Prevalence and Correlates of the Family Polyvictimization: An Updated Meta-Analysis

Abstract

This study aims to (a) provide updated prevalence estimates of family polyvictimization by synthesizing recent literature, (b) examine the effect sizes of how one type of family victimization impacts other types, and (c) investigate correlates of family polyvictimization. Literature databases were searched for publications from May 2018 onward. A total of 49 studies, producing 51 estimates, met the inclusion criteria and were included in this updated analysis, providing a substantial sample size of 543,196 for the general synthesis of effect sizes. Combined with a previous meta-analysis, 89 estimates were included to synthesize overall effect sizes and estimate associations among different types of family victimization and their correlates. Among the 51 new estimates, the event rate of family polyvictimization was 27.0%. When combined with the previous meta-analysis (18.3%), the overall event rate across 89 estimates decreases slightly to 23.1%. This updated meta-analysis includes more studies from continents such as Africa and South America, which consistently report higher rates of polyvictimization. Findings also show that the prevalence of family victimization co-occurrence is higher in clinical samples compared to the general population. The higher prevalence of family victimization co-occurrence in clinical samples highlights the increased likelihood of identifying a second victim in the same family once the first victim comes to the attention of health or social services. The overrepresentation of child polyvictimization and the co-occurrence of intimate partner violence and child abuse suggest that more members from extended families should be included in family violence studies.

Keywords

violence abuse maltreatment family polyvictimization meta-analysis

Introduction

Family violence is a complex and pervasive issue, often involving multiple forms of victimization that co-occur within the same household. In our previous meta-analysis (Chan et al., 2021), we highlighted the significance of examining the co-occurrence of various types of family violence, reporting an overall event rate of 18.3%. Notably, co-occurrence rates were substantially higher in clinical populations (36.0%) compared to the general population (9.7%), underscoring the heightened vulnerability among families already engaged with protective or support services. Importantly, the identification of an initial victim increased the likelihood of uncovering additional victims within the same family by sixfold. These findings underscore the urgent need for early detection strategies to identify victims who might otherwise remain hidden, and they point to the importance of understanding family polyvictimization for the development of effective prevention and intervention efforts.

Despite these insights, our earlier review was limited by the small number of eligible studies, restricting our ability to explore sub-groups, moderators, and family-level risk factors associated with polyvictimization. Since then, the field has seen a notable increase in research on the co-occurrence of violence and polyvictimization, presenting a timely opportunity to update and expand upon previous findings.

A further challenge in advancing this field is the inconsistent use of terminology. While some researchers use “polyvictimization” to refer to multiple forms of victimization experienced by a single individual, our focus is on family polyvictimization—the experience of multiple types of victimization reported by different members within the same family (Chan, 2017; Chan et al., 2021). This distinction is critical, as family polyvictimization encompasses a broader range of victimization patterns, including but not limited to intimate partner violence (IPV), child abuse and neglect (CAN), elder abuse (EA), and in-law abuse (ILA), among others. Understanding family polyvictimization requires attention to both the diversity of victimization types and the relationships between victims and perpetrators within the family context.

The current definition of family polyvictimization emphasizes the presence of multiple victims within the same family, setting it apart from situations where a single individual experiences multiple forms of violence—such as repeat, recurrent, or multiple victimization, as well as child polyvictimization. Several studies define repeat or recurrent victimization as instances where a single victim reports experiencing repeated incidents of family violence (Davis et al., 2021; Goodlin & Dunn, 2010). Other research uses the term polyvictimization to refer to a single victim, whether a child or an adult woman, who has experienced multiple types of victimization or trauma, including physical, psychological, or sexual abuse, and neglect (Bridges-Curry & Newton, 2024; Burden & Roddy, 2024). While repeat or multiple victimization reflects the severity and complexity of an individual’s experiences, it does not necessarily indicate the involvement of additional victims, nor does it specify whether the perpetrators are family members.

There is an expanding body of research on child polyvictimization, frequently assessed using the Juvenile Victimization Questionnaire (JVQ; Finkelhor et al., 2007) or its revised version, the JVQ-R2 (Finkelhor et al., 2011). The JVQ captures multiple forms of victimization experienced by a child, often involving different perpetrators and occurring in various contexts. Beyond simply measuring, the prevalence and correlates of child polyvictimization, the JVQ can provide valuable insights into the involvement of multiple perpetrators, diverse contexts, and potentially even the presence of multiple victims within the same family. For example, it assesses experiences such as child maltreatment (including physical, psychological, and sexual abuse, as well as neglect by parents or family caretakers), sibling violence, witnessing domestic violence, and witnessing a parent assault a sibling. However, to date, no study has explicitly coded data according to the concept of family polyvictimization. Nevertheless, the use of the JVQ may indirectly capture cases involving multiple family victims, and such studies are included in this meta-analysis.

In this meta-analysis, family victimization is defined as the identification of multiple victims within a single family, which may include a child, an adult parent, or an elderly member. Depending on the specific perpetrator-victim relationships, these cases may be categorized as IPV, CAN, EA, ILA, and so forth. Understanding family polyvictimization is crucial for identifying which families are at risk of having multiple victims, whether harmed by one or several perpetrators. It is therefore important to focus on family-level risk factors, rather than solely individual-level ones, when studying family polyvictimization. In practice, recognizing the presence of one victim should prompt proactive screening for additional victims by health, social welfare, or criminal justice professionals. However, early identification of hidden victims of family violence remains a significant challenge.

Given the recent growth in research on child and family polyvictimization, there is a clear need to synthesize the latest evidence, update prevalence estimates, and examine the correlates and impact of different forms of family victimization. By updating our meta-analysis, we aim to advance the field by (a) providing updated prevalence estimates of family polyvictimization, (b) quantifying the effect sizes of associations between different types of family victimization, and (c) identifying key correlates and risk factors at the family level. This work will contribute to a more nuanced understanding of family violence and inform the development of more effective prevention and intervention strategies.

Method

Literature Search Strategy

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to present our findings. The studies included in this meta-analysis were identified through searches in databases such as the Applied Social Sciences Index and Abstracts, ERIC, MEDLINE, PsycINFO, Sociological Abstracts, and Social Service Abstracts. In line with the definition of family victimization as the identification of multiple victims within a single family—which may include a child, an adult parent, or an elderly member—the literature search was conducted using combinations of the following keywords: polyvictimization, multiple victimization; child abuse, child victim, child neglect, violence against child, child maltreatment, bully; partner violence, partner abuse, spouse abuse, partner aggression, couple violence, couple abuse, couple aggression, domestic violence; elder neglect, elder abuse; and ILA. We reviewed the reference lists of eligible articles and relevant review papers to identify studies and reports that were not captured in the initial database search.

Study Selection

The two authors performed the data extraction and evaluation process. Three primary criteria were used for the inclusion of articles in this meta-analysis: (a) published between May 2018 and March 2025 (i.e., after the period covered by our previous meta-analysis; Chan et al., 2021); (b) reported at least two types of family victimization, including but not limited to CAN, IPV, EA, and ILA; and (c) reported the co-occurrence of victimization among members of the same household. Articles were excluded if they: (a) reported only a single type of family victimization; (b) reported victimization occurring at different life stages for the same individual; (c) did not provide sufficient data for effect size calculation; or (d) focused on intervention evaluations or reviewed topics unrelated to family victimization. All studies meeting these criteria were considered eligible, regardless of context, participants’ age, gender, sexual orientation, religion, language, or cultural background.

The search identified a total of 5,848 records. After removing duplicates, 3,423 records remained. Of these, 3,257 records were excluded for the following reasons: 308 focused on events that occurred in childhood, 88 were intervention studies without prevalence data, 844 addressed non-family violence, 635 were qualitative studies (e.g., commentaries, legal reports), 1,206 examined a single type of victimization without co-occurrence, and 176 were review studies lacking empirical data. During full-text screening for eligibility, 117 articles were excluded from the meta-analysis due to insufficient data (n = 86), or not involving members of the same household (n = 31). As a result, 49 studies were included in the analysis. One article (Dong & Wang, 2019) provided three estimates reporting the co-occurrence of CAN and IPV, CAN and EA, and IPV and EA. These estimates were computed independently for effect size calculation. In total, we listed 51 estimates from the 49 articles. Details of the article selection process are presented in Figure 1.

Figure 1.

Summary of the search results and article selection procedures.

