Table 4.
Associations between conflicts of interests (COIs) of authors of abstracts of randomized controlled trials (RCTs) presented at ARVO conferences during the years 2001–2004 and likelihood of publication of the RCTs, overall model (model 1), and interaction model, stratified by statistical significance of results for the main outcome (model 2)
| Author COIs | Model 1** | Model 2*** | |||
|---|---|---|---|---|---|
| Relative risks (RR) (95 % CI) | RR (95 % CI) among abstracts with main outcome not statistically significant | RR (95 % CI) among abstracts with main outcome statistically significant | RR (95 % CI) among abstracts with statistical significance of main outcome not reported | P value of F test of interaction **** | |
| COIs of first author (years 2002–2004 only) | N = 400 | N = 87 | N = 79 | N = 234 | |
| At least one COI | 1.31 (1.04–1.64) | 1.13 (0.75–1.72) | 1.22 (0.82–1.81) | 0.93 (0.63–1.37) | 0.73 |
| Financial support* | 1.50 (1.19–1.90) | 1.44 (0.95–2.17) | 1.37 (0.92–2.06) | 1.11 (0.74–1.67) | 0.75 |
| Personal finance interest* | 1.17 (0.44–3.14) | – | – | – | - |
| Employee of business with interest* | 0.54 (0.23–1.28) | 0.64 (0.20–2.03) | 0.96 (0.04–3.92) | 0.21 (0.03–1.38) | 0.64 |
| Consultant of business with interest* | 1.10 (0.78–1.55) | 0.87 (0.41–1.86) | 1.08 (0.58–2.02) | 0.86 (0.52–1.40) | 0.69 |
| Inventor/developer with patent* | 0.88 (0.36–2.16) | – | – | – | - |
| Received gifts within the past year* | 1.42 (1.05–1.92) | 1.47 (0.87–2.46) | 1.49 (0.97–2.28) | 0.91 (0.52–1.60) | 0.92 |
| COIs of last author (years 2002–2004 only) | N = 400 | N = 87 | N = 79 | N = 234 | |
| At least one COI | 1.06 (0.82–1.37) | 0.93 (0.59–1.47) | 1.06 (0.67–1.70) | 0.75 (0.50–1.14) | 0.85 |
| Financial support* | 0.84 (0.53–1.34) | 0.73 (0.36–1.50) | 0.81 (0.34–1.97) | 0.58 (0.25–1.36) | 0.96 |
| Personal finance interest* | – | – | – | – | - |
| Employee of business with interest* | 1.06 (0.77–1.47) | 0.89 (0.45–1.75) | 1.08 (0.58–2.02) | 0.82 (0.51–1.31) | 0.76 |
| Consultant of business with interest* | 1.32 (0.86–2.03) | 1.73 (1.13–2.64) | 1.56 (0.85–2.87) | 0.52 (0.15–1.76) | 0.36 |
| Inventor/developer with patent* | 0.94 (0.32–2.75) | – | – | – | - |
| Received gifts within the past year* | 1.18 (0.69–2.01) | 1.52 (0.83–2.78) | 1.01 (0.25–4.12) | 0.76 (0.30–1.91) | 0.70 |
| Not reported | 1.18 (0.66–2.10) | – | – | – | - |
| Not applicable (abstract had only one author) | 1.00 (0.54–1.86) | 1.64 (1.00–2.68) | 1.28 (0.72–2.29) | 1.03 (0.50–2.12) | 0.68 |
| COIs of any author (years 2001–2004) | N = 513 | N = 111 | N = 117 | N = 285 | |
| At least one COI | 1.13 (0.93–1.38) | 1.01 (0.71–1.45) | 1.09 (0.77–1.56) | 0.81 (0.59–1.11) | 0.72 |
| Financial support* | 1.16 (0.93–1.45) | 1.02 (0.68–1.53) | 1.18 (0.82–1.71) | 0.81 (0.55–1.18) | 0.83 |
| Personal finance interest* | 0.74 (0.24–2.31) | – | – | – | - |
| Employee of business with interest* | 1.07 (0.82–1.38) | 0.99 (0.62–1.58) | 1.13 (0.66–1.94) | 0.80 (0.55–1.17) | 0.87 |
| Consultant of business with interest* | 1.18 (0.91–1.52) | 0.94 (0.52–1.69) | 1.25 (0.86–1.81) | 0.83 (0.55–1.25) | 0.68 |
| Inventor/developer with patent* | 0.86 (0.43–1.71) | – | – | – | - |
| Received gifts within the past year* | 1.35 (1.05–1.72) | 1.53 (1.03–2.27) | 1.36 (0.90–2.05) | 0.95 (0.64–1.42) | 0.85 |
* More than one option could apply to each abstract
** Model 1 – Overall model
*** Model 2 – Interaction model – Results are stratified by whether the main outcome results were not statistically, statistically significant, or not reported
**** F-test of interaction tests the overall statistical significance of the interaction between COI and all interaction terms in model 2. We added two interaction terms in model 2: (1) whether or not results for the main outcome in the abstract were statistically significant and (2) whether or not statistical significance of results for the main outcome was reported in the abstract
Data (RRs and 95 % CIs) reported in bold text are statistically significant at the 5 % level