0	10.8	1.09 (0.98 – 1.21)
No	79.4	10.1	Ref
Unknown	15.7	9.1	10.3 (0.92 – 1.16)
Race
White	67.2	10.3	Ref
African American	9.3	10.6	1.01 (0.94 – 1.09)
Other	5.6	8.4	0.95 (0.86 – 1.05)
Unknown	17.9	9.1	0.88 (0.79 – 0.98)†
Discharge disposition
Home	92.0	9.5	0.97 (0.89 – 1.05)
Not home	8.0	15.6	Ref
Other/Unknown	0.1	27.2	2.27 (1.33 – 3.86)
Treatment
Length of hospitalization in days (mean, s.e.)	5.9 (0.01)		0.90 (0.89 – 0.90)†
Tumor location
Left	41.5	10.2	Ref
Right	36.9	9.5	1.01 (0.95 – 1.07)
Transverse	21.6	10.4	1.01 (0.96 – 1.08)
Colostomy or ileostomy
Yes	12.1	15.1	1.39 (1.30 – 1.48)†
No	87.9	9.3	Ref

^†95% confidence interval did not cross 1.0

Discharge disposition Other/Unknown includes left AMA

^*Adjusted for all demographic factors, treatment-related factors, comorbidities, behavior-related factors, complications, and biomarkers.

Reducing readmissions

The logistic model with all risk factors fit well based on the AUC (0.79) and McFadden’s R square (0.178). In the same adjusted model (Table 2), several modifiable risk factors increased the odds of readmission (p<0.05).

Table 2.

Frequency, percentage readmitted, and adjusted association with 30-day readmission for potentially modifiable risk factors (n=111,691).^*

Potentially modifiable risk factor	Percentage of the study population (100.0%)	Unadjusted % readmitted	Adjusted Odds ratio (95% confidence interval)
Treatment
Type of surgery
Open colectomy	34.2	13.0	1.31 (1.24 – 1.38)†
Minimally invasive surgery	65.4	8.4	Ref
Other/Unknown	0.4	13.2	1.42 (1.06 – 1.90)
Mechanical bowel preparation*
Yes	61.1	9.6	1.06 (1.00 – 1.12)
No	26.7	11.1	Ref
Unknown	12.2	9.4	1.01 (0.92 – 1.12)
Preoperative antibiotic use
Yes	42.3	9.3	Ref
No	46.4	11.0	1.12 (1.06 – 1.19)†
Unknown	11.3	9.1	0.95 (0.86 – 1.5)
Comorbidity
Functional status
Partially/totally dependent	2.4	16.4	1.45 (1.28 – 1.63)†
Independent	97.4	9.8	Ref
Unknown	0.3	9.5	0.75 (0.48 – 1.17)
Behavior
BMI, kg/m2
<18.5	2.4	12.6	1.21 (1.06 – 1.39)†
18.5–24.9	29.1	9.7	Ref
25.0–29.9	33.4	9.6	0.96 (0.91 – 1.02)
30.0–39.9	28.1	10.1	0.97 (0.92 – 1.04)
40.0+	5.7	11.5	0.97 (0.88 – 1.08)
Unknown	1.3	10.5	0.96 (0.79 – 1.17)
Smoking
Yes	13.2	11.8	1.11 (1.04 – 1.19)†
No	86.8	9.7	Ref
Complications
Sepsis	2.1	49.9	3.96 (3.57 – 4.41)†
Pulmonary embolism	0.6	51.1	6.69 (5.55 – 8.07)†
Myocardial infarction	0.6	35.5	3.57 (2.97 – 4.29)†
Stroke	0.2	48.8	7.86 (5.91 – 10.46)†
Surgical site infection	3.4	22.0	2.50 (2.29 – 2.74)†
Wound infection	0.7	16.1	1.10 (0.88 – 1.39)
Deep wound infection	0.5	52.6	6.73 (5.51 – 8.20)†
Anastomotic leak	2.2	59.3	5.41 (4.86 – 6.01)†
Prolonged ileus	12.2	25.3	3.48 (3.28 – 3.70)†
C. Difficile	0.7	37.6	5.66 (4.82 – 6.65)†
Biomarkers
Albumin elevated
Yes	6.5	15.9	1.34 (1.23 – 1.46)†
No	67.3	9.9	Ref
Unknown	26.2	8.6	0.91 (0.83 – 1.00)
Anemia
Severe	1.7	13.0	1.24 (0.86 – 1.77)
Moderate	7.5	13.0	1.29 (1.19 – 1.41)
Mild	34.3	10.9	1.12 (1.06 – 1.18)
No anemia	53.1	9.0	Ref
Unknown	3.5	7.6	1.24 (0.86 – 1.77)

