The 2844 participants included in the study were 1,465 women (51.5%) and 1,379 men (48.5%), with an average age (SD) of 59 6 ( 11.9) years.Participants with self-reported OD were more likely to be female, low annual family income, former smokers, hypertensive, dyslipidemia, non-diabetic, non-prediabetes and to drink more alcohol.Adults who had an Excellent/ Very Good/ Good Health self-evaluation had significantly.
greater reported frequency of olfactory alteration.In addition, a higher frequency of olfactory alteration has been reported in people with a history of severe loss of consciousness following a tonsillectomy caused by a persistent cold or flu and a serious head or face injury.Short sleep duration patients have higher chance of olfactory dysfunction.The subjects' baseline characteristics are summarized in Table 1.Univariate analysis demonstrated that age, annual family income, smoking, work activity, BMI, self-reported dyslipidemia, tonsillectomy, persistent cold or flu lasted 12 months? loss of consciousness from head injury, broken nose or other serious injury to face or skull were associated with migraines (Table 2).Table 3 presents the association between sleep duration and self-reported OD. The odds ratios (95% confidence intervals) of self-reported OD was different between mid-range and long sleep durations compared to short sleep durations. Adjusted OR values for self-reported ODs and duration of sleep are 0.77 (0.61 ~ 0.99, p = 0.039) and 0.69 (0.53 ~ 0.9, p = 0.006), respectively. This illustrated that OD incidence rose by 7% for every 1 h increase in sleep duration.
Table 1 Baseline characteristics of the study participants
Characteristics
|
Self-Reported Olfactory Dysfunction
|
All participants
(n = 2844)
|
NO
(n = 2169)
|
YES
(n = 675)
|
p
|
Gender n (%)
|
|
|
0.84
|
Male
|
1379 (48.5)
|
1054 (48.6)
|
325 (48.1)
|
|
Female
|
1465 (51.5)
|
1115 (51.4)
|
350 (51.9)
|
|
Age(years), Mean ± SD
|
59.6 ± 11.9
|
59.3 ± 12.0
|
60.5 ± 11.8
|
0.031
|
Race/ethnicity, n (%)
|
|
|
0.02
|
Mexican American
|
224 ( 7.9)
|
168 (7.7)
|
56 (8.3)
|
|
Other Hispanic
|
307 (10.8)
|
232 (10.7)
|
75 (11.1)
|
|
Non-Hispanic White
|
1133 (39.8)
|
835 (38.5)
|
298 (44.1)
|
|
Non-Hispanic Black
|
800 (28.1)
|
624 (28.8)
|
176 (26.1)
|
|
Other Race
|
380 (13.4)
|
310 (14.3)
|
70 (10.4)
|
|
Education,n (%)
|
|
|
0.291
|
below high school
|
732 (25.7)
|
544 (25.1)
|
188 (27.9)
|
|
high school
|
626 (22.0)
|
476 (21.9)
|
150 (22.2)
|
|
some college or above
|
1486 (52.3)
|
1149 (53)
|
337 (49.9)
|
|
Marital status,n (%)
|
|
|
0.223
|
married
|
2422 (85.2)
|
1859 (85.7)
|
563 (83.4)
|
|
living with partner
|
123 ( 4.3)
|
94 (4.3)
|
29 (4.3)
|
|
not married
|
299 (10.5)
|
216 (10)
|
83 (12.3)
|
|
Annual family income (dollar), Median (IQR)
|
7.0 (4.0, 12.0)
|
7.0 (5.0, 13.0)
|
6.0 (4.0, 10.0)
|
< 0.001
|
Smoking status, n (%)
|
|
|
0.003
|
Former smokers
|
2309 (81.2)
|
1787 (82.4)
|
522 (77.3)
|
|
Current smoking
|
535 (18.8)
|
382 (17.6)
|
153 (22.7)
|
|
Alcohol status, n (%)
|
|
|
0.484
|
Yes
|
1996 (70.2)
|
1515 (69.8)
|
481 (71.3)
|
|
No
|
848 (29.8)
|
654 (30.2)
|
194 (28.7)
|
|
Work activity, n (%)
|
|
|
0.138
|
Inactive
|
1881 (66.1)
|
1455 (67.1)
|
426 (63.1)
|
|
Moderate
|
865 (30.4)
|
639 (29.5)
|
226 (33.5)
|
|
Vigorous
|
98 ( 3.4)
|
75 (3.5)
|
23 (3.4)
|
|
Physical activity, n (%)
|
|
|
0.134
|
Inactive
|
1625 (57.1)
|
1217 (56.1)
|
408 (60.4)
|
|
Moderate
|
1104 (38.8)
|
861 (39.7)
|
243 (36)
|
|
Vigorous
|
115 ( 4.0)
|
91 (4.2)
|
24 (3.6)
|
|
BMI (kg/m2 ), Mean ± SD
|
29.5 ± 6.8
|
29.3 ± 6.8
|
29.9 ± 6.9
|
