Panel 6: STRATEGIES TO IMPROVE THE USE OF VTE PROPHYLAXIS

Surgeon General's Workshop on Deep Vein Thrombosis

1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12

Return to Menu

SLIDE 1: STRATEGIES TO IMPROVE THE USE OF VTE PROPHYLAXIS

Samuel Z. Goldhaber, MD
Cardiovascular Division
Brigham and Women’s Hospital
Professor of Medicine
Harvard Medical School
May 9, 2006

Return to Top

SLIDE 2: CHANGING MD BEHAVIOR

Guidelines alone do not suffice
Strategies for change
- Surgeon General’s DVT Workshop
- Physician champions
- Registries
- Coalitions (Coalition to Prevent DVT)
- Litigation
- Electronic alerts

Return to Top

SLIDE 3: QUALITY IMPROVEMENT INITIATIVE TO IMPROVE CLINICAL PRACTICE

Randomized controlled trial to issue or withhold electronic alerts to MDs whose high-risk patients were not receiving DVT prophylaxis

Return to Top

SLIDE 4: COMPUTER PROGRAM

We developed a computer program linked to the patient database that screened the system daily to identify high-risk patients.
We included consecutive high-risk patients on medical and surgical services who were not receiving DVT prophylaxis.

Return to Top

SLIDE 5: DEFINITION: “HIGH RISK”

VTE risk score > points:

Cancer 3 (ICD codes)
Prior VTE 3 (ICD codes)
Hypercoagulability 3 (Leiden, ACLA)
Major surgery 2 (> 60 minutes)
Bed rest 1 (“bed rest” order)
Advanced age 1 (> 70 years)
Obesity 1 (BMI > 29 kg/m²)
HRT/OC 1 (order entry)

Return to Top

SLIDE 6: RANDOMIZATION

VTE risk score > 4 No prophylaxis N = 2506

with arrows pointing to:

INTERVENTION Single alert n = 1255
CONTROL No alert n = 1251

Return to Top

SLIDE 7: ALERT MESSAGE

Screen shot of an alert indicating patient is at high risk for deep vein thrombosis, according to the BWH guidelines.

Return to Top

SLIDE 8: BASELINE CHARACTERISTICS

Median age: 62.5 years
Medical services: 83%
Surgical services: 17%
Comorbidities
- Cancer: 80%
- Hypertension: 34%
- Infection: 30%
- Prior VTE: 20%

Return to Top

SLIDE 9: ORDERS: DVT PROPHYLAXIS

	Intervention n = 1255	Control n = 1251	P value
Overall	421 (33.5)	182 (14.5)	< 0.001
Mechanical	125 (10.0)	19 (1.5)	< 0.001
GCS	52 (4.1)	7 (0.6)	< 0.001
IPC	73 (5.8)	12 (1.0)	< 0.001
Pharmacologic	296 (23.6)	163 (13.0)	< 0.001
Heparin	213 (17.0)	81 (6.5)	< 0.001
Warfarin	28 (2.2)	41 (3.3)	0.11
Enoxaparin	55 (4.4)	41 (3.3)	0.18
(Kucher N, et al. NEJM 2005;352:969-977)

Return to Top

SLIDE 10: PRIMARY END POINT: 90 DAYS

	Intervention (n = 1255)	Control (n = 1251)	Hazard ratio (95% CI)	P
Total VTE	61 (4.9)	103 (8.2)	0.59 (0.43–0.81)	0.001
Acute PE	14 (1.1)	35 (2.8)	0.40 (0.21–0.74)	0.004
Prox. DVT	10 (0.8)	23 (1.8)	0.47 (0.20–1.09)	0.08
Distal DVT	5 (0.4)	12 (1.0)	0.42 (0.15–1.18)	0.10
UE DVT	32 (2.5)	33 (2.6)	0.97 (0.60–1.58)	0.90

Return to Top

SLIDE 11: PRIMARY END POINT

Line chart showing intervention and control primary end points. The primary end point of symptomatic and imaging-confirmed DVT or pulmonary embolism occurred in 61 intervention-group and 103 control-group patients. At 90 days, the intervention strategy reduced the risk of DVT or pulmonary embolism by 41%

Return to Top

SLIDE 12: RISK SCORE AND PROPHYLAXIS

	Present Cohort n(%)	Historical Intervention n(%)	p
Patients at risk of VTE with alert sent	411 (7%)	1255 (18%)	p <0.001
Services
Medical	330 (80)	1038 (82)	p=0.30
Surgical	81 (19)	217 (17)	p=0.30
Cumulative Risk Score
4	266 (64)	792 (63)	p=0.60
5	101 (24)	327 (26)	p=0.60
6	37 (9)	110 (9)	p=0.96
7	7 (1)	22 (2)	p=0.88
8	0 (0)	4 (0.3)	p=0.57
Prophylaxis Prescribed	166 (40%)	421 (34%)	p=0.01

Return to Top

SLIDE 13: VTE PROPHYLAXIS: THE FUTURE

Can a 3-screen alert improve physician response (34% in RCT and 40% in cohort)?
Should we use an “opt in” vs. “opt out” strategy for alerts?
Can human alerts function as well or better than electronic alerts?

Return to Top

Return to Menu