Warning: fopen(/home/virtual/colon/journal/upload/ip_log/ip_log_2025-02.txt): failed to open stream: Permission denied in /home/virtual/lib/view_data.php on line 95 Warning: fwrite() expects parameter 1 to be resource, boolean given in /home/virtual/lib/view_data.php on line 96 Comparison of colorectal cancer surgery patients in intensive care between rural and metropolitan hospitals in Australia: a national cohort study
Skip Navigation
Skip to contents

Ann Coloproctol : Annals of Coloproctology

OPEN ACCESS
SEARCH
Search

Articles

Page Path
HOME > Ann Coloproctol > Ahead of print > Article
Original Article
Comparison of colorectal cancer surgery patients in intensive care between rural and metropolitan hospitals in Australia: a national cohort study
Jessica A. Paynter1orcid, Zakary Doherty1,2orcid, Chun Hin Angus Lee1orcid, Kirby R. Qin1orcid, Janelle Brennan1orcid, David Pilcher2,3,4orcid

DOI: https://doi.org/10.3393/ac.2024.00269.0038
Published online: January 24, 2025

1Department of Surgery, Monash University School of Rural Health, Bendigo, VIC, Australia

2Department of Intensive Care, Alfred Health, Melbourne, VIC, Australia

3Australian and New Zealand Intensive Care Society (ANZICS), Centre for Outcome and Resource Evaluation, Melbourne, VIC, Australia

4Australian and New Zealand Intensive Care Research Centre (ANZICS-RC), Monash University School of Public Health and Preventive Medicine, Melbourne, VIC, Australia

Correspondence to: Jessica A. Paynter, MBBS (Hons), BMedSci, MSurg Monash University School of Rural Health, Bldg 20/26 Mercy St, Bendigo, VIC 3550, Australia
• Received: April 21, 2024   • Revised: June 24, 2024   • Accepted: July 28, 2024

