Among 4890 participants, 999 (20.4%) women had insufficient gestational weight gain, 2059 (42.1%) women had adequate GWG and 1832 (37.7%) women had excessive GWG. Among people who were underweight and normal weight before pregnancy, about half of women had adequate GWG (53.5% and 43.4% respectively), while excessive GWG occupied the majority in those who were overweight and obese (64.0% and 73.2% respectively) (Fig1). Table 1 presents the demographic information of participants by pre-pregnancy BMI categories. No significant differences were observed between the four groups in terms of place of residence, occupation, ethnic, smoking status and drinking status (All p > 0.05). Women with obese pre-pregnancy BMI tended to be younger and had a higher percentage of excessive GWG (All p < 0.05). Significant differences could be also observed in way of conception, annual household income, education and parity. Most women conceived naturally and underweight women were more likely to have relatively higher income, education and most were nulliparous before this pregnancy.
Table 1Characteristics of participants by pre-pregnancy BMI
Variables
|
Total
(n = 4890)
|
Pre-BMI underweight
(n = 665)
|
Pre-BMI normal
(n = 3474)
|
Pre-BMI overweight
(n = 639)
|
Pre-BMI obesity
(n = 112)
|
F/c2
|
p
|
Age, Mean ±SD
|
30.59±3.57
|
31.94±3.88
|
32.65±4.10
|
32.71±4.23
|
30.59±3.57
|
34.848
|
< 0.001
|
Gestational weight gain, Mean ±SD
|
39.21±0.99
|
39.26±1.02
|
39.19±0.98
|
38.89±0.88
|
39.21±0.99
|
5.282
|
0.001
|
Gestational weight gain , n (%)
|
|
|
|
|
|
382.345
|
< 0.001
|
insufficient
|
999 (20.4)
|
194 (29.2)
|
741 (21.3)
|
58 (9.1)
|
6 (5.4)
|
|
|
adequate
|
2059 (42.1)
|
356 (53.5)
|
1507 (43.4)
|
172 (26.9)
|
24 (21.4)
|
|
|
excessive
|
1832 (37.5)
|
115 (17.3)
|
1226 (35.3)
|
409 (64)
|
82 (73.2)
|
|
|
Way of conception, n (%)
|
|
|
|
|
|
30.376
|
< 0.001
|
Assisted reproductive technology
|
859 (17.6)
|
72 (10.8)
|
627 (18)
|
141 (22.1)
|
19 (17)
|
|
|
Conceived Naturally
|
4031 (82.4)
|
593 (89.2)
|
2847 (82)
|
498 (77.9)
|
93 (83)
|
|
|
Place of residence, n (%)
|
|
|
|
|
|
-
|
0.405
|
Rural
|
218 (4.5)
|
38 (5.7)
|
148 (4.3)
|
27 (4.2)
|
5 (4.5)
|
|
|
Urban
|
4672 (95.5)
|
627 (94.3)
|
3326 (95.7)
|
612 (95.8)
|
107 (95.5)
|
|
|
Annual household income (10,000 RMB (1410 USD)/year), n (%)
|
|
|
|
|
|
44.903
|
< 0.001
|
~10
|
1542 (31.5)
|
205 (30.8)
|
1028 (29.6)
|
