Abstract

Objective: To determine prevalence rates of non-partner and partner violence (IPV) in men and women from a population-based study.

Methods: We recruited 2,887 randomly selected respondents (1,464 women and 1,423 men) from three regions of New Zealand between 2017 and 2019. Face-to-face interviews using a questionnaire adapted from the WHO multi-country study on violence against women was used for data collection.

Results: Physical violence by non-partners was most commonly experienced by men (39.9% lifetime exposure) compared with 11.9% of women. More women (8.2%) experienced lifetime non-partner sexual violence compared with men (2.2%). About 29% of men and women reported at least one act of physical-IPV in their lifetime, and about 12.4% of women and 2.1% of men reported at least one act of lifetime sexual IPV. More women than men reported serious injuries, fear, and physical and mental health impacts following IPV experience.

Conclusions: These findings indicate high prevalence of interpersonal violence exposure in the population, with marked gender differences in the types and impacts of violence reported.

Implications for public health: Study results call for the urgent implementation of violence prevention programs, and funding for both services to rehabilitate people who have perpetrated violence and services to support recovery of those affected.

Interpersonal violence is a global health issue and includes violence by strangers and acquaintances as well as intimate partner violence (IPV).1, 2 Understanding the prevalence and pattern of violence exposure at the population level is fundamental to developing and implementing public health policies and responses.3, 4 Frequently, the available evidence for understanding interpersonal violence utilises data from the most extreme cases,5-7 or data from administrative sources.8

Population-based surveys have emerged as the gold standard for assessing the prevalence of violence against women.9 However, as the primary purpose of these studies has been to assess IPV, they provide limited information on interpersonal physical violence by non-partners.10, 11 Other population-based violence surveys do not disaggregate prevalence data based on non-partner or partner violence.12 Additionally, many previous studies have focused on women of reproductive age,13 so there is a paucity of data on violence exposure among older adults.14

While studies frequently do not survey men,15, 16 studies that have explored gender differences in interpersonal violence tend to focus on IPV exposure. Population-based studies from high income countries have found relatively similar prevalence rates for physical and psychological IPV in men and women.13, 17-19 Others have noted that there are substantial gender differences in the frequency, severity and impact (including injuries and fear) of IPV. Women report experience of more frequent and more severe acts of physical violence,20 more sexual violence, and greater physical and mental health consequences.17 Some researchers suggest that men may experience more psychological IPV.21, 22

In this study, we utilised data from a population-based face-to-face survey to assess the prevalence of violence against women and men, by non-partners and by intimate partners. The study included adults aged 16 years and over. Types of interpersonal violence explored included: non-partner physical and sexual violence, and physical, sexual and psychological IPV. In addition, the study reports the gender of the person who perpetrated the violence, and, for IPV, compares the frequency, severity, and the self-reported impact of IPV on the physical and mental health of the respondents.

Methods

The 2019 New Zealand Family Violence Survey is a cross-sectional population-based study. Respondents were recruited from three regions: Waikato, Auckland and Northland for 24 months, between 2017 and 2019. Respondents were men and women aged 16 years and older who lived in the household for at least one month, slept four or more nights a week in the house, and could speak conversational English.

Ethics approval was received from The University of Auckland Human Participants Ethics Committee (reference number 2015/ 018244).

Sampling: Random sampling was conducted using primary sampling units (PSU) (using meshblock boundaries, the smallest geographical units used by Statistics New Zealand). Every second and sixth house was selected from a random starting point within each PSU. Rest homes, retirement villages, and short-term residential institutions were excluded. Women and men were recruited from different PSUs for safety reasons, and only one person per household was randomly selected as a respondent, to ensure the content of the interview was private. In households with more than one eligible resident, the participant was randomly selected from a random number sheet.

Included sample

Of 9,568 households approached, 1,532 were ineligible due to speaking a language other than English (n=110), vacancy or inaccessibility of dwelling (n=760), and absence of household members for an extended period (n=662). Of the remaining 8,306, 1,804 households refused to participate (22.2%). Of the 6,232 households who agreed to participate, 1,162 individuals were ineligible (n=1070) or incapacitated (n=92), or spoke a language other than English (n=109), and another 251 were not at home (despite up to seven visits from the interviewer). Of the remaining 4,710 eligible individuals, 1,767 (37.1%) refused, leaving 2,944 respondents (62.5%; n=1,495 women, and n=1,449 men) who agreed to participate (62.27%). Fifty-five respondents did not complete the interview, leaving 2,888 completed interviews (n=1,464, women, n=1,423 men, n=1 other [excluded from subsequent analyses]). Of these, 2,786 respondents were ever-partnered, defined as ever married, cohabiting, or currently in a sexual or dating relationship (n=1,431 women, n=1,355 men). Weighting variables were available for 2,746 ever-partnered participants (n=1,414 women, n=1,332 men).