© 2025 The Korean Society of Coloproctology

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 195 Views
  • 7 Download
  • Purpose
    A small proportion of colorectal cancer (CRC) surgical patients will require an admission to an intensive care unit (ICU) within the early postoperative period. This study aimed to compare the characteristics and outcomes of patients admitted to an ICU following CRC surgery per hospital type (metropolitan vs. rural) over a decade in Australia.
  • Methods
    A retrospective cohort analysis was undertaken of all adult patients admitted to a participating Australian ICUs following CRC surgery between January 2011 and December 2021. The primary outcome was in-hospital mortality.
  • Results
    Over the 10-year period, 19,611 patients were treated in 122 metropolitan ICUs and 4,108 patients were treated in 42 rural ICUs. Rural ICUs had a lower proportion of annual admissions following CRC surgery (20 vs. 36, P<0.001). Patients admitted to a rural ICU were more likely to have undergone emergency CRC surgery compared to those admitted to a metropolitan cohort (28.5% vs. 13.8%, P<0.001). There was no difference in in-hospital mortality between metropolitan and rural hospitals (odds ratio [OR], 1.03; 95% confidence interval [CI], 0.73–1.35; P=0.500). There was a general trend for lower mortality in later years of the study with the odds of death in the final year of the study (2021) almost half that of the first study year (OR, 0.52; 95% CI, 0.34–0.80; P=0.003).
  • Conclusion
    There was no difference between in-hospital mortality outcomes for CRC surgical patients requiring ICU admission between metropolitan and rural hospitals. These findings may contribute to discussions regarding rural scope of colorectal practice within Australia and globally.
Surgery for colorectal cancer (CRC) is the cornerstone of curative treatment, and approximately 9% of these patients require admission to an intensive care unit (ICU) in the early postoperative period [1]. In recent years, there has been a trend towards centralization of CRC surgery services to metropolitan centers, with international literature indicating that high-volume centers have improved patient outcomes due to improved ancillary services such as ICUs, despite previous Australian studies showing no significant differences in outcomes between rural and metropolitan centers [24]. Centralization is proposed primarily due to the resources required for the following: (1) complex pelvic exenterations or hyperthermic intraperitoneal chemotherapy; and (2) the lack of postoperative ICU support available in smaller hospitals. The move to centralization may disadvantage some rural Australian patients who require equitable delivery of high-quality CRC surgery close to home [5, 6]. Care delivered away from home may be associated with worse outcomes for some patient groups, especially those who require readmission to a center where the surgery was not performed, as well as imposing psychological and financial distress [7].
Given Australia's geographic challenges, this study aimed to compare the demographics and patient outcomes for patients admitted to ICUs in Australia following CRC surgery between metropolitan and rural hospitals. This analysis may determine whether there is any difference in short-term outcomes, which could provide insight into how we can better provide equitable CRC surgical care to all Australians, irrespective of rurality.
Ethics statement
This project received ethics approval from the Monash University Human Research Ethics Committee (No. 36437) and approval from the governance committee of the Australian and New Zealand Intensive Care Society (ANZICS) Centre for Outcome and Resource Evaluation. Informed consent was not required due to the retrospective nature of the study. This retrospective project was conducted in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement (Supplementary Material 1).
Study design
This was a retrospective cohort study that used the ANZICS Adult Patient Database (APD).
Data source
The ANZICS APD is a binational registry that contains de-identified information on admissions to 98% of adult ICUs in Australia. Data collection is conducted at each participating ICU site by trained staff.
Study setting
All Australian sites that reported to the ANZICS APD during the study period (2011–2021) were included. The hospital type classification (rural, metropolitan) is shown in Supplementary Table 1, as per the geographical location of the hospital. Throughout Australia, CRC surgery is performed by either a general or colorectal surgeon at both rural and metropolitan locations. Patients are then primarily cared for by an intensive care specialist during their ICU admission, with the operative surgeon consulting, and subsequently resuming primary care on discharge from the ICU. Planned admission to the ICU for surgery occurs if the need for ICU admission is anticipated preoperatively or prior to induction of anesthesia, or if there is a planned transfer to an ICU following emergency surgery at another hospital. Planned ICU admissions are not standardized nationally, and result from a combined decision between the surgeon and intensivist.
Study population
This study included all adult patients (≥18 years) admitted to participating Australian ICUs between 2011 and 2021 (inclusive) with a diagnosis code of “colorectal cancer surgery.” This encompassed both planned and unplanned admissions. In the instance of the same patient having multiple ICU admissions within the same hospitalization only the first was included. Patients admitted for palliation or organ donation were excluded.
Variables
Patient demographic variables included in this analysis consisted of age at the time of ICU admission, sex, and the presence of chronic comorbidities. Admission variables included the source of admission to both the hospital and ICU, whether the admission was planned or unplanned, the use of invasive ventilation, ICU and hospital length of stay, the severity of illness quantified by the Acute Physiology and Chronic Health Evaluation (APACHE) III-J score and Australian and New Zealand Risk of Death (ANZROD) mortality prediction model, and the discharge destination from the ICU and hospital overall. The APACHE III-J score is an international standardized measure used to quantify illness severity for each patient admitted to an ICU [8]. ANZROD is a highly discriminatory mortality prediction model, specifically designed and calibrated for the Australian setting [9]. Both scores are calculated using each individual patient’s physical and biochemical markers from the first 24 hours of their ICU admission, and also account for age, chronic comorbidities, source of admission, elective surgical status, and whether the ICU admission was planned or not. ANZROD also adjusts for the presence of treatment limitations at ICU admission. The Index of Relative Socio-economic Advantage and Disadvantage (IRSAD) was also used to determine an individual’s level of economic and social advantage, with a lower score indicating greater disadvantage [10]. Distance was calculated by taking the center of the patient’s postcode and calculating the shortest distance to the hospital. Rurality was classified according to the Australian Statistical Geography Standard-Remoteness Area, 3rd edition [11].