259 (40.5)
|
50 (44.6)
|
|
|
20~30
|
1491 (30.5)
|
191 (28.7)
|
1082 (31.1)
|
190 (29.7)
|
28 (25)
|
|
|
30~
|
1857 (38)
|
269 (40.5)
|
1364 (39.3)
|
190 (29.7)
|
34 (30.4)
|
|
|
Occupation, n (%)
|
|
|
|
|
|
1.963
|
0.580
|
Employed
|
4376 (89.5)
|
591 (88.9)
|
3122 (89.9)
|
564 (88.3)
|
99 (88.4)
|
|
|
Unemployed
|
514 (10.5)
|
74 (11.1)
|
352 (10.1)
|
75 (11.7)
|
13 (11.6)
|
|
|
Education, n (%)
|
|
|
|
|
|
35.197
|
< 0.001
|
~College/vocational
|
1187 (24.3)
|
138 (20.8)
|
821 (23.6)
|
189 (29.6)
|
39 (34.8)
|
|
|
Undergraduate
|
2566 (52.5)
|
382 (57.4)
|
1793 (51.6)
|
334 (52.3)
|
57 (50.9)
|
|
|
Postgraduate~
|
1137 (23.3)
|
145 (21.8)
|
860 (24.8)
|
116 (18.2)
|
16 (14.3)
|
|
|
Parity, n (%)
|
|
|
|
|
|
19.191
|
< 0.001
|
Nulliparous
|
3976 (81.3)
|
581 (87.4)
|
2799 (80.6)
|
507 (79.3)
|
89 (79.5)
|
|
|
Multiparous
|
914 (18.7)
|
84 (12.6)
|
675 (19.4)
|
132 (20.7)
|
23 (20.5)
|
|
|
Ethnic, n (%)
|
|
|
|
|
|
-
|
0.858
|
Han
|
4797 (98.1)
|
655 (98.5)
|
3404 (98)
|
628 (98.3)
|
110 (98.2)
|
|
|
non-Han
|
93 (1.9)
|
10 (1.5)
|
70 (2)
|
11 (1.7)
|
2 (1.8)
|
|
|
aSmoking status, n (%)
|
|
|
|
|
|
-
|
0.110
|
no
|
4772 (97.7)
|
642 (96.5)
|
3400 (98)
|
622 (97.5)
|
108 (97.3)
|
|
|
yes
|
111 (2.3)
|
23 (3.5)
|
69 (2)
|
16 (2.5)
|
3 (2.7)
|
|
|
bDrinking status, n (%)
|
|
|
|
|
|
3.927
|
0.270
|
no
|
2999 (61.4)
|
390 (58.6)
|
2141 (61.7)
|
404 (63.3)
|
64 (57.7)
|
|
|
yes
|
1884 (38.6)
|
275 (41.4)
|
1328 (38.3)
|
234 (36.7)
|
47 (42.3)
|
|
|
Abbreviations: SD, standard deviation, BMI, body mass index, GWG, gestational weight gain,
aSmoking during the pregnancy or during the year before pregnancy. b Alcohol consumption during the pregnancy or during the year before pregnancy.
The results of the comparisons of scale scores (CES-D score, SAS score and PSS-10 score) among different subgroups of GWG by total sample and pre-pregnancy BMI were shown in table 2. It was apparent that women with excessive GWG had higher SAS score in both total sample and normal weight women (p <0.001). Similar trend could be also observed in the percentage of SAS score (≥50) (7.2% in total sample and 7.8% in normal pre-pregnancy BMI). In addition, PSS-10 score in those who were normal weight is higher in excessive GWG group, with a higher likelihood of PSS-10 score (≥14) in both total and normal sample. On the other hand, the CES-D score is not associated with GWG or BMI (all p >0.05).
