Sage Journals: Discover world-class research

Abstract

Background and objective:

The aim of this study was to describe short-term changes in morbidity and mortality associated with the implementation of screening for colorectal cancer in Denmark.

Methods:

Prospective cohort study with inclusion of all patients aged 50–75 years treated for colorectal cancer between 1 March 2014 and 31 December 2015 in Denmark. Adjusted hazard ratios were calculated for 30 and 90 days mortality using Cox Regression. We made two adjusted models—a “basic” adjusted for screening status, sex, age, smoking, alcohol consumption, and cancer type and an “advanced” that also included body mass index and American society of Anesthesiologists score in analyses. Relative risks were calculated for postoperative surgical and medical complications.

Results:

In total, 5348 patients were included. In the “basic model,” adjusted risk of 30 and 90 days total mortality was reduced in the screen-detected group (p < 0.01, HR = 0.43, CI = 0.24–0.76) and (p < 0.01, HR = 0.45, CI = 0.30–0.69). In the “advanced model,” only 90 days total mortality was significantly reduced in the screen-detected group (p = 0.01, HR 0.59, CI = 0.39–0.90). No significant changes were found with regard to surgical and medical complications, respectively, (p = 0.05 (CI = 0.76–1.00) and p = 0.47(CI = 0.74–1.15)).

Conclusions:

This nationwide study showed that screening for colorectal cancer was associated with a lower 90 days total mortality although no significant improvements were seen with regard to morbidity.

Keywords

screening CRC surgery short-term outcomes

Introduction

Colorectal cancer (CRC) is a common cancer with more than 1.4 million new cases worldwide each year and nearly 700,000 dying of the disease every year.¹ CRC develops through the adenoma-carcinoma sequence and fulfills the World Health Organization (WHO) screening criteria.^2,3 In CRC, screening is known to reduce the disease-specific mortality with 16% using the guaiac fecal occult blood test (gFOBT)⁴ possibly due to early detection of cases, removal of benign adenomas, and a reduced need for emergency surgery.^5,6 Little is known regarding and mortality following the implementation of CRC screening. However, screening could lead to improved outcomes on both.^7,8 In Denmark, screening for CRC was implemented nationwide on 1 March 2014 and all Danes aged 50–75 years are invited. The cut-off value for the fecal immune test (FIT) is 100 µg hemoglobin/L. Following a positive FIT test, patients are offered fast-track diagnostics and treatment in according to the National Cancer Plans to reduce diagnostic and therapeutic delays.^9,10

Surgery is the only curative option in CRC. This includes an inborn risk of morbidity and mortality with several well-identified risk factors, that is, American Society of Anesthesiologists physical state score (ASA) and location of CRC.^11,12 The optimal strategy is elective resection where the mortality rate is below 2%¹⁰, although endoscopic approaches can be used in selected cases.¹³ Endoscopy may be advantageous due to lower complication rates.¹³ In general, these studies indicate advantages associated with screening, that is, less emergency surgery and fewer complications but has methodological shortcomings, such as exclusion of emergency cases, no control group, no endoscopic therapy, and no reporting on 90 days mortality.^6,8,14
–16 As previously stated, screening for CRC leads to the finding of lower disease stages⁴ which could make endoscopic treatment a viable option¹³ and reduce the need for resections with high morbidity rates following surgery.^17
–19 It could therefore be argued that screening may also improve short-term outcomes following treatment for CRC. We decided to conduct this study to describe short-term changes in morbidity and mortality associated with the implementation of screening for CRC in Denmark.

Material and Methods

Study population

In Denmark, treatment for CRC is provided solely by the public healthcare system. All patients diagnosed with CRC between 1 March 2014 and 31 December 2015 were primarily included in this prospective cohort study. In Denmark, it is mandatory to report cases of CRC to the Danish Colorectal Cancer Database (DCCG). We included patients between 50 and 75 years of age at the time of detection as this is the age group invited to screening for CRC in Denmark. Further exclusion criteria were no CRC treatment that is due to refusal of treatment, disseminated disease leading to unresectable disease, no surgical procedure specified, unknown screening status, and other unspecified reasons. We also excluded patients who died before treatment could be initiated. Due to few cases, no analyses were performed with regard to complications after endoscopic treatment as final procedure.