Data Extraction

The eligible articles were numbered and summarized in a spreadsheet to record the following aspects: (a) Publication information (authors, publication year, and study site); (b) Violence-related factors (types of victimization and measures); (c) Study characteristics (sampling method, setting, sample size, age distribution, gender ratio, ethnicity, and informants); and (d) Individual and family factors associated with polyvictimization. Data were extracted from the included studies for the outcome measures, including descriptive data (event rate and sample sizes) and effect sizes (e.g., Cohen’s d and odds ratios [ORs]). The 49 studies were randomly distributed among the authors for independent assessment to ensure coding reliability.

Quality Assessment

The authors independently assessed the quality of the 49 studies using a checklist covering 5 key aspects, rated on a binary “1/0” scale. Studies were evaluated based on whether they: (a) reported sample recruitment procedures, (b) described the profiles or characteristics of the sample, (c) used validated measures or scales, (d) performed appropriate statistical analyses, and (e) provided sufficient data for analysis. Scores ranged from 0 to 5, with higher scores indicating better quality for inclusion in the meta-analysis. The average score of the included studies was 4.78, indicating their overall high quality. No study was excluded due to unsatisfactory quality.

Statistical Analysis

First, we calculated pooled estimates of the co-occurrence rates of family victimization and polyvictimization using 51 estimates from the 49 studies screened during this study period (May 2018 to March 2025). As an update to the meta-analysis on this topic, we combined these 51 estimates with our previous meta-analysis (Chan et al., 2021), which included 38 studies published before April 2018, resulting in pooled estimates based on a total of 89 estimates. The 89 estimates were further categorized by sample type: (a) community samples, primarily selected through probability sampling in household or school settings; and (b) clinical samples, typically drawn via convenience sampling from child protective services (CPS), medical records, or police databases. Due to heterogeneity in victimization types and contexts across studies, we employed a random-effects model (Borenstein et al., 2010). Results were visualized using forest plots with 95% confidence intervals (CI) for each study. Second, we computed pooled ORs with 95% CIs for individual and family correlates of family polyvictimization. Heterogeneity was assessed using Cochran’s Q statistic with corresponding p-value and I² statistic to quantify the variance proportions. To assess potential publication bias, a funnel plot was generated, supplemented by several statistical tests including the Classic Fail-Safe N, Egger’s regression test, and Duval and Tweedie’s trim-and-fill method. The Comprehensive Meta-Analysis software (version 4; Biostat, Inc.) was used to conduct all statistical analyses.

Results

Study Characteristics

As shown in Figure 1, the final pool for this meta-analysis consisted of 49 studies screened during this study period, all of which provided ORs on the co-occurrence of victimization or associated factors. The characteristics of the studies are summarized in Table 1. Sample sizes ranged from 90 to 305,867, together providing a substantial sample size of 543,196 for the general synthesis of effect sizes. This represents a significant increase compared to the 99,956 participants included in the previous review, along with an increase in the number of studies from 38 to 49. Two studies (#1–2) reported family polyvictimization (CAN, IPV, EA, and additional forms), 13 studies (#3–15) reported child polyvictimization, 31 studies (#16–46) reported the co-occurrence of CAN and IPV, 1 study (#47) on the co-occurrence of CAN and IPV, CAN and EA, and IPV and EA. Two studies (#48–49) reported the co-occurrence of IPV and ILA.

Table 1.

Characteristics of the Studies Included in the Overall Prevalence Synthesis.