0.049
|
Hypertension, n (%)
|
|
|
0.556
|
No
|
945 (33.2)
|
727 (33.5)
|
218 (32.3)
|
|
Yes
|
1899 (66.8)
|
1442 (66.5)
|
457 (67.7)
|
|
Dyslipidemia, n (%)
|
|
|
< 0.001
|
No
|
1114 (39.2)
|
893 (41.2)
|
221 (32.7)
|
|
Yes
|
1730 (60.8)
|
1276 (58.8)
|
454 (67.3)
|
|
Diabetes, n (%)
|
|
|
0.232
|
No
|
2135 (75.1)
|
1640 (75.6)
|
495 (73.3)
|
|
Yes
|
709 (24.9)
|
529 (24.4)
|
180 (26.7)
|
|
Prediabetes, n (%)
|
|
|
0.783
|
No
|
2558 (89.9)
|
1949 (89.9)
|
609 (90.2)
|
|
Yes
|
286 (10.1)
|
220 (10.1)
|
66 (9.8)
|
|
Self-reported health condition , n (%)
|
|
|
< 0.001
|
Excellent, very good, good
|
2094 (73.6)
|
1654 (76.3)
|
440 (65.2)
|
|
Fair, Poor
|
750 (26.4)
|
515 (23.7)
|
235 (34.8)
|
|
Cold/flu for >1 month, n (%)
|
|
|
< 0.001
|
Yes
|
179 ( 6.3)
|
109 (5)
|
70 (10.4)
|
|
No
|
2665 (93.7)
|
2060 (95)
|
605 (89.6)
|
|
Tonsils removed, n (%)
|
|
|
0.031
|
Yes
|
703 (24.7)
|
515 (23.7)
|
188 (27.9)
|
|
No
|
2141 (75.3)
|
1654 (76.3)
|
487 (72.1)
|
|
Loss of consciousness from head injury, n (%)
|
|
|
< 0.001
|
Yes
|
361 (12.7)
|
238 (11)
|
123 (18.2)
|
|
No
|
2483 (87.3)
|
1931 (89)
|
552 (81.8)
|
|
Serious head/face injury, n (%)
|
|
|
< 0.001
|
Yes
|
418 (14.7)
|
285 (13.1)
|
133 (19.7)
|
|
No
|
2426 (85.3)
|
1884 (86.9)
|
542 (80.3)
|
|
Sleep duration
(hours), n (%)
|
|
|
0.002
|
< 6
|
487 (17.1)
|
342 (15.8)
|
145(21.5)
|
|
6 - 8
|
1389 (48.8)
|
1071(49.4)
|
318 (47.1)
|
|
≥ 8
|
968( 34.0)
|
756(34.9)
|
212(31.4)
|
|
Abbreviations: BMI, body mass index.
Table 2
Unlvariate analysis for smell dysfunction.
Variable
|
OR_95CI
|
P_value
|
Gender
|
|
|
Male
|
Reference
|
|
Female
|
1.02 (0.86 ~ 1.21)
|
0.84
|
Age
|
1.01 (1 ~ 1.02)
|
0.031
|
Race/ethnicity
|
|
|
Mexican American
|
Reference
|
|
Other Hispanic
|
0.97 (0.65 ~ 1.45)
|
0.88
|
Non-Hispanic White
|
1.07 (0.77 ~ 1.49)
|
0.685
|
Non-Hispanic Black
|
0.85 (0.6 ~ 1.2)
|
0.343
|
Other Race
|
0.68 (0.45 ~ 1.01)
|
0.055
|
Education
|
|
|
below high school
|
Reference
|
|
high school
|
0.91 (0.71 ~ 1.17)
|
0.465
|
some college or above
|
0.85 (0.69 ~ 1.04)
|
0.118
|
Marital status
|
|
|
married
|
Reference
|
|
living with partner
|
1.02 (0.66 ~ 1.56)
|
0.932
|
not married
|
1.27 (0.97 ~ 1.66)
|
0.084
|
Annual family income
|
0.96 (0.94 ~ 0.98)
|
< 0.001
|
Smoking status
|
|
|
Former smokers
|
Reference
|
|
Current smoking
|
1.37 (1.11 ~ 1.69)
|
0.003
|
Alcohol status
|
|
|
Yes
|
Reference
|
|
No
|
0.93 (0.77 ~ 1.13)
|
0.484
|
Work activity
|
|
|
Inactive
|
Reference
|
|
Moderate
|
1.21 (1 ~ 1.46)
|
0.047
|
Vigorous
|
1.05 (0.65 ~ 1.69)
|
0.85
|
Physical activity
|
|
|
Inactive
|
Reference
|
|
Moderate
|
0.84 (0.7 ~ 1.01)
|
0.063
|
Vigorous
|
0.79 (0.49 ~ 1.25)
|
0.31
|
BMI
|
1.01 (1 ~ 1.03)
|
0.049
|
Hypertension
|
|
|
No
|
Reference
|
|
Yes
|
1.06 (0.88 ~ 1.27)
|
0.556
|
Dyslipidemia
|
|
|
No
|
Reference
|
|
Yes
|
1.44 (1.2 ~ 1.72)
|
< 0.001
|
Diabetes
|
|
|
No
|
Reference
|
|
Yes
|
1.13 (0.93 ~ 1.37)
|
0.232
|
Prediabetes
|
|
|
No
|
Reference
|
|
Yes
|
0.96 (0.72 ~ 1.28)
|
0.783
|
Self-reported health condition
|
|
|
Excellent, very good, good
|
Reference
|
|
Fair, Poor
|
1.72 (1.42 ~ 2.07)
|
< 0.001
|
Cold/flu for > 1 month
|
|
|
Yes
|
Reference
|
|
No
|
0.46 (0.33 ~ 0.63)
|
< 0.001
|
Tonsils removed
|
|
|
Yes
|
Reference
|
|
No
|
0.81 (0.66 ~ 0.98)
|
0.031
|
Head injury
|
|
|
Yes
|
Reference
|
|
No
|
0.55 (0.44 ~ 0.7)
|
< 0.001
|
Serious head/face injury
|
|
|
Yes
|
Reference
|
|
No
|
0.62 (0.49 ~ 0.77)
|
< 0.001
|
Sleep duration
|
0.92(0.87 ~ 0.98)
|
0.007
|
Table 3
Association between sleep duration and olfactory dysfunction in multiple regression model.
