Skip to main content
. 2017 Mar 22;109(8):djw323. doi: 10.1093/jnci/djw323

Table 1.

Characteristics of included studies

Variables Cancer studies* Cancer surgical studies
No. of studies/total No. (%) No. of studies/total No. (%)
Cancer type
 Gastrointestinal cancer 7/33 (21.2) 118/306 (38.6)
 Lung cancer 7/33 (21.2) 89/306 (29.1)
 Genitourinary cancer 6/33 (18.2) 67/306 (21.9)
 Breast cancer 5/33 (15.2) 20/306 (6.5)
 Thyroid cancer 0 5/306 (1.6)
 Head and neck cancers 0 3/306 (1.0)
 Hematopoietic and lymphoid cancers 3/33 (9.1) 0
 Skin cancer 1/33 (3.0) 0
 Nervous system cancer 0 1/306 (0.3)
 Advanced/metastatic cancer 3/33 (9.1) 0
 Others 1/33 (3.0) 3/306 (1.0)
No. of treatment groups
 2 32/33 (97.0) 277/306 (90.5)
 3 1/33 (3.0) 21/306 (6.9)
 ≥4 0 8/306 (2.6)
No. of patients enrolled
 Median 4515 699
 Q1 to Q3 1392 to 20600 307 to 2783
 Not mentioned 0 3/306 (1.0)
Propensity score methods type
 Propensity score matching 16/33 (48.5) 207/306 (67.6)
 Propensity score weighting 9/33 (27.3) 20/306 (6.5)
 Propensity score stratification 1/33 (3.0) 27/306 (8.8)
 Covariate adjustment using propensity score 1/33 (3.0) 31/306 (10.1)
 More than one type 6/33 (18.2) 21/306 (6.9)
The proportion of matched sample size†
 <25% 2/21 (9.5) 35/219 (16.0)
 25%–<50% 6/21 (28.6) 80/219 (36.5)
 50%–<75% 3/21 (14.3) 53/219 (24.2)
 75%–<100% 6/21 (28.6) 14/219 (6.4)
 Not mentioned 4/21 (19.0) 12/219 (5.5)
 Only reported the matched sample 0 25/219 (11.4)
*

 Cancer studies in top medical and cancer journals in 2014 and 2015. Nine studies were included in both cancer studies and cancer surgical studies.

The reporting of the proportion of matched sample size was evaluated in studies utilizing matching.