Study	Violence Type			Measures (items)	Setting	Site	Sample					Informant	Correlates		Q
Study	First	Second	Third	Measures (items)	Setting	Site	Sampling method	Sample size	Female (%)	Age group and Mean age	Ethnicity	Informant	Individual	Family	Q
1. Chan et al. (2019)	IPV	CAN	EA	FPS (11)	Household	Hong Kong and Shanghai, China	Non-probability (convenience)	445 adults	Not reported	Fathers ~40.0 years (SD = 9.52), Mothers ~38.5 years (SD = 7.17); Children mean age = 9.0 years (SD = 2.91)	Chinese	Proxy report (one adult parent per household)	/	/	5
2. Chen and Chan (2025)	F-Poly	/	/	FPS (12)	Household	Hong Kong	Probability (two-stage stratified random)	1,400 adults	75.4%	Mean age 39.75 (SD = 6.99)	Chinese	Self-report	Family members’ disability and chronic disease, stress	Male dominance, Violence socialization	5
3. Al-Eissa et al. (2019)	C-Poly	/	/	ICAST-CH (36)	School	Saudi Arabia	Probability (multistage stratified)	16,010 students	49.2%	15 to 18 years (mean age 16.8 ± 0.9)	Saudi Arabian	Self-report	Being female	Living with extended family; in urban areas	5
4. Almeida et al. (2020)	C-Poly	/	/	JVQ (34)	School	Portugal	Non-probability (convenience)	849 adolescents	54.1%	12 to 17 years (mean age 13.70 ± 1.43)	Portuguese	Self-report	Being older	/	5
5. Buffarini et al. (2024)	C-Poly	/	/	JVQ (34)	Household	Brazil	Population-based birth cohort	3,993 children	49.4%	4 years (mean age 3.8 ± 0.2)	Brazilian	Proxy report (Maternal)	/	Maternal ACE; low maternal education; absence of father; parental antisocial behavior; maternal depression; violent neighborhood	5
6. Chen and Chan (2022)	C-Poly	/	/	JVQ	School	Hong Kong	Probability (two-stage stratified random)	5,567 children and their caregivers	53.2%	3 to19 years (mean age = 6.26)	Chinese	Self-report (<10) Proxy report (10+)	Being male	Having more siblings; residential mobility; less family support	5
7. Peña Cárdenas et al. (2022)	C-Poly	/	/	JVQ	School	Mexico	Non-probability (convenience)	874 students	53.2%	12 to17 years (mean age = 14.30)	Mexican	Self-report	Being female	/	5
8. Fernández-Artamendi et al. (2020)	C-Poly	/	/	JVQ	Residential Child Care	Spain	Non-probability (convenience)	321 adolescents	34.9%	11 to18 (mean age 15.25 ± 1.34)	Spanish	Self-report	Alcohol abuse	/	5
9. Méndez-López and Pereda (2019)	C-Poly	/	/	JVQ	School	Mexico	Non-probability (convenience)	1,068 adolescents	47.2%	14.45 ± 1.54	Mexican	Self-report	Being older	/	5
10. Pinto-Cortez et al. (2018)	C-Poly	/	/	JVQ	School	Chile	Non-probability (convenience)	706 adolescents	50.8%	15 years and 8 months (SD = 1.32)	Predominantly Hispanic (65.7%)	Self-report	/	/	5
11. Pinto-Cortez et al. (2021)	C-Poly	/	/	JVQ	School	Chile	Non-probability (convenience)	718 university students	53.1%	17 to 28 years (mean age 21.6 ± 4.11)	Chilean	Self-report	Being male		5
12. Pinto-Cortez et al. (2022)	C-Poly	/	/	JVQ	School	Chile	Probability (multistage)	9,648 children and youth	50.3%	12 to 18 years (mean age 14.72 ± 1.52)	Predominantly Hispanic (86.7%)	Self-report	/	/	5
13. Suárez-Soto et al. (2018)	C-Poly	/	/	JVQ	Child welfare and juvenile justice systems	Spain	Non-probability (convenience)	227 adolescents	36.1%	12 to 17 years (mean age 15.24 ± 1.56)	Predominantly 58.6% born in Spain	Self-report	/	/	5
14. Wemmers et al. (2018)	C-Poly	/	/	JVQ	Household	Canada	Probability (Random digit dial)	1,400 youths	50.3%	12 to 17 years	Predominantly White (85.5%)	Self-report	Being older; Being male		5
15. Zhou et al. (2024)	C-Poly	/	/	JVQ-R2	School	China	Probability (stratified cluster random)	2,091 students	52.75%	12 to 15 years mean = 13.51 ± 0.882)	Chinese	Self-report	/	/	5
16. Butala et al. (2023)	CAN	IPV	/	ICD-10-CM; child safety assessments (CSAs)	Emergency departments	USA	Non-probability (convenience)	227 encounters	94.3%	18 to 60 years	Not specified	Medical records	/	/	5
17. Al-Eissa et al. (2020)	CAN	IPV	/	ICAST-CH	School	Saudi Arabia	Probability (multistage-stratified)	16,939 students	49%	15 to 18 years	Saudi Arabian	Self-report	/	/	5
18. Brown et al. (2021)	CAN	IPV	/	CTS2; CTSPC	Household	USA	Probability (stratified cluster)	1,987 children	49.8%	0 to 18 years	Predominantly White	Proxy (maternal report)	/	/	5
19. Brühl et al. (2023)	CAN	IPV	/	ICAST-TC; CTS2	Household	Romania	Non-probability (convenience)	813 mothers	Phase 1: 56.3%; Phase 2: 39.9%	Phase 1: 5.75 ± 2.06; Phase 2: 5.64 ± 1.96	Romanian	Proxy (maternal report)	Caregiver’s history of CM, child externalizing behavior problems	/	5
20. Herbert et al. (2023)	CAN	IPV	/	ALSPAC	Household	UK	Population Study	15,589 fetuses	Not specified	Not specified	Children born in the United Kingdom	Parent and child reports	/	/	4
21. Joh-Carnella et al. (2023)	CAN	IPV	/	OIS-2018	Panel data	Canada	Probability (stratified random)	7,590 CM-related investigations	Not specified	0 to 17 years	Canadian	Child welfare workers	/	/	5
22. Bott et al. (2022)	CAN	IPV	/	CTS2	Household	Colombia, Mexico, Peru	Probability (multistage)	Colombia: 13,518; Mexico: 38,097; Peru: 17,156 (total 6,8771)	Not reported	Women aged 15 to 49	Indigenous and non-indigenous	Self-report	/	/	5
23. Falb et al. (2023)	CAN	IPV	/	WHO	Household	Republic of Congo	Cluster randomized controlled trial	394 participants	50.3%	Women: 32.0 years (SD = 8.7); Men: 36.9 years (SD = 8.7)	Not specified	Self-report	/	/	5
24. Tiyyagura et al. (2023)	CAN	IPV	/	Medical records	Child advocacy center, emergency department, pediatric office	USA	Prospective observational	90 children	41.1%	<3 years old	42.2% Hispanic, 41.1% Black	Medical records	/	/	4
25. Tiyyagura et al. (2023)	CAN	IPV	/	Medical records	Emergency Department	USA	Non-probability (convenience)	2,890 abused children	44.3%	Mean age = 14.1 months	White 57.4%	Medical records	/	/	4
26. Gonzalez et al. (2020)	CAN	IPV	/	CIS-2008	Child welfare services	Canada	Probability (multistage)	2,184 children	50.0%	Mean age: 6.05 years (SD = 4.45)	63.9% White	Child welfare workers	/	/	5
27. Gregory et al. (2023)	CAN	IPV	/	Triage screening, Social work screening, PAST	Pediatric Emergency Department	USA	Non-probability (convenience)	329 caregivers	Not reported	Children: 1 month to 16 years, median age 8.7 months	Predominantly White	Caregiver report	/	/	5
28. Hultmann et al. (2022)	CAN	IPV	/	LITE questionnaire	Child and Adolescent Mental Health Services	Sweden	Non-probability (convenience)	578 patients	49.6%	9 to 17 years, mean age 13.49 years (SD = 2.45)	Predominantly Swedish, 61.4%	Self-report	PTSD	/	4
29. Gul et al. (2020)	CAN	IPV	/	Self-constructed	Online	Turkey	Non-probability (convenience)	336 women	100%	Mean age: 36.14 ± 8.23 years	Turkish	Self-report	/	/	4
30. Buffarini et al. (2021)	CAN	IPV	/	WHO; JVQ	Household	Brazil	Cohort study	3,533 mothers and 3,723 children	Not reported	Children mean age: 45.5 months (SD = 2.6)	Brazilian	Maternal report	/	Violent neighborhood, young maternal age, paternal antisocial behavior, absence of father at home,Maternal depression	5
31. Skafida et al. (2022)	CAN	IPV	/	Self-constructed	Household	UK	Probability (stratified random)	5,217 2004–2005 newborns	Not reported	Children aged 2 to 7 years, 10 months during first survey	Predominantly Scottish	Maternal and paternal reports, children’s self-reports	/	/	4
32. Lacey et al. (2021)	CAN	IPV	/	Self-constructed	Household	USA	Probability (multistage probability)	3,277 women	100%	Mean age = 42.5 years	African American and Caribbean Black	Self-report	/	/	4
33. Kita et al. (2021)	CAN	IPV	/	VAW (9);self-constructed for CAN (14)	Household	Japan	Non-probability (convenience)	822 pregnant women, 636 completed baseline, 445 completed follow-up	100%	Mothers mean age 32.45 ± 4.64 children mean age: 1 to 3 months postnatal	Japanese	Self-report	/	/	5
34. Ahmadabadi et al. (2018)	CAN	IPV	/	Self-constructed for IPV (7); CTQ	Household	Australia	Probability (cohort)	2,064 mothers and children	59.0%	Mothers mean age = 39.8 ± 4.9); children mean age = 30.3 ± 1.13)	Predominantly Caucasian	Self-report and mother report	/	/	5
35. Blangis et al. (2024)	CAN	IPV	/	ICD-10	Nationwide register (EPI-MERES)	France	Population Study	2,994 abused infants	Not specified	Newborn	100% born in France	Register data	/	/	4
36. Mielityinen et al. (2024)	CAN	IPV	/	CTSPC; self-constructed for IPV	School	Finland	Probability (two-phase stratified sampling)	6,825 children	51.4%	12 to 16 years	Predominantly Finnish	Self-report	/	/	5
37. Leppäkoski et al. (2021)	CAN	IPV	/	Self-constructed	Child health clinic	Finland	Non-probability (convenience)	8,720 children	72%	4 years old	Finnish	Parent report	/	/	4
38. Rebbe et al. (2021)	CAN	IPV	/	CPS records	CPS records	USA	Population study	305,867 unique report-child dyads	Not reported	Children under 5 years	22.0% White, 13.6% Black, 51.5% Hispanic	CPS reports	/	/	4
39. Song et al. (2022)	CAN	IPV	/	CTSPC; CTS2	Household	South Korea	Probability (multi-stage systematic)	5,688 married individuals	67.19%	Adults aged 19+	Korean	Self-report	/	/	5
40. Miranda et al. (2021)	CAN	IPV	/	Chilean JVQ (20)	School	Chile	Population Study	19,684 adolescents	50.4%	Grades 7 to 11 (~12–18 years)	Chilean	Self-report	/	/	5
41. Chen (2024)	CAN	IPV	/	CTS2; CTSPC	Household	USA	Probability (two-stage stratified cluster)	1,569 children	45.88%	Children 1to 6 mean 3.96 ± 1.59	46.26% White	Self-report	Children’s aggressive behavior, anxiety/depression	/	5
42. Black et al. (2021)	CAN	IPV	/	Maltreatment Assessment Form	Child welfare agencies	Canada	Probability (random sample of 17 child welfare agencies)	5,265 child welfare investigations	Not specified	Children aged 0 to 15 years	Not specified	Child protection workers (CPWs)	/	/	4
43. Langevin et al. (2021)	CAN	IPV	/	CTS2; CPS records	Clinical settings (specialized intervention centers)	Canada	Non-probability (convenience)	997 sexually abused children (3 to 14 years old) and their mothers	100%	Children aged 3 to 14 years	Not reported	Mothers, teachers/educators, CPS records	/	/	5
44. Ogbonnaya et al. (2019)	CAN	IPV	/	CTS2; CTSPC	Child welfare services	USA	Population study	965 mothers involved with child welfare	100%	Mothers; mean age 31.47 ± 0.31); children; mean age 7.63 ± 0.19)	53.75% White	Self-report and caseworker report	/	/	5
45. Carlson et al. (2020)	CAN	IPV	/	WHO; ICAST	School	Uganda	Probability (two-stage stratified)	535 adolescent-caregiver dyads	40.91%	Adolescents: 13.85 to 14.50; caregivers: 40.9 ± 10.8	Not specified	Caregiver and adolescent reports	Being female	/	5
46. Chan et al. (2023)	CAN	IPV	/	CTS2; CTSPC	Refuge centers	Hong Kong	Non-probability (convenience)	260 battered women	100%	Women: 40.6 ± 11.4; children: 7.39 ± 4.62)	Predominantly Chinese	Self-report	Being younger, unemployed, chronic disease	/	5
47. Dong and Wang (2019)	CAN and IPV; CAN and EA; IPV and EA			HSEAST	Household	USA	Population study	3,157 older adults	57.9%	60 years and older mean = 72.8 ± 8.3	Chinese	Self-report	/	/	5
48. Jewkes et al. (2019)	IPV	ILA	/	WHO; self-constructed for ILA	Household	Afghanistan	Non-probability (volunteer)	1,463 women	100%	18 to 48 years	Afghan	Self-report	Depression, suicidal thoughts, PTSD, disability	/	5
49. Choi (2022)	IPV	ILA	/	CTS2	Household	Hong Kong	Probability (Stratified random)	260 women	100%	18 to 59 years	Chinese	Self-report	/	/	5