|
Crude
|
Model I
|
Model II
|
Model III
|
OR(95%CI)
|
P value
|
OR(95%CI)
|
P value
|
OR(95%CI)
|
P value
|
OR(95%CI)
|
P value
|
Sleep duration
|
0.92 (0.87 ~ 0.98)
|
0.007
|
0.91 (0.86 ~ 0.97)
|
0.003
|
0.92 (0.86 ~ 0.98)
|
0.009
|
0.93 (0.87 ~ 0.99)
|
0.023
|
Hours of Sleep
|
|
|
|
|
|
|
|
|
< 6
|
Ref
|
|
Ref
|
|
Ref
|
|
Ref
|
|
6 ~ 8
|
0.7 (0.56 ~ 0.88)
|
0.003
|
0.72(0.57 ~ 0.91)
|
0.006
|
0.74 (0.59 ~ 0.95)
|
0.016
|
0.77 (0.61 ~ 0.99)
|
0.039
|
≥ 8
|
0.66(0.52 ~ 0.85)
|
0.001
|
0.64(0.49 ~ 0.82)
|
0.001
|
0.66(0.51 ~ 0.85)
|
0.001
|
0.69 (0.53 ~ 0.9)
|
0.006
|
Trend test
|
|
0.003
|
|
0.001
|
|
0.002
|
|
0.008
|
Model I:adjusted for gender,age,race/ethnicity,education,marital status,annual family income,smoking status,alcohol status,work activity,physical activity and BMI;
Model II:adjusted for model I plus hypertension,dyslipidemia,diabetes,prediabetes and self-reported health condition;
Model III:adjusted for model II plus cold/flu for >1 month,tonsils removed,head injury and serious head/face injury.
Fig.2 presents a non-linear association between self-reported OD and sleep duration after adjusting for potential confounding factors.In our study, the P-value for the non-linear test was less than 0.028 (Table 4).We found an inflection point at about 7 hours.On the left side of the inflection point, the OR was 0.89 (OR: 0.89, 95%CI: 0.801–0.996, p < 0.028) .On the right side of the inflection point, the OR was 1. 19 (OR:1. 19, 95%CI: 0.97– 1.455, p = 0.096).It suggests that the risk of self-reported OD is reduced by 0. 11% with every 1 hour increase in everyday.Then longer than 7 hours is not significant between OD and sleep duration.This means that the risk of OD no longer decreases with increasing sleep hours.
Table 4
Threshold effect analysis of sleep duration on olfactory dysfunction.
Outcome
|
OR (95% CI)
|
P value
|
One - line linear regression model
|
0.93 (0.87 ~ 0.99)
|
0.023
|
Two - piecewise linear regression model
|
|
|
Sleep duration < 7 hours
|
0.89(0.801 ~ 0.996)
|
0.042
|
Sleep duration ≥ 7 hours
|
1.19 (0.97 ~ 1.455)
|
0.096
|
Log - likelihood ratio test
|
< 0.028
|
|
Notes:adjusted for gender,age,race/ethnicity,education,marital status,annual family income,smoking status,alcohol status,work activity,physical activity,BMI,hypertension,dyslipidemia,diabetes,prediabetes,self-reported health condition,cold/flu for >1 month,tonsils removed,head injury and serious head/face injury.
In the Fig.2, the solid line indicates the estimated risk of self-reported OD, and the dotted lines represent point-wise 95% confidence interval adjusted for gender, age, race/ethnicity, education, marital status, annual family income, smoking status, alcohol status, work activity, physical activity, BMI, hypertension, dyslipidemia, diabetes, prediabetes, self-reported health condition, cold/flu for > 1 month, tonsils removed, loss of consciousness from head injury;broken nose or other serious injury to face or skull.