Table 3.
Associations of visit 5 characteristics with visit 5 anosmia prevalence and visit 6 anosmia risk.
| Visit 5 Anosmia | Visit 6 Anosmia | |
|---|---|---|
| Characteristics | Prevalence Ratio (95% CI) a |
Incidence Risk Ratio (95% CI) b |
| Age (per 10-year higher) c | 1.965 (1.740, 2.219) | 2.124 (1.701, 2.651) |
| Male c | 1.689 (1.470, 1.941) | 1.405 (1.119, 1.764) |
| Education c | ||
| >High School | Referent | Referent |
| High School | 1.014 (0.865, 1.189) | 1.479 (1.159, 1.888) |
| <High School | 1.302 (1.090, 1.555) | 1.531 (1.099, 2.134) |
| Study Site–Race c | ||
| Washington–White | Referent | Referent |
| Forsyth–Black | 1.719 (1.007, 2.933) | 2.296 (1.117, 4.719) |
| Forsyth–White | 0.893 (0.706, 1.129) | 1.458 (1.050, 2.024) |
| Jackson–Black | 2.354 (1.978, 2.802) | 2.506 (1.824, 3.445) |
| Minneapolis–White | 0.903 (0.730, 1.117) | 1.023 (0.732, 1.430) |
| Cigarette Smoking c | ||
| Never Smoker | Referent | Referent |
| Former Smoker | 0.979 (0.844, 1.135) | 1.172 (0.919, 1.494) |
| Current Smoker | 0.895 (0.654, 1.225) | 1.448 (0.922, 2.275) |
| Alcohol Consumption c | ||
| Never Drinker | Referent | Referent |
| Former Drinker | 1.013 (0.844, 1.216) | 1.167 (0.859, 1.585) |
| Current Drinker | 0.940 (0.769, 1.150) | 0.881 (0.646, 1.201) |
| APOE ε4 (any ε4 allele presence) d | 1.254 (1.088, 1.446) | 1.330 (1.053, 1.680) |
| BMI (kg/m2, per 1SD) e | 0.922 (0.848, 1.002) | 1.021 (0.908, 1.147) |
| Obesity Status (BMI (kg/m2)) c | ||
| <25 | Referent | Referent |
| 25–<30 | 0.865 (0.733, 1.019) | 1.131 (0.840, 1.523) |
| ≥30 | 0.752 (0.624, 0.905) | 1.215 (0.888, 1.663) |
| Physical Activity (Total MET-Min/week, per 1SD) e | 0.893 (0.821, 0.973) | 0.941 (0.838, 1.056) |
| Serum Total Cholesterol (mg/dL, per 1SD) f | 0.929 (0.855, 1.009) | 0.939 (0.831, 1.061) |
| Diabetes f | 1.163 (0.999, 1.355) | 1.037 (0.811, 1.327) |
| Hypertension f | 0.830 (0.701, 0.983) | 1.050 (0.800, 1.378) |
| Myocardial Infarction History g | 0.959 (0.739, 1.245) | 1.550 (1.000, 2.403) |
| CHD History g | 1.177 (0.956, 1.451) | 0.785 (0.531, 1.160) |
| Stroke History g | 1.019 (0.755, 1.375) | 0.477 (0.194, 1.174) |
| C-Reactive Protein (mg/L, per 1SD) h | 0.990 (0.915, 1.070) | 1.036 (0.878, 1.222) |
| Serum Vitamin B12 (pg/mL, per 1SD) h | 0.976 (0.886, 1.074) | 0.706 (0.510, 0.977) |
| Serum Vitamin B12 Categories h | ||
| 1st Quartile | Referent | Referent |
| 2nd Quartile | 0.845 (0.647, 1.104) | 1.054 (0.646, 1.721) |
| 3rd Quartile | 1.027 (0.794, 1.328) | 1.049 (0.618, 1.781) |
| 4th Quartile | 1.014 (0.779, 1.319) | 0.650 (0.326, 1.297) |
| Blood Hemoglobin (g/dL, per 1SD) h | 0.962 (0.896, 1.032) | 0.961 (0.873, 1.058) |
| Anemia h | 1.109 (0.940, 1.308) | 0.939 (0.704, 1.252) |
Abbreviations: BMI, body mass index; CHD, coronary heart disease; SD, standard deviation. One SD unit: BMI: 5.7 kg/m2; physical activity: 838 total MET-Min/week; serum total cholesterol: 42 mg/dL; C-reactive protein: 8.2 mg/L; serum vitamin B12: 532 pg/mL; blood hemoglobin: 1.5 g/dL. a: Obtained from regression models examining cross-sectional associations between visit 5 factors and anosmia prevalence among 6053 visit 5 participants. b: Obtained from regression models examining the associations between visit 5 factors and risk of anosmia among participants who attended both visits 5 and 6. c: Obtained from Model 1: age, sex, site–race, education, cigarette smoking, and alcohol consumption. d: Model 2: adjusted for covariates in Model 1. e: Obtained from Model 3: Model 1 covariates, BMI, and physical activity. f: Obtained from Model 4: Model 3 covariates, total cholesterol, diabetes, and hypertension. g: Obtained from Model 5: Model 4 covariates, myocardial infarction history, CHD, and stroke. h: Model 6: adjusted for covariates in Model 4. Note: we repeated Model 1 for incident anosmia analysis by incorporating the inverse probability of attrition weights to correct for potential bias due to selective attrition from visit 5 to visit 6. The results were similar to those without incorporating the inverse probability weights.