Note. F-Poly = Family polyvictimization; C-Poly = Child polyvictimization; CAN = child abuse and neglect; IPV = intimate partner violence; EA = elder abuse; ILA = in-law abuse; Q = quality score. ACE = adverse childhood experiences; ALSPAC = Avon Longitudinal Study of Parents and Children; CIS-2008 = Canadian Incidence Study of Reported Child Abuse and Neglect; CPS = Child Protective Services; CTQ = Childhood Trauma Questionnaire; CTS2 = Conflict Tactics Scale; CTSPC = Conflict Tactics Scale Parent-Child version; FPS = Family Polyvictimization Screen; HSEAST = Hwalek-Sengstok Elder Abuse Screening Test; ICAST-CH = ISPCAN Child Abuse and Neglect Screening Tool-Child: Home version; ICAST-TC = ISPCAN Child Abuse Screening Tool-Trial scale; ICD = International Classification of Diseases; JVQ = Juvenile Victimization Questionnaire; LITE = Lifetime Incidence of Traumatic Events; OIS-2018 = Ontario Incidence Study of Reported Child Abuse and Neglect 2018; PAST = Psychosocial Assessment Screening Tool; PTSD = post-traumatic stress disorder; SD = standard deviation; VAW = violence against women; WHO = Multi-Country Study on Women’s Health and Domestic Violence against Women.

Table 1 Column “measures” shows that the measures used across studies to assess family polyvictimization and related constructs include a combination of standardized tools and self-constructed instruments. The most commonly used measures across the studies include the JVQ and its variants (e.g., JVQ-R2, Chilean JVQ), which were employed in 14 studies. The JVQ assesses exposure to multiple forms of victimization, including physical abuse, sexual abuse, peer/sibling victimization, and witnessing family violence (e.g., “Has anyone ever hit or attacked you on purpose with a stick or other hard object?”). The Conflict Tactics Scales (CTS2) and its Parent-Child version (CTSPC) were employed in 10 studies. These tools measure interpersonal violence in families, such as physical assault, psychological aggression, and negotiation strategies (e.g., “How often did you push or shove your partner during an argument?” or “How often did you spank your child with your hand?”). The International Child Abuse Screening Tool (ICAST-CH and ICAST-TC) was used in four studies. ICAST measures physical, emotional, and sexual abuse, as well as neglect (e.g., “Has someone ever deliberately burned you with a cigarette, lighter, or something else hot?”). A smaller number of studies relied on self-constructed measures tailored to specific topics: Choi (2022) and Jewkes et al. (2019) used self-constructed tools to assess ILA in Afghanistan and Turkey, respectively. Ahmadabadi et al. (2018) and Langevin et al. (2021) developed instruments for IPV and its overlap with child abuse. Items in these measures included culturally specific questions, such as “Has your mother-in-law ever prevented you from leaving the house without permission?.” Additionally, three studies incorporated medical records (Chen, 2024; Tiyyagura et al., 2023) to provide objective data on injuries and abuse-related healthcare visits. Population-based studies such as ALSPAC (UK) and OIS-2018 (Canada) used comprehensive longitudinal datasets to assess adverse childhood experiences and family polyvictimization.

The studies are conducted across various global regions, including North America (USA, Canada, Mexico), South America (Brazil, Chile, Colombia, Peru), Europe (Portugal, Spain, UK, Romania, France, Finland, Sweden), Asia (mainland China, Hong Kong, Japan, South Korea, Saudi Arabia, Afghanistan, Turkey), Africa (Republic of Congo, Uganda), and Oceania (Australia). Compared to the previous meta-analysis, this review does not include studies from Vietnam, Taiwan, Tajikistan, or India, but incorporates 19 additional countries across various global regions, including North America (Mexico), South America (Brazil, Chile, Colombia, Peru), Europe (Portugal, UK, Romania, France, Finland, Sweden), Asia (Japan, South Korea, Saudi Arabia, Afghanistan, Turkey), Africa (Republic of Congo, Uganda), and Oceania (Australia). This expanded geographical distribution highlights the growing global recognition of family polyvictimization as an important issue across diverse cultural contexts.

Sample sizes ranged from small community-based studies to large-scale national surveys, offering a broad scope of representation. Participants included various age groups, such as children, adolescents, adults, and older adults, with gender representation generally balanced, though some studies focused on specific groups, such as women or children. The informants recruited across studies include a mix of proxy reports in 5 studies (e.g., parents, caregivers, maternal and paternal reports, child welfare workers, teachers, and caseworkers), self-reports in 31 studies (from children, adolescents, or adults), and official records in 9 studies (e.g., medical records, CPS reports, register data). The studies employed a mix of sampling methods, including both probability and non-probability approaches. Among these, convenience sampling was the most frequently used method in 18 studies. Probability-based methods, such as stratified, multistage, or cluster sampling, were also commonly employed in 21 studies to ensure representativeness. A smaller subset of studies used population-based sampling techniques, including birth cohorts and large-scale population studies.

Overall Synthesis of Co-occurrence Rates

The current meta-analysis (Current MA) comprises 49 studies, among which 1 study (Dong & Wang, 2019) reported 3 estimates including CAN and IPV, IPV and EA, and CAN and EA. Figure 2 and Table 2 show that these 49 studies produced 51 estimates in the overall synthesis, reporting an overall event rate of 27.0% (95% CI [22.5, 32.1]), with significant heterogeneity (Q = 48084.249, p < .001, I² = 99.896). When combined with the previous meta-analysis, which reported an event rate of 18.3% [14.3, 23.2] (Chan et al., 2021), across a total of 89 estimates, the overall event rate decreases slightly to 23.1% [20.0, 26.6]. This analysis also shows significant heterogeneity (Q = 54,911.015, p < .001, I² = 99.840).

Figure 2.

Co-occurrence rates of family violence among the included studies. Heterogeneity: Q = 48084.249, df = 50, p = .000, I² = 99.896.

Table 2.

Event Rates of Different Types of Polyvictimization.

MA Version	No. of estimates	Event rate % [95% CI]	Heterogeneity
MA Version	No. of estimates	Event rate % [95% CI]	Q	df	p	I ²
Current MA
Overall	51	27.0 [22.5, 32.1]	48084.249	50	<.001	99.896
Combined MA
Overall	89	23.1 [20.0, 26.6]	54911.015	88	<.001	99.840

Note. MA = meta-analysis; CI = confidence interval.

Subgroup Comparisons of Rates of Polyvictimization

Table 3 shows the subgroup comparison using a total of 89 estimates (combined MA). The event rate of family polyvictimization is 11.3% (95% CI [1.8, 46.8]), based on 3 estimates, with heterogeneity (Q = 781.923, p < .001, I² = 99.744). Similarly, for child polyvictimization, the event rate is 35.1% [22.5, 50.1], based on 19 estimates, with heterogeneity (Q = 17565.631, p < .001, I² = 99.898).

Table 3.

Subgroup Comparisons of Event Rates of Polyvictimization.