Outcomes
The primary outcome was in-hospital mortality. The secondary outcomes were ICU length of stay and hospital length of stay.
Statistical analysis
Categorical data were presented as frequencies with percentages. All continuous variables were nonparametric, and medians with interquartile ranges were reported. Comparisons between outcomes according to hospital type were made using the Kruskal-Wallis rank sum test and Pearson chi-square test as indicated.
Multivariable logistic regression was used to investigate the relationship between survival to discharge and hospital type. All included variables were deemed clinically relevant and therefore included irrespective of their association in the univariable analysis. All continuous variables met the linearity assumption and were not adjusted. A sensitivity analysis was conducted with the exclusion of patients who had a discharge destination of another ICU, as it was not known whether they survived this subsequent ICU admission. Odds ratios (ORs) are reported with 95% confidence intervals (CIs). A 2-sided P-value of <0.05 was used at the level of significance. All analyses were conducted using R ver. 1.3.959 (R Foundation for Statistical Computing).
There were 19,611 patients (82.7%) treated in 122 metropolitan ICUs (61 public, 61 private) and 4,108 patients (17.3%) treated in 42 rural ICUs (37 public, 5 private) from January 2011 to December 2021. In 2011, 88 of the 136 units contributing to the APD had admissions that met our inclusion criteria. In 2021, this proportion increased to 150 of 178 [12] (Supplementary Table 2).
Cohort characteristics
The characteristics of the cohort are described in Table 1. The median age of the rural cohort was higher than that of the metropolitan cohort (74.6 years vs. 73.1 years, P<0.001). Patients in the rural cohort were more likely to have undergone emergency CRC surgery (27.1% vs. 13.8%, P<0.001).
A higher proportion of the rural cohort identified as Indigenous compared to the metropolitan cohort (3.7% vs. 1.6%, P<0.001). The rural cohort contained a greater proportion of patients with respiratory and cardiovascular disease, yet the metropolitan cohort included a higher proportion of those with immunosuppressive diseases and those receiving immunosuppressive therapy. The median APACHE III-J and ANZROD scores were similar in both cohorts. Metropolitan ICUs cared for a greater proportion of ventilated patients (13.5% vs. 9.0%, P<0.001). Patients admitted to rural ICUs had a lower median IRSAD score (949 vs. 1,010, P<0.001), indicating a higher level of social disadvantage. A greater proportion of unplanned ICU admissions was observed in the rural cohort than in the metropolitan cohort following CRC surgery (30.8% vs. 16.7%, P<0.001). ICU admission via hospital transfer was more common in the metropolitan cohort than in the rural cohort (1.5% vs. 0.8%, P<0.001). Contrastingly, ICU admission via the ward was more widespread rurally (2.1% vs. 1.2%, P<0.001). The rural cohort lived further away from their admitted ICU (33.3 km vs. 11.6 km, P<0.001). Patients in the rural cohort were also more likely to have treatment limitations in place (6.8% vs. 2.1%, P<0.001).
The median number of annual admissions was higher in metropolitan hospitals (36 vs. 20, P<0.001). Metropolitan sites saw a mean increase of 0.69% in admissions annually (P=0.002), compared to rural sites where a mean decrease of 3.3% (P<0.001) was observed (Supplementary Fig. 1).
Survival to hospital discharge
The total mortality in the cohort was 575 (2.4%), with 442 deaths (2.3%) occurring in the metropolitan cohort and 133 deaths (3.2%) in the rural cohort. Compared to the reference group (metropolitan), there was no significant difference in in-hospital mortality for rural hospital patients following CRC surgery when controlling for all listed variables (Table 2). Similarly, no significant difference was noted in in-hospital mortality between patients who resided in a major city and those who had a rural residence. As the distance between a patient’s residence and the ICU increased, their in-hospital mortality decreased. There was a general trend for lower in-hospital mortality in more recent years of the study, with the odds of death in the final year of the study (2021) being almost half that in the first study year. Lower in-hospital mortality was also identified in privately funded ICUs following CRC surgery. The results of the univariable analysis are also included in Supplementary Table 3. The sensitivity analysis that excluded patients with a discharge destination of “transfer to other ICU” had equivalent findings to those summarized above (Supplementary Table 4).
Secondary outcomes
The rural cohort had longer ICU stays (2.1 days vs. 1.0 days, P<0.001) but shorter overall hospital stays (9.3 days vs. 10.0 days, P<0.001) than the metropolitan cohort (Table 3). The ICU readmission rate was similar between rural and metropolitan ICUs (6.4% vs. 6.1%, P=0.500).
This study of 23,719 patients admitted to 164 Australian ICUs over a 10-year period showed no significant difference in short-term survival outcomes between those admitted to rural ICUs and those admitted to metropolitan ICUs. The adjusted in-hospital mortality rate for these patients decreased over the 10-year study period. Whilst 28% of Australians live in rural Australia, only 17.3% of this cohort were admitted to a rural ICU following their CRC surgery [13]. This may suggest that not all rural Australians receive CRC surgery and postoperative care as close to home as possible.
Admission as a result of hospital transfer was more common in metropolitan ICUs than in rural ICUs. This may be a reflection of smaller rural hospitals (without ICU capacity) transferring patients either pre- or post-CRC surgery to larger, metropolitan hospitals with ICU capacity due to the surgery being emergent or a complication occurring. A higher proportion of ICU admissions from the ward for CRC surgical patients was seen in rural hospitals. A reason for this might be that rural hospital surgical wards have a lower acuity of care, and therefore a lower threshold for ICU transfer with patient deterioration. Rural hospitals had a higher proportion of both emergency CRC operations and unplanned ICU admissions. This could reflect a greater proportion of emergency CRC operations occurring in the setting of acute CRC obstruction, perforation, or bleeding. Both short-term outcomes and long-term survival have been shown to be poorer following emergency surgery for CRC than after elective surgery [14]. It is known that rural Australians have less access to CRC screening services and general practitioner appointments, which may have contributed to this [15, 16]. The metropolitan ICUs may have had proportionally higher planned ICU admissions due to planned transfers occurring from other hospitals without ICUs. Patients admitted to rural hospital ICUs following CRC surgery suffered from higher levels of socioeconomic disadvantage, which is known to be associated with poorer CRC survival [17]. There was no difference in mortality prediction scores; however, patients in metropolitan ICUs had a higher rate of ventilation, and immunosuppressive treatment and diseases, both of which have been shown to be associated with poorer long-term survival following CRC surgery [18, 19].
