AbstractBackground and ObjectivesSeveral studies have suggested that blood donors have lower risk of gastrointestinal and breast cancers, whereas some have indicated an increased risk of haematological cancers. We examined these associations by appropriately adjusting the ‘healthy donor effect’ (HDE).Materials and MethodsWe examined the risk of gastrointestinal/colorectal, breast and haematological cancers in regular high‐frequency whole blood (WB) donors using the Sax Institute's 45 and Up Study data linked with blood donation and other health‐related data. We calculated 5‐year cancer risks, risk differences and risk ratios. To mitigate HDE, we used 5‐year qualification period to select the exposure groups, and applied statistical adjustments using inverse probability weighting, along with other advanced doubly robust g‐methods.ResultsWe identified 2867 (42.4%) as regular high‐frequency and 3888 (57.6%) as low‐frequency donors. The inverse probability weighted 5‐year risk difference between high and low‐frequency donors for gastrointestinal/colorectal cancer was 0.2% (95% CI, −0.1% to 0.5%) with a risk ratio of 1.25 (0.83–1.68). For breast cancer, the risk difference was −0.2% (−0.9% to 0.4%), with a risk ratio of 0.87 (0.48–1.26). Regarding haematological cancers, the risk difference was 0.0% (−0.3% to 0.5%) with a risk ratio of 0.97 (0.55–1.40). Our doubly robust estimators targeted minimum loss‐based estimator (TMLE) and sequentially doubly robust (SDR) estimator, yielded similar results, but none of the findings were statistically significant.ConclusionAfter applying methods to mitigate the HDE, we did not find any statistically significant differences in the risk of gastrointestinal/colorectal, breast and haematological cancers between regular high‐frequency and low‐frequency WB donors.
Read full abstract