Table 2 Scale scores of pregnant women in subgroups of GWG by total sample and pre-pregnancy BMI
Variables
|
n
|
GWG
insufficient
|
GWG
adequate
|
GWG
excessive
|
p
|
Total
|
4890
|
|
|
|
|
CES-D score
|
|
8.90 ± 7.50
|
9.20 ± 7.63
|
9.40 ± 8.07
|
0.251
|
<16
|
|
807 (80.8)
|
1647 (80.0)
|
1444 (78.8)
|
0.427
|
≥16
|
|
192 (19.2)
|
412 (20.0)
|
388 (21.2)
|
|
SAS score
|
|
35.78 ± 7.06
|
35.97 ± 6.74
|
36.85 ± 7.44
|
<0.001
|
<50
|
|
943 (94.4)
|
1960 (95.2)
|
1700 (92.8)
|
0.006
|
≥50
|
|
56 ( 5.6)
|
99 ( 4.8)
|
132 ( 7.2)
|
|
PSS-10 score
|
|
9.44 ± 5.54
|
9.73 ± 5.47
|
9.95 ± 5.79
|
0.070
|
<14
|
|
780 (78.1)
|
1562 (75.9)
|
1346 (73.5)
|
0.020
|
≥14
|
|
219 (21.9)
|
497 (24.1)
|
486 (26.5)
|
|
Pre-BMI
normal
|
3474
|
|
|
|
|
CES-D score
|
|
8.73 ± 7.53
|
9.15 ± 7.65
|
9.60 ± 8.09
|
0.052
|
<16
|
|
605 (81.6)
|
1212 (80.4)
|
953 (77.7)
|
0.076
|
≥16
|
|
136 (18.4)
|
295 (19.6)
|
273 (22.3)
|
|
SAS score
|
|
35.62 ± 7.10
|
35.79 ± 6.68
|
36.94 ± 7.66
|
<0.001
|
<50
|
|
700 (94.5)
|
1438 (95.4)
|
1130 (92.2)
|
0.001
|
≥50
|
|
41 ( 5.5)
|
69 ( 4.6)
|
96 ( 7.8)
|
|
PSS-10 score
|
|
9.31 ± 5.57
|
9.66 ± 5.43
|
10.06 ± 5.79
|
0.015
|
<14
|
|
586 (79.1)
|
1151 (76.4)
|
888 (72.4)
|
0.002
|
≥14
|
|
155 (20.9)
|
356 (23.6)
|
338 (27.6)
|
|
Pre-BMI underweight
|
665
|
|
|
|
|
CES-D score
|
|
9.68 ± 7.38
|
9.72 ± 7.45
|
9.06 ± 7.72
|
0.716
|
<16
|
|
148 (76.3)
|
276 (77.5)
|
92 (80.0)
|
0.751
|
≥16
|
|
46 (23.7)
|
80 (22.5)
|
23 (20.0)
|
|
SAS score
|
|
36.52 ± 6.88
|
36.84 ± 6.95
|
36.77 ± 7.25
|
0.871
|
<50
|
|
184 (94.8)
|
335 (94.1)
|
108 (93.9)
|
0.921
|
≥50
|
|
10 ( 5.2)
|
21 ( 5.9)
|
7 ( 6.1)
|
|
PSS-10 score
|
|
9.61 ± 5.58
|
10.26 ± 5.42
|
9.73 ± 6.19
|
0.370
|
<14
|
|
143 (73.7)
|
260 (73.0)
|
84 (73.0)
|
0.984
|
≥14
|
|
51 (26.3)
|
96 (27.0)
|
31 (27.0)
|
|
Pre-BMI
overweight
|
639
|
|
|
|
|
CES-D score
|
|
8.72 ± 7.52
|
8.89 ± 7.76
|
9.08 ± 8.14
|
0.928
|
<16
|
|
49 (84.5)
|
138 (80.2)
|
331 (80.9)
|
0.770
|
≥16
|
|
9 (15.5)
|
34 (19.8)
|
78 (19.1)
|
|
SAS score
|
|
35.10 ± 7.10
|
36.10 ± 6.66
|
36.79 ± 7.22
|
0.181
|
<50
|
|
53 (91.4)
|
164 (95.3)
|
382 (93.4)
|
0.472
|
≥50
|
|
5 ( 8.6)
|
8 ( 4.7)
|
27 ( 6.6)
|
|
PSS-10 score
|
|
10.79 ± 4.99
|
9.62 ± 5.59
|
9.71 ± 5.66
|
0.281
|
<14
|
|
45 (77.6)
|
130 (75.6)
|
311 (76.0)
|
0.953
|
≥14
|
|
13 (22.4)
|
42 (24.4)
|
98 (24.0)
|
|
Pre-BMI
obesity
|
112
|
|
|
|
|
CES-D score
|
|
6.50 ± 5.79
|
7.29 ± 7.82
|
8.59 ± 7.78
|
0.621
|
<16
|
|
5 ( 83.3)
|
21 (87.5)
|
68 (82.9)
|
0.900
|
≥16
|
|
1 ( 16.7)
|
3 (12.5)
|
14 (17.1)
|
|
SAS score
|
|
37.33 ± 6.83
|
33.50 ± 6.47
|
35.83 ± 5.26
|
0.274
|
<50
|
|
6 (100.0)
|
23 (95.8)
|
80 (97.6)
|
0.611
|
≥50
|
|
0 ( 0.0)
|
1 ( 4.2)
|
2 ( 2.4)
|
|
PSS-10 score
|
|
6.67 ± 4.18
|
7.04 ± 6.84
|
9.80 ± 5.86
|
0.112
|
<14
|
|