No significant difference was seen in in-hospital mortality amongst CRC surgical patients admitted to ICUs between metropolitan and rural hospitals. This may align with other Australian studies that have assessed hospital geographical remoteness in relation to overall survival following CRC resection, and found no differences in survival between rural and metropolitan hospitals [2, 3]. Despite higher proportions of emergency CRC surgery and unplanned ICU admissions in rural hospitals, there was no significant difference in short-term mortality outcomes in Australian ICU-admitted patients who underwent CRC surgery.
An increased distance from a patient’s home to the admitted hospital was associated with significantly higher survival for patients receiving ICU care following CRC surgery. This may reflect survivorship bias, whereby only those who were assessed to likely have a good outcome were transferred from their home location to an ICU to receive active support for their CRC surgery. This could also reflect centralization already occurring in Australia. Consequently, those who may have been assessed and deemed not to have a potentially favorable outcome may not have been transferred to a hospital with an ICU.
During the 10-year study period, a decrease in in-hospital mortality was observed. This may have been due to an increased proportion of metropolitan ICU admissions occurring during this time, reflecting centralization occurring in Australia. Other factors that may have contributed to this include improvements in ICU care and operative and anesthetic techniques, such as increased uptake of laparoscopic-assisted CRC surgery, the use of acute pain services, the implementation of enhanced recovery after surgery protocols, and stoma and wound care improvements [6, 20]. Additionally, selection bias may have also played a role in Australia, given the increased participation within the National Bowel Screening Program from 2014 to 2020 resulting in earlier diagnosis and an associated decrease in complications of malignancy [15]. The finding of improved survival in private ICUs was in keeping with a prior Australian study that highlighted more favorable private CRC surgical outcomes due to sociodemographic factors, such as a younger population with a higher index of socioeconomic advantage and more economic resources [21]. The sensitivity analysis excluding patients with a discharge destination of “transfer to other ICU” also found no difference between rural and metropolitan outcomes. The primary reason for this second analysis was to control for the possibility that rural sites were more likely to transfer CRC patients to metropolitan ICUs for escalation of care. This excluded the possibility that transfers out of rural ICUs may have been more likely to survive the index admission (by being transferred), but died at the transfer hospital, hence artificially increasing the rural ICU survival.
The findings of this study have implications for the surgical and ICU care of CRC patients, particularly in rural Australia, and for critically unwell patients. Our results provide evidence, that when controlling for severity of illness and other variables, there is no significant difference in short-term mortality between rural and metropolitan sites. This may support ongoing rural surgical care for unwell CRC surgical patients. This study also highlights that in-hospital mortality for all CRC surgical patients admitted to an Australian ICU improved across the decade, reflecting substantial improvements in care.
This study has several strengths. The study was national, describing CRC surgical patients from most hospitals in Australia, and it therefore likely captured a large proportion of the patients nationwide that were admitted to ICUs after their surgery. The study also analyzed data over a 10-year period, enabling an investigation of long-term trends. Finally, the cohort was not limited to specific surgical procedures; instead, it provided a broad overview of all CRC patients requiring admission to Australian ICUs.
Nonetheless, some limitations need to be considered. The study was retrospective and observational, and therefore, it was impacted by the presence of confounders. Firstly, the decision to admit a patient to an Australian ICU is not a standardized decision, and indications vary by institution. Different Australian hospitals and peri-operative departments have different criteria for the admission of postoperative patients to the ICU. This factor cannot be accounted for within this study. Secondly, this study only included CRC patients admitted to an ICU during surgical admissions. It is recognized that approximately 90% of CRC surgical patients do not require an ICU admission, and our findings may not apply to them [1]. The database also did not delineate specifics of type of CRC surgery. This study was unable to assess outcomes beyond hospital discharge or long-term outcomes, such as quality of life or long-term survival. It is known that ICU admission and CRC surgery are associated with reduced quality of life in subsequent years; therefore, reviewing survival outcomes in isolation has limitations [22, 23]. Finally, the data collected did not differentiate between the outcomes of nonspecialized general surgeons and specialist colorectal surgeons, although this delineation exists in Australia. Specialist colorectal surgeons are fellowship-trained and generally take on more complex, higher risk cases, which may ultimately result in more postoperative ICU admissions.
In conclusion, over the 10-year study period, we found an overall decrease in-hospital mortality after CRC surgery across Australian ICUs. Additionally, we identified similar in-hospital mortality after CRC surgery across rural and metropolitan ICUs. This suggests that the short-term outcomes after CRC surgery may not be impacted by critical care capabilities. These findings may contribute to discussions regarding the rural scope of colorectal practice within Australia and globally.

