Background
Domestic violence (DV) refers to all forms of violence against family members by means of beating, binding, maiming, abuse, restriction of personal freedom, abandonment and sexual abuse [
1,
2]. DV is not only a serious social problem but also a public health problem, and has been reported in both developed and developing countries [
3,
4]. Pregnancy is considered as a special period that DV starts or exacerbates [
5,
6]. Some researchers found that pregnant women were 2.7 to 3.9 times more likely to suffer from physical violence and twice as likely to suffer from sexual violence compared with women who are not pregnant [
7]. What’s more, because of the increased vulnerability women experience during pregnancy, DV not only will do harm to physical and mental health (disability, depression, substance abuse, etc.) [
8,
9], but also may cause adverse pregnancy outcomes (premature birth, abortion, etc.), which seriously affect the health of the next generation [
10].
Due to the different national cultures and the tools of assessment, studies conducted in pregnant women have reported that the prevalence of DV varies from 0.9 to 20% [
11]. The prevalence of DV among pregnant women is even higher in some developing countries or underdeveloped regions, the reported prevalence of DV was 28.2% in Iran [
12], 51% in Pakistan [
13], 45.1% in Peru [
14], 37% in Kenya [
15]. DV include physical, mental and sexual violence. Physical violence was the second leading cause of trauma during pregnancy [
16], and the prevalence varied from 0.9 to 21% worldwide [
17‐
19]. A review including 18 studies from 1996 to 2009 reported that the prevalence of mental violence during pregnancy was 1.5 to 36% and that of sexual violence was 1.0 to 3.9% [
20].
In previous studies, many factors were associated with the occurrence of DV among pregnant women. First, social instability was the potential risk factor associated with violence. For instance, some studies showed that compared to pregnant women who lived with their partners, had stable marriages and income, higher risk of DV tend to be reported among low-income, separated and divorced pregnant women [
21‐
23]. Additionally, violence during pregnancy has been associated with some pregnancy characteristics, such as multiple deliveries [
24], lack of prenatal care [
25] and unintended pregnancy [
5,
21]. Moreover, there has been some indication that violence during pregnancy is associated with unhealthy lifestyle behaviours, such as drinking, smoking, unhealthy diet and substance abuse [
26]. Some characteristics of pregnant women’s intimate partners, such as alcohol misuse, jealousy, stress [
26], and unemployment [
27,
28] might be risk factors of DV during pregnancy. Based on the current literature, we found that the relationship between family factors and DV among pregnant women has not received much attention. Few studies have explored the relationship between family characteristics and DV from the various perspectives of family characteristics in China. Therefore, this study focuses on the association between DV among pregnant women and four family characteristics: family relationships, family structure, family-related stressors and family function, to facilitate DV intervention for pregnant women.
Results
The mean age of the 813 participants was 28.98 ± 4.52 years. Most of them were ethnic Han (
n = 798, 98.2%), had stable marital status (
n = 728, 89.5%) and were employed (
n = 599, 73.7%). More than half of them had college / university degree or above education level (
n = 472, 58.1%). A total of 9% had a monthly household income of ≤3000, 36.9% of 3001–5000, 33.5% of 5001–8000, and 20.7% of > 8001. The majority had medical insurance (
n = 592,72.8%). 3 (0.4%) and 80 (9.8%) had smoking and drinking habits, respectively. Regarding pregnant women’s partner, 487 (59.9%) had college / university degree or above education level, and 799 (98.3%) were employed. A total of 288 (35.4%) and 314 (38.6%) had smoking and drinking habits, respectively. (Table
2).
During pregnancy, 127 (15.62%) participants suffered from DV in at least one form. Mental violence was the most serious (
n = 90, 11.07%), followed by physical and sexual violence, which only occurred in 8 (0.98%) and 7(0.86%), respectively. A total of 25 (3.08%) participants reported that they had experienced two types of violence (physical and mental violence), and 44 (5.41%) participants reported that they were afraid of the person who hurt them. (Table
1).
Table 1Distribution of different types of domestic violence (n = 813)
Only physical violence | 8 (0.98) |
Only mental violence | 90(11.07) |
Only sexual violence | 7 (0.86) |
Double violence (both physical and mental violence) | 25(3.08) |
Differences in general characteristics and family factors were compared between the two groups
According to Tables
2 and
3, nine variables showed statistically significant differences between the DV group and the no DV group (p<0.05): whether they had medical insurance, whether they had a drinking habit, whether they had experienced multiple abortions/sterility, the relations with the mother-in-law, relations with other family members, whether they had medium household debt, whether they were separated from their partner, and whether their partner had extramarital affairs.