Data collection

Data collection occurred through private face-to-face interviews, with no child over the age of two years present. Interviewers received training in the interview protocol, including safety and ethics considerations. At the conclusion of the interview, regardless of violence disclosure status, all respondents were provided with a referral card with contact details for local support services, so that they could access further support if required. Quality assurance included interview audits, regular meetings and review of completed interviews. Interviews took place in respondents’ homes or other private locations. All respondents provided written informed consent before the interview commenced, and were reminded of their right to refuse to answer or stop the interview at any time before the questions on violence experience commenced.

The questionnaire was based on the WHO Multi-Country Study on Violence Against Women (VAW).23 The 12-domain questionnaire included sections on respondents’ socio-demographic characteristics, current and previous partnership status, general health, any experience of violence including non-partner PV (two questions) and IPV (five questions for physical IPV, three for sexual IPV and four for psychological IPV). Impact of the violence experience was also assessed. The questionnaire was pre-tested with a convenience sample. Further details can be found in the methods paper.24

Outcome variable measures

Outcome variables were defined as a ‘yes’ responses to the following questions:

Non-partner violence (NPV).

NPV Physical Violence (NPV-PV): ‘Since the age of 15, has anyone ever hit, beaten or done anything else to hurt you physically?’ (for respondents with current or past partner: other than your partner).

NPV sexual violence (NPV-SV): ‘Since the age of 15, has anyone ever forced you to have sex or to perform a sexual act when you did not want to, for example, by threatening you, holding you down or putting you in a situation that you could not say no)?’ (for respondents with current or past partner: other than your partner).

IPV. To measure physical, sexual, and psychological IPV, respondents were asked if any current or previous partner had ever done any of the following acts:

IPV physical violence (IPV-PV): ‘Has any partner ever a) slapped you or thrown something at you that could hurt you? b) pushed or shoved you or pulled your hair? c) hit you with their fist or with something else that could hurt you? d) kicked, dragged or beaten you up? e) choked or burnt you on purpose? or f) threatened to use or actually used a gun, knife, or other weapon against you?’. The first two questions were categorised as ‘moderate’ and the last four questions were categorised as severe IPV-PV.

IPV sexual violence (IPV-SV): a) ‘Has any partner ever forced you to have sexual intercourse when you did not want to, for example by threatening you or holding you down?’ b) ‘Did you ever have sexual intercourse you did not want to because you were afraid of what your current or any other partner might do if you refused?,’ c) ‘Did your current partner or any other partner ever force you to do anything else sexual that you did not want or that you found degrading or humiliating?’

IPV psychological abuse (IPV-PA): ‘Has any current or previous partner ever: a) Insulted you or made you feel bad about yourself? b) Said or did something that made you feel humiliated in front of other people? c) Did things that made you feel scared or intimidated? d) Threatened to harm you or someone you care about? e) Destroyed things that are important to you?. We report on prevalence of two or more acts of IPV-PA to distinguish this from a one-off incident, as there is no consensus on how to measure IPV-PA.25 The first two questions were categorised as moderate PA acts and the last three questions were categorised as severe acts of IPV-PA.

Recency and frequency of IPV. Following each IPV question to which respondents gave a ‘yes’ answer, they were asked if 1) it had happened within the past 12-months, or before, to determine lifetime prevalence, and 2) whether it happened once, a few times, or many times. In addition, for each group of questions (IPV-PV, IPV-SV, IPV-PA), for those who reported at least one act, respondents were asked which partner did these things (current, most recent, previous). Multiple answers were possible.

Impact of IPV. Respondents who reported at least one experience of IPV-PV or IPV-SV were asked if the experience resulted in any injury, and whether they required clinical care or hospitalisation for this injury. Those who reported any experience of IPV were asked if these experiences affected their physical or mental health (two separate questions). Additionally, these respondents were asked if they had ever sought help from service providers, i.e. police, social services, legal advisors, hospitals/clinics, mental health services, general practitioners/primary care physician, courts, women's refuge/shelter, local leader, religious leader.

Perpetrator gender for NPV and IPV. For reported violence experienced, respondents were asked to indicate if the perpetrator/s were female/male/both.

Statistical analyses

Descriptive data analyses were performed in SAS statistical package version 9.5 (SAS Institute, Cary, NC, USA).

Analyses included all respondents who completed the questionnaire for NPV-PV and NPV-SV (n=2,887), and were limited to ever-partnered respondents for IPV outcomes (n=1,431 women, and n=1,355 men). Frequency, weighted percentages and 95% confidence intervals (95%CIs) were calculated for each gender separately. Chi-square test statistics (Fisher exact test) were used to compare prevalence rates between genders.

