Except for cannabis related diagnoses, their effect size (change in the incidence rate ratio ∆IRR) is of a similar magnitude as opioid-related diagnoses which they included for falsification testing and said it did not change.
Am I misinterpreting their statistics?
> Cannabis-related diagnoses (∆IRR 1.138, 95% CI 1.066 to 1.214, p<0.001),
> AF (∆IRR 1.023, 95% CI 1.010 to 1.035, p<0.001),
> HF (∆IRR 1.037, 95% CI 1.029 to 1.046, p<0.001),
> MI (∆IRR 1.031, 95% CI 1.014 to 1.048, p<0.001),
> ischaemic stroke (∆IRR 1.011, 95% CI 0.999 to 1.024, p=0.078),
> VTE (∆IRR 1.009, 95% CI 0.999 to 1.020, p=0.066)
> and VT (∆IRR 1.019, 95% CI 0.993 to 1.046, p=0.161).
> For falsification testing, there was no change in hospitalisations for
> urinary stones (∆IRR 1.006, 95% CI 0.986 to 1.027, p=0.535) or
> opioid-related diagnoses (∆IRR 0.966, 95% CI 0.927 to 1.006, p=0.091).
Am I misinterpreting their statistics?
> Cannabis-related diagnoses (∆IRR 1.138, 95% CI 1.066 to 1.214, p<0.001),
> AF (∆IRR 1.023, 95% CI 1.010 to 1.035, p<0.001),
> HF (∆IRR 1.037, 95% CI 1.029 to 1.046, p<0.001),
> MI (∆IRR 1.031, 95% CI 1.014 to 1.048, p<0.001),
> ischaemic stroke (∆IRR 1.011, 95% CI 0.999 to 1.024, p=0.078),
> VTE (∆IRR 1.009, 95% CI 0.999 to 1.020, p=0.066)
> and VT (∆IRR 1.019, 95% CI 0.993 to 1.046, p=0.161).
> For falsification testing, there was no change in hospitalisations for
> urinary stones (∆IRR 1.006, 95% CI 0.986 to 1.027, p=0.535) or
> opioid-related diagnoses (∆IRR 0.966, 95% CI 0.927 to 1.006, p=0.091).