Conflict of interest

No conflict of interest relevant to this article was reported.

Funding

None.

Acknowledgments

The authors wish to acknowledge all staff involved in data collection at the participating intensive care units. A full list is provided in Supplementary Material 2.

Author contributions

Conceptualization: JAP, ZD, JB, CHAL; Data curation: ZD, DP; Formal analysis: ZD; Investigation: JAP, ZD, JB; Methodology: JAP, KRQ, CHAL, JB; Project administration: JAP, ZD, DP, CHAL, JB; Resources: JAP, ZD, CHAL; Software: ZD; Supervision: JB, DP, CHAL; Validation: ZD, KRQ; Visualization: JAP, CHAL; Writing–original draft: JAP; Writing–review & editing: all authors. All authors read and approved the final manuscript.

Supplementary Table 1.

Classification of hospital type according to the Australian Statistical Geography Standard-Remoteness Area, 3rd edition
ac-2024-00269-0038-Supplementary-Table-1.pdf

Supplementary Table 2.

Intensive care unit admissions for colorectal cancer surgical patients per hospital type, per year
ac-2024-00269-0038-Supplementary-Table-2.pdf

Supplementary Table 3.

Univariable regression analysis for mortality (n=23,603)
ac-2024-00269-0038-Supplementary-Table-3.pdf

Supplementary Table 4.

Multivariable regression sensitivity analysis for mortality excluding patients with a discharge destination of “transfer to other ICU”
ac-2024-00269-0038-Supplementary-Table-4.pdf

Supplementary Fig. 1.

Annual admissions changes across the decade according to hospital type
ac-2024-00269-0038-Supplementary-Fig-1.pdf

Supplementary Material 1.

STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement checklist.
ac-2024-00269-0038-Supplementary-Material-1.pdf

Supplementary Material 2.