6 (100.0)
|
21 (87.5)
|
63 (76.8)
|
0.308
|
≥14
|
|
0 ( 0.0)
|
3 (12.5)
|
19 (23.2)
|
|
Abbreviations: CES-D, Center for Epidemiologic Studies Depression Scale, SAS, Self-Rating Anxiety Scale,PSS-10, 10-item version of Perceived Stress Scale ,BMI, body mass index,, GWG, gestational weight gain.
Low correlations were showed between pre-pregnancy BMI, GWG and three scale scores, respectively. There was not reported any significant association between PSS-10 score, SAS score and pre-pregnancy BMI (all p>0.05). A low negative correlation was found between CES-D score and pre-pregnancy BMI (rho = -0.03, p < 0.05).GWG of women was on average low positively correlated with CES-D score, SAS score and PSS-10 score (rho = 0.03, 0.07, 0.04 respectively, p< 0.05). A strong correlation was reported between CES-D score and SAS score (rho = 0.77, p < 0.05), CES-D score and PSS-10 score (rho = 0.74, p < 0.05), as well as SAS score and PSS-10 score (rho = 0.63, p < 0.05). This means that three scale scores are positively correlated. (see Fig. 2 for details)
We performed logistic regression to explore the relationship between GWG and the risk of maternal mental disorders (Table 3). The adequate gestational weight gain is the reference group. The relationship between CES-D and GWG was not significant (insufficient group: OR = 0.951, 95% CI: 0.785, 1.151, excessive group: OR = 1.179, 95% CI: 0.842, 1.651). The same result was also found in the relationship between PSS-10 and GWG (insufficient group: OR = 0.88, 95% CI: 0.734, 1.055, excessive group: OR = 1.133, 95% CI: 0.98, 1.309). While excessive weight gain during pregnancy was associated with the risk of exceeding SAS score thresholds in the entire sample (OR= 1.535, 95% CI: 1.173, 2.008). After adjustment for potential confounders, excessive weight gain during pregnancy was also associated with a higher risk of anxiety symptoms in the entire sample (OR= 1.479, 95% CI: 1.128, 1.938).
Another logistic regression was conducted to explore the relationship between GWG and the risk of depression, anxiety, and stress by pre-pregnancy BMI. No significant relationship was found in subgroups of underweight, overweight or obesity. In normal BMI women, insufficient and excessive GWG were not associated with CES-D. Excessive GWG significantly and more obviously increases the risk of anxiety in both unadjusted model and adjusted model (unadjusted model: OR = 1.767, 95% CI: 1.284, 2.431, adjusted model: OR = 1.668, 95% CI: 1.209, 2.302). On the other side, it could be observed that excessive GWG is significantly associated with the risk of exceeding PSS-10 score thresholds in the unadjusted model (OR = 1.227, 95% CI: 1.033, 1.459) but not in the adjusted model (OR = 1.151, 95% CI: 0.966, 1.372).