Table 2General characteristics of the study sample in this study. (DV vs No DV)
Age | | | | 0.189 | 0.910 |
<26 | 35 (27.6) | 202 (29.4) | 237 (29.2) | | |
27–32 | 67 (52.8) | 354 (51.6) | 421 (51.8) | | |
>33 | 25 (19.7) | 130 (19.0) | 155 (19.1) | | |
Ethnicity | | | | 0.690 | 0.406 |
Minority | 4 (3.1) | 11 (1.6) | 15 (1.8) | | |
Han ethnic | 123 (96.9) | 675 (98.4) | 798 (98.2) | | |
Marital status | | | | 0.739 | 0.390 |
Stable | 111 (87.4) | 617 (89.9) | 728 (89.5) | | |
Unstable | 16 (12.6) | 69 (10.1) | 85 (10.5) | | |
Education | | | | 1.064 | 0.302 |
Senior school / technical school or less | 48 (37.8) | 293 (42.7) | 341 (41.9) | | |
College / university degree or above | 79 (62.2) | 393 (57.3) | 472 (58.1) | | |
Occupation | | | | 0.944 | 0.331 |
Employed | 98 (77.2) | 501 (73) | 599 (73.7) | | |
Unemployed | 29 (22.8) | 185 (27) | 214 (26.3) | | |
Education (partner) | | | | 1.064 | 0.302 |
Senior school / technical school or less | 48 (37.8) | 293 (42.7) | 341 (41.9) | | |
College / university degree or above | 79 (62.2) | 393 (57.3) | 472 (58.1) | | |
Occupation (partner) | | | | 0.019 | 0.890 |
Employed | 125 (98.4) | 674 (98.3) | 799 (98.3) | | |
Unemployed | 2 (1.6) | 12 (1.7) | 14 (1.7) | | |
Monthly per capita household income | | | | 4.025 | 0.259 |
≤ 3000 | 12 (9.4) | 61 (8.9) | 73 (9.0) | | |
3001–5000 | 44 (34.6) | 256 (37.3) | 300 (36.9) | | |
5001–8000 | 51 (40.2) | 221 (32.2) | 272 (33.5) | | |
>8001 | 20 (15.7) | 148 (21.6) | 168 (20.7) | | |
Medical insurance | | | | 4.275 | 0.039 |
Yes | 102 (80.3) | 490 (71.4) | 592 (72.8) | | |
No | 25 (19.7) | 196 (28.6) | 221 (27.2) | | |
Smoking habit | | | | 0.002 | 0.960 |
Yes | 1 (0.8) | 2 (0.3) | 3 (0.4) | | |
No | 126 (99.2) | 684 (99.7) | 810 (99.6) | | |
Drinking habit | | | | 9.499 | 0.002 |
Yes | 22 (17.3) | 58 (8.5) | 80 (9.8) | | |
No | 105 (82.7) | 628 (91.5) | 733 (90.2) | | |
Smoking habit (partner) | | | | 0.042 | 0.838 |
Yes | 46 (36.2) | 242 (35.3) | 288 (35.4) | | |
No | 81 (63.8) | 444 (64.7) | 525 (64.6) | | |
Drinking habit (partner) | | | | 1.393 | 0.238 |
Yes | 55 (43.3) | 259 (37.8) | 314 (38.6) | | |
No | 72 (56.7) | 427 (62.2) | 499 (61.4) | | |
Mode of pregnancy | | | | 1.113 | 0.291 |
Natural conception | 124 (97.6) | 656 (95.6) | 780 (95.9) | | |
Artificial impregnation | 3 (2.4) | 30 (4.4) | 33 (4.1) | | |
History of delivery | | | | 0.388 | 0.533 |
≤ 1 | 66 (89.2) | 347 (86.5) | 413 (86.9) | | |
>1 | 8 (10.8) | 54 (13.5) | 62 (13.1) | | |
Pregnancy intention | | | | 0.958 | 0.328 |
Intended | 70 (55.1) | 410 (59.8) | 480 (59.0) | | |
Unintended | 57 (44.9) | 276 (40.2) | 333 (41.0) | | |
Multiple abortions/sterility | | | | 11.476 | 0.001 |
Yes | 16 (12.6) | 33 (4.8) | 49 (6.0) | | |
No | 111 (874) | 653 (95.2) | 764 (94.0) | | |
Regular antenatal examination | | | | 1.836 | 0.175 |
Yes | 120 (94.5) | 623 (90.8) | 743 (91.4) | | |
No | 7 (5.5) | 63 (9.2) | 70 (8.6) | | |
Pregestational diseases | | | | 0.676 | 0.411 |
Yes | 3 (2.4) | 7 (1.0) | 10 (1.2) | | |
No | 124 (97.6) | 679 (90.0) | 803 (98.8) | | |
Pregnancy complications | | | | 3.264 | 0.071 |
Yes | 19 (15.0) | 66 (9.6) | 85 (10.5) | | |
No | 108 (85.0) | 620 (90.4) | 728 (89.5) | | |
Table 3Family characteristics of the study sample in this study. (DV vs No DV)
Family relationships |
Relationship with mother-in-law | | | | 45.411 | 0.000 |
Good | 106 (83.5) | 668 (97.4) | 774 | | |
Bad | 21 (16.5) | 18 (2.6) | 39 | | |
Relationship with other families | | | | 49.391 | 0.000 |
Good | 104 (81.9) | 666 (97.1) | 770 | | |
Bad | 23 (18.1) | 20 (2.9) | 43 | | |
Family structure |
Number of cohabitation (include oneself) | | | | 0.058 | 0.809 |
1–2 | 26 (20.5) | 147 (21.4) | 173 | | |
>3 | 101 (79.5) | 539 (78.6) | 640 | | |
Live with eldership | | | | 0.813 | 0.367 |
Yes | 88 (69.3) | 447 (65.2) | 535 | | |
No | 39 (30.7) | 239 (34.8) | 278 | | |
Family related stressors |
Medium household debt | | | | 8.857 | 0.003 |
Yes | 20 (15.7) | 52 (7.6) | 72 | | |
No | 107 (84.3) | 624 (92.4) | 741 | | |
Separation | | | | 8.095 | 0.004 |
Yes | 29 (22.8) | 90 (13.1) | 119 | | |
No | 98 (77.2) | 596 (85.9) | 694 | | |
Partner’s extramarital affairs | | | | 18.347 | 0.000 |
Yes | 9 (7.1) | 8 (1.2) | 17 | | |
No | 118 (92.9) | 678 (98.8) | 796 | | |
Family function |
