Page 487 - An Evidence Review of Active Surveillance in Men With Localized Prostate Cancer

Appendix Table C4.2. Comparison between watchful waiting and radical prostatectomy

Author, Year Study name Comparison Outcome Followup Sample size Results Factors included in the
[Pubmed ID] /Database definition/ (yr) per group model
measurement
Study design instrument
Prostate-cancer
specific
mortality
Bill-Axelson 213 SPCG-4 RP vs. WW Death from 12.8 RP: 347 RR=0.62 (0.44, 0.87); P = None (RCT)
2011 prostate cancer WW: 348 0.01
21542742

RCT
Hadley 207 SEER- RP vs. Death from Survival RP: 11,936; Unweighted regression PS: age, race/ethnicity,
2010 Medicare conservative prostate cancer was conservative analysis: marital status, tumor
20944078 management from SEER observed management: HR=0.62 (0.50, 0.79); P characteristics, previous
records. for up to 5879 <0.001 health problems (based
Retrospective 12 years; PS reweighted analysis on NCI combined
cohort mean [calculated using IPTW: HR=0.63 (0.55, comorbidity index and
survival based on the 0.71); P <0.001 Medicare
time free proportion of PS reweighted analysis reimbursements in the
of all patients using SMRW: HR=0.72 12 months before
cause- treated with (0.57, 0.91); P <0.001 diagnosis), year of
death = each IV regression using the diagnosis. These
83.0 mo; modality, for previous year’s local area variables were included
median the overall treatment pattern for in all multivariable
survival population] conservative management models.
time from as an instrument:
diagnosis HR=1.37 (0.15, 12.5); P = Instrumental variable:
to Dec 0.78) the lagged (previous
st
31 , 2008 year’s) local area
(censoring treatment pattern for
date) = 78 conservative
mo (IQR = management.
48 mo)

C-161