Subgroup	Subgroup	No. of estimates	Event rate % [95% CI]	Heterogeneity
Subgroup	Subgroup	No. of estimates	Event rate % [95% CI]	Q	df	p	I ²
Types of co-occurrence	Family polyvictimization	3	11.3 [1.8, 46.8]	781.923	2	<.001	99.744
	Child polyvictimization	19	35.1 [22.5, 50.1]	17565.631	18	<.001	99.898
	CAN and IPV	59	22.7 [19.8, 25.9]	25158.053	58	<.001	99.769
	IPV and ILA	6	6.3 [3.2, 12.0]	160.402	5	<.001	96.883
	EA and IPV	1	48.8 [47.1, 50.5]	0.000	0	1.000	0.000
	CAN and EA	1	25.2 [23.7, 26.7]	0.000	0	1.000	0.000
Study design	Community	57	18.9 [15.0, 23.5]	43628.487	56	<.001	99.872
Study design	Clinical	32	31.7 [26.3, 37.6]	6748.508	31	<.001	99.541
Site	Household	41	18.5 [13.9, 24.2]	28597.086	40	< 001	99.860
	School	17	35.9 [27.8, 45.0]	8498.504	16	<.001	99.812
	CPS	17	30.1 [24.0, 36.9]	2365.739	16	<.001	99.324
	Hospital	8	10.9 [6.5, 17.7]	548.376	7	<.001	98.724
	Residential care	6	26.3 [10.5, 52.0]	322.226	5	<.001	98.448
Informant	Self	48	24.7 [19.2, 31.3]	39447.612	47	<.001	99.881
	Both	10	19.5 [13.3, 27.8]	2134.866	9	<.001	99.578
	Proxy	21	22.9 [17.0, 30.1]	4592.461	20	<.001	99.565
	Record	10	18.7 [12.7, 26.7]	1087.690	9	<.001	99.173
Measures	AAS	1	18.3 [14.3, 23.1]	0.000	0	1.000	0.000
	CTQ	1	4.9 [4.0, 5.9]	0.000	0	1.000	0.000
	CTS	24	20.4 [14.9, 27.3]	7869.861	23	<.001	99.708
	FPS	2	22.1 [4.0, 65.9]	220.935	1	<.001	99.547
	HSEAST	2	36.2 [17.0, 61.1]	366.843	1	<.001	99.727
	ICAST	5	48.8 [34.1, 63.7]	2197.614	4	<.001	99.818
	ICD	1	4.5 [3.8, 5.3]	0.000	0	1.000	0.000
	JVQ	21	28.0 [19.2, 39.1]	14013.739	20	<.001	99.857
	WHO	2	23.7 [1.9, 83.3]	385.465	1	<.001	99.741
	Others	30	21.3 [18.4, 24.5]	4837.938	29	<.001	99.401
Continent	Africa	3	37.2 [21.7, 55.9]	132.134	2	<.001	98.486
	Asia	24	16.1 [10.1, 24.8]	21342.571	23	<.001	99.892
	Europe	12	29.9 [22.6, 38.3]	2617.495	11	<.001	99.580
	Multiple	1	14.0 [13.7, 14.3]	0.000	0	1.000	0.000
	North America	42	24.1 [20.8, 27.7]	6660.586	41	<.001	99.384
	Oceania	1	4.9 [4.0, 5.9]	0.000	0	1.000	0.000
	South America	6	37.3 [17.8, 61.9]	6104.312	5	<.001	99.918

Note. CAN = child abuse and neglect; IPV = intimate partner violence; EA = elder abuse; ILA = in-law abuse; AAS = Abuse Assessment Screen; CPS = Child Protective Services; CTQ = Childhood Trauma Questionnaire; CTS = Conflict Tactics Scale; FPS = family polyvictimization screen; HSEAST = Hwalek-Sengstok Elder Abuse Screening Test; ICAST = ISPCAN Child Abuse and Neglect Screening Tool; ICD = International Classification of Diseases; JVQ = Juvenile Victimization Questionnaire; WHO = Multi-Country Study on Women’s Health and Domestic Violence against Women; CI = confidence interval.

For the co-occurrence of two types of family violence, the most common form is CAN and IPV. The event rate of CAN and IPV co-occurrence is 22.7% (95% CI [19.8, 25.9]), based on 59 estimates, with significant heterogeneity (Q = 25158.053, p < .001, I² = 99.769). For IPV and ILA, the event rate is 6.3% [3.2, 12.0], based on 6 estimates, with significant heterogeneity (Q = 160.402, p < .001, I² = 96.883). For EA and IPV as well as EA and CAN, only 1 estimate in each case was recorded.

Comparisons of Rates in the Characteristics of Studies

There are several study characteristics that show higher rates of co-occurrence, such as the investigation of clinical samples (31.7%, 95% CI [26.3, 37.6]), particularly recruited from CPS (30.1% [24.0, 36.9]). For the community samples (18.9% [15.0, 23.5]), the school-based studies showed the higher rates of co-occurrence (35.9% [27.8, 45.0]).

The informants do not seem to produce very different event rates. The rates obtained from the self-reported studies (24.7%), followed by proxy-only studies (22.9%), and both (19.5%) were comparable. The record-based study, which relies on reports from professionals after investigations, appears to have a slightly lower rate (18.7%), but this difference may not be significant. For the measures employed, there are no clear patterns related to the rates of co-occurrence.

Notably, while North America previously accounted for the majority of studies in our earlier analyses (Chan et al., 2021), this updated version includes more studies from other continents, particularly from Africa, South America, and Asia. These regions consistently report higher event rates of polyvictimization compared to North American studies. For instance, South America (37.3% [17.8, 61.9]) and Africa (37.2% [21.7, 55.9]) are the regions with the highest event rates. North America (24.1% [20.8, 27.7]) and Europe (29.9% [22.6, 38.3]) report intermediate rates, with Asia reporting a lower event rate in the combined analysis (16.1% [10.1, 24.8]). This shift in representation underscores the importance of including diverse geographical contexts, as studies from higher-rate regions may reflect differences in cultural, social, or systemic factors contributing to polyvictimization. Heterogeneity is significant across all continents (I² > 97%).

Odds Ratios of Factors Associated with Polyvictimization

Table 4 lists the ORs on several factors associated with polyvictimization. The factors are comparable in the ORs, with relatively higher values in disability (OR = 3.000, 95% CI [2.023, 4.448]) and violence approval (OR = 3.319 [1.897, 5.807]). However, there is a great variation among the studies investigating the factors. For each factor, only a few studies, ranging from 2 to 7, provided analyses. It makes it difficult to obtain widely studied factors for comparison.

Table 4.

Odds Ratios of the Factors Associated with Polyvictimization.

Associated Factor	No. of Estimates	Random Effect Size [95% CI]	Q	df	p	I ²
Absence of father at home	2	1.316** [1.109, 1.561]	0.162	1	.687	0.000
Age (older)	7	1.343*** [1.176, 1.534]	7.269	6	.297	17.462
Alcohol abuse	3	2.268*** [1.966, 2.616]	0.858	2	.651	0.000
Being female	5	1.480** [1.146, 1.911]	13.745	4	.008	70.899
Being male	6	1.372** [1.126, 1.671]	20.682	5	.001	75.824
Caregiver’s ACE	2	2.633** [1.437, 4.823]	5.214	1	.022	80.822
Depression	7	2.682*** [1.860, 3.867]	75.675	6	.000	92.071
Disability	2	3.000*** [2.023, 4.448]	0.159	1	.691	0.000
Paternal antisocial behavior	2	2.576* [1.161, 5.711]	18.112	1	.000	94.479
PTSD	4	2.647** [1.391, 5.036]	53.113	3	.000	94.352
SES below average	3	1.810*** [1.406, 2.331]	1.861	2	.394	0.000
Single parent	4	2.376*** [1.859, 3.035]	3.216	3	.360	6.705
Stress	2	1.726 [0.897, 3.318]	4.895	1	.027	79.570
Substance abuse	2	1.380 [0.772, 2.468]	1.243	1	.265	19.570
Violence approval	2	3.319*** [1.897, 5.807]	2.124	1	.145	52.913
Violent neighborhood	2	1.876*** [1.508, 2.335]	0.003	1	.953	0.000
Young maternal age	2	1.859 [0.969, 3.564]	3.898	1	.048	74.348

Note. ACE = adverse childhood experiences; SES = socioeconomic status; PTSD = post-traumatic stress disorder; CI = confidence interval.

p < .05. **p < .01. ***p < .001.

Publication Bias

The funnel plot (Figure 3) shows a relatively symmetrical distribution of studies around the pooled effect size, with no substantial asymmetry indicative of publication bias. The Classic Fail-Safe N indicates that 1,023 additional studies with null results would be required to render the observed effects non-significant (p > .05). This high number, relative to the 51 observed studies, suggests that the results are robust against publication bias. Egger’s regression test resulted in a non-significant intercept (6.96, p = .206), suggesting no evidence of small-study effects. The Duval and Tweedie’s trim-and-fill method estimated that adjusting for two potentially missing studies would have a minimal impact on the pooled effect size, with the fixed-effects estimate changing from 25.86% to 25.79% and the random-effects estimate decreasing slightly from 27.05% to 25.09%. These small adjustments indicate that the results are stable and unlikely to be heavily influenced by publication bias. In all, the statistical tests and visual inspection of the funnel plot provide no strong evidence of publication bias, supporting the reliability of the meta-analysis results.

Figure 3.

Funnel plot of standard error by logit event rate.

