Mental Health Effects of Intimate Partner Violence (IPV) Across Cultures: A Coherent Mechanism for SUICIDALITY

1
Mental Health Effects of Intimate Partner
Violence (IPV) Across Cultures A Coherent
Mechanism for SUICIDALITY

• Gregory Luke Larkin, MD, MS, MSPH, FACEP
• Professor of Surgery, Emergency Medicine,
  Public Health
• The University of Texas Southwestern Medical
  Center at Dallas
• Director of Academic Development
• Medical Director
• Violence Intervention Prevention Center
• Parkland Health Hospital System
• Chair, Section of Injury Control and Prevention
• American College of Emergency Physicians
• Dallas, Texas USA

2
Injury Related Mortality, World 1998
3
Leading Causes of Death 15-44 yrs.
Both Sexes Male Female High Income Countries Low/Middle Income Countries
HIV/AIDS HIV/AIDS HIV/AIDS Road Traffic Injuries HIV/AIDS
Road Traffic Injuries Road Traffic Injuries Self Inflicted Injuries Self Inflicted Injuries Road Traffic Injuries
Interpersonal Violence Interpersonal Violence TB Interpersonal Violence Interpersonal Violence
Self Inflicted Injuries Self Inflicted Injuries War Injuries Heart Disease Self Inflicted Injuries
TB TB Road Traffic Injuries HIV/AIDS TB
4
Leading Causes of Death 15-44 yrs. (continued)
AFRICA EUROPE South East ASIA/India CHINA USA
HIV/AIDS Road Traffic Injuries HIV/AIDS Self Inflicted Injuries Road Traffic Injuries
Interpersonal Violence Self Inflicted Injuries TB Road Traffic Injuries Self Inflicted Injuries
War Injuries Interpersonal Violence Road Traffic Injuries Liver Cancer HIV/AIDS
TB Heart Disease Self Inflicted Injuries Cerebrovascular Disease Interpersonal Violence
Road Traffic Injuries War Injuries Fires TB Heart Disease
5
Intimate Partner Violence (IPV)

• Largest contributor of the injury burden to women
  in the world (4.4 million adult women in US/yr)
• Over 10 billion/yr lost productivity, criminal
  justice interventions, physical mental
  healthcare cost annually
• Approx one of every 3 women experiences IPV in
  her lifetime (number varies cross cultures)
• High prevalence in the ED/AE setting
• Over 1 million AE visits/year
• 25 AE patients lifetime exposure
• Emphasis on screening identification
• Little focus on cross cultural opportunities to
  date

6
Asian Invasion IPV in US

• In US, Asian population growing at 4.5 every
  year
• Over represented in domestic violence cases
  example Massachusetts
• Over 10 of IPV cases are Asian
• Asians only 2.4 of population in Massachusetts

7
South Asian women, psychological distress
self-harm lessons for primary care trusts
(Chew-Graham et al., 2002)

• Needs of South Asian womem suffering distress
  mental health problems may lead to self-harm
  suicide
• Qualitative study (focus groups) of South Asian
  women in Manchester, UK
• Systemic risks for mental distress
• Social, political economic pressures
• Domestic violence
• Poverty
• Language problems
• Family childrens issues
• Health
• Women saw self-harm as a usable coping strategy
• Services accessed only at a point of desperation

8
Epidemiology of IPV in China?

• Chinese people comprise 21 of the worlds
  population (1.3 billion)
• Little is know regarding the prevalence, risk
  factors health outcomes of IPV in China
• Few studies have been conducted in China
• Current laws are inadequate for meeting the needs
  of women who are abused
• Studies of IPV in Chinese are needed to increase
  the Chinese publics recognition of his social
  problem its impact on womens rights health
• US State Dept estimates (1995) one-fifth of
  Chinese wives are beaten
• No original source or data cited

9
Violence Against Women in China

• Chinese custom requires that women submit to
  husbands
• Endure humiliation for the families children
• Dont wash your dirty linen in public
• Tradition for Chinese women to suffer
• Public rarely pays attention to issues of IPV

10
IPV in China

• Dramatic economic social changes since the late
  1970s
• Third World Womens Conference in 1985
• Fourth World Womens Conference in Beijing in
  1996
• 1993, a group at Chinese Womens College
  conducted face-to-face interviews with 30 female
  victims of violence in Pinggu
• But investigators did not release the results to
  Western society

11
Epidemiology of IPV in China

• Survey on marriage quality in Beijing, 21 of
  2,118 respondents indicated they had been beaten
  by their husbands (Sun 1995)
• Shanghai in 1992, 61.5 of 3,899 reported abuse
  cases were for intimates.
• 1999, Womens Studies Information Center in
  Beijing surveyed 4,000 women in 4 cities in
  China, 1,000 participants from each city

12
Epidemiology of IPV in China

• Masses attitudes toward IPV in China
• Four cities for inclusion
• The capital
• Southeastern coastal areas
• Areas with high ethnic minority density
• Northern poor rural areas
• Violent acts reportedly happened more frequently
  in the economically advanced SE region
• High minority density, high incidence of severe
  violence (wife in iron underpants)
• Investigators have not released the results to
  Western society (Sun, 2002)

13
Chinese Womens Position in Society Family

• Subordinate/Hierarchical
• Three Obediences
• Before marriage, woman follows obeys father
• After marriage, she follows obeys husband
• After death of husband, she follows obeys son
  (Arsella, DeVos, Hsu, 1985 Lum, 1988)
• China is a country with little individual
  privacy, but IPV is stubbornly protected as
  inviolably private is often overlooked
  ignored.