All analyses were conducted with survey procedures to allow for stratification by location (3 regions), clustering by primary sampling units, and weighting to account for the number of eligible participants in the household.

Results

Table 1 compares demographic characteristics of respondents with the whole NZ population from the 2018 Census. The survey sample had a smaller proportion of respondents in the 15–19 and 20–29 year age ranges, and more respondents aged 60–79 years old compared with the general population. Respondents had slightly higher individual and household incomes relative to the general population. Ethnic composition of the sample was closely comparable to the general population.

Table 1. 2019 NZ family violence study sample demographic characteristics compared with the general New Zealand population.

NZ population

Survey respondents

Gender

Female (%)

Male (%)

Total (%)

Female n (%)

Male n (%)

Total n (%)

1,927,086 (50.5)

1,888,929 (49.5)

3,816,015

1,445 (50.8)

1,400 (49.2)

2,845

Age groupsa

15–19b

20–29

30–39

40–49

50–59

60–69

70–79

≥ 80

7.9

17.4

16.5

16.6

16.6

13.6

6.2

5.2

8.3

18.4

16.2

15.9

16.0

13.2

8.3

3.7

8.1

17.9

16.4

16.2

16.3

13.4

7.2

4.50

41 (5.1)

138 (12.1)

224 (14.5)

253 (18.5)

277 (18.8)

265 (16.7)

184 (10.6)

63 (3.7)

55 (6.7)

149 (14.6)

212 (14.9)

260 (16.9)

276 (19.7)

235 (14.4)

150 (9.1)

61 (3.6)

96 (5.9)

287 (13.4)

436 (14.7)

513 (17.7)

553 (19.3)

500 (15.6)

334 (9.8)

124 (3.7)

Ethnicitya,c, a,c

Māori

Pasifika

Asian

MELAA

European

14.6

7.1

13.2

1.3

62.8

16.5

6.7

14.5

1.4

59.6

14.6

7.2

13.3

1.3

62.4

186 (14.5)

70 (7.7)

159 (12.2)

22 (1.3)

1008 (64.2)

127 (9.3)

88 (8.1)

213 (19.1)

22 (1.6)

948 (61.8)

313 (11.9)

158 (7.9)

372 (15.7)

44 (1.5)

1956 (63.0)

Individual incomea,d, a,d

Zero

<$25,000

$25,001–$50,000

$50,001–100,000

>$100,000

8.2

41.3

27.0

19.4

4.2

5.4

29.4

24.8

29.0

11.3

6.8

35.5

25.9

24.1

7.6

94 (7.2)

466 (35.7)

380 (26.9)

341 (25.3)

74 (4.8)

76 (6.2)

224 (18.9)

280 (21.6)

478 (33.6)

302 (19.7)

170 (6.7)

697 (27.1)

660 (24.2)

819 (29.6)

376 (12.4)

Household incomea,e, a,e

<$50,000

$50,000–$100,000

>$100,000

NA

NA

NA

NA

NA

NA

33.6

29.3

37.1

389 (29.9)

455 (36.1)

427 (33.9)

268 (20.3)

421 (31.8)

624 (47.9)

657 (24.9)

876 (33.9)

1051 (41.2)

Area level deprivation tertilesf

Least deprived

Moderately deprived

Most deprived

31.1

39.7

29.0

31.6

39.8

28.5

30.0

40.0

30.0

423 (27.2)

586 (39.3)

435 (33.5)

359 (24.4)

634 (44.3)

407 (31.3)

782 (25.8)

1220 (41.8)

842 (32.4)

Notes: a: Census 2018 for the population. b: Survey respondent's age groups started from 16 years whereas it is from 15 years in Census data. c: Available data for ethnicity in Census= 4,696,913 individuals. ‘Others’ are integrated into one of the five categories in the survey database whereas it is a separate category in Census data (comprised of 1.25% of males and 0.92% of females in the population). d: Available data for individual income in Census 2018 is from 3,755,730 individuals. e: Available data for household income in Census 2018 is from 1,526,958 households. f: 2013 IMD Indices of Multiple deprivation37 NA: Not Available

Lifetime prevalence rates for NPV-PV since the age of 15 years were 11.9% (95%CI=10.1, 13.7) for women and 39.9% (95%CI=36.7, 43.0) for men. Men were the main perpetrators of NPV-PV in both genders, accounting for 57.2% and 91.6% of reported experiences in women and men, respectively. Frequency: about 58% of women reported more than one NPV-PV experience, compared with 71% of men. For women, NPV-PV was mainly perpetrated by parents (36.0%), followed by relatives/family members (27.5%). For men, NPV-PV was mainly perpetrated by a stranger (52.4%), followed by someone from school (45.7) (Table 2).

