Study Population
Out of the 15,560 participants who completed the interviews, 6,328 were excluded because they were under 20 years old. Therefore, 9,232 participants aged 20 and older were included. Further data screening excluded 1,794 participants for the following reasons: 22 due to missing gallstone data, 8 due to missing trouble sleeping data, and 1,764 due to missing other covariate data. Ultimately, 7,438 participants were included in the final analysis (Figure 1).
Figure 1 Study flow chart.
Baseline Characteristics
A total of 7,438 adult participants were included, with 6,658 not having gallstones (no gallstones group) and 780 having gallstones (gallstones group). The average age of the overall study population was 50.6 years (SD 17.5 years), with 51.9% being female. Table 1 shows the demographic characteristics of the two groups. Compared to the no gallstones group, participants in the gallstones group were older on average (58.0 years vs. 49.7 years, P < 0.001), had a higher proportion of females (71.7% vs. 49.6%, P < 0.001), and had higher BMI (mean 33.3 vs. 29.6, P < 0.001). The proportion of participants with diabetes and hypertension was also higher in the gallstones group (26.7% vs. 14.4% and 54% vs. 35.6%, respectively, P < 0.001).
Table1 Baseline characteristics of participants in the NHANES 2017–2020 cycle.
Variables
|
Total (n = 7438)
|
No Gallstone (n = 6658)
|
Gallstone (n = 780)
|
P-value
|
statistic
|
Sex, n (%)
|
|
|
|
< 0.001
|
136.077
|
Male
|
3576 (48.1)
|
3355 (50.4)
|
221 (28.3)
|
|
|
Female
|
3862 (51.9)
|
3303 (49.6)
|
559 (71.7)
|
|
|
Age (years)
|
50.6 ± 17.5
|
49.7 ± 17.5
|
58.0 ± 15.8
|
< 0.001
|
161.216
|
Race, n (%)
|
|
|
|
< 0.001
|
31.813
|
Non-Hispanic white
|
2650 (35.6)
|
2315 (34.8)
|
335 (42.9)
|
|
|
Non-Hispanic black
|
1872 (25.2)
|
1726 (25.9)
|
146 (18.7)
|
|
|
Mexican American
|
911 (12.2)
|
802 (12)
|
109 (14)
|
|
|
Others
|
2005 (27.0)
|
1815 (27.3)
|
190 (24.4)
|
|
|
Education (years), n (%)
|
|
|
|
0.079
|
5.065
|
<9
|
558 ( 7.5)
|
513 (7.7)
|
45 (5.8)
|
|
|
9-12
|
2602 (35.0)
|
2310 (34.7)
|
292 (37.4)
|
|
|
>12
|
4278 (57.5)
|
3835 (57.6)
|
443 (56.8)
|
|
|
Marital status, n (%)
|
|
|
|
0.174
|
1.845
|
Married or living with a partner
|
4361 (58.6)
|
3886 (58.4)
|
475 (60.9)
|
|
|
Living alone
|
3077 (41.4)
|
2772 (41.6)
|
305 (39.1)
|
|
|
BMI (kg/m2)
|
30.0 ± 7.4
|
29.6 ± 7.2
|
33.3 ± 8.5
|
< 0.001
|
175.466
|
Cholesterol (mg/dL)
|
4.8 ± 1.1
|
4.8 ± 1.1
|
4.8 ± 1.1
|
0.333
|
0.938
|
Diabetes, n (%)
|
|
|
|
< 0.001
|
80.126
|
No
|
6274 (84.4)
|
5702 (85.6)
|
572 (73.3)
|
|
|
Yes
|
1164 (15.6)
|
956 (14.4)
|
208 (26.7)
|
|
|
Hypertension, n (%)
|
|
|
|
< 0.001
|
100.413
|
No
|
4646 (62.5)
|
4287 (64.4)
|
359 (46)
|
|
|
Yes
|
2792 (37.5)
|
2371 (35.6)
|
421 (54)
|
|
|
Smoking status, n (%)
|
|
|
|
0.004
|
8.075
|
No
|
4339 (58.3)
|
3921 (58.9)
|
418 (53.6)
|
|
|
Yes
|
3099 (41.7)
|
2737 (41.1)
|
362 (46.4)
|
|
|
Drinking status, n (%)
|
|
|
|
< 0.001
|
36.387
|
No
|
