  |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
26-November-2004, Vol 117 No 1206
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Violence against women in New Zealand: prevalence and health
consequences
Janet Fanslow, Elizabeth Robinson
Internationally, violence has become recognised as a
significant contributor to ill-health.1 In New
Zealand (NZ), this recognition has been accompanied by significant policy
attention. Reducing violence in interpersonal relationships is a priority
objective of the NZ Health strategy;2 and the
Ministry of Social Development is working on implementing
Te Rito, a family violence prevention
strategy.3
While these documents are framed to recognise the multiple
types of violence, both documents recognise that a significant proportion of
violence is directed at women, and that much of this violence occurs in the
context of intimate relationships.
Furthermore, these policy initiatives have been driven by
health consumers’ recognition of the importance of addressing physical and
sexual violence as a high priority for health
gain,4 by international research documenting
the health consequences of violence,5 lobbying
from non-governmental organisations (NGOs), and by NZ studies suggesting that
intimate partner violence is likely to be highly prevalent within
NZ.6
The present study reports on the conduct of a large-scale,
population-based study of NZ women, using an internationally standardised
questionnaire. It documents lifetime prevalence of violence against women, and
outlines some of the health consequences associated with violence by intimate
partners.
MethodsQuestionnaire
development/translation—The base questionnaire was developed by the
Core Technical Team of the WHO Multi-Country Study on Violence Against Women,
following extensive review of the literature and consultation with
experts.7 The 13-domain questionnaire was
reviewed by experts within NZ (researchers, governmental representatives, Maori
advisers, and advocates), who suggested minor modifications to increase its
appropriateness for the NZ context. The revised questionnaire, with 302 possible
items, was pilot tested to determine its understandability and acceptability by
NZ respondents. As Mandarin/Cantonese speakers were the largest group who could
not complete the questionnaire in English, the questionnaire was translated into
Simple Chinese.
‘Intimate partners’ included male current
or ex-partners that the women were married to or had lived with, or current
regular male sexual partners. Physical violence was defined as having been
slapped or had something thrown at them which could hurt them—or having
been pushed, shoved, or had their hair pulled (grouped as ‘moderate
violence’ for later analyses); and those who had been hit with a fist or
something else, had been kicked, dragged, or beaten up, had been choked or burnt
on purpose, or been threatened with (and/or had used against them) a gun, knife,
or other weapon (termed ‘severe violence’ in later analyses).
Sexual violence was defined as having experienced one
or more of the following acts: being physically forced to have sexual
intercourse when the woman did not want to; having sexual intercourse because
she was afraid of what her partner might do, or being forced to do something
sexual that she found degrading or humiliating. The SRQ is a validated
instrument used to screen for emotional
distress.8 At the conclusion of the interview,
all respondents were asked ‘I have asked you many difficult things. How
has talking about these things made you feel?’
Study
population—The study population was women aged 18–64 years,
who were usually resident in Auckland or one rural region (north Waikato), and
who resided in private homes.
Study
location—The Auckland urban area was defined by the Territorial
Authority Units (TLAs): Auckland City, Manukau City, Waitakere City, North Shore
City. The Waikato area consisted of the four TLAs: Hauraki, Matamata-Piako,
Waikato, and Waipa Districts.
Sampling
strategy—A population-based cluster-sampling scheme with a fixed
number of dwellings per cluster was used. The target sample size was 1480 in
each region (2,960 total), based on a prevalence estimate of 15%, an 80%
response rate, and design effect of 1.5. Meshblocks were the primary sampling
units (PSUs), and were used to provide starting points for the selection of
households. The probability that a PSU was included was proportional to the
number of dwellings in that PSU. The starting point consisted of a randomly
selected street and street number within each PSU, provided by Statistics NZ.
Interviewers approached (using a predetermined procedure) 10 households in each
PSU, beginning from the designated starting point
In Auckland, interviewers approached every
4th house, in the Waikato interviewers
approached every 2nd house. Non-residential and
short-term residential properties were excluded from the count. In households
with more than one eligible respondent, one woman was randomly selected, for
safety and confidentiality reasons. If the woman selected was available to talk,
consent was sought and an interview arranged, otherwise contact details were
obtained and further attempts made to set up an interview.
The households visited and the outcomes of all visits
were recorded. To maximise the chance of obtaining an interview, a minimum of
three return visits were made to each household at different times on different
days. In practice, some interviewers made
up to nine repeat
visits.
Data
management—All questionnaires were checked for completeness, and
participants were re-contacted to obtain missing data. All data were
double-entered in the Epi-Info software application, checked, and corrected if
necessary.
