Impact of Critical Illness on Subsequent Cancer Care and Outcomes, in Patients With Colorectal Cancer

What happens to patients with colorectal cancer, after intensive care?

Why this MD?

Intensive Care Single CCT -> Research SSY
Approached Kathryn for a data focused project
Outcomes of critical illness
Cancer an interesting cohort

Practicalities

This is the cohort we had data for: CORECT-R

Relevance

Most common solid organ cancer in critical care (44% of solid organ cancer admissions)

MD ≠ Thesis

Open Questions Following Critical Illness

Colorectal Cancer Care
- Receipt, Quality, Intensity
Long Term Survival

Introduction

1

Bayesian Statistics

2

Methods

3

Demographics

4

Quality Performance Indicators

5

Systemic Anti-Cancer Therapy

6

Survival

7

Critical Illness

8

Reflection & Conclusion

9

Introduction

1

Bayesian Statistics

2

Methods

3

Demographics

4

Quality Performance Indicators

5

Systemic Anti-Cancer Therapy

6

Survival

7

Critical Illness

8

Reflection & Conclusion

9

Introduction

3rd most common cancer in Scotland
Most common cancer in critical care (44%)

Perspective

Oncologist
Surgeon
Intensivist
Patient

Stage

0 / I / II LR
II HR / III
IV

Statistics

Bayesian (NHST/Regression)
Single Variable Screening -> Causal

Methods

National Safe Haven /CORECT-R
SMR01/SMR06/SICSAG
QPI/CHEMOCARE

Demographics

Emergency vs Elective
Ward Vs Critical Care

Quality Performance Indicators

12 QPIs with 21 targets

Pre-Operative
Peri-Operative
Post-Operative

QPI 11 - Post Operative Chemotherapy

Those QPIs pertaining to pre‐operative management include: • Radiological staging (QPI 1i and 1ii) • Pre‐operative imaging (QPI 2) • Pre‐treatment multi‐disciplinary meeting (QPI 3) The peri‐operative QPIs include: • Anticipatory stoma care (QPI 4) • Adequate lymph node yield (QPI 5) • Neo‐adjuvant therapy in appropriate patients (QPI 6) • Adequate surgical margins (QPI 7i and 7ii) • Re‐operation rates (QPI 8) • Anastomotic dehiscence rates (QPI 9i and 9ii) 5.1. Introduction 148 The post‐operative QPIs include: • Post‐operative mortality rates at 30 and 90 days (QPI 10i Elective, QPI10i Emer‐ gency, QPI 10ii Elective and QPI 10ii Emergency) • Use of adjuvant chemotherapy in appropriate patients (QPI 11‐Stage III and QPI 11‐High‐Risk Stage II) • Mortality rates at 30 and 90 days post chemo‐ or radio‐therapy (QPI 12i‐30, QPI 12i‐90 and QPI 12ii‐30

For patients admitted electively, those who received organ support (com‐ pared against those monitored in critical care), had a 44% reduction in odds of timely adjuvant chemotherapy (OR 0.56, 95% Credible Interval 0.42‐0.75, 99.9% posterior probability). For patients admitted emergently, those who received organ support (compared to those monitored in critical care), had a 58% reduction in odds of timely adjuvant chemotherapy (0.42 0.27‐0.66, 99.9%). 5.3. Results 172 Among patients treated at a ward level, impact on the odds of attaining QPI 11 was dependent on the nature of hospital admission. Patients treated at a ward level elec‐ tively, when compared to those monitored in critical care, may have an estimated 21% increased odds of attaining QPI 11, (OR 1.21, 0.9‐1.6, 87% posterior probabil‐ ity). Patients treated in the ward emergently, when compared to those monitored in critical care, had a 39% reduction in odds of attaining QPI 11 (0.63, 0.43‐0.93, 98%).

SACT & SURVIVAL

Critical Illness

Intensivist Perspective

Cohorts

Still Elective Vs Emergency
Now Just Critical Care:
1. No Organ Support
2. Single Day Organ Support
3. Multiple Days Organ Support

Questions (Conditional On Hospital Survival)

Adjuvant SACT Receipt
Overall Survival

Statistics

Still Bayes
Now Causal
DAG From SNAP-3

To address question one, among the total cohort of patients admitted to critical care, eligible for SACT, there was a 53% probability of receiving adjuvant SACT in the 12 weeks following surgical resection. For elective patients with critical illness, estimated HR of SACT receipt was reduced to 0.59 (95% CrI 0.48‐0.71). This resulted in apredicted probability of 41% of adjuvant SACT receipt for elective patients with critical illness. For emergency patients with critical illness, estimated HR of SACT receipt was reduced to 0.43 (95% CrI 0.31‐0.59). This resulted in a predicted probability of 28% of adjuvant SACT receipt for emergency patients with critical illness.

Elective ‐ Multiple contrasted against Elective ‐ Monitoring, had estimated median HR 0.59 (95% CrI 0.48‐0.71).

Emergency ‐ Multiple contrasted against Emergency ‐ Monitoring, had estimated median HR 0.43 (95% CrI 0.31‐0.59).

Each of these estimates were associated with close to 100% posterior probability of being a clinically meaningful difference.

Conclusions

Quality of Care
Planned Care Receipt
Long Term Outcomes

My MD in 20 minutes

Impact of Critical Illness on Subsequent Cancer Care and Outcomes, in Patients With Colorectal Cancer

What happens to patients with colorectal cancer, after intensive care?

Why this MD?

Practicalities

Relevance

MD ≠ Thesis

Open Questions Following Critical Illness

Introduction

Perspective

Stage

Statistics

Methods

Demographics

Quality Performance Indicators

SACT & SURVIVAL

Critical Illness

Cohorts

Questions (Conditional On Hospital Survival)

Statistics

Conclusions