List of participating intensive care units.
ac-2024-00269-0038-Supplementary-Material-2.pdf
Supplementary materials are available from https://doi.org/10.3393/ac.2024.00269.0038.
Table 1.
Characteristics of the study cohort (n=23,719)
Characteristic Overall (n= 23,719) Metropolitan (n=19,611) Rural/regional (n=4,108) P-valuea
No. of ICUs 164 122 42 -
Age (yr) 73.5 (64.2–81.4) 73.1 (63.8–81.3) 74.6 (66.2–81.5) <0.001
Sex 0.500
 Male 13,088 (55.2) 10,801 (55.1) 2,287 (55.6)
 Female 10,620 (44.8) 8,800 (44.9) 1,820 (44.3)
 Unknown 11 (0.0) 10 (0.0) 1 (0.0)
Patient identified as Indigenous 463 (2.0) 311 (1.6) 152 (3.7) <0.001
 Unknown 2,863 (12.1) 2,364 (12.1) 499 (12.1)
Remoteness level of patient's residence <0.001
 Major city 13,604 (57.4) 13,328 (68.0) 276 (6.7)
 Inner regional 6,433 (27.1) 4,262 (21.7) 2,171 (52.8)
 Outer regional 3,152 (13.3) 1,628 (8.3) 1,524 (37.1)
 Remote/very remote 530 (2.2) 393 (2.0) 137 (3.3)
Distance from ICU to patient's residence (km) 13.0 (5.9–38.0) 11.6 (5.7–29.0) 33.3 (9.1–63.4) <0.001
IRSAD score 993 (948–1,053) 1,010 (960–1,064) 949 (917–973) <0.001
 Unknown 21 (0.1) 19 (0.1) 2 (0.0)
Respiratory disease 1,307 (5.5) 1,004 (5.1) 303 (7.4) <0.001
Cardiovascular disease 2,121 (8.9) 1,642 (8.4) 479 (11.7) <0.001
Liver disease 142 (0.6) 121 (0.6) 21 (0.5) 0.400
Renal disease 481 (2.0) 389 (2.0) 92 (2.2) 0.300
Immunosuppressive disease 976 (4.1) 925 (4.7) 51 (1.2) <0.001
Immunosuppressive therapy 1,018 (4.3) 915 (4.7) 103 (2.5) <0.001
Hospital funding model <0.001
 Private 12,372 (52.2) 11,773 (60.0) 599 (14.6)
 Public 11,347 (47.8) 7,838 (40.0) 3,509 (85.4)
Source of hospital admission <0.001
 Home 21,983 (92.7) 18,259 (93.1) 3,724 (90.7)
 Hospital transfer 1,310 (5.5) 980 (5.0) 330 (8.0)
 RACF/palliative care 109 (0.5) 75 (0.4) 34 (0.8)
 Unknown 317 (1.3) 297 (1.5) 20 (0.5)
Source of ICU admission <0.001
 Operating room 22,980 (96.9) 19,046 (97.1) 3,934 (95.8)
 Ward 314 (1.3) 226 (1.2) 88 (2.1)
 Emergency department 105 (0.4) 51 (0.3) 54 (1.3)
 Hospital transfer 320 (1.3) 288 (1.5) 32 (0.8)
 ICU transfer 0 (0) 0 (0) 0 (0)
Treatment limitation <0.001
 Full 22,618 (95.4) 18,851 (96.1) 3,767 (91.7)
 Limitation of care 692 (2.9) 413 (2.1) 279 (6.8)
 Unknown 409 (1.7) 347 (1.8) 62 (1.5)
Type of admission to ICU <0.001
 Planned 19,057 (80.3) 16,280 (83.0) 2,777 (67.6)
 Unplanned 4,540 (19.1) 3,275 (16.7) 1,265 (30.8)
 Unknown 122 (0.5) 56 (0.3) 66 (1.6)
Type of surgery <0.001
 Elective 19,800 (83.5) 16,886 (86.1) 2,934 (71.4)
 Emergency 3,823 (16.1) 2,710 (13.8) 1,113 (27.1)
 Unknown 96 (0.4) 35 (0.2) 61 (1.5)
APACHE III-J score 45 (36–55) 45 (36–55) 45 (36–55) >0.999
 Unknown 39 (0.2) 35 (0.2) 4 (0.1)
ANZROD score
 Median (IQR) 1.07 (0.59–2.23) 1.07 (0.59–2.20) 1.06 (0.59–2.35) 0.500
 Mean±SD 2.39±4.97 2.31±4.69 2.76±6.12 0.500
 Unknown 39 (0.2) 35 (0.2) 4 (0.1) -
Ventilated during ICU admission 3,001 (12.7) 2,635 (13.5) 366 (9.0) <0.001
 Unknown 58 (0.2) 30 (0.2) 28 (0.7)
Annual no. of admissions for colorectal cancer surgery per site 32 (18–52) 36 (19–55) 20 (12–29) <0.001

Values are presented as median (IQR), number (%), or mean±SD, unless otherwise indicated. Percentages may not total 100 due to rounding.

ICU, intensive care unit; IRSAD, Index of Relative Socio-economic Advantage and Disadvantage; RACF, residential aged care facility; APACHE, Acute Physiology and Chronic Health Evaluation; ANZROD, Australian and New Zealand Risk of Death; IQR, interquartile range; SD, standard deviation.

aWilcoxon rank sum test, Pearson chi-square test, or Fisher exact test.

Table 2.
Logistic regression analysis for mortality
Variable OR 95% CI P-value
Hospital type
 Metropolitan - - -
 Rural 0.93 0.71–1.22 0.600
Hospital funding model
 Private - - -
 Public 1.61 1.30–2.01 <0.001
Annual no. of admissions for colorectal cancer surgerya 0.97 0.93–1.01 0.130
Remoteness level of patient's residence
 Major city - - -
 Inner regional 1.20 0.93–1.55 0.200
 Outer regional 1.32 0.92–1.88 0.120
 Remote/very remote 1.17 0.49–2.44 0.700
Distance from ICU to patient's residenceb (km) 0.86 0.80–0.93 <0.001
ANZROD model 1.10 1.10–1.11 <0.001
Year of admission
 2011 - - -
 2012 0.80 0.52–1.21 0.300
 2013 1.12 0.75–1.67 0.600
 2014 0.65 0.41–1.01 0.057
 2015 1.00 0.67–1.49 >0.900
 2016 0.81 0.53–1.22 0.300
 2017 0.86 0.57–1.28 0.400
 2018 0.80 0.54–1.19 0.300
 2019 0.55 0.36–0.84 0.005
 2020 0.73 0.49–1.09 0.120
 2021 0.52 0.34–0.80 0.003

OR, odds ratio; CI, confidence interval; ICU, intensive care unit; ANZROD, Australian and New Zealand Risk of Death.

aFor the annual number of admissions for colorectal surgery, each OR represents a +10 increase.

bLog-adjusted.