Patients were extracted from the DCCG and Danish bowel screening Database (DTS). This study was approved by the Danish data protection agency (no. 2012-58-0004). No approval from ethic committees was necessary.

Patients were tracked using the Danish computerized, unique central personal registration number (CPR). Status (dead/alive) and exact date of death were ascertained through linkage of the patients’s CPR with the Danish Civil registration system.²⁰ All cancers were histologically confirmed and staged according to the Union for International Cancer Control (UICC) stages.

Data, extraction, variable, and management

The following data were extracted from DCCG: sex; status of screening (yes/no); UICC stages (I–IV); alcohol intake divided into 0, 1–14, above 15 units per week and missing values, respectively (1 unit = 12 g alcohol); smoking (non-smokers, active smokers, previous smokers, missing values); complications (Clavien-Dindo grades 1–3a, 3b–5, and unknown)²¹; ASA score (I–V); tumor site (colon/rectum); goal of treatment (curative, compromised resection, or palliative); treatment approach grouped according to surgical magnitude (laparotomy, minimally invasive (i.e. laparoscopy or robotic), and endoscopic (i.e. stent, EMR, or TEM)); emergency surgery; postoperative complications; and 30-day and 90-day mortality. Body mass index (BMI) was grouped as follows: below 19, 20–24.99, 25–29.99, 30–34.99, and 35 and above. Age was grouped into the following groups: 50–59, 60–69, and 70–75 years, respectively. Data from DTS were compared to DCCG to ensure validity of the screening variable in DCCG.

Outcome measures

Main outcomes were 30/90 days total mortality and postoperative complications dichotomized according to the Clavien-Dindo classification where a stratification was done at grade 3b as this is a complication that requires intervention under general anesthesia.²¹ Besides this, the severity of complications was classified as the most severe medical and/or surgical complication a patient suffered postoperatively. The exposure was whether the CRC was detected through screening.

Statistics

Comparison of the 3535 non-screened-detected patients to the 1813 screen-detected patients was done using Pearson’s chi-square and Fisher’s exact tests. Adjusted hazard ratios (HR) were calculated for 30 and 90 days mortality using Cox Regression. As screening works by finding the cancer at an earlier stage, we would take away the screening effect if adjusting for stage. Similarly detecting the cancer at an earlier stage by screening will result in better BMI and ASA. This would take away some of the screening effect if adjusting for BMI and ASA. We therefore, made one model that did not include these variables (the basic model). The variables in the “basic model” can though not adjust for the healthy user bias. To get this in to the model, we included BMI and ASA in an “advanced model” to adjust for some of the healthy user bias, knowing that this will underestimate the screening effect.^8,14 It should though be noticed that no adjustment was made for alcohol in medical complications due to few complications which made adjustment impossible. Using SAS version 9.4 (SAS Institute, NC, USA), logistic regression with the above-mentioned variables, we used to calculate adjusted relative risks (RR) for postoperative surgical complications and postoperative medical complications for each variable. To achieve convergence for the analysis for postoperative medical complications, we had to use the method proposed by Deddens et al.²² A p-value of 0.05 or lower was considered statistically significant.

Results

A total of 5348 patients were included in the study. There were 1813 (33.8%) in the screened-detected group and 3535 (66.2%) in the non-screened group. For analysis regarding complications, a total of 4896 patients undergoing surgery were included excluding the 452 treated endoscopically of which 257 were in the screened-detected group (Fig. 1 and Table 2).

Fig. 1.

Flowchart of study.

There were several demographic differences between the groups in unadjusted analysis. However, 30 and 90 days mortality was significantly lower in the screen-detected group (p = 0.01 and p < 0.001) (Table 1).

Table 1.

Demographic variables.