Analyses—The
sampling scheme was taken into account in all analyses (by using survey
procedures in SAS v9 software). Prevalences are presented with 95% confidence
intervals, and are presented separately for the two study
locations—because sampling was representative of those regions, rather
than representative of New Zealand as a whole.
Logistic regression models (including age, NZDep2001,
ethnicity, educational status, household income, and location) were used to
investigate the association between lifetime physical violence and health
outcomes. Interactions between the location and violence were investigated to
see whether the effect of violence on the outcome differed for the two
locations. (Except for hospitalisation, this was not found to be so, and the
data were analysed with the main effect of location included in the model.)
For analyses related to association between intimate
partner violence and health, ever-partnered respondents were grouped into three
levels:
Safety
and ethical considerations—The safety of respondents and
interviewers, and the confidentiality of information, were important
considerations in the collection of these data. All interviews were conducted in
private (no children over the age of 2 years were present), and all
participants, regardless of whether they disclosed abuse or not, were provided
with a list of support agencies. In addition, ethical and safety recommendations
for research on intimate partner violence (developed by the World Health
Organization [WHO], and approved by The Scientific and Ethical Review Group of
the UNDP/UNFPA/WHO/World Bank Special Programme of Research, Development and
Research Training in Human Reproduction) were strictly followed as part of the
conduct of the present study.9 Ethics approval
was granted by the Human Subjects Ethics Committee of the University of Auckland
(Ref number: 2002/199).
ResultsIn total, 6,174 addresses were
selected. Of these addresses, 57 did not have a dwelling (ineligible pre
contact). 784 (12.8%) of households refused to participate—or indefinitely
postponed, did not speak English or Mandarin/Cantonese, or were unable to be
contacted. Of the remaining 5,333 houses, 1563 did not have eligible women
(ineligible post contact).
From the 3,770 households with eligible women, 2,855 women
aged 18–64 years were interviewed. In Auckland, 1,411 interviews (98%)
were conducted in English, and 29 (2%) were conducted in Mandarin or Cantonese.
All interviews in north Waikato were conducted in English. An 88.3% household
response rate, and 75.8% eligible woman response rate was obtained, resulting in
an overall response rate of 66.9%.
Table 1 presents demographic characteristics of New Zealand,
its regions, and the sampled area. It indicates that the distribution of ages in
our sample differed slightly from the population. This distribution is an
artefact of the sampling strategy, in which only one woman per household was
selected. Sample percentages for the older age groups, ethnicity, and marital
status were comparable with the regional distributions.
At least one act of physical violence inflicted by
non-partners in their lifetime was reported by 15% of participants in the
Auckland area and 17% in north Waikato, while sexual violence by non-partners
was reported by 9% and 12% of women in Auckland and north Waikato respectively
(Table 2).
Table 1. Demographic characteristics of females in New
Zealand, its regions, and the sampled area
NA=not available; NZ=New
Zealand; AKL=Auckland; WAI=Waikato.
Table 2. Physical and
sexual violence by non-partners reported by all women*
*Greater than or equal to 15
years old.
Of the 2,855 women who completed the full questionnaire,
2,744 were ever partnered and 111 had never had partners. Of the 2,744 women who
were ever partnered, 67 did not have current partners but did have(in the
past) a male partner they did not live with. Three cases had missing data
related to the partner status. As a result, we report data from a total sample
of 2,674 ever partnered women.
Thirty-three percent of participants in Auckland, and 39% in
north Waikato, reported that they had experienced at least one act of physical
and/or sexual violence by an intimate partner in their lifetime. Experience of
physical and/or sexual violence by an intimate partner within the previous 12
months was reported by approximately 5% of respondents (Table 3). Of those who
had experienced moderate or severe physical violence, 42.4 % (n=362), had also
experienced sexual violence.
Table 3. Prevalence of intimate partner violence
reported by ever partnered women
*5 women chose not to answer
questions on IPV (intimate personal violence).
Compared with women who had not experienced physical
violence by a partner, women with a lifetime experience of moderate or severe
physical IPV were significantly more likely to have consulted a healthcare
provider within the previous 4 weeks because they themselves were sick. Of these
women, 75% had consulted a general practitioner, and 16% had consulted a
pharmacist.
In Auckland, women who had experienced severe violence were
more than twice as likely to have been hospitalised within the previous 12
months compared with women who had not experienced any physical violence (Table
4).
Compared with women who had not experienced physical
violence by a partner, women who had experienced moderate physical violence were
over 2.5 times more likely to report current symptoms of emotional distress and
suicidal thoughts in their lifetime, while women who had experienced severe
physical violence were almost 4 times more likely to report these
effects.