4127 (55.5)
|
3615 (54.3)
|
512 (65.6)
|
|
|
Yes
|
3311 (44.5)
|
3043 (45.7)
|
268 (34.4)
|
|
|
LTSB, n (%)
|
|
|
|
0.194
|
1.686
|
No
|
6288 (84.5)
|
5641 (84.7)
|
647 (82.9)
|
|
|
Yes
|
1150 (15.5)
|
1017 (15.3)
|
133 (17.1)
|
|
|
Trouble sleeping, n (%)
|
|
|
|
< 0.001
|
92.876
|
No
|
5312 (71.4)
|
4870 (73.1)
|
442 (56.7)
|
|
|
Yes
|
2126 (28.6)
|
1788 (26.9)
|
338 (43.3)
|
|
|
Sleep duration, n (%)
|
|
|
|
0.953
|
0.096
|
≤ 6 h
|
1432 (19.3)
|
1281 (19.2)
|
151 (19.4)
|
|
|
>6 and <9 h
|
4314 (58.0)
|
3859 (58)
|
455 (58.3)
|
|
|
≥9 h
|
1692 (22.7)
|
1518 (22.8)
|
174 (22.3)
|
|
|
Abbreviations: BMI, Body Mass Index; LTSB, long‑term sedentary behavior
Relationship Between Trouble Sleeping and Gallstones
Univariable analysis results (Table 2) showed that participants with trouble sleeping had a 2.08 times higher risk of gallstones than those without trouble sleeping (95% CI 1.79-2.42, P < 0.001). Additionally, age, sex, BMI, diabetes, hypertension, smoking, and drinking were significantly associated with the risk of gallstones.
Multivariable logistic regression results are shown in Table 3. In the unadjusted model, trouble sleeping was positively associated with the risk of gallstones (OR, 2.08; 95% CI, 1.79-2.42). In Model 1, adjusting for age and sex, the results were similar (OR, 1.79; 95% CI, 1.53-2.09). In Model 2, adjusting for age, sex, BMI, and hypertension (confounders selected based on their associations with the outcomes of interest or a change in effect estimate of more than 10%), the positive association remained (OR, 1.56; 95% CI, 1.33-1.83). In Model 3, adjusting for age, sex, race, education, marital status, BMI, cholesterol, diabetes, hypertension, smoking status, drinking status, LTSB, and sleep duration, the positive association persisted (OR, 1.49; 95% CI, 1.26-1.76).
Table 2 Association of covariates and gallstones
Variable
|
OR (95% CI)
|
P-value
|
Sex
|
|
|
Male
|
1 (Reference)
|
|
Female
|
2.57 (2.18~3.02)
|
<0.001
|
Age (years)
|
1.03 (1.02~1.03)
|
<0.001
|
Race
|
|
|
Non-Hispanic white
|
1 (Reference)
|
|
Non-Hispanic black
|
0.58 (0.48~0.72)
|
<0.001
|
Mexican American
|
0.94 (0.75~1.18)
|
0.594
|
Others
|
0.72 (0.6~0.87)
|
0.001
|
Education (years)
|
|
|
<9
|
1 (Reference)
|
|
9-12
|
1.44 (1.04~2)
|
0.029
|
>12
|
1.32 (0.96~1.81)
|
0.092
|
Marital status
|
|
|
Married or living with a partner
|
1 (Reference)
|
|
Living alone
|
0.9 (0.77~1.05)
|
0.175
|
BMI (kg/m2)
|
1.06 (1.05~1.07)
|
<0.001
|
Cholesterol(mg/dL)
|
0.97 (0.9~1.04)
|
0.333
|
Diabetes
|
|
|
No
|
1 (Reference)
|
|
Yes
|
2.17 (1.82~2.58)
|
<0.001
|
Hypertension
|
|
|
No
|
1 (Reference)
|
|
Yes
|
2.12 (1.83~2.46)
|
<0.001
|
Smoking status
|
|
|
No
|
1 (Reference)
|
|
Yes
|
1.24 (1.07~1.44)
|
0.005
|
Drinking status
|
|
|
No
|
1 (Reference)
|
|
Yes
|
0.62 (0.53~0.73)
|
<0.001
|
LTSB
|
|
|
No
|
1 (Reference)
|
|
Yes
|
1.14 (0.94~1.39)
|
0.194
|
Trouble sleeping
|
|
|
No
|
1 (Reference)
|
|
Yes
|
2.08 (1.79~2.42)
|
<0.001
|
Sleep duration
