Hematologic variables associated with HMIP-2 genotype
| Variable . | Median (interquartile range) . | Individuals studied, no. . | Significant* confounders . | HMIP-2 effect (β) of allele 2 (P) . | ΔVARHMIP-2, % . |
|---|---|---|---|---|---|
| RBC | 4.5 × 1012/L (4.27-4.74) | 1789 | Sex | −0.07424 (<.001) | 0.6 |
| MCV | 91.7 fl (88.7-95) | 1761 | Age | 0.9202 (<.001) | 1.7 |
| MCH | 29.9 pg (28.8-30.9) | 1724 | Age, sex, dataset | 15.5833 (<.001) | 2.1 |
| WBC | 5.8 × 109/L (4.8-7.1) | 1788 | Dataset | −0.02279 (.040)† | 0.8 |
| Plt | 232 × 109/L (193-274) | 1784 | Sex, dataset | 0.03908 (<.001) | 0.6 |
| Mon | 0.33 × 109/L (0.26-0.41) | 1416 | Age, dataset | −0.04163 (.004) | 1.6 |
| Variable . | Median (interquartile range) . | Individuals studied, no. . | Significant* confounders . | HMIP-2 effect (β) of allele 2 (P) . | ΔVARHMIP-2, % . |
|---|---|---|---|---|---|
| RBC | 4.5 × 1012/L (4.27-4.74) | 1789 | Sex | −0.07424 (<.001) | 0.6 |
| MCV | 91.7 fl (88.7-95) | 1761 | Age | 0.9202 (<.001) | 1.7 |
| MCH | 29.9 pg (28.8-30.9) | 1724 | Age, sex, dataset | 15.5833 (<.001) | 2.1 |
| WBC | 5.8 × 109/L (4.8-7.1) | 1788 | Dataset | −0.02279 (.040)† | 0.8 |
| Plt | 232 × 109/L (193-274) | 1784 | Sex, dataset | 0.03908 (<.001) | 0.6 |
| Mon | 0.33 × 109/L (0.26-0.41) | 1416 | Age, dataset | −0.04163 (.004) | 1.6 |
The β parameter indicates the strength and direction of HMIP-2 haplotype 2. No association was seen with traits hemoglobin concentration (Hb), packed red cell volume (PCV), mean red cell hemoglobin concentration (MCHC), red cell distribution width (RDW), platelet distribution width (PDW), neutrophil, lymphocyte, eosinophil, and basophil counts.
ΔVARHMIP-2 indicates decrease in trait variance after fitting the HMIP-2effect; RBC, red blood cell count; MCV, mean red cell volume; MCH, mean red cell hemoglobin; WBC, white blood cell count; Plt, platelet count; and Mon, monocyte count.
Shown here when P < .05. Irrespective of their significance, the effects of all 3 potential confounders (age, sex, and dataset) were considered in the regression model.
White cell counts are not significantly associated when considering the multiple statistical testing implications of examining 6 traits.