Prof. Lee Shepstone's presentation from Osteoporosis 2016: The SCOOP study – Do we now have a rationale to screen for osteoporosis?

1. National Osteoporosis Conference 7th November 2016
The SCOOP study – do we now have a
rationale to screen for osteoporosis ?

2. E Lenaghan
C Cooper
S Clarke
R Fordham
N Gittoes
I Harvey
N Harvey
R Holland
A Howe
JA Kanis
T Marshall
TW O’Neill
T Peters
D Torgerson
E McCloskey
& the SCOOP study team

3.  A pragmatic randomised controlled trial of the
effectiveness and cost effectiveness of screening older
women for the prevention of fractures – the SCOOP
study.
 Is a community based screening programme effective
and cost-effective in reducing fractures in older women
in the UK?

4.  Cost of Osteoporosis in the UK :
- 536 000 new fragility fractures in 2010
- 79 000 hip fractures
- Around £3.5 billion each year and rising.
 Currently no community based systematic
screening in the UK.
 Would it be beneficial?

5.  Participants :
(i) Female
(ii) Aged 70 to 85
(iii) Not on prescription anti-osteoporosis medication
(iv) Not deemed inappropriate for research study
 Identified from primary care.
 Seven geographical regions of the UK.
 First randomisation April 2008.

6.  Screening :
(1) FRAX questionnaire by post
- 10 year probability of hip fracture
(2) DXA scans in those above risk threshold
(3) FRAX probability re-calculated with BMD
- using femoral neck T-score
(4) GPs informed of those at high risk
- previous UK thresholds (2004)

7.  Outcomes :
Primary :
i) Proportion of individuals experiencing an osteoporosis-
related fracture.
Secondary :
ii) Hip fractures
iii) Mortality
iv) Anxiety (State-Trait Anxiety Inventory)
 Five-year follow-up period.

8. Participant Flow
Women 70–85 yrs from
100 primary care practices
(n = 52,033)
Eligible subjects
(n = 38,031)
Consenting participants
(n = 13,029)
Randomisation
(n = 12,495)
SCREENING
n = 6,233
CONTROL
n = 6,250
Post-Randomisation Exclusions
(n=12)

9. Randomised
(N = 12 483)
Responding
Non-Consenters
Has a degree? Yes 2536 (20.3%) 1080 ( 9.9%)
Social Class I
II
IIIN
IIIM
IV
V
1256 (10.2%)
3698 (30.1%)
2109 (17.2%)
3249 (26.5%)
1471 (12.0%)
494 ( 4.0%)
570 ( 5.8%)
2206 (22.4%)
1651 (16.8%)
3196 (32.5%)
1476 (15.0%)
739 ( 7.5%)
Fallen in past year? Yes 3445 (27.6%) 2186 (19.9%)
Broken bone since
50?
Yes 2863 (22.9%) 1859 (17.0%)
Parents broken hip? Yes 1162 ( 9.3%) 536 ( 5.3%)
Smoker? Yes 581 ( 4.7%) 826 ( 7.4%)
Moderate Drinker? Yes 444 ( 3.6%) 383 ( 3.4%)
Age Mean (SD) 75.4 (4.15) 76.8 (5.84)
BMI Mean (SD) 26.6 (4.73) 26.1 (4.90)
Baseline Variables - Randomised versus Responding Non-Consenters

10. Intervention
(N=6233)
Control
(N=6250)
Fallen in past year? Yes 1744 (28.0%) 1700 (27.2%)
Broken bone since 50? Yes 1399 (22.4%) 1463 (23.4%)
Parents broken hip? Yes 585 ( 9.4%) 577 ( 9.2%)
Smoker? Yes 290 ( 4.7%) 290 ( 4.6%)
Moderate Drinker? Yes 219 ( 3.5%) 225 ( 3.6%)
Glucocorticoid Use? Yes 316 ( 5.1%) 312 ( 5.0%)
Rheumatoid Arthritis? Yes 426 ( 6.8%) 410 ( 6.6%)
Secondary Causes of
OP?
Yes 1483 (23.8%) 1408 (22.5%)
Age (at response) Mean (SD) 75.4 (4.16) 75.5 (4.14)
BMI Mean (SD) 26.7 (4.71) 26.7 (4.75)
FRAX 10 year HIP
Probability
Mean (SD) 8.5% (7.4%) 8.5% (7.3%)
FRAX 10 year Major
OP Probability
Mean (SD) 19.3% (8.9%) 19.3% (8.8%)
Baseline FRAX Variables - Intervention versus Control

11. SCREENING participants
n = 6,233
TO DXA
n = 3,064 (49%)
LOW RISK subjects
n = 3,171 (51%)
HIGH RISK subjects
n = 898
LOW RISK subjects
n = 1,917
RISK not Calculated
n = 247
HIGH RISK LOW RISK NOT CALCULATED
n = 898 14% n = 5,088 82% n = 247 4%
Screening Figures

12. 15.3
13.8
13.3 13.4
13.9
4.2
6.6
8.1
9.3
10.1
1 2 3 4 5
%
Year of follow-up
Percentage of participants with prescription anti-
osteoporosis medication
Screening Control
Medication Figures

13. Results
Anxiety (STAI)
0
2
4
6
8
10
12
14
0 6 12 24 36 48 60
Screening - LOW RISK
Screening - HIGH RISK
Control
Repeated
Measures
Analysis
Group
Group*Time
p=0.515
p=0.942

14. 12.9 13.6
1
%
Osteoporotic-related Fractures
Hazard Ratio : 0.93 95% CI : 0.85 to 1.03
p=0.199
Results

15. 2.6
3.5
1
%
Hip Fractures
Hazard Ratio : 0.72 95% CI : 0.59 to 0.89
p=0.002
Results

16. 8.8 8.4
1
%
5-year Mortality
Hazard Ratio : 1.05 95% CI : 0.93 to 1.19
p=0.433
Results

17. Discussion Points
Is there a selection bias ?
(8.6% mortality vs expected 19.0%)
Why hips and not all fractures ?
28% reduction in hip fractures ?

18. Conclusions from SCOOP :
1) A community based UK screening programme is
feasible, generally well received.
2) No evidence that the overall rate of fractures or
mortality could be significantly reduced.
3) Evidence that hip fractures could be reduced.