Suicide attempts were also more common for those who had
experienced physical IPV compared with those who had not (moderate violence: 3
times more likely; severe violence: almost 8 times more likely) (Table 5).
Lifetime experience of intimate partner violence was
significantly associated with a range of current (within the past 4 weeks)
effects on health, including: self-perceived poor health, problems with
activities of daily living, and other physical health indicators. A
‘dose-response’ effect was noted, with women who reported
experiencing more severe physical violence by an intimate partner having
stronger risks of current ill-health than women who experienced moderate
physical violence by an intimate partner. However, even the group who had
reported experiencing ‘moderate’ physical violence were at
significantly elevated risk of health problems, compared with women who had not
experienced physical violence by an intimate partner (Table 6).
Table 4. Women who had
contact with healthcare professionals, or were hospitalised
* ‘No physical
violence’ group contains a small proportion of women who had experienced
sexual violence (n=101, 5.6%); †Logistic regression models included age,
NZDep2001, ethnicity, educational status, household income. Model for
“consulted health professionals in last 4 weeks” also included
location.
Table 5. Mental health effects of violence on
women
SRQ=self-reporting questionnaire; * ‘No physical
violence’ group contains a small proportion of women who had experienced
sexual violence (n=101, 5.6%); †Logistic regression models included age,
NZDep2001, ethnicity, educational status, household income, and
location.
Table 6. Associations of lifetime physical violence and
health outcomes reported by ever partnered women
* ‘No physical
violence’ group contains a small proportion of women who had experienced
sexual violence (n=101, 5.6%); †Logistic regression models included age,
NZDep2001, ethnicity, educational status, household income, and
location.
Lifetime experience of intimate partner violence was also
significantly associated with usage of medication (either prescription or
over-the-counter) within the past 4 weeks. Women who experienced moderate
physical violence or severe physical violence were both approximately twice as
likely to use medication to relieve physical or mental symptoms (Table 7).
Table 7. Associations of lifetime physical violence and
medication usage reported by ever partnered women
* ‘No physical
violence’ group contains a small proportion of women who had experienced
sexual violence (n=101, 5.6%); †Logistic regression models included age,
nzdep2001, ethnicity, educational status, household income and
location.
The acceptability of doing a survey on this topic was
demonstrated by the high proportion of women who reported feeling fine/good, or
the same after completion of the questionnaire. This demonstrates that, with
appropriate attention to staff training and safety and ethics considerations,
studies on this topic can be done in a way that does not contribute to stress
for the majority of women. The majority of those women who reported feeling
bad/worse indicated that it was difficult to re-visit previous bad experiences
(Table 8).
Table 8: How participants
felt after completing the survey
DiscussionThis study presents the results of
a recent population-based study of violence against women, conducted according
to an internationally developed standard of measurement and rigorous data
collection. Overall,
the results indicated that many women experience violence in a lifetime.
For those women aged 15 and over, at least one act of
physical violence inflicted by non-partners was reported by approximately 1 in 6
participants, while sexual violence was reported by approximately 1 in 10 women.
Approximately 1 in 3 ever-partnered women reported that they had experienced at
least one act of physical and/or sexual violence by an intimate partner, and
experience of physical and/or sexual violence by a current or previous intimate
partner within the previous 12 months was reported by approximately 5% of
respondents. Thus, these results indicate that the majority of violence against
women was perpetrated by current or former male partners.
These results concur with rates of intimate partner violence
reported by other studies, such as the New Zealand National Survey of Crime
Victims (NZNSCV), which reported that 26.4% of women had been physically abused
by an intimate partner in their lifetime, and 3.0% had experienced physical
violence by a current partner within the previous
year.13 The slightly lower rates obtained by
the NZNSCV may be due to inclusion of women
aged over 65 years, who may be less likely to disclose IPV, and/or
methodological differences (eg, use of a computer-based survey), and inclusion
of questions about IPV in a ‘crime’
context.14
Our results are also consistent with other NZ cohort
studies, such as the 1995 Hitting Home Survey, in which a nationally
representative sample of men reported that 35% had been physically violent to an
intimate partner in their lifetime.6
While causation cannot be determined from a cross-sectional
survey, the temporal relationship (ie, lifetime exposure and current health),
the strength of associations (odds ratios ranging from 1.3–7.6), and the
dose-response relationship between experience of moderate versus severe violence
all strongly support the notion of a causal link between IPV and ill-health in
women.15 Furthermore, the criteria of
plausibility and consistency are supported by numerous other studies that have
documented the health consequences of IPV.16
Collectively, the weight of this evidence supports the view
that lifetime experience of IPV is a major contributor to women’s
ill-health, and may underpin a broad range of health outcomes. Furthermore, when
combined with the information that approximately 40% of women with a lifetime
experience of IPV had presented to a healthcare provider (usually a GP) within
the previous 4 weeks, the findings have considerable implications for healthcare
delivery.