	Non-screened 3535 patients, n (%)	Screened 1813 patients, n (%)	p-value
Age (years)			<0.001
50–59	628 (17.8)	312 (17.2)
60–69	1653 (46.8)	699 (38.6)
70–75	1254 (35.5)	802 (44.2)
Sex			<0.001
Male	1990 (56.3)	1098 (60.6)
UICC stage			<0.001
Stage I	548 (15.5)	540 (29.8)
Stage II	955 (27.0)	416 (22.9)
Stage III	899 (25.4)	414 (22.8)
Stage IV	484 (13.7)	82 (4.5)
Unknown, missing T	13 (0.4)	12 (0.7)
Unknown, missing N	168 (4.8)	238 (13.1)
Unknown, missing M	54 (1.5)	41 (2.3)
Unknown, neoadjuvant therapy	414 (11.7)	70 (3.9)
Location of cancer			0.003
Colon	2253 (63.7)	1313 (72.4)
Rectum	1282 (36.3)	500 (27.6)
BMI			<0.001
<19	143 (4.3)	32 (1.8)
19–24.99	1426 (40.3)	590 (32.5)
25–29.99	1227 (34.7)	723 (39.9)
30–34.99	468 (13.2)	314 (17.3)
35>	190 (5.4)	101 (5.6)
Unknown BMI	81 (2.3)	53 (2.9)
ASA			<0.001
ASA I–II	2911 (82.3)	1576 (86.9)
ASA III–V	589 (16.7)	209 (11.9)
ASA unknown	35 (1.0)	28 (1.5)
Smoking			0.003
Non-smokers	1262 (35.7)	641 (35.4)
Active smokers	707 (20.0)	304 (16.8)
Previous smokers	1280 (36.2)	680 (37.5)
Missing value	286 (8.1)	188 (10.4)
Alcohol consumption (units)			<0.001
0	701 (19.8)	283 (15.6)
1–14	2105 (59.6)	1102 (60.8)
>15	481 (13.6)	274 (15.1)
Missing values	248 (7.0)	154 (8.5)
Mortality (days)
30	61 (1.7)	15 (0.8)	0.01
90	109 (3.1)	28 (1.5)	<0.001

UICC: Union for International Cancer Control; BMI: body mass index; ASA: American Society of Anesthesiologists score.

Numbers are absolute and in parentheses are percentages. 1 unit of alcohol = 12 g.

Bold values are significant at p < 0.05.

The therapeutic procedures also differed with more endoscopic and minimally invasive surgery, fewer emergency resections, less supplemental resections of adjacent organs, and more treatment with curative intent in the screen-detected group compared to the non-screened group (Table 2).

Table 2.

Overview of procedures.

	Non-screened 3535 patients, n (%)	Screened 1813 patients, n (%)	p-value
Priority			<0.001
Emergency	320 (9.1)	13 (0.7)
Elective	3213 (90.9)	1798 (99.3)
Surgical approach			<0.001
Laparoscopy/robotic	2234 (63.2)	1267 (69.9)
Laparotomy	1106 (31.3)	289 (15.9)
Endoscopy/TEM/Stent	195 (5.5)	257 (14.0)
Supplemental resection^a	456 (12.9)	85 (4.7)	<0.001
Goal of treatment			<0.001
Curative	3298 (93.3)	1776 (98)
Compromised resection^b	167 (4.7)	13 (0.7)
Palliative	50 (1.4)	15 (0.8)
Unknown	20 (0.6)	9 (0.5)

Supplemental resection: resection of an adjacent organ to ensure radical resection.

Compromised resection: a curative resection with lymph node dissection at D1 level.

Bold values are significant at p < 0.05.

The incidence of surgical complications was significantly lower in the screen-detected group compared to the non-screened group (p = 0.02), while no significant difference in the severity of surgical complications was found (p = 0.06). With regard to medical complications, no significant differences were found in neither incidence (p = 0.14) nor severity (p = 0.58) (Table 3). For severity of individual complications and detailed overview of individual procedures, please refer to the Supplemental Appendices 1 and 2, respectively.

Table 3.

Overview of complications.

