Table 2. Subgroup Analysis of Observational Studies and Randomized Controlled Trials Stratified by Initial Source of Investigation.
| Characteristic | Observational Studies (N = 91) | Randomized Controlled Trials (N = 39) | ||||
| Newspapers (N = 56) | Medical Journals (N = 35) | P-value | Newspapers (N = 13) | Medical Journals (N = 26) | P-value | |
| Journal impact factor* | 5.4 (3.9–14.5) | 16.7 (16.7–30.0) | <0.001 | 11.5 (4.7–53.3) | 33.8 (30.0–53.3) | 0.09 |
| Participants, n* | 1984 (173–57491) | 21136 (1655–264758) | 0.029 | 420 (84–1020) | 568 (312–1723) | 0.30 |
| Pharmaceutical funding, n (%) | 3 (6) | 1 (3) | 0.64 | 2 (15) | 10 (38) | 0.27 |
| Studies assessing mortality, n (%) | 13 (23) | 9 (26) | 0.79 | 3 (23) | 8 (31) | 0.72 |
| Cross-sectional studies, n (%) | 40 (71) | 11 (31) | <0.001 | 2 (15) | 0 (0) | 0.11 |
| Length of follow-up in longitudinal studies, y* | 5.50 (2.50–10.13) | 3.50 (0.08–10.10) | 0.33 | 0.42 (0.23–1.00) | 1.00 (0.25–1.00) | 0.36 |
Presented as median (25th–75th percentile) because data was not normally distributed.