^†95% confidence interval did not cross 1.0

^*Adjusted for all demographic factors, treatment-related factors, comorbidities, behavior-related factors, complications, and biomarkers.

The factors with the largest impact on readmission were prolonged ileus and switching from open to MIS procedures (28.0% RR) and prevention of anastomotic leaks (6.2% relative reduction). (Figure 1). Alternatively, one patient might be prevented from being readmitted by switching 36 patients to MIS and preventing prolonged ileus (Figure 2). Additionally, readmission might be reduced by 5.4% by antibiotics use, 5.3% by treating preoperative anemia, followed by preventing sepsis (4.5%), and surgical site infections (3.5%). Other less impactful risk factors included preventing C. Difficile (1.8%), pulmonary embolisms (1.7%), deep infections (1.5%), myocardial infarctions (1.1%), smoking cessation (1.1%), stroke (0.8%), and improving functional status (0.8%). Other risk factors were not significantly associated with readmission in the multivariable model, thus, increasing its use would not significantly improve readmissions either.

Figure 1.

Relative reduction (%) in 30-day readmission (95% confidence interval) by modifiable risk factor.

Figure 2.

Number of patients needed to intervene to prevent one patient from readmission within 30-days (95% confidence interval) by modifiable risk factor.

In sensitivity analyses, the results were generally similar after excluding the year 2020 because of the potential effect of COVID-19 on hospitalizations or when excluding patients with length of stay longer than 75 percentile (8 days) to examine immortal time bias.

Discussion

This is the first study to quantify the potential effect of changes in modifiable risk factors to reduce 30-day readmission in colon cancer patients at the population level. The reduction in 30-day readmission would be largest by preventing both prolonged ileus and by switching open colectomies to MIS (28.0% relative reduction) and preventing anastomotic leaks (6.2% relative reduction).

Strategies focused on other complications such as pulmonary embolisms, myocardial infarction, stroke, and surgical site infections would have a more limited effect on the readmission rate because the rates of these complications are low despite their high risk of readmission. Thus, reducing readmissions will require a concerted focus on improving several key risk factors. Readmission rates are difficult to improve(28) and despite our examination of numerous well-established risk factors, relatively few opportunities were identified that would significantly impact readmission. Nonmedical factors such as unmet needs related to social determinants of health, a history of homelessness, or substance use disorder should be examined.(29)

The largest potential improvements in readmission were by focusing on prolonged ileus in conjunction with switching from open colectomies to MIS. Other studies also showed that prolonged ileus increased the risk of readmission following colectomy.(26, 30) Potential risk factors (not including confounders such as an infection) may include smoking, weight loss, preoperative oral antibiotics, mechanical bowel preparation, and surgical approach,(26) but more studies are needed. While it can be challenging to identify a single source of ileus in the absence of a confounder, several interventions have been identified that may decrease the incidence of prolonged ileus. A recent meta-analysis of 18 randomized controlled trials showed that chewing gum and Alvimopan each resulted in a 60% lower risk of prolonged postoperative ileus.(31) Alvimopan is an inhibitor of the μ-opioid receptor blocking the peripheral effect of opioid drugs on the GI tract. Alvimopan use was also associated with a decrease in readmissions(32) although cost may be an issue prohibiting widespread use. While the American Society for Enhanced Recovery and Perioperative Quality Initiative recommends the implementation of Enhanced Recovery Protocols, which aim to optimize organ function following surgical stress to maintain homeostasis and accelerate postoperative recovery, including early return of bowel function(33) a recent review described no effect on readmission.(34)