	Non-screened 3340 patients, n (%)	Screened 1556 patients, n (%)	p-value
Surgical complications
Surgical complication	617 (18.5)	238 (15.3)	0.02
Reoperation without findings	14 (0.4)	7 (0.4)	0.94
Severity of surgical complications			0.06
Clavien-Dindo 1–3a	158 (4.7)	58 (3.7)
Clavien-Dindo 3b–5	472 (14.1)	187 (12.0)
Unknown Clavien-Dindo grade	1 (0.0)	1 (0.1)
Medical complications			0.14
Medical complication	297 (8.8)	119 (7.6)
Severity of medical complications			0.58
Clavien-Dindo 1–3a	168 (5.0)	68 (4.4)
Clavien-Dindo 3b–5	124 (3.7)	50 (3.2)
Unknown Clavien-Dindo grade	3 (0.1)	1 (0.1)

Bold values are significant at p < 0.05.

In the “basic model,” adjusted risk of 30- and 90-day total mortality was reduced in the screen-detected group (p < 0.01, HR = 0.43, CI = 0.24–0.76) and (p < 0.01, HR = 0.45, CI = 0.30–0.69) (Table 4). In the “advanced model,” including BMI and ASA score, 30 days total mortality was not significantly reduced in the screened group compared to the non-screened group (p = 0.07) but 90 days total mortality was significantly reduced in the screened group (p = 0.01, HR = 0.59, CI = 0.39–0.90) (Table 4).

Table 4.

HR for 30 and 90 days total mortality.

		30 days mortality				90 days mortality
		p-value	HR ratio	HR lower	HR upper	p-value	HR ratio	HR lower	HR upper
Basic model
Screening	Yes	<0.01	0.43	0.24	0.76	<0.01	0.45	0.30	0.69
Age (years)	60–69	0.18	2.06	0.71	5.97	0.21	1.55	0.78	3.10
Age (years)	70–75	<0.01	4.46	1.60	12.4	<0.01	3.08	1.60	5.94
Sex	Female	0.36	0.80	0.49	1.30	0.49	0.88	0.62	1.26
Smoking	Active/former	0.03	1.80	1.06	3.07	<0.01	1.95	1.32	2.88
Units alcohol	1–14	0.09	0.59	0.32	1.09	0.10	0.68	0.43	1.07
Units alcohol	>15	0.76	1.12	0.78	5.12	0.79	1.08	0.62	1.88
Cancer type	Rectum	0.87	0.96	0.59	1.55	0.35	0.84	0.58	1.21
Advanced model
Screening	Yes	0.07	0.58	0.33	1.04	0.01	0.59	0.39	0.90
Age (years)	60–69	0.30	1.75	0.60	5.11	0.42	1.32	0.66	2.65
Age (years)	70–75	0.06	2.71	0.97	7.60	0.04	1.98	1.02	3.86
Sex	Female	0.62	0.88	0.54	1.44	0.81	0.96	0.66	1.37
Smoking	Active/former	0.38	1.28	0.73	2.23	0.41	1.40	0.94	2.10
Units alcohol	1–14	0.42	0.77	0.42	1.43	0.55	0.87	0.55	1.37
Units alcohol	>15	0.45	1.32	0.64	2.74	0.41	1.26	0.72	2.21
Cancer type	Rectum	0.97	1.01	0.62	1.63	0.48	0.87	0.60	1.27
ASA score	ASA score III–V	0.01	9.11	5.52	15.04	0.01	7.53	5.24	10.83
BMI	<19	0.57	1.30	0.53	3.18	0.29	1.41	0.74	2.67
BMI	⩾35	0.94	0.97	0.42	2.42	0.43	0.76	0.38	1.50
BMI	25–29.99	0.15	0.65	0.36	1.16	0.04	0.65	0.42	0.99
BMI	30–34.99	0.17	0.58	0.26	1.27	0.05	0.55	0.31	1.00

HR: hazard ratios; ASA: American society of Anesthesiologists score; BMI: body mass index.

Comparisons made for hazard ratios (HR): Screening: no versus yes; Age: 50–59 years versus other groups; Sex: male versus female; Smoking: No smoking versus active and previous smokers; Units of alcohol (1 unit = 12 g) none versus other groups; Cancer type: colon versus rectum; ASA score I–II versus III–V and BMI 20–24.99 versus other groups.

Bold values are significant at p < 0.05.

