Skip to main content

Table 3 Interpretation characteristics of statistical analysis in title or abstract of the included article

From: Reporting of abstracts in studies that used routinely collected data for exploring drug treatment effects: a cross-sectional survey

Reporting item

Total

journal rank

(n = 222)

Top 5 general medicine (n = 19)

Other journals (n = 203)

Predefined hypothesis a, n (%)

44 (19.8)

6 (31.6)

38 (18.7)

Conclusion in a framed manner, n (%)

 Population

176 (79.3)

13 (68.4)

163 (80.3)

 Exposure

206 (92.8)

16 (84.2)

190 (93.6)

 Comparator

100 (45.1)

10 (52.6)

90 (44.3)

 Outcomes

203 (91.4)

19 (100.0)

184 (90.6)

Direction of the treatment effect

 Positive

170 (76.6)

17 (89.5)

153 (75.4)

 Negative

61 (27.5)

4 (21.1)

57 (28.1)

Consistency with predefined hypothesis

 Consistent

29 (67.4)

3 (50.0)

26 (70.3)

 Not consistent

8 (18.6)

1 (16.7)

7 (18.9)

 Partly consistent

6 (14.0)

2 (33.3)

4 (10.8)

Claim of the treatment effect

 Strongb

39 (17.6)

4 (21.1)

35 (17.2)

 Moderatec

143 (64.4)

11 (57.9)

132 (65.0)

 Weakd

40 (18.0)

4 (21.1)

36 (17.7)

  1. a Specific wording that indicates the direction of risk between exposure and outcome (e.g., increase, decrease) will be considered a predefined hypothesis
  2. b Authors convey a conviction that the treatment effect truly exists
  3. c Authors convey a belief that the treatment effect possibly exists
  4. d Authors suggest a treatment effect but convey uncertainty about whether such an effect exists