Healthcare providers, and GPs in particular, need to be
aware that substantial proportions of their female patients are likely to have
experienced IPV in their lifetime, and that such violence can have broad-ranging
health effects that are not restricted to injuries. While we have more to learn
about the best way for healthcare providers to respond, GPs and other healthcare
providers are likely to need skills in appropriately identifying current and
past victims of IPV (eg, through routine inquiry). Because of the high
co-occurrence of physical and sexual violence (42.4% of those women who
experienced physical violence had also experienced sexual violence), healthcare
providers may need to assess for both of these types of violence.
Additionally, while women who are currently victims of IPV
may require immediate referrals to specialist services for IPV or crisis support
services, recognition of the underlying connection between historical IPV and
current health is important, so that this can be discussed explicitly with the
client, and appropriate treatment and referral options that adequately address
the role of IPV can be agreed on.
The association between women’s experience of IPV and
increased use of medications may be understandable, given that women who
experienced intimate partner violence were also more likely to experience pain,
depression, and sleep problems. Thus, there may be circumstances in which
medications assist in the appropriate clinical management of symptoms associated
with these problems.
Women may also self-manage their health problems using
over-the-counter (OTC) medications. However, there are documented instances
where medications such as mild tranquilisers or pain medications are prescribed
for victims of IPV, yet have the potential to make her more vulnerable to
further assault.17 Unless prescription is
taking place in the context of physician knowledge about the client’s
experience of IPV, the principle of non-maleficence can be breached.
The reason for the regional differences in the association
between IPV and hospitalisation is unclear. One possible interpretation is that
the decade of advocacy and training work related to IPV that has been conducted
within the hospital and DHB systems in Auckland have contributed to increased
awareness of the health consequences of this type of violence, and have altered
response, at least for the more severe
cases.18-20 An alternative explanation is that
Waikato has an overall difference in service provision, reflected by generally
higher admission rates for all women. Further investigation is needed to
determine why these differential admission rates exist.
Limitations of the study include the use of a questionnaire
designed for assessment of health effects in developing countries, which did not
include all health indicators that might be relevant within a developed country.
Future papers planned from this study include analyses related to: other health
consequences associated with IPV (eg, injury, reproductive health consequences,
alcohol and drug use), emotional violence by intimate partners, violence by
other perpetrators inflicted on women as children and/or adults, and exploration
of the possible independence or interaction between physical and sexual violence
on health.
The prevalence rates from this study and the strong
associations with multiple physical and mental health effects suggest that
intimate partner violence may be as significant a factor as poverty in terms of
contributing to ill-health. As such, it warrants a considerable and sustained
investment in policy attention and other resources for prevention.
Ministry of Health initiatives (such as the Family Violence
Prevention Project, and the Toolkit for the prevention of interpersonal
violence) are important initial contributions to the field—but will
require sustained funding over time, and broad coverage across healthcare
settings, if they are to achieve their goal of ensuring that health care
providers’ adequately identify, assess, and respond to victims.
However, beyond facilitating better responses to victims,
the high lifetime prevalence of all forms of violence against women indicates
that we must direct serious effort to primary prevention of violence, and target
the perpetrators of violence. If our goal is to alleviate the health
consequences and other burdens of violence against women, we must work to
eliminate the violence.
Author information:
Janet L Fanslow, Social and Community Health, School of Population Health,
University of Auckland, Auckland; Elizabeth M Robinson, Biostatistician,
Biostatistics Unit, School of Population Health, University Of
Auckland.
Acknowledgements:
Funding for this project was provided by the Health Research Council of
New Zealand (Grant 02/207).
We gratefully acknowledge the women who participated in this
study, as well as the Project Manager: Cherie Lovell; Project Assistants: Clare
Murphy and Margaret (Meg) Tenny; Data Manager: Vivien Lovell; Auckland and
Waikato Interview Teams; and data entry staff. We also thank the Advisory Group,
who provided important support during the conduct of this study.
This study replicates the WHO Multi-Country Study on
Violence Against Women (WHO/EIP/GPE/99.3).
Correspondence:
Janet Fanslow, Social and Community Health, School of Population Health,
University of Auckland, Private Bag 92019, Auckland. Fax: (09) 303 5932; email:
j.fanslow@auckland.ac.nz
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|