Table 2. Reported prevalence of non-partner physical and sexual violence since the age of 15, the 2019 NZ family violence study.

Women

Men

p-value for Chi-square

Since 15 years

nc(%)

95% (CIs)

Since 15 years

nc(%)

95% (CIs)

Physical violence

178 (11.9)

10.09, 13.73

569 (39.9)

36.74, 43.08

<0.001

Gender of the perpetratora

Male

103 (57.2)

48.89, 65.53

525 (91.6)

89.00, 94.24

<0.001

Female

59 (31.7)

24.32, 39.14

13 (2.2)

0.85, 3.61

Both

14 (11.1)

5.28, 16.84

31 (6.1)

3.68, 8.43

How many timesa

Once

73 (42.4)

33.50, 51.26

163 (29.2)

25.30, 33.12

0.005

2–5

69 (37.6)

29.54, 45.69

298 (52.7)

48.26, 57.08

>5

35 (20.0)

13.08, 26.92

105 (18.1)

14.45, 21.78

Who did this to you?a,b, a,b

Parents

65 (36.0)

28.22, 43.82

55 (9.8)

7.00, 12.55

<0.001

Relatives/family

45 (27.5)

20.04, 34.93

45 (8.5)

5.71, 11.32

<0.001

A friend or neighbour

12 (8.0)

3.42, 12.70

47 (8.5)

5.87, 11.17

0.87

Someone at school or work

26 (13.7)

8.45, 19.04

258 (45.7)

40.93, 50.41

<0.001

Stranger/other

40 (22.7)

16.14, 29.36

305 (52.4)

47.62, 57.12

<0.001

Sexual violence

126 (8.2)

6.78, 9.65

30 (2.2)

1.35, 3.10

<0.001

Gender of the perpetratora

Male

110 (97.7)

93.17, 100.00

20 (69.2)

51.17, 87.29

–

Female

–d

–

9 (30.7)

12.71, 48.83

Both

∗

0, 6.82

–

–

How many timesa

Once

70 (55.8)

46.34, 65.25

12 (47.4)

26.90, 67.83

0.38

2–5

33 (29.7)

20.38, 39.03

8 (26.3)

7.89, 44.84

>5

16 (14.5)

7.18, 21.81

8 (28.6)

11.84, 45.30

Who did this to you?a,b, a,b

Parents

12 (11.7)

4.86, 18.58

–

–

NA

Relatives/family

22 (20.7)

11.94, 29.44

∗

0.24, 24.76

0.33

A friend or neighbour

30 (24.8)

16.34, 33.31

8 (35.0)

13.99, 50.01

0.34

Someone at school or work

17 (13.8)

6.98, 20.60

6 (17.5)

4.36, 30.64

0.60

Stranger/other

42 (31.7)

22.65, 40.79

13 (37.5)

20.37, 54.63

0.54

Notes: a: Percentages are calculated for those with an experience of physical or sexual violence b: Total % exceeds 100 as some experienced more than one event by different perpetrators. c: Weighted percentages and 95%Confidence intervals for percentages are calculated. d: There is no observation for cells with a -. e: For the sake of privacy, cells with fewer than 5 respondents were suppressed and shown as ∗.

The prevalence rate of NPV-SV since the age of 15 years was four times higher in women (8.2%, 95% CI=6.8, 9.6) compared with men (2.2%, 95%CI=1.3, 3.1). Men were reported as the perpetrators of 97.7% of reported NPV-SV against women, and 69.2% of NPV-SV against men. Strangers were the main perpetrators of NPV-SV, accounting for around 32% of reported NPV_SV experiences in women and 37.5% of it in men. For women, for 11.7% of NPV-SV a father/step-father was reported as the perpetrator. No men reported sexual violence from a father/step-father so the prevalence of sexual violence by these perpetrators cannot be estimated(Table 2).

Table 3 shows lifetime and 12-month IPV prevalence reported by ever-partnered men and women. The lifetime prevalence rates for IPV-PV were 28.0% for women (26.9% moderate, 7.6% severe) and 29.4% for men (28.2% moderate, 9.4% severe). Frequency: 64.7% of women versus 56.3% of men reported more than one experience of IPV-PV (p of difference<0.001). Past 12-month physical IPV prevalence rate was 2.4% (2.1% moderate, 1.3% severe) for women compared with 4.9% (4.8% moderate, 2.1% severe) for men.

Table 3. Reported Lifetime and past 12-month prevalence of physical or sexual violence by an intimate partner by gender, the 2019 NZ family violence study.

Since the age of 15

Past 12-month

Women

Men

Women

Men