In adjusted analysis using the “advanced model,” risk of surgical and medical complications was not reduced in the screened group compared to the non-screened group, respectively, (p = 0.05, RR = 0.87 (CI = 0.76–1.00) and p = 0.47, RR = 0.93 (CI = 0.74–1.15)) (Table 5).

Table 5.

Association between demographic variables and postoperative complications.

Variable	Surgical complications
Variable	RR	p-value	Confidence interval
Screening	0.87	0.05	0.76–1.00
Age (years)
- 60–69	0.90	0.21	0.75–1.06
- 70–75	0.88	0.13	0.73–1.04
Female	0.65	<0.01	0.56–0.74
Smoking
- Active smoking	1.19	0.02	1.02–1.37
- Missing value	1.19	0.20	0.91–1.57
Alcohol consumption (units)
- 0–14	0.89	0.13	0.76–1.04
- >15	0.87	0.18	0.70–1.07
- Missing value	0.97	0.86	0.71–1.34
Cancer type
- Rectum	1.60	<0.01	1.42–1.80
ASA score
- III–V	1.34	0.02	1.15–1.55
- Missing	0.84	0.69	0.34–2.05
BMI
- <19	0.79	0.25	0.53–1.18
- 25–29.99	1.04	0.58	0.90–1.20
- 30–34.99	1.13	0.19	0.94–1.34
- ⩾35	1.40	<0.01	1.11–1.74
- Missing value	0.81	0.48	0.46–1.44
Variable	Medical complications
Variable	RR	p-value	Confidence interval
Screening	0.93	0.47	0.74–1.15
Age (years)
- 60–69	1.44	0.04	1.01–2.04
- 70–75	1.89	<0.01	1.34–2.67
Female	0.77	<0.05	0.63–0.94
Smoking
- Active smoking	1.22	0.07	0.98–1.53
- Missing value	1.20	0.30	0.85–1.67
Cancer type
- Rectum	1.03	0.76	0.85–0.125
ASA score
- III—V	2.75	<0.01	2.26–3.35
- Missing value	1.51	0.46	0.51–4.47
BMI
- >19	1.37	0.13	0.91–2.07
- 25–29.99	0.97	0.78	0.78–1.21
- 30–34.99	0.90	0.47	0.67–1.20
- >35	0.99	0.97	0.68–1.46
- Missing	1.29	0.44	0.67–2.50

ASA: American society of Anesthesiologists score; BMI: body mass index.

Comparisons made for RR: Screening: no versus yes; Age: 50–59 years versus other groups; Sex: male versus female; Smoking: No smoking versus other groups; Units of alcohol (1 unit = 12 g) none versus other groups; Cancer type: colon versus rectum; ASA score I–II versus other groups, and BMI 20–24.99 versus other groups.

Bold values are significant at p < 0.05.

Discussion

The main finding of this study is a lower 90 days total mortality in the screen-detected group. No improvements with regard to complications were found in adjusted analysis. Besides this, several changes in the therapeutic approach were found, that is, more curative treatment, more elective treatment, and more endoscopic treatment.

In this nationwide population-based study, there were more patients in the screen-detected group above 70 years of age compared to the non-screened group due to the way invitations were issued in the CRC screening program. This resulted in many patients above 70 years of age in order not to miss cases of CRCs among patients approaching the end of the screening program (9). In general, increasing age is a strong risk factor for complications and mortality following surgery for CRC.¹⁹

We have tried to adjust for the “healthy user effect” which is a possible bias in screening studies by including BMI and ASA score in the “advanced model” as screening works through the detection of earlier cancers and thereby patients are in a better general condition before undergoing treatment. Doing this, we only saw an effect of screening on 90 days mortality perhaps due to the elimination of some of the screening effect.