14
Societal Rationale/5 Common Reasons for IPV Among
Chinese (Xie, 1992 Xu et al., 2000)

• 1. Male chauvinism sexism used to make ones
  wife obey (Gallin,1992Xie, 1992)
• As less traditional economic gender role norms
  accompany increasing westernization, men may try
  to reassert their traditional prerogatives with
  violence (Gallin, 1992)
• 2. Women may be beaten for giving birth to a girl
  or not producing children (XU et al., 2000)
• Her fault/ her crime
• Bad progeny supports Husbands right to divorce
  or take a concubine
• More severe in after institution of one child per
  family policy in 1979
• 3. A man with a lover may use abuse/violence to
  force wife into divorce
• 4. Man has some vice, gambling or alcohol
• 5. Sex related

15
Chinese Literature on Prevalence of IPV in China

• Wuhan,1983 43 of 760 divorce cases wife
  initiated because of husbands chauvinism and/or
  abuse (Xie, 1992)
• Tianjin, 1984100 randomly selected divorce
  cases 41 IPV 1986, 51 IPV (wife beating)
• 1991, 106 divorced women interviewed (Pi, 1991)
• 46 women asked for divorce because they could
  no longer stand the beating endured from husband
  
• 70 reported being beaten by husbands at some
  time during their marriage
• 1997, Qindao Womens Federation (QDWF)
• Of 2,348 QDWF cases reviewed 1991-1997, 15
  related to IPV (Tan, 1997)
• 1996, Of 205 divorces 25.3 were caused by IPV
  (Tan, 1997)
• 1998, Voice of America (VOA)
• wife beating causes 70-80 of divorces in
  China.

16
Chinese Literature on Prevalence of IPV in China
(cont)

• Above statistics cannot be generalized
• divorce cases increased from 300,000 (0.3) in
  1980 to 1,113,000 (1.1) in 1996 (VOA 1998)
• None of the studies had operational definitions
  of domestic violence
• Peer-reviewed Hong Kong prevalence studies
  published (Leung, Leung, Lam, Ho, 1999
  Tang, 1994, 1999)
• Tang, 1994 N482 undergraduate students CTS
• 14 of parents used physical violence against
  each other
• 75 used verbal aggression during the prior year
• Tang, 1999
• 67 reported at least one incident of verbal
  aggression,
• 10 experienced at least one incident of physical
  aggression by husband/partner during prior year.

17
Chinese Literature on Prevalence of IPV in China
(cont)

• Leung et al (1999) Abused Assessment Screen (AAS)
  convenience sample of pregnant women (N631) OB
  clinic in Hong Kong
• first known IPV study in a health care setting in
  China
• 18 women reported history of abuse
• 5 reported IPV while pregnant
• 16 reported abuse in the year prior
• 9 reported sexual abuse in the year prior
• No assessment of AAS instrument cultural
  appropriateness
• Mainland China (Xu, Zhu, Campbell, 2000)
• only 3 sample of 30 women reporting physical
  violence,
• 10 reported forced sex,
• 7 reporting verbal abuse

18
Likely Explanations for Inaccurate IPV Estimates

• Socialized to accept physical emotional abuse
• Doctors and patients adverse to reporting
• Victims Fear of retaliation
• Not a priority yet, but that can change
• Media coverage
• Advocate by putting a face on the problem
• Start the process of gathering hard data

19
Two HK Population Studies That Identify Risk
Factors for IPV

• Leung et al.s (1999) study of pregnant Chinese
  women in Hong Kong found that unplanned pregnancy
  was one of the main risk factors for abuse
• Husbands/partners unemployed or manual workers,
  permanent local resident, immigrants were at risk
  for IPV
• Tang (1999) education level, difference in
  education, occupation, family, number of children
  were not related to various forms of wife abuse
  In Hong Kong
• Couples age, age difference, length of marriage
  marital satisfaction correlates wife abuse

20
Barriers to Obtaining Help

• Violence in home viewed as family matter
• Loss of face
• Leaving ones spouse can result in ostracism
• Loss of community respect, connection to only
  existing support system
• Conditioned to fear uniformed authorities
• Less likely to report accounts of IPV to the
  police
• Language /educational/literacy skills
• Trapped in violent potentially violent homes

21
Barriers to Obtaining Help (cont)

• Many are note even aware domestic violence is a
  crime because of cultural norms related to women
• Lack of community support services
• no womens shelters in China
• limited domestic violence hotlines
• Eg. Local Womens Federation
• Having affairs/lovers outside of home
• currently in fashion in China (as elsewhere!)
• increases spead of IPV as well as HIV/AIDS
  (Xu
  1999VOA, 1998)

