Critical Care Research Network CCRNet

1
Critical Care Research Network (CCR-Net)

• Indians Sappers in the ICU How to tell the
  score

2
Do You Wonder

• How outcomes of treatment in your ICU compare
  with other peer hospitals?
• How resource utilization in your ICU compare with
  other peer hospitals?
• How to identify your quality improvement issues
  from the base line information?
• Whether someone else has an ICU policy or
  procedure addressing an issue you face?

3
Critical Care Research Network

• What is CCR-Net?
• Results from early projects
• Overview of current projects
• Future directions

4
Critical Care Research Network

• Academic
• research
• need for relevance

• Community
• quality and efficiency

• Research Projects
• data review
• literature analysis
• project execution

• End User
• formulate question
• participate in design

• Research Transfer
• annual conference
• workshops
• website

5
The goal of evidence based healthcare

• for the right patient
• doing the right thing
• at the right time
• in the right setting
• by the right health care provider
• using the right resources

Clinical Effectiveness
6
"How to" Benchmark

• Primary data acquisition to evaluate outcomes
• Clinical trials
• Surveillance studies
• Primary data acquisition to answer other
  questions
• Registries and data banks
• Summarizing existing information
• Literature reviews
• Meta-analysis
• Group Judgment
• Consensus conference
• Mathematical modeling

7
Original Southwestern Ontario and Current CCR-Net
MDS
17 community hospitals 6 teaching hospitals (7
ICUs)

• BRITISH
• COLUMBIA

X
8
Clinical Evaluation CycleLevel of evidence for
each step
Informal needs assessment (5)
Community-based research What are we doing now?
(5 level 1 for evaluation of CPG)
Develop improvement strategy (CPG) (1) Implement
strategy (CPG) (1)
Feedback How well are we doing? (5)
Identifying need for improvement Can we do
better? (5)
9
CCR-Net projectsWhat are we doing now?
What are we doing now? Minimum dataset Nutrition
survey WLS survey Ventilator survey ED ICU
admissions
What strategies measurable objectives should we
set?
How well are we doing?
Can we do better?
10
Withdrawal of life supportDifferences between
hospitals
70
60
50
40
30
20
10
0

• AT

WDLS
WHLS
11
Withdrawal of life supportVariability between
hospitals

• 1

12
Rationale for the Minimal Data Set

• Database (registry) - core project from 1st SOCCR
  Conference (April 1994)
• Pragmatic approach
• feasible
• identify common issues and generate hypotheses
• monitor data validity following implementation
• core for other studies

13
CCR-Net projectsHow well are we doing?
What are we doing now?
How well are we doing? Long stay
patient Readmissions Bed utilization
What strategies measurable objectives should we
set?
Can we do better?
14
Hospital MortalityStandardized Mortality Ratio
15
Long stay patientsObjectives

• To prospectively determine cause of and resources
  consumed by long term stay ICU patients in
  community and teaching hospitals
• To measure outcome, including quality of life in
  these patients

16
ResultsShort stay versus prolonged stay patients
Note no more than 37measures per observation
were missing except APACHE II where scores were
not available for 1,791 short stay patients and
46 prolonged stay patients and for Risk where no
percentage was available for 2,131 short stay
patients and 65 long stay patients. The pattern
of these results remains the same by hospital
type except that in community hospitals the
prolonged stay patients tend to be older but not
significantly so and there is no significant
difference in operative status. mean(SD)
17
Average TISS for all days by group
The most frequent group classification for each
patient. Numbers in parenthesis are total days in
each series. The sample for each observation
ranged from 4 patient days for TH, non-active to
2688 for total, MOD.
18
ICU and Hospital MortalityOverall and by Group
Bars indicate ICU mortality error bars
indicate additional hospital mortality. There
were no patients in the non-active treatment
group. Note relatively greater additional
hospital mortality in ventilation group.
Differences between CH and TH all non-significant.
19
MEAN SF-36 SCORES AT 30 d AND MEAN US GENERAL
POPULATION (95 confidence intervals)
20
Long stay patientsClinical Evaluation Cycle
What are we doing now?
What strategies measurable objectives should we
set?
How well are we doing?
Can we do better? Weaning centres? Care maps?
21
Readmission to ICU

• Quality indicator?
• time frame no difference lt or gt 24 hours
• Risk factors
• higher APACHE, longer LOS on 1st admission
• Respiratory, GI at highest risk for readmission
• Outcomes
• mortality higher than predicted by APACHE

22
Readmission to ICUCauses
23
Readmission to ICUMortality
24
ICU beds required for days in yearCommunity
hospitals
25
Illness severity does not change with bed
availability
26
ICU discharges increase with demand for ICU beds
27

Does ICU bed availability affect patient outcome?
Percentage of patients who developed outcome from
those at risk (absolute numbers in parentheses)
Teaching
Hospital
Community
Hospital
Empty beds
???
???
???
???
ICU Readmission
8.6 (3)
5.5 (140)
3.1 (19)
4.3 (66)
p-value
0.43
0.22


Hospital death
19.4 (6)
5.5 (129)
6.8 (35)
5.7 (79)
p-value

0.007

0.40
28
CCR-Net projectsCan we do better?
What are we doing now?
What strategies measurable objectives should we
set?
How well are we doing?
Can we do better? Nutrition guidelines study
29
Nutritional SupportCan guidelines improve
practice?

• Study to document appropriateness of nutritional
  support

30
Challenges

• Improving utilization of current data
• analysis and education
• increase membership to broaden comparisons
• Resource utilization and case costing
• how do you define appropriate ICU use?
• Identification of Quality Indicators
• nutritional support, pressure ulcers,
  readmissions, reintubations, nosocomial
  pneumonia, adjusted LOS and mortality

31
ICU Organization Conceptual Framework
32
Implementation of CCR-NetSource of funding
The Richard and Jean Ivey Fund
SSHRC
Physicians Services Inc. Foundation
Industry Nestle Siemens
33
Implementation of CCR-NetAccess to data

• CCR-Net Steering Committee
• William Sibbald, director
• Claudio Martin, associate director
• project co-ordination and central facility
• Data Advisory Committee
• oversee and review requests for access to
  database (confidentiality, ethics)
• membership drawn from LHSC and CCR-Net members

34
Implementation of CCR-NetCost and benefits to
members

• Services provided
• data collection software (MDS, TISS, ventilation)
• standard quarterly reports of activity (self and
  peer hospital comparisons)
• ad hoc reports on request
• No cost for these services as sufficient funding
  is obtained from sponsors
• Hospitals responsible for cost of data collection
  (funded studies have been able to provide token
  support to participants)

35
CAP
36
(No Transcript)