Skip to main content

Table 4 study characteristics in pharmacological cohort studies and in non pharmacological cohort studies

From: Reporting of heterogeneity of treatment effect in cohort studies: a review of the literature

Study characteristics

Pharmacological cohort studies (n = 89)

Non pharmacological cohort studies (n = 61)

p-value*

Sample size, Median(Interquartile range)

3434(490–20,482)

1041(151–2714)

<0.001

Length of follow up (years), Median(Interquartile range)

4(1.7–8.3)

3(1–6)

0.14

Date of publication

  

0.67

 2000–2013

44(49)

27(45)

 

 2014–2016

45(51)

34(55)

 

Journal type, n(%)

  

0.74

 High impact factor journals (HIJ)

45(51.1)

29(48)

 

 Low impact factor journals (LIJ)

44(48.9)

32(52)

 

Study area, n(%)

   

 Non pharmacological

61(100)

 

 Pharmacological

89(100)

 

Source of funding, n(%)

  

0.04

 Industry

61(68)

39(63)

 

 Other

15(17)

13(22)

 

 NA

13(15)

9(15)

 

Type of primary outcome, n(%)

  

0.37

 Time to event

42(47)

24(39)

 

 Binary

30(33)

19(31)

 

 Continuous

17(20)

18(30)

 

Main effect of primary outcome, n(%)

  

0.55

 Statistically significant

53(60)

42(68)

 

 Not statistically significant

36(40)

19(32)

 

Newcastle-Ottawa Quality Assessment Score, n(%)

  

0.5

 0–3

3(4)

2(3)

 

 4–6

52(58)

36(59)

 

 7–8

34(38)

23(38)

 

Analysis of predictive factors on side effects, n(%)

1(2)

1(2)

0.49

HTE reporting

63(71)

24(39)

<0.001

Prespecification of HTE, n(%)

59(66)

24(100)

0.59

Type of HTE analysis, n(%)

  

0.40

 Subgroup

2(3)

0(0)

 

 Adjusted

50(79)

16(67)

 

 Propension

11(18)

8(33)

 

Predictive variables studied, n(%)

   

 Age

19(30)

10(42)

0.19

 Sexe

11(17)

4(17)

1

 Social level

2(3)

2(8)

0.33

 Genetics

9(14)

3(12)

1

 Treatment

53(84)

13(54)

0.002

 Comorbidities

19(30)

11(46)

0.13

 Severity of disease

9(14)

3(12)

0.18

Prognostic variables studied, n(%)

30(48)

21(34)

0.77

Analysis of prognostic factors on side effect, n(%)

1(2)

1(2)

0.52

Interaction test, n(%)

20(22)

7(11)

0.27

  1. *to determinate p-value we used the Wilcoxon rank sum test for the analysis of continuous data and the chi square or fisher test for binary data