The Essentials of Biostatistics for Physicians, Nurses, and Clinicians

90 CHAPTER 6 Hypothesis Testing

Table 6.4
Cutting Balloon Angioplasty Combined p - Value Meta - Analysis by
Fisher ’ s Test

Study Cutting balloon

restenosis proportion

Conventional

balloon restenosis

proportion

p - value − 2 ln( U )

Grt 173/551 170/559 0.7455 0.5874
Molstad 5/30 8/31 0.5339 1.2551
Inoue 7/32 13/32 0.1769 3.4641
Kondo 22/95 40/95 0.0083 9.5830
Ergene 14/51 22/47 0.0483 6.0606

Nozaki 26/98 40/93 0.022 7.6334
Suzuki 104/357 86/188 0.001 13.8155
Combined 0.000107 42.3994

Table 6.3

Balloon Angioplasty Restenosis Rates

by Country

Country Restenosis rate % (no. of

failures/no. of patients)

A 40% (18/45)

B 41% (58/143)

C 29% (20/70)

D 29% (51/177)

E 22% (26/116)

The client did have several published studies of the use of the
cutting balloon for angioplasty. The hope is that combining this data
with the pooled results in the clinical trial, a clinically signifi cant
improvement over the conventional rate of 40% could be shown and
used to improve the case for approval of the treatment.
I conducted a meta - analysis using six independent studies with
the cutting balloon along with the company ’ s clinical trial (referred to
as GRT). Table 6.4 shows the meta - analysis using Fisher ’ s p - value