However, 90 days mortality, which is a better marker for postoperative mortality, has not been used previously in screening literature.^6,8,14,15,23 We also found that more patients could be treated with curative intent in the screened group, more could be treated with endoscopic resections, and that emergency surgery was almost eliminated in the screening group which is also new compared to other studies.^8,24
–26 This is important as emergency treatment is associated with high mortality rates and morbidity rates^17,18,27, whereas endoscopic treatment is associated with lower mortality and morbidity compared with surgery.^15,28

The methodological limitations of this study include a setup that leaves room for selection bias and healthy user bias.²⁹ These biases were minimized by including all patients in the screening population to limit selection bias. Besides this, we also used two different statistical models to minimize healthy user bias. To our knowledge, healthy user bias has not been addressed by any previous studies on short-term outcomes after screening for CRC. It can also be argued that adjustments for TNM stage should be made. However, other studies on short-term outcomes following the implementation of screening do not adjust for TNM stage either.^8,14,26 Strengths of this study include its nationwide population-based design with very few missing variables and inclusion of all cases and 90 days mortality, which has been lacking in previous studies.^8,15,24,30 It can be argued that complications for endoscopically treated patients should be included but this would not be meaningful as we have only 452 of these procedures and very few complications arise following endoscopic procedures.^24,28 This is underlined by another study with a larger sample size of 2984 screening colonoscopies with 1 patient dying and 44 patients with complications¹⁵ Therefore, the results in our study represent everyday practice in Denmark where endoscopically treated patients may be treated incidentally, but surgically treated patients are generally treated by specialists.¹⁰

In conclusion, we found that the treatment of CRC following screening is associated with more curative treatment, elective treatment, and improvements with regard to number of endoscopic resections and lower 90 days total mortality, despite inclusion of many patients above 70 years in the screened group and inclusion of emergency cases.

Supplemental Material

sj-pdf-1-sjs-10.1177_14574969211019824 – Supplemental material for Impact of screening on short-term mortality and morbidity following treatment for colorectal cancer

Supplemental material, sj-pdf-1-sjs-10.1177_14574969211019824 for Impact of screening on short-term mortality and morbidity following treatment for colorectal cancer by Michael Wilhelmsen, Sisse Helle Njor, Ole Roikjær, Morten Rasmussen and Ismail Gögenur in Scandinavian Journal of Surgery

Supplemental Material

sj-pdf-2-sjs-10.1177_14574969211019824 – Supplemental material for Impact of screening on short-term mortality and morbidity following treatment for colorectal cancer

Supplemental material, sj-pdf-2-sjs-10.1177_14574969211019824 for Impact of screening on short-term mortality and morbidity following treatment for colorectal cancer by Michael Wilhelmsen, Sisse Helle Njor, Ole Roikjær, Morten Rasmussen and Ismail Gögenur in Scandinavian Journal of Surgery

Footnotes

Acknowledgements

The authors thank the Danish Colorectal Cancer Database (DCCG) and the Danish Colorectal Cancer Screening Database.

ORCID iD

Michael Wilhelmsen

Supplemental material

Supplemental material for this article is available online.

References

Ferlay

Soerjomataram

Dikshit

, et al: Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015;136:E359–E386.

Wilson

JMG

Jungner

: WHO principles and practice of screening 1968. Public health papers no. 34. Available at: https://apps.who.int/iris/bitstream/handle/10665/37650/WHO_PHP_34.pdf (accessed 20 August 2020).

Vogelstein

Fearon

Hamilton

, et al: Genetic alterations during colorectal-tumor development. N Engl J Med 1988;319:525–532.

Hewitson

Glasziou

Irwig

, et al: Screening for colorectal cancer using the faecal occult blood test, Hemoccult. Cochrane Database Syst Rev 2007;2007:CD001216.

Kubisch

Crispin

Mansmann

, et al: Screening for colorectal cancer is associated with lower disease stage: A population-based study. Clin Gastroenterol Hepatol 2016;14(11):1612.e3–1618.e3.

Askari

Nachiappan

Currie

, et al: Who requires emergency surgery for colorectal cancer and can national screening programmes reduce this need? Int J Surg 2017;42:60–68.

Amri

Bordeianou

Sylla

, et al: Impact of screening colonoscopy on outcomes in colon cancer surgery. JAMA Surg 2013;148(8):747–754.

De Neree Tot Babberich

MPM

Vermeer

NCA

Wouters

MWJM

, et al: Postoperative outcomes of screen-detected vs non-screen-detected colorectal cancer in the Netherlands. JAMA Surg 2018;153:e183567.