Discussion

The updated meta-analysis includes a significantly larger sample size of 543,196 participants, compared to 99,956 in the previous meta-analysis, for the general synthesis of effect sizes (Table 5). The effect size in the current meta-analysis is larger than the previous one (27.0% vs. 18.3%). With the 87 studies providing 89 estimates, the overall event rate of co-occurrence is 23.1%. The increase in the co-occurrence rate may be related to the geographical regions where the studies were conducted. While North America accounted for the majority of studies in our earlier meta-analysis (Chan et al., 2021), this updated version includes more studies from other continents, such as Africa and South America. These regions consistently report higher event rates of polyvictimization compared to North American studies. The differences in co-occurrence rates between regions are challenging to explain definitively but may be influenced by variations in family structures. For instance, regions with a higher prevalence of extended family households, including grandparents or other relatives living in the same household, may report more types of family violence. This could increase the likelihood of identifying and reporting co-occurring victimization, contributing to higher event rates.

Table 5.

Summary of Critical Findings.

Category	Critical Findings
Overall event rates	• Among the 51 new estimates, the overall event rate was 27.0%. Combined with the previous meta-analysis (18.3%), the overall event rate across 89 estimates decreases slightly to 23.1%.
Subtypes of co-occurrence	• Family polyvictimization event rate: 11.3%. Child polyvictimization event rate: 35.1%, primarily measured using JVQ/JVQ-R2.• CAN and IPV co-occurrence: 22.7%. IPV and ILA co-occurrence: 6.3%.
Geographical differences	• This updated version includes more studies from other continents, particularly from Africa, South America, and Asia. Higher co-occurrence rates observed in Africa and South America compared to North America.
Clinical versus general populations	•Co-occurrence rates: 31.7% (clinical samples) versus 18.9% (general population).
Informants and measures	• Type of informants and measures did not significantly affect event rates for screening purposes.

Note. IPV = intimate partner violence; CAN = child abuse and neglect; ILA = in-law abuse.

In North America, the nuclear family model is quite prevalent, particularly in the United States and Canada. A significant percentage of households consist of nuclear families, although there has been a rise in single-parent households and non-traditional family structures in recent years (Mitchell & Li, 2023; Organisation for Economic Co-operation and Development, 2016; Sassler, & Lichter, 2020). In South America, while nuclear families are common, extended family living arrangements are also prevalent (Cabib et al., 2023; Esteve et al., 2022; Reynolds et al., 2018). Cultural norms often emphasize close family ties, leading to a mix of nuclear and extended family households. In African countries, extended families are more common than nuclear families. This is influenced by cultural traditions that emphasize communal living and support systems (Nwanmuoh et al., 2024; Okeke-Ihejirika et al., 2022). Asia presents a varied picture. In countries like Japan and South Korea, nuclear families are common, especially in urban areas (Jiang et al., 2024). This is similar to Hong Kong, a metropolitan Chinese society. According to our most recent population study, 43.2% of Hong Kong families live with at least one grandparent in the same household (Chen & Chan, 2025). In South Asia and parts of Southeast Asia, extended families are more prevalent due to cultural norms and economic factors (Dommaraju, 2023; Lajom & Sibunruang, 2024). While nuclear families are a common household structure globally, the prevalence varies widely depending on regional, cultural, economic, and social contexts. We do not intend to draw a conclusion and raise the possibility of examining the family structure, rather than focusing on parents and children, when investigating multiple types of family violence.

Notably, there is a mild increase in the number of studies on family polyvictimization, with an event rate of 11.3% (95% CI [1.8, 46.8]). The number of studies on child polyvictimization has increased significantly, with an event rate of 35.1% [22.5, 50.1]. Our findings show that the prevalence of family victimization co-occurrence is higher in clinical samples compared to the general population (31.0% vs. 18.9%). This result highlights the greater likelihood of identifying a second victim in the same family once the first victim comes to the attention of health or social services.

Sub-group analyses indicate that the type of informants and measures do not seem to affect the event rates. Most studies reported the occurrence of victimization for screening purposes, allowing for the computation of co-occurrence rates. While variations in the types of informants and measures do not impact screening, this does not imply that they have no effect for assessment purposes, which may depend on the quality of responses.

Limitations and Future Research

One of the study objectives is to investigate the correlates of family polyvictimization. Several factors have been identified as significantly associated with polyvictimization. However, the variations in ORs are not significant, likely due to the considerable variation among the studies examining these factors. For each factor, only a few studies, ranging from 2 to 7, provided analyses, making it challenging to identify widely studied factors for comparison. Therefore, interpretation should be cautious when comparing the ORs, as the differences may not reliably indicate the relative importance of the factors.

Among the various types of family polyvictimization, the major proportions are found in child polyvictimization (21%) and the co-occurrence of CAN and IPV (66%). Child polyvictimization is primarily measured using the JVQ or JVQ-R2, which can include multiple types of family victimization, such as child maltreatment, sibling violence, witnessing domestic violence, and witnessing a parent’s assault on a sibling. The JVQ can potentially be coded for family polyvictimization if perpetrators are specified and multiple victims are identified (e.g., child, sibling, or parents). However, it cannot identify other victims like the elderly or adult children. Studies using the JVQ do not typically code for multiple victims. Some items are grouped into broader categories, such as peer or sibling, and “parent” in Module G of JVQ-R2 may refer to parents, stepparents, parents’ intimate partners, foster parents, or grandparents. To maximize the JVQ’s potential in measuring family polyvictimization, researchers may need to adjust item responses to align with their research goals.

The investigation of the co-occurrence of CAN and IPV is understandably more manageable as it focuses on the nuclear family structure. However, families often live in extended arrangements, not only in South American, African, or Asian cultures. In the United States, for example, African American grandparents are underrepresented in studies (Kelley et al., 2013). Investigations should consider the dynamics of family members who play a role in the occurrence of CAN or IPV. If CAN occurs in the context of family polyvictimization, welfare plans should be tailored to engage other family members to protect child victims and ultimately end family violence.

The strengths of this review include updating the prevalence estimates of family polyvictimization by incorporating a greater number of studies and covering larger, more diverse populations from various regions. As a result, the effect sizes computed are more representative of a broader population. The findings also indicate that the prevalence of overlapping family victimization is higher in clinical samples, underscoring the increased likelihood of identifying a second victim within the same family once the first victim comes to the attention of health or social services.

Implications for Research, Practice, and Policy

Future research should prioritize the investigation of family-level risk factors for polyvictimization, with particular attention to diverse family structures and cultural contexts (Table 6). There is a need to adapt and expand existing assessment tools, such as the JVQ, to measure family polyvictimization beyond child victims, including elderly family members and adult children. Additionally, more studies are needed to address gaps in the literature concerning underrepresented populations, such as African American grandparents and extended family members. Examining the co-occurrence of victimization within extended family arrangements will provide a deeper understanding of the dynamics and patterns of family violence.

Table 6.

Implications for Practice, Policy, and Research.

Category	Implications
Research	• Investigate family-level risk factors for polyvictimization, focusing on diverse family structures and cultural contexts.• Adjust and expand the JVQ to measure family polyvictimization beyond children (e.g., elderly, adult children).• Increase studies on underrepresented populations, such as African American grandparents and extended family members.• Examine the co-occurrence of victimization in extended family arrangements to understand the dynamics of family violence.
Practice	• Health and social services should proactively screen for additional victims in families where one victim is identified.• Tailor welfare plans to engage extended family members in cases of family polyvictimization.
Policy	• Promote policies that support early identification and intervention for family polyvictimization, especially in clinical settings.• Develop culturally sensitive policies addressing varying family structures (e.g., extended vs. nuclear families).

Note. JVQ = Juvenile Victimization Questionnaire.

In practice, health and social service providers should proactively screen for additional victims whenever a single victim is identified within a family. This approach can help uncover hidden cases of polyvictimization and ensure that all affected individuals receive appropriate support. Furthermore, welfare plans should be tailored to engage extended family members, recognizing their potential roles in both the risk and protection of family members in cases of polyvictimization.

At the policy level, there is a pressing need to promote policies that support the early identification and intervention of family polyvictimization, particularly within clinical settings where the risk is higher. Policymakers should also focus on developing culturally sensitive policies that address the needs of families with varying structures, such as extended versus nuclear families, to ensure that interventions are both effective and inclusive.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work described in this paper was supported by a fellowship award from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. PolyU/SRFS2223-5H01).

ORCID iDs

Ko Ling Chan

Qiqi Chen

Author Biographies

Ko Ling Chan, PhD, is a chair professor at the Department of Applied Social Sciences at The Hong Kong Polytechnic University. He specializes in research and practice with family violence and child and family polyvictimization.