|
|
|
≤ 6 h
|
1 (Reference)
|
|
>6 and <9 h
|
1 (0.82~1.22)
|
0.998
|
≥9 h
|
0.97 (0.77~1.22)
|
0.812
|
Abbreviations: BMI, Body Mass Index; LTSB, long‑term sedentary behavior
Table 3 Association between trouble sleeping and gallstones
Variable
|
Non-adjusted
|
|
Model 1
|
|
Model 2
|
|
Model 3
|
|
OR (95%CI)
|
P-value
|
|
OR (95%CI)
|
P-value
|
|
OR (95%CI)
|
P-value
|
|
OR (95%CI)
|
P-value
|
Sleep
|
2.08 (1.79~2.42)
|
<0.001
|
|
1.79 (1.53~2.09)
|
<0.001
|
|
1.56 (1.33~1.83)
|
<0.001
|
|
1.49 (1.26~1.76)
|
<0.001
|
Abbreviations: LTSB, long‑term sedentary behavior
Model 1: adjusted for sex, age.
Model 2: adjusted for age, sex, BMI, and hypertension.
Model 3: age, sex, race, education, marital status, BMI, cholesterol, diabetes, hypertension, smoking status, drinking status, LTSB, and sleep duration.
Subgroup Analysis
Subgroup analyses were conducted to determine if the association between trouble sleeping and gallstones was consistent across several subgroups. When stratified by age, sex, obesity, diabetes, hypertension, and sleep duration, consistent positive associations were observed (Figure 2), with an interaction observed in the age subgroup, indicating that the impact of trouble sleeping on gallstones is stable and unaffected by changes in covariates.
Figure 2 Subgroup analyses for the association of trouble sleeping and gallstones.
Sensitivity Analysis
To validate the robustness of the results, multiple imputation, weighted analysis, and PSM analyses were conducted (Table 4). Among the 9,232 participants, excluding those with missing data on gallstones and trouble sleeping, multiple imputation was performed for the remaining 9,202 participants with missing covariates. Participants with trouble sleeping had a 1.44 times higher risk of gallstones than those without trouble sleeping (95% CI 1.24-1.67, P < 0.001). In the weighted analysis, the positive association remained (OR, 1.49; 95% CI, 1.0-2.2). After PSM, 2,074 well-matched pairs were formed, with no significant differences between the matched groups. The risk ratio of gallstones after PSM was 1.38 (95% CI: 1.15-1.65, P < 0.001).
Table 4 Sensitivity analyses
Variable
|
Non-adjusted
|
|
Model 1
|
|
Model 2
|
|
Model 3
|
|
OR (95%CI)
|
P-value
|
|
OR (95%CI)
|
P-value
|
|
OR (95%CI)
|
P-value
|
|
OR (95%CI)
|
P-value
|
Multiple imputation
|
2.05 (1.79~2.35)
|
<0.001
|
|
1.76 (1.53~2.02)
|
<0.001
|
|
1.51 (1.31~1.74)
|
<0.001
|
|
1.44 (1.24~1.67)
|
<0.001
|
weighted analysis
|
2.091.78 (1.49~2.92)
|
<0.001
|
|
1.78 (1.26~2.53)
|
0.002
|
|
1.54 (1.08~2.21)
|
0.02
|
|
1.49 (1.0~2.2)
|
0.048
|
propensity score matching
|
-
|
-
|
|
-
|
-
|
|
-
|
-
|
|
1.38 (1.15~1.65)
|
<0.001
|
Abbreviations: LTSB, long‑term sedentary behavior
Model 1: adjusted for sex, age.
Model 2: adjusted for age, sex, BMI, and hypertension.
Model 3: age, sex, race, education, marital status, BMI, cholesterol, diabetes, hypertension, smoking status, drinking status, LTSB, and sleep duration.