Table 3.
Outcomes
Outcome Overall (n= 23,719) Metropolitan (n=19,611) Rural/regional (n=4,108) P-valuea
Survival to hospital discharge <0.001
 Survived 23,067 (97.3) 19,104 (97.4) 3,963.0 (96.5)
 Died 575 (2.4) 442 (2.3) 133 (3.2)
 Unknown 77 (0.3) 65 (0.3) 12 (0.3)
Hospital discharge destination <0.001
 Death 575 (2.4) 442 (2.3) 133 (3.2)
 Home 19,377 (81.7) 16,199 (82.6) 3,178 (77.4)
 Hospital transfer 1,485 (6.3) 1,015 (5.2) 470 (11.4)
 RACF/palliative care 1,315 (5.6) 1,093 (5.6) 222 (5.4)
 Rehabilitation 751 (3.2) 676 (3.4) 75 (1.8)
 Other 139 (0.6) 121 (0.6) 18 (0.4)
 Unknown 77 (0.3) 65 (0.3) 12 (0.3)
ICU length of stay (day) 1.1 (0.8–2.6) 1.0 (0.8–2.0) 2.1 (1.1–3.9) <0.001
 Unknown 35 (0.1) 33 (0.2) 2 (0.0)
Hospital length of stay (day) 10.0 (6.9–16.0) 10.0 (6.9–16.2) 9.3 (6.4–14.9) <0.001
 Unknown 98 (0.4) 76 (0.4) 22 (0.5)
Readmitted to ICU 1,463 (6.2) 1,200 (6.1) 263 (6.4) 0.500
ICU discharge destination 0.031
 Died 149 (0.6) 112 (0.6) 37 (0.9)
 Ward 23,098 (97.4) 19,112 (97.5) 3,986 (97.0)
 Hospital transfer 417 (1.8) 337 (1.7) 80 (1.9)
 Unknown 55 (0.2) 50 (0.3) 5 (0.1)

Values are presented as number (%) or median (interquartile range). Percentages may not total 100 due to rounding.

RACF, residential aged care facility; ICU, intensive care unit.

aWilcoxon rank sum test or Pearson chi-squared test.

  • 1. Dale CD, McLoone P, Sloan B, Kinsella J, Morrison D, Puxty K, et al. Critical care provision after colorectal cancer surgery. BMC Anesthesiol 2016;16:94.ArticlePubMedPMCPDF
  • 2. Udayasiri DK, MacCallum C, Silva ND, Skandarajah A, Hayes IP. Impact of hospital geographic remoteness on short-term outcomes after colorectal cancer resection using state-wide administrative data. ANZ J Surg 2020;90:1328–34. ArticlePubMedPDF
  • 3. Udayasiri DK, MacCallum C, Da Silva N, Skandarajah A, Hayes IP. Impact of hospital geographic remoteness on overall survival after colorectal cancer resection using state-wide administrative data. ANZ J Surg 2020;90:1321–7. ArticlePubMedPDF
  • 4. Huo YR, Phan K, Morris DL, Liauw W. Systematic review and a meta-analysis of hospital and surgeon volume/outcome relationships in colorectal cancer surgery. J Gastrointest Oncol 2017;8:534–46. ArticlePubMedPMC
  • 5. Clancy B. Royal Australasian College of Surgeons Rural Health Equity Strategic Action Plan: excellence through equity. ANZ J Surg 2022;92:1990–4. ArticlePubMedPDF
  • 6. Wichmann MW, McCullough TK, Beukes E, Gunning T, Maddern GJ. Outcomes of laparoscopic colorectal cancer surgery in a non-metropolitan Australian hospital: audit of 120 patients with long-term follow-up results. ANZ J Surg 2022;92:3214–8. ArticlePubMedPDF
  • 7. Tsai TC, Orav EJ, Jha AK. Care fragmentation in the postdischarge period: surgical readmissions, distance of travel, and postoperative mortality. JAMA Surg 2015;150:59–64. ArticlePubMed
  • 8. Knaus WA, Wagner DP, Draper EA, Zimmerman JE, Bergner M, Bastos PG, et al. The APACHE III prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest 1991;100:1619–36. ArticlePubMed
  • 9. Pilcher D, Paul E, Bailey M, Huckson S. The Australian and New Zealand Risk of Death (ANZROD) model: getting mortality prediction right for intensive care units. Crit Care Resusc 2014;16:3–4. ArticlePubMed
  • 10. Australian Bureau of Statistics. 2033.0.55.001: census of population and housing: Socio-Economic Indexes for Areas (SEIFA), Australia, 2016: IRSAD (Index of Relative Socio-economic Advantage and Disadvantage) [Internet]. Australian Bureau of Statistics; 2018 [updated 2023 Mar 21; cited 2024 Jan 31]. Available from: https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/2033.0.55.001~2016~Main%20Features~IRSAD~20
  • 11. Australian Government Department of Health and Aged Care. Australian Statistical Geography Standard: remoteness area [Internet]. Australian Government Department of Health and Aged Care; 2021 [cited 2024 Jan 31]. Available from: https://www.health.gov.au/topics/rural-health-workforce/classifications/asgs-ra
  • 12. Australian and New Zealand Intensive Care Society (ANZICS). ANZICS Centre for Outcome and Resource Evaluation 2020 Report Melbourne, Australia. ANZICS; 2021.PDF
  • 13. Australian Institute Health and Welfare (AIWH). Rural and remote health [Internet]. AIWH; [updated 2024 Apr 30; cited 2024 May 2]. Available from: https://www.aihw.gov.au/reports/rural-remote-australians/rural-and-remote-health
  • 14. Zhou H, Jin Y, Wang J, Chen G, Chen J, Yu S. Comparison of short-term surgical outcomes and long-term survival between emergency and elective surgery for colorectal cancer: a systematic review and meta-analysis. Int J Colorectal Dis 2023;38:41.ArticlePubMedPDF
  • 15. Australian Institute of Health and Welfare (AIWH). National Bowel Cancer Screening Program: monitoring report 2023. AIWH; 2023.PDF
  • 16. Thorn H, Olley R. Barriers and facilitators to accessing medical services in rural and remote Australia: a systematic review. Asia Pac J Health Manag 2023;18:20–9. ArticlePDF
  • 17. Tham NL, Skandarajah A, Hayes IP. Socioeconomic disadvantage and its impact on colorectal cancer in Australia: a scoping review. ANZ J Surg 2022;92:2808–15. ArticlePubMedPMCPDF
  • 18. Khoury W, Lavery IC, Kiran RP. Effects of chronic immunosuppression on long-term oncologic outcomes for colorectal cancer patients undergoing surgery. Ann Surg 2011;253:323–7. ArticlePubMed
  • 19. Wilson RJ, Yates DR, Walkington JP, Davies SJ. Ventilatory inefficiency adversely affects outcomes and longer-term survival after planned colorectal cancer surgery. Br J Anaesth 2019;123:238–45. ArticlePubMedPMC
  • 20. Australian and New Zealand College of Anaesthetists (ANZCA). PS41(G) Position statement on acute pain management 2023. ANZCA; 2023.PDF
  • 21. Morris M, Iacopetta B, Platell C. Comparing survival outcomes for patients with colorectal cancer treated in public and private hospitals. Med J Aust 2007;186:296–300. ArticlePubMedPDF
  • 22. Timmers TK, Verhofstad MH, Moons KG, van Beeck EF, Leenen LP. Long-term quality of life after surgical intensive care admission. Arch Surg 2011;146:412–8. ArticlePubMed
  • 23. Tang H, Besson A, Deftereos I, Mahon B, Cho J, Faragher I, et al. The health-related quality of life changes following surgery in patients with colorectal cancer: a longitudinal study. ANZ J Surg 2022;92:1461–5. ArticlePubMedPDF