Qiqi Chen, PhD, is an assistant professor at the Department of Applied Social Sciences at The Hong Kong Polytechnic University. Her research interests encompass child and family polyvictimization, bullying and cyberbullying prevention.

References

*Ahmadabadi

Najman

J. M.

Williams

G. M.

Clavarino

A. M.

d’Abbs

Abajobir

A. A.

(2018). Maternal intimate partner violence victimization and child maltreatment. Child Abuse & Neglect, 82, 23–33.

*Al-Eissa

M. A.

Saleheen

H. N.

Almuneef

(2020). Examining the relationship between witnessing intimate partner violence and victimization among children in Saudi Arabia. Journal of Interpersonal Violence, 35(5–6), 1334–1350.

*Al-Eissa

M. A.

Saleheen

H. N.

Almuneef

Al-Sulaiman

AlBuhairan

F. S.

(2019). Poly-victimization among secondary high school students in Saudi Arabia. Journal of Child and Family Studies, 28, 2078–2085.

*Almeida

T. C.

Ramos

Brito

Cardoso

(2020). The juvenile victimization questionnaire: Psychometric properties and poly-victimization among Portuguese youth. Children and Youth Services Review, 113, 105001.

*Black

Saini

Fallon

Deljavan

Theoduloz

(2021). The intersection of child welfare, intimate partner violence and child custody disputes: Secondary data analysis of the Ontario incidence study of reported child abuse and neglect. Journal of Public Child Welfare, 15(4), 473–486.

*Blangis

Drouin

Launay

Miranda

Zureik

Cohen

J. F.

Weill

Dray-Spira

Chalumeau

(2024). Maternal, prenatal and postnatal risk factors for early child physical abuse: A French nationwide cohort study. The Lancet Regional Health–Europe, 4 2, 100921.

Borenstein

Hedges

L. V.

Higgins

J. P.

Rothstein

H. R.

(2010). A basic introduction to fixed-effect and random-effects models for meta-analysis. Research Synthesis Methods, 1(2), 97–111.

*Bott

Ruiz-Celis

A. P.

Mendoza

J. A.

Guedes

(2022). Correlates of co-occurring physical child punishment and physical intimate partner violence in Colombia, Mexico and Peru. BMC Public Health, 22(1), 2195.

Bridges-Curry

Newton

T. L.

(2024). Polyvictimization and emotion regulation in daily life: An ecological momentary assessment study of young adults. Psychological Trauma: Theory, Research, Practice, and Policy. Advance online publication. https://doi.org/10.1037/tra0001745

10.

*Brown

S. M.

Rhoades

G. K.

Marti

C. N.

Lewis

(2021). The co-occurrence of child maltreatment and intimate partner violence in families: Effects on children’s externalizing behavior problems. Child Maltreatment, 26(4), 363–375.

11.

*Brühl

Ward

C. L.

Lachman

J. M.

Foran

H. M.

Raleva

Baban

Heinrichs

(2023). Co-occurrence of intimate partner violence against mothers and maltreatment of their children with behavioral problems in Eastern Europe. Violence Against Women, 29(12–13), 2439–2463.

12.

*Buffarini

Coll

C. V.

Degli Esposti

Murray

(2024). Unique and shared risk factors for early childhood victimisation and polyvictimisation in a Brazilian population-based birth cohort. The Lancet Regional Health–Americas, 32, 100715.

13.

*Buffarini

Coll

C. V.

Moffitt

da Silveira

M. F.

Barros

Murray

(2021). Intimate partner violence against women and child maltreatment in a Brazilian birth cohort study: Co-occurrence and shared risk factors. BMJ Global Health, 6(4), Article e004306.

14.

Burden

Roddy

A. L.

(2024). The effects of polyvictimization and cultural connectedness on indigenous youth’s mental health and negative coping mechanisms. Victims & Offenders. Advance online publication. https://doi.org/10.1080/15564886.2024.2371476

15.

*Butala

Asnes

Gaither

Leventhal

J. M.

O’Malley

Jubanyik

Aydin

Tiyyagura

(2023). Child safety assessments during a caregiver’s evaluation in emergency departments after intimate partner violence. Academic Emergency Medicine, 30(1), 23–31.

16.

Cabib

Araos

Palma

Yopo Diaz

(2023). Long-term trajectories of intergenerational coresidence: Evidence of living arrangements patterns in Chile. Family Relations, 72(3), 838–856.

17.

*Carlson

Namy

Norcini Pala

Wainberg

M. L.

Michau

Nakuti

Knight

Allen

Ikenberg

Naker

Devries

(2020). Violence against children and intimate partner violence against women: overlap and common contributing factors among caregiver-adolescent dyads. BMC Public Health, 20, 124.

18.

Chan

K. L.

(2017). Family poly-victimization and elevated levels of addiction and psychopathology among parents in a Chinese household sample. Journal of Interpersonal Violence, 32, 2433–2452.

19.

*Chan

K. L.

Chen

C. K.

Chen

X. Y.

Tang

(2023). Who is at high risk for child abuse and neglect: Risk assessment among battered women using shelter services. International Journal of Environmental Research and Public Health, 20(1), 833.

20.

Chan

K. L.

Chen

(2021). Prevalence and correlates of the co-occurrence of family violence: A meta-analysis on family polyvictimization. Trauma, Violence, & Abuse, 22(2), 289–305.

21.

*Chan

K. L.

Chen

C. K.

(2019). Screening for multiple types of family violence: Development and validation of the family polyvictimization screen. Frontiers in Public Health, 7, 282.

22.

*Chen

Chan

K. L.

(2022). The effects of social and cultural factors on child poly-victimization in Hong Kong. Journal of Interpersonal Violence, 37(11–12), NP9277–NP9294.

23.

*Chen

Chan

K. L.

(2025). Early identification of multiple victims in a family: Family–based computation of prevalence. Journal of Interpersonal Violence. Advance online publication. https://doi.org.10.1177/08862605251353501

24.

*Chen

(2024). The effects of child maltreatment and exposure to intimate partner violence on child behavioral outcomes: An examination of the severity, frequency, and co-occurrence [Doctoral dissertation, Rutgers The State University of New Jersey, School of Graduate Studies].

25.

*Choi

A. W. M.

(2022). In-law psychological aggression and its impacts on women′ s health. Aggressive Behavior, 48(2), 163–172.

26.

Davis

R. C.

Petersen

Weisburd

Taylor

(2021). Updated protocol: Effects of second responder programs on repeat incidents of family abuse: An updated systematic review and meta-analysis. Campbell Systematic Reviews 17(4), Article e1200.

27.

Dommaraju

(2023). Transformation of the family structure in Southeast Asia: Trends and implications. In Tan

K. S. Y.

Chan

S. K. L.

(Eds.), Populations and precarity during the covid-19 pandemic: Southeast Asian perspectives (pp. 12–27). ISEAS Publishing.

28.

*Dong

Wang

(2019). Associations of child maltreatment and intimate partner violence with elder abuse in a US Chinese population. JAMA Internal Medicine, 179(7), 889–896.

29.

Esteve

Castro-Martín

Castro Torres

A. F.

(2022). Families in Latin America: Trends, singularities, and contextual factors. Annual Review of Sociology, 48(1), 485–505.

30.

*Falb

Blackwell

Hategekimana

J. D. D.

Roth

O’Connor

(2023). Preventing co-occurring intimate partner violence and child abuse in eastern Democratic Republic of Congo: The role of family functioning and programmatic reflections. Journal of Interpersonal Violence, 38(1–2), 183–211.

31.

*Fernández-Artamendi

Águila-Otero

Del Valle

J. F.

Bravo

(2020). Victimization and substance use among adolescents in residential child care. Child Abuse & Neglect, 104, 104484.

32.

Finkelhor

Hamby

Turner

Ormrod

(2011). The juvenile victimization questionnaire: 2nd Revision (JVQ-R2). Crimes Against Children Research Center.

33.

Finkelhor

Ormrod

R. K.

Turner

H. A.

(2007). Poly-victimization: A neglected component in child victimization. Child Abuse & Neglect, 31(1), 7–26. https://doi.org/10.1016/j.chiabu.2006.06.008

34.

*Gonzalez

MacMillan

Tanaka

Jack

S. M.

Tonmyr

(2020). Child welfare responses linked to subtypes of exposure to intimate partner violence: Evidence from the Canadian incidence study of reported child abuse and neglect. Journal of Interpersonal Violence, 35(13–14), 2607–2623.

35.

