Skip to main content

Table 6 Observed and expected risk ratios of having a total joint replacement associated with smoking: one-way sensitivity analysis accounting for uncertainty in the classification of smoking exposure

From: Smoking and primary total hip or knee replacement due to osteoarthritis in 54,288 elderly men and women

  Level of uncertainty! Sensitivity!! Risk ratio (95% CI) P value
Observed risk ratio - - 0.62 (0.54 – 0.72) <0.001
Expected risk ratios based on one-way sensitivity analyses!!! 1% .99 0.66 (0.58 – 0.75) <0.001
  5% .95 0.76 (0.68 – 0.84) <0.001
  10% .90 0.82 (0.75 – 0.90) <0.001
  20% .80 0.89 (0.82 – 0.95) 0.001
  25% .75 0.90 (0.84 – 0.97) 0.004
  30% .70 0.92 (0.86 – 0.98) 0.009
  40% .60 0.93 (0.88 – 0.99) 0.032
  50% .50 0.94 (0.89 – 1.01) 0.073
  1. !Uncertainty in the classification of smoking. The values represent hypothetical percentages of study participants potentially declining to reveal that they indeed did smoke.
  2. !!Sensitivity is defined as the probability of a true smoker being classified as a smoker.
  3. !!!The one way sensitivity analysis evaluated the risk ratio under various sensitivity measures while holding the specificity fixed.