Njor

Friis-Hansen

Andersen

, et al: Three years of colorectal cancer screening in Denmark. Cancer Epidemiol 2018;57:39–44.

10.

Iversen

Ingeholm

Gögenur

, et al: Major reduction in 30-day mortality after elective colorectal cancer surgery: A nationwide population-based study in Denmark 2001-2011. Ann Surg Oncol 2014;21(7):2267–2273.

11.

Fazio

Tekkis

Remzi

, et al: Assessment of operative risk in colorectal cancer surgery: The Cleveland clinic foundation colorectal cancer model. Dis Colon Rectum 2004;47(12):2015–2024.

12.

Tekkis

Poloniecki

Thompson

, et al: Operative mortality in colorectal cancer: Prospective national study. BMJ 2003;327:1196–1199.

13.

Aarons

Shanmugan

Bleier

JIS

: Management of malignant colon polyps: Current status and controversies. World J Gastroenterol 2014;20:16178–16183.

14.

Sebastian

Courtier

Macià

, et al: The impact of screening on short-term outcome after surgery for colorectal cancer. Rev Esp Enferm Dig 2017;109(7):485–490.

15.

Saraste

Martling

Nilsson

, et al: Complications after colonoscopy and surgery in a population-based colorectal cancer screening programme. J Med Screen 2016;23(3):135–140.

16.

Adam

Turner

Sun

, et al: The appropriateness of 30-day mortality as a quality metric in colorectal cancer surgery. Am J Surg 2018;215(1):66–70.

17.

Mohan

Evans

Larkin

, et al: Multivisceral resection in colorectal cancer: A systematic review. Ann Surg Oncol 2013;20(9):2929–2936.

18.

Klaver

CEL

Gietelink

Bemelman

, et al: Locally advanced colon cancer: Evaluation of current clinical practice and treatment outcomes at the population level. J Natl Compr Canc Netw 2017;15(2):181–190.

19.

Walker

Finan

van der Meulen

: Model for risk adjustment of postoperative mortality in patients with colorectal cancer. Br J Surg 2015;102(3):269–280.

20.

Pedersen

: The Danish Civil registration system. Scand J Public Health 2011;39:22–25.

21.

Dindo

Demartines

Clavien

P-A

: Classification of surgical complications: A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240(2):205–213.

22.

Deddens

Pedersen

Xuidong

: Estimation of prevalence ratios when PROC GENMOD does not converge. SUGI 28, 2003. Available at: http://www2.sas.com/proceedings/sugi28/270-28.pdf (accessed 20 August 2020).

23.

Vermeer

NCA

Snijders

Holman

, et al: Colorectal cancer screening: Systematic review of screen-related morbidity and mortality. Cancer Treat Rev 2017;54:87–98.

24.

Kewenter

, Brevinge H: Endoscopic and surgical complications of work-up in screening for colorectal cancer. Dis Colon Rectum 1996;39(6):676–680.

25.

Vermeer

NCA

De Neree Tot Babberich

MPM

Fockens

, et al: Multicentre study of surgical referral and outcomes of patients with benign colorectal lesions. BJS Open 2019;3(5):687–695.

26.

Vermeer

NCA

Bahadoer

Bastiaannet

, et al: Introduction of a colorectal cancer screening programme: Results from a single-centre study. Colorectal Dis 2018;20(9):O239–O247.

27.

Iversen

Bülow

Christensen

, et al: Postoperative medical complications are the main cause of early death after emergency surgery for colonic cancer. Br J Surg 2008;95(8):1012–1019.

28.

Reumkens

Rondagh

EJA

Bakker

, et al: Post-colonoscopy complications: A systematic review, time trends, and meta-analysis of population-based studies. Am J Gastroenterol 2016;111(8):1092–1101.

29.

Sung

: A review of the healthy worker effect in occupational epidemiology. Occup Med 1999;49:225–229.

30.

Kronborg

Fenger

Worm

, et al: Causes of death during the first 5 years of a randomized trial of mass screening for colorectal cancer with fecal occult blood test. Scand J Gastroenterol 1992;27(1):47–52.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB

0.03 MB