Goodlin

W. E.

Dunn

C. S.

(2010). Three patterns of domestic violence in households: Single victimization, repeat victimization, and co-occurring victimization. Journal of Family Violence, 25, 107–122.

36.

*Gregory

Fingarson

Pierce

M. C.

Budde

Lorenz

Charleston

Rosado

(2023). Examining cases of child physical abuse evaluations to identify opportunities to improve intimate partner violence screening in pediatric emergency departments. Journal of Interpersonal Violence, 38(21–22), 11429–11444.

37.

*Gul

Gul

Kara

(2020). Intimate partner violence (IPV) types are common among Turkish women from high socioeconomic status and have differing effects on child abuse and contentment with life. Northern Clinics of Istanbul, 7(4), 359–365.

38.

*Herbert

Feder

Gilbert

Powell

Howarth

Morris

(2023). Child maltreatment and parental domestic violence and abuse, co-occurrence and the effect on lifetime outcomes in the Avon Longitudinal Study of Parents and Children (ALSPAC). SSM-Population Health, 24, 101555.

39.

*Hultmann

Broberg

A. G.

Axberg

(2022). Child psychiatric patients exposed to intimate partner violence and/or abuse: The impact of double exposure. Journal of Interpersonal Violence, 37(11–12), NP8611–NP8631.

40.

*Jewkes

Corboz

Gibbs

(2019). Violence against Afghan women by husbands, mothers-in-law and siblings-in-law/siblings: Risk markers and health consequences in an analysis of the baseline of a randomised controlled trial. PLoS One, 14(2), Article e0211361.

41.

Jiang

(2024). Caregiving in Asia: Priority areas for research, policy, and practice to support family caregivers. Health Care Science, 3(6), 374–382.

42.

*Joh-Carnella

Livingston

Stoddart

Fallon

(2023, September). Child welfare investigations of exposure to intimate partner violence referred by medical professionals in Ontario: A uniquely vulnerable population?. Healthcare, 11(18), 2599.

43.

Kelley

S. J.

Whitley

D. M.

Campos

P. E.

(2013). African American caregiving grandmothers: Results of an intervention to improve health indicators and health promotion behaviors. Journal of Family Nursing, 19, 53–73. https://doi.org/10.1177/1074840712462135.

44.

*Kita

Chan

K. L.

Tobe

Hayashi

Umeshita

Matsunaga

Uehara

Kamibeppu

(2021). A follow-up study on the continuity and spillover effects of intimate partner violence during pregnancy on postnatal child abuse. Journal of Interpersonal Violence, 36(13–14), NP6904–NP6927.

45.

*Lacey

K. K.

Shahid

H. R.

Jeremiah

R. D.

(2021). Intimate partner violence and the role of child maltreatment and neighborhood violence: A retrospective study of African American and US Caribbean black women. International Journal of Environmental Research and Public Health, 18(5), 2245.

46.

*Langevin

Hébert

Wallace

(2021). The intersection of intimate partner violence and childhood sexual abuse in mother-child dyads. Child Abuse & Neglect, 120, 105218.

47.

Lajom

J. A. L.

Sibunruang

(2024). The rise of flexible working in Southeast Asia: Insights from Thailand and the Philippines. In Chan

X. W.

Shang

(Eds.), Work-life research in the Asia-Pacific: Implications for justice, equity, diversity, and inclusion (pp. 23–45). Springer Nature Switzerland.

48.

*Leppäkoski

Vuorenmaa

Paavilainen

(2021). Psychological and physical abuse towards four-year-old children as reported by their parents: A national Finnish survey. Child Abuse & Neglect, 118, 105127.

49.

*Méndez-López

Pereda

(2019). Victimization and poly-victimization in a community sample of Mexican adolescents. Child Abuse & Neglect, 96, 104100.

50.

*Mielityinen

L. A.

Ellonen

Fagerlund

(2024). Is physical intimate partner violence a risk factor for physical child maltreatment in a nationally representative sample of finnish school children? Journal of Interpersonal Violence, 40(17–18), 4325–4345.

51.

*Miranda

J. K.

Crockett

M. A.

Vera-Pavez

J. I.

(2021). The co-occurrence of intimate partner violence exposure with other victimizations: A nationally representative survey of Chilean adolescents. Child Abuse & Neglect, 117, 105046.

52.

Mitchell

B. A.

(2023). Diversity in Canadian families. The Changing Faces of Families, 15, 15–38.

53.

Nwanmuoh

E. E.

Dibua

E. C.

Friday

E. C.

(2024). Implication of extended family culture in African nations on youth development: Evidence from Nigeria. International Journal of Public Administration and Management Research, 10(2), 82–90.

54.

Okeke-Ihejirika

Punjani

N. S.

Salami

(2022). African immigrant’s women experiences on extended family relations. International Journal of Environmental Research and Public Health, 19(14), 8487.

55.

*Ogbonnaya

I. N.

Keeney

A. J.

Villodas

M. T.

(2019). The role of co-occurring intimate partner violence, alcohol use, drug use, and depressive symptoms on disciplinary practices of mothers involved with child welfare. Child Abuse & Neglect, 90, 76–87.

56.

Organisation for Economic Co-operation and Development. (2016). Family database: Family size and household composition. https://www.oecd.org/en/data/datasets/oecd-family-database.html

57.

*Peña Cárdenas

Pinto Cortez

Zamorano González

Villarreal Sotelo

Hernandez Rodríguez

G. I.

(2022). Victimization and polyvictimization in children and adolescents in a Violent City at Northern Border of Mexico. Journal of Aggression, Maltreatment & Trauma, 31(10), 1337–1358.

58.

*Pinto-Cortez

Guerra Vio

Barocas

Pereda

(2022). Victimization and poly-victimization in a national representative sample of children and youth: The case of Chile. Journal of Aggression, Maltreatment & Trauma, 31(1), 3–21.

59.

*Pinto-Cortez

Gutiérrez-Echegoyen

Henríquez

(2021). Child victimization and polyvictimization among young adults in Northern Chile. Journal of Interpersonal Violence, 36(5–6), 2008–2030.

60.

*Pinto-Cortez

Pereda

Álvarez-Lister

M. S.

(2018). Child victimization and poly-victimization in a community sample of adolescents in northern Chile. Journal of Aggression, Maltreatment & Trauma, 27(9), 983–1002.

61.

*Rebbe

Eastman

A. L.

Adhia

Foust

Putnam-Hornstein

(2021). Co-reporting of child maltreatment and intimate partner violence: The likelihood of substantiations and foster care placements. Child Maltreatment, 26(4), 431–440.

62.

Reynolds

S. A.

Fernald

L. C.

Deardorff

Behrman

J. R.

(2018). Family structure and child development in Chile: A longitudinal analysis of household transitions involving fathers and grandparents. Demographic Research, 38, 1777.

63.

Sassler

Lichter

D. T.

(2020). Cohabitation and marriage: Complexity and diversity in union-formation patterns. Journal of Marriage and Family, 82(1), 35–61.

64.

*Song

Yoon

Cho

(2022). The association between polyvictimization in childhood and intimate partner violence and child abuse in adulthood. Journal of Interpersonal Violence, 37(9–10), 6009–6033.

65.

*Suárez-Soto

Guilera

Pereda

(2018). Victimization and suicidality among adolescents in child and youth-serving systems in Spain. Children and Youth Services Review, 91, 383–389.

66.

*Skafida

Morrison

Devaney

(2022). Intimate partner violence and child maltreatment in Scotland–Insights from nationally representative longitudinal survey data. Child Abuse & Neglect, 132, 105784.

67.

*Tiyyagura

Christian

Berger

Lindberg

, & ExSTRA Investigators. (2018). Occult abusive injuries in children brought for care after intimate partner violence: An exploratory study. Child Abuse & Neglect, 79, 136–143.

68.

*Tiyyagura

Leventhal

J. M.

Crawley

Asnes

A. G.

(2023). Frequency of injuries in a voluntary program evaluating young children exposed to intimate partner violence. Child Abuse & Neglect, 144, 106385.

69.

*Wemmers

J. A.

Cyr

Chamberland

Lessard

Collin-Vézina

Clement

M. E.

(2018). From victimization to criminalization: General strain theory and the relationship between poly-victimization and delinquency. Victims & Offenders, 13(4), 542–557.

70.

*Zhou

Luo

Chang

(2024). Patterns of poly-victimization among early adolescents: A latent class and two-wave latent transition analysis. Psychology Research and Behavior Management, 17, 3341–3354.