Family APGAR index | | | | 40.096 | 0.000 |
Low level | 25 (19.7) | 41 (6) | 66 | | |
Middle level | 52 (40.9) | 204 (29.7) | 256 | | |
High level | 50 (39.4) | 441 (64.3) | 491 | | |
The result of multivariate binary logistic regression analysis
These six variables were adjusted as confounding variables (medical insurance, drinking habit, multiple abortions/sterility, age, ethnicity, marital status). After adjustment, the results of multivariate binary logistic regression showed that the risk factors for DV were tensions with their mother-in-law (OR: 2.85; 95% CI: 1.29 to 6.30), tensions with other family members (OR: 3.30; 95% CI: 1.57 to 6.93), medium household debt (OR: 2.17; 95% CI: 1.18 to 4.00), middle and low family APGARI (OR: 2.01; 95% CI: 1.30 to 3.13, OR: 4.01; 95% CI: 2.09 to 7.69) (Table
4).
Table 4Multivariate binary logistic regression analysis of family factors associated with DV among pregnant women
Family relationships |
Relationship with mother-in-law |
Good | 1.00 | 1.00 |
Bad | 2.71 (1.24, 5.90) | 2.85 (1.29, 6.30) |
Relationship with other families |
Good | 1.00 | 1.00 |
Bad | 3.40(1.63,7.09) | 3.30 (1.57, 6.93) |
Family structure |
Number of cohabitation (include oneself) |
1–2 | 1.34 (0.71, 2.51) | 1.00 |
>3 | 1.00 | 1.19 (0.63, 2.26) |
Live with eldership |
Yes | 1.29 (0.83, 2.00) | 1.30 (0.83, 2.03) |
No | 1.00 | 1.00 |
Family related stressors |
Medium Household debt |
Yes | 2.03 (1.10, 3.74) | 2.17 (1.18, 4.00) |
No | 1.00 | 1.00 |
Separation |
Yes | 1.55 (0.93, 2.60) | 1.61 (0.96, 2.70) |
No | 1.00 | 1.00 |
Partner’s extramarital affairs |
Yes | 3.06 (1.01, 9.30) | 2.37 (0.74, 7.54) |
No | 1.00 | 1.00 |
Family function |
Family APGAR index |
Low level | 3.52 (1.85, 6.69) | 4.01 (2.09, 7.69) |
Middle level | 1.94 (1.25, 3.01) | 2.01 (1.30, 3.13) |
High level | 1.00 | 1.00 |
Discussion
In this study, 127 pregnant women were identified as victims of DV (15.62%). Of these participants, 90 experienced mental violence (11.07%), 8 experienced physical violence (0.98%), 7 experienced sexual violence (0.86%), and 25 reported experiencing two forms of violence (3.08%). The prevalence of DV in this study was similar to that of another study carried out in Changsha City, Hunan Provinces, China (the positive rate was 11.3%) [
1]. However, the prevalence in this study was lower than those in some other countries. In a Portuguese study, 43.4% reported a history of DV during pregnancy [
35]. A hospital-based cross-sectional study carried out in Northwest Ethiopia reported that 58.7% were victims of DV during pregnancy [
24]. The possible reasons for the lower positive rate in this study may be that people from different regions and cultural backgrounds have different understandings of DV, leading the fact that some women who are suffering from violence do not regard these behaviours as DV [
27]. In Eastern culture, women were more likely to be silent to maintain the stability of the family and to avoid more serious violence, which might have led to a decline in the positive reporting rate [
36]. Furthermore, the outcome assessment tool might the possible reason for the differences in the prevalence of DV. The study conducted in Changsha used the Abuse Assessment Screen (AAS), whereas the studies in Portugal and Ethiopia used the WHO tool to assess the outcome variable. Comparing these two scales, we can find that the WHO tool is more specific and detailed in describing violent behaviours, such as occasional verbal and physical conflicts, which are easy to be associated with and identified as domestic violence, thus showing a higher positive rate. In terms of the types of DV, mental violence was the most serious, which was similar to findings in Guatemala and Iran [
37,
38]. The possible reason for this phenomenon is that physical and sexual violence was often accompanied by mental violence, and mental violence almost always precedes physical violence, or could be independent of physical and sexual violence [
39,
40]. Most of the perpetrators are the intimate partner of pregnant women (husband, ex-husband, boyfriend), which was consistent with the study by Mtonga in Zambia [
41]. This study did not explore the characteristics of different perpetrator, and further research in this aspect would be needed.