The American Society of Colon and Rectal Surgeons recommend the use of MIS when expertise is available.(15) In most circumstances, MIS is preferred including among obese/nonobese patients, in T4a/T4b patients, and regardless of the location of the tumor.(35) However, minimally invasive approach is not always feasible. The NSQIP data did not include any variables that allowed us to determine if the use of MIS was dependent on surgeon and/or patient factors. MIS, including laparoscopic and robotic surgery, has become increasingly popular as a surgical approach in recent years due to its benefits on patient outcomes, including lower blood loss, earlier return of bowel function, improved patient comfort, faster recovery times, shorter hospital stay, and reduced readmissions.(36) Our results show the importance of switching from open colectomies to MIS and its impact on prolonged ileus.

Although much lower, the third largest potential improvement in readmission reduction (6.2%) was by preventing anastomotic leaks, which increase the risk of readmission.(37) Adequate blood supply, lack of tension on the anastomosis, and a negative intraoperative leak test are key intraoperative factors that can be modified to prevent leakage.(38) Other modifiable risk factors played less of a role in potentially affecting 30-day readmission. Addressing anemia could improve readmission by 5.3%. Preoperative anemia in colon cancer patients can be due to chronic bleeding, anemia of chronic disease, nutritional deficiency, and preoperative chemotherapy treatment. Treatment of preoperative anemia with iron may result in reduced complication risks and possibly readmission.(39)

Sepsis is one of the costliest reasons for readmission following colorectal surgery (40) and the fifth largest potential improvement in readmission (4.5%). The key to improved outcomes from sepsis is early identification and prompt management.(41) Some studies observed improved patient outcomes by implementing a sepsis prediction model,(42, 43) but others did not.(44) Unlike others, who suggested that readmission reduction should focus on preventing postoperative complications such as surgical site infections,(2) our results show that, despite the high risk of readmission, the improvement in readmissions by preventing such infections is lower than focusing on preventing prolonged ileus and anastomotic leakage and increasing the use of MIS. Our approach accounts for both the risk of readmission and the frequency distribution of a specific risk factor. In addition, our results show that the relative reduction in readmission is only 0.8% when improving functional status in patients. Although mechanical bowel preparation, especially in conjunction with antibiotics, is recommended (16) our results showed that mechanical bowel preparation would not reduce readmissions.

A strength of our study is its inclusion of readmission from multiple hospitals, the use of numerous risk factors for readmission, and our novel approach of quantifying the effect of changes in modifiable risk factors on 30-day readmission. Previous studies have used regression models to identify risk factors for readmission, but this approach provides little information about the extent to which readmission might change if the underlying population distribution of these risk factors were modified.(45) Using the counterfactual framework effectively standardizes the effect of potential changes in the modifiable risk factors, allowing for the direct comparison of their values and determining which changes will have the greatest impact. This approach provides information that is relevant for local health system leaders, policy makers, quality improvement staff, and clinicians by providing quantitative results that estimate effects of modifying risk factors based on both the risk factor prevalence and its association with readmission.

We also recognize the limitations of our study. First, NSQIP-defined readmissions likely underestimate the readmission rate based on the CMS definition since NSQIP collects data up to 30 days after the resection, which is slightly different than CMS, which defines readmissions up to 30 days after discharge from the index procedure.(10) Second, even though we adjusted for all risk factors identified through a recent systematic review and the model fits well,(5) we cannot rule out bias due to unmeasured or mismeasured confounders. For example, data about risk factors related to hospitals (e.g., volume), social determinants of health (e.g., income), psychosocial risk factors (e.g., depression), and types of additional treatments (e.g., type or duration of neoadjuvant therapy, type of anastomosis) were not available.(46) In addition, we recognize that some risk factors are interrelated as part of a tangled web of causation and that caution should be exercised in this retrospective study. Third, reason for readmission was not available. Furthermore, we recognize that NSQIP typically includes data from larger hospitals and is not a nationally representative sample, likely underrepresenting smaller and community hospitals.

Conclusion

Among limited opportunities identified, preventing prolonged ileus, increasing the use of MIS, and preventing anastomotic leaks might maximize efforts to reduce readmissions within 30 days following colon cancer surgery.