Figure & Data

References

    Citations

    Citations to this article as recorded by  

      • PubReader PubReader
      • Cite this Article
        Cite this Article
        export Copy Download
        Close
        Download Citation
        Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

        Format:
        • RIS — For EndNote, ProCite, RefWorks, and most other reference management software
        • BibTeX — For JabRef, BibDesk, and other BibTeX-specific software
        Include:
        • Citation for the content below
        Comparison of colorectal cancer surgery patients in intensive care between rural and metropolitan hospitals in Australia: a national cohort study
        Close
      • XML DownloadXML Download
      Comparison of colorectal cancer surgery patients in intensive care between rural and metropolitan hospitals in Australia: a national cohort study
      Comparison of colorectal cancer surgery patients in intensive care between rural and metropolitan hospitals in Australia: a national cohort study
      Characteristic Overall (n= 23,719) Metropolitan (n=19,611) Rural/regional (n=4,108) P-valuea
      No. of ICUs 164 122 42 -
      Age (yr) 73.5 (64.2–81.4) 73.1 (63.8–81.3) 74.6 (66.2–81.5) <0.001
      Sex 0.500
       Male 13,088 (55.2) 10,801 (55.1) 2,287 (55.6)
       Female 10,620 (44.8) 8,800 (44.9) 1,820 (44.3)
       Unknown 11 (0.0) 10 (0.0) 1 (0.0)
      Patient identified as Indigenous 463 (2.0) 311 (1.6) 152 (3.7) <0.001
       Unknown 2,863 (12.1) 2,364 (12.1) 499 (12.1)
      Remoteness level of patient's residence <0.001
       Major city 13,604 (57.4) 13,328 (68.0) 276 (6.7)
       Inner regional 6,433 (27.1) 4,262 (21.7) 2,171 (52.8)
       Outer regional 3,152 (13.3) 1,628 (8.3) 1,524 (37.1)
       Remote/very remote 530 (2.2) 393 (2.0) 137 (3.3)
      Distance from ICU to patient's residence (km) 13.0 (5.9–38.0) 11.6 (5.7–29.0) 33.3 (9.1–63.4) <0.001
      IRSAD score 993 (948–1,053) 1,010 (960–1,064) 949 (917–973) <0.001
       Unknown 21 (0.1) 19 (0.1) 2 (0.0)
      Respiratory disease 1,307 (5.5) 1,004 (5.1) 303 (7.4) <0.001
      Cardiovascular disease 2,121 (8.9) 1,642 (8.4) 479 (11.7) <0.001
      Liver disease 142 (0.6) 121 (0.6) 21 (0.5) 0.400
      Renal disease 481 (2.0) 389 (2.0) 92 (2.2) 0.300
      Immunosuppressive disease 976 (4.1) 925 (4.7) 51 (1.2) <0.001
      Immunosuppressive therapy 1,018 (4.3) 915 (4.7) 103 (2.5) <0.001
      Hospital funding model <0.001
       Private 12,372 (52.2) 11,773 (60.0) 599 (14.6)
       Public 11,347 (47.8) 7,838 (40.0) 3,509 (85.4)
      Source of hospital admission <0.001
       Home 21,983 (92.7) 18,259 (93.1) 3,724 (90.7)
       Hospital transfer 1,310 (5.5) 980 (5.0) 330 (8.0)
       RACF/palliative care 109 (0.5) 75 (0.4) 34 (0.8)
       Unknown 317 (1.3) 297 (1.5) 20 (0.5)
      Source of ICU admission <0.001
       Operating room 22,980 (96.9) 19,046 (97.1) 3,934 (95.8)
       Ward 314 (1.3) 226 (1.2) 88 (2.1)
       Emergency department 105 (0.4) 51 (0.3) 54 (1.3)
       Hospital transfer 320 (1.3) 288 (1.5) 32 (0.8)
       ICU transfer 0 (0) 0 (0) 0 (0)
      Treatment limitation <0.001
       Full 22,618 (95.4) 18,851 (96.1) 3,767 (91.7)
       Limitation of care 692 (2.9) 413 (2.1) 279 (6.8)
       Unknown 409 (1.7) 347 (1.8) 62 (1.5)
      Type of admission to ICU <0.001
       Planned 19,057 (80.3) 16,280 (83.0) 2,777 (67.6)
       Unplanned 4,540 (19.1) 3,275 (16.7) 1,265 (30.8)
       Unknown 122 (0.5) 56 (0.3) 66 (1.6)
      Type of surgery <0.001
       Elective 19,800 (83.5) 16,886 (86.1) 2,934 (71.4)
       Emergency 3,823 (16.1) 2,710 (13.8) 1,113 (27.1)
       Unknown 96 (0.4) 35 (0.2) 61 (1.5)
      APACHE III-J score 45 (36–55) 45 (36–55) 45 (36–55) >0.999
       Unknown 39 (0.2) 35 (0.2) 4 (0.1)
      ANZROD score
       Median (IQR) 1.07 (0.59–2.23) 1.07 (0.59–2.20) 1.06 (0.59–2.35) 0.500
       Mean±SD 2.39±4.97 2.31±4.69 2.76±6.12 0.500
       Unknown 39 (0.2) 35 (0.2) 4 (0.1) -
      Ventilated during ICU admission 3,001 (12.7) 2,635 (13.5) 366 (9.0) <0.001
       Unknown 58 (0.2) 30 (0.2) 28 (0.7)
      Annual no. of admissions for colorectal cancer surgery per site 32 (18–52) 36 (19–55) 20 (12–29) <0.001
      Variable OR 95% CI P-value
      Hospital type
       Metropolitan - - -
       Rural 0.93 0.71–1.22 0.600
      Hospital funding model
       Private - - -
       Public 1.61 1.30–2.01 <0.001
      Annual no. of admissions for colorectal cancer surgerya 0.97 0.93–1.01 0.130
      Remoteness level of patient's residence
       Major city - - -
       Inner regional 1.20 0.93–1.55 0.200
       Outer regional 1.32 0.92–1.88 0.120
       Remote/very remote 1.17 0.49–2.44 0.700
      Distance from ICU to patient's residenceb (km) 0.86 0.80–0.93 <0.001
      ANZROD model 1.10 1.10–1.11 <0.001
      Year of admission
       2011 - - -
       2012 0.80 0.52–1.21 0.300
       2013 1.12 0.75–1.67 0.600
       2014 0.65 0.41–1.01 0.057
       2015 1.00 0.67–1.49 >0.900
       2016 0.81 0.53–1.22 0.300
       2017 0.86 0.57–1.28 0.400
       2018 0.80 0.54–1.19 0.300
       2019 0.55 0.36–0.84 0.005
       2020 0.73 0.49–1.09 0.120
       2021 0.52 0.34–0.80 0.003
      Outcome Overall (n= 23,719) Metropolitan (n=19,611) Rural/regional (n=4,108) P-valuea
      Survival to hospital discharge <0.001
       Survived 23,067 (97.3) 19,104 (97.4) 3,963.0 (96.5)
       Died 575 (2.4) 442 (2.3) 133 (3.2)
       Unknown 77 (0.3) 65 (0.3) 12 (0.3)
      Hospital discharge destination <0.001
       Death 575 (2.4) 442 (2.3) 133 (3.2)
       Home 19,377 (81.7) 16,199 (82.6) 3,178 (77.4)
       Hospital transfer 1,485 (6.3) 1,015 (5.2) 470 (11.4)
       RACF/palliative care 1,315 (5.6) 1,093 (5.6) 222 (5.4)
       Rehabilitation 751 (3.2) 676 (3.4) 75 (1.8)
       Other 139 (0.6) 121 (0.6) 18 (0.4)
       Unknown 77 (0.3) 65 (0.3) 12 (0.3)
      ICU length of stay (day) 1.1 (0.8–2.6) 1.0 (0.8–2.0) 2.1 (1.1–3.9) <0.001
       Unknown 35 (0.1) 33 (0.2) 2 (0.0)
      Hospital length of stay (day) 10.0 (6.9–16.0) 10.0 (6.9–16.2) 9.3 (6.4–14.9) <0.001
       Unknown 98 (0.4) 76 (0.4) 22 (0.5)
      Readmitted to ICU 1,463 (6.2) 1,200 (6.1) 263 (6.4) 0.500
      ICU discharge destination 0.031
       Died 149 (0.6) 112 (0.6) 37 (0.9)
       Ward 23,098 (97.4) 19,112 (97.5) 3,986 (97.0)
       Hospital transfer 417 (1.8) 337 (1.7) 80 (1.9)
       Unknown 55 (0.2) 50 (0.3) 5 (0.1)
      Table 1. Characteristics of the study cohort (n=23,719)

      Values are presented as median (IQR), number (%), or mean±SD, unless otherwise indicated. Percentages may not total 100 due to rounding.

      ICU, intensive care unit; IRSAD, Index of Relative Socio-economic Advantage and Disadvantage; RACF, residential aged care facility; APACHE, Acute Physiology and Chronic Health Evaluation; ANZROD, Australian and New Zealand Risk of Death; IQR, interquartile range; SD, standard deviation.

      Wilcoxon rank sum test, Pearson chi-square test, or Fisher exact test.

      Table 2. Logistic regression analysis for mortality

      OR, odds ratio; CI, confidence interval; ICU, intensive care unit; ANZROD, Australian and New Zealand Risk of Death.

      For the annual number of admissions for colorectal surgery, each OR represents a +10 increase.

      Log-adjusted.

      Table 3. Outcomes

      Values are presented as number (%) or median (interquartile range). Percentages may not total 100 due to rounding.

      RACF, residential aged care facility; ICU, intensive care unit.

      Wilcoxon rank sum test or Pearson chi-squared test.


      Ann Coloproctol : Annals of Coloproctology Twitter Facebook
      TOP