Table 3 Odds ratios and 95% confidence intervals for the association among maternal pre-pregnancy BMI, GWG and the risk of exceeding scale thresholds in total sample
Variables
|
CES-D
|
SAS
|
PSS-10
|
GWG
insufficient
|
GWG adequate
|
GWG
excessive
|
GWG
insufficient
|
GWG adequate
|
GWG
excessive
|
GWG
insufficient
|
GWG adequate
|
GWG
excessive
|
Total
|
|
|
|
|
|
|
|
|
|
Model 1a
|
0.951 (0.785,1.151)
|
Reference
|
1.075 (0.92,1.257)
|
1.179 (0.842,1.651)
|
Reference
|
1.535 (1.173,2.008)
|
0.88 (0.734,1.055)
|
Reference
|
1.133 (0.98,1.309)
|
Model 2b
|
0.961 (0.793,1.164)
|
Reference
|
1.052 (0.899,1.231)
|
1.2 (0.855,1.683)
|
Reference
|
1.479 (1.128,1.938)
|
0.905 (0.754,1.087)
|
Reference
|
1.073 (0.927,1.243)
|
Pre-BMI normal
|
|
|
|
|
|
|
|
|
|
Model 1
|
0.928 (0.741,1.163)
|
Reference
|
1.179 (0.98,1.419)
|
1.222 (0.821,1.817)
|
Reference
|
1.767 (1.284,2.431)
|
0.849 (0.686,1.051)
|
Reference
|
1.227 (1.033,1.459)
|
Model 2
|
0.939 (0.748,1.177)
|
Reference
|
1.14 (0.946,1.374)
|
1.248 (0.838,1.86)
|
Reference
|
1.668 (1.209,2.302)
|
0.875 (0.705,1.086)
|
Reference
|
1.151 (0.966,1.372)
|
Pre-BMI underweight
|
|
|
|
|
|
|
|
|
|
Model 1
|
1.072 (0.709,1.622)
|
Reference
|
0.862 (0.513,1.451)
|
0.867 (0.4,1.88)
|
Reference
|
1.034 (0.428,2.499)
|
0.966 (0.65,1.436)
|
Reference
|
0.9995 (0.6224,1.6052)
|
Model 2
|
1.083 (0.709,1.655)
|
Reference
|
0.836 (0.493,1.418)
|
0.85 (0.386,1.871)
|
Reference
|
1.055 (0.433,2.571)
|
0.969 (0.644,1.459)
|
Reference
|
0.954 (0.588,1.548)
|
Pre-BMI overweight
|
|
|
|
|
|
|
|
|
|
Model 1
|
0.663 (0.287,1.529)
|
Reference
|
0.956 (0.61,1.499)
|
1.971 (0.618,6.288)
|
Reference
|
1.449 (0.645,3.257)
|
0.915 (0.45,1.86)
|
Reference
|
0.975 (0.644,1.478)
|
Model 2
|
0.687 (0.293,1.612)
|
Reference
|
0.946 (0.6,1.49)
|
2.145 (0.658,6.993)
|
Reference
|
1.459 (0.646,3.294)
|
0.919 (0.445,1.899)
|
Reference
|
0.942 (0.619,1.435)
|
Pre-BMI obesity
|
|
|
|
|
|
|
|
|
|
Model 1
|
1.75 (0.143,21.384)
|
Reference
|
1.441 (0.378,5.501)
|
0 (0,Inf)
|
Reference
|
0.575 (0.05,6.629)
|
0 (0,Inf)
|
Reference
|
2.111 (0.567,7.855)
|
Model 2
|
2.829 (0.162,49.469)
|
Reference
|
1.368 (0.324,5.784)
|
0 (0,Inf)
|
Reference
|
0.229 (0.014,3.746)
|
0 (0,Inf)
|
Reference
|
1.504 (0.379,5.958)
|
Abbreviations: CES-D, Center for Epidemiologic Studies Depression Scale, SAS, Self-Rating Anxiety Scale,PSS-10, 10-item version of Perceived Stress Scale, BMI, body mass index,, GWG, gestational weight gain.
a: Model 1 unadjusted logistic regression model
b: Model 2 was adjusted for Way of conception, Annual household income, Occupation, Education, Smoking status and Drinking status