The findings showed that tensions with family members, especially the mother-in-law, was significantly related to the occurrence of DV during pregnancy (p<0.05). The risk of DV was 1.85 times higher in pregnant women who have strained relation with their mother-in-law than that in good relation group (OR:2.85; 95%CI: 1.29 to 6.30), these findings were almost consistent with those reported by Chen in Hong Kong, China [
42]. In China, discord between family members, especially contradictions between mother-in-law and daughter-in-law, are important factors of family conflict. Due to the differences in fertility and health concepts between women and their mothers-in-law [
43], contradictions occur frequently, especially during pregnancy. On the other hand, it is also possible that after the conflict between husband and wife, it is common for a mother-in-law to follow the will of her son by disciplining a wife, resulting in tensions between mother-in-law and daughter-in-law [
42].
In this study, medium household debt were associated with the occurrence of DV (p<0.05). The risk in participants who had medium household debt was 2.17 times as much as that in the reference group (OR: 2.17; 95% CI: 1.18 to 4.00). Thus, pregnant women with medium household debt were more likely to suffer from DV. This was consistent with results from the research by Tsui in Hong Kong [
44]. During pregnancy, with the increasing household expenses accompanied by the loss of working ability of the pregnant woman, pregnant women will more dependent on their husbands in economic, which will easily leads to disagreement and violence [
24].
Although the differences in extramarital affairs between the two groups were not statistically significant after adjustment, we cannot ignore the possible association between this factor and the occurrence of DV. Because this association has been approved by Peedicayil et al. [
45]. The extramarital affairs of women’s partner often occurred during the pregnancy [
46]. During pregnancy, women paid more attention to their own physiological changes and the foetal growth, and they sometimes ignore the emotional and sexual needs of their male partners, which can cause family conflicts and violence to some extent [
47].
The results of this study suggest that family dysfunction was a risk factor for DV, this finding was similar to the study by Tuesca in Colombia [
48]. The lower the family APGARI indicates the more serious the family dysfunction. The risk for participants who with low and middle family APGARI was 4 and 2 times as much as that of pregnant women with a high family APGARI (OR: 4.01; 95% CI: 2.09 to 7.69; OR: 2.01; 95% CI: 1.30 to 3.13), respectively. Family dysfunction reflected that pregnant women could not acquire enough care, love and help from their families and even that the family might be the source of bodily injury and mental pressure.
In this study, the selection of the sample is representative. The principle of randomization was followed, and women in the third trimester of pregnancy were selected as the subjects of the survey. Women in the third trimester of pregnancy have experienced nearly the entire pregnancy process and can report fully on their experiences of abuse throughout pregnancy with less recall bias. Equally important, pregnant women have different mental activities and physiological characteristics at different stages of pregnancy. Compared with all periods of pregnant women, the choice of women in the third trimester can reduce the influence of confounding factors such as gestational age. This study provides a comprehensive discussion of the relationship between family factors and pregnant women’s DV from various dimensions, which has certain innovative value in the prevention of DV. The present study also had certain limitations. This study was cross-sectional, thus, there were limitations in identifying causal relationships. Second, because of the self-reported design, we cannot deny the reporting bias. Last, there are some limitations to the AAS scale: AAS cannot distinguish the severity of DV and injured body parts, which may reduce the investigative effect of the scale, so that some risk characteristics cannot be identified. However, this study promoted an understanding of the relationship between family factors and DV and further confirmed the importance of family factors.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.