22
Why Do IPV Research in China?

• IPV research has lead to important health
  public policy change in the west only (criminal)
• IPV largely ignored in China
• Prevalence of violence among Chinese populations
  not yet established
• Lack of culturally appropriate instruments for
  assessment of IPV against Chinese women,
  cross-cultural comparisons impossible
• Literature addressed the reasons IPV against
  Chinese women, no research has tested those
  reasons as risk factors
• Two studies from HK identified some
  sociodemographic risk factors of IPV but results
  were inconsistent and failed to assess factors
  unique to HK culture
• WHO/CDC others may support measures of
  prevalence, risk factors, health outcomes among
  different cultures

23
Key Points of Research Reviews

• Intimate partner violence (IPV) exits among
  Chinese, yet little known about extent or nature
  of IPV in Chinese either in China or US
• Chinese authorities have tended to overlook
  ignore IPV
• Violence against women by husband is considered
  one of the few categories of life protected as
  inviolably private
• In traditional Chinese families, the family
  structure is hierarchical husband assumes
  highest authority
• Even though literature has cited 5 societal
  rationales for IPV among Chinese women, not
  research studies have been conducted to test or
  validate those premises.

24
Key Points of Research Reviews (cont)

• Available data on IPV prevalence in Mainland
  China is mainly from physical abuse among divorce
  cases
• National data in US do not include
  Chinese-Americans as a separate group.
• Inconsistencies in sampling, design, IPV
  definitions, and measurements have made it
  difficult to compare IPV prevalence
  cross-culturally.
• Battered Chinese women face unique barriers to
  obtaining help
• Even though Western societies have reported
  adverse health outcomes associated with IPV for
  many years, but no study to date has examined
  health ramifications of IPV among Chinese women

25
Implications for Policy, Practice Research

• In order to affect policy change in China in
  terms of IPV womens rights issues, the Chinese
  society governments awareness of IPV must be
  increased
• IPV Research is central to increasing spousal
  assault awareness
• As well as research, public education through
  mass media educational courses is essential to
  increase public awareness of IPV
• Researchers need to investigate the unique
  culturally related issues (e.g. barriers, risk
  factors) of IPV among Chinese populations
• Culturally valid instruments with the same
  operational definitions of IPV are needed in
  order to make cross-cultural comparisons of IPV
• Health care providers need to screen for IPV,
  conduct culturally appropriate assessments of
  IPV, and measure mental physical health impact
  on patients.

26
SUICIDE A Worldwide Epidemic

• There is only one serious philosophical
  question, and that is suicide.
• Albert Camus, in The Myth of
  Sisyphus
• Definition Suicide is voluntarily and
  deliberately taking ones own life.

27
Suicide Global Burden

• Approximately 1M people per year
• Among top 10 causes of death in every country
• 8th in the US
• 1 of the 3 leading causes of death for 15-35 yr
  age
• Each suicide affects at least 6 other people
• 1.8 of the total world wide burden of disease
• 2.3 in high income countries (DALYs)
• 1.7 in low income countries
• Equal to burden due to wars homicide
• Twice the burden of diabetes

28
Suicide Epidemiology

• 5 overall attempt sometime during lifetime
  (Kessler 1999)
• 18 attempts per each completion
• 10-15 of attempters ultimately complete
• Estimated 12 month incidence is 1
• (Crosby et al 1999).

29
Suicide Epidemiology

• Top 5 cause of death 15 - 19 yrs.
• First or second in many countries
• Less common under 15 yrs but this younger group
  increasing
• Boys die more often than girls in most
  countries.... Except China
• Rural females outnumber males in China
• Female male approximately equal in urban China
• Elsewhere, boys use more violent means (hanging,
  guns, explosives)

30
Difference Between Attempters Non-attempters on
Risk Factors

• Compared with non-attempters, attempters
  reported
• Higher stressful events
• Childhood trauma
• Psychological distress
• Depression
• Hopelessness
• Alcohol use
• Drug use

31
Difference Between Attempters Non-attempters on
Protective Factors

• Attempters significantly lower scores on
• Hopefulness
• Self-efficacy
• Coping skills
• Family/friend support
• Effectiveness obtaining material resources
• Spiritual well-being

32
Risk Factors for Suicide in USA
Kaslow et al 2000

• Compared to non-attempters, African American
  attempters more likely to report
• Global psychological distress
• Symptoms of PTSD
• Hopelessness
• Drug abuse
• Relationship discord
• Childhood maltreatment
• Poor interpersonal conflict resolution
• Low level of self-efficacy

33
Wild Swans Three Daughters of China (Chang
1992)

• Jing Wang dreads her inevitable, arranged
  marriage to Rui Xiao. After an unsuccessful
  attempt to escape with the man she wishes to
  marry, the reluctant bride is held captive
  beaten by her prospective husband father-in-law
  who feel humiliation lose face at her
  rejection. After several days of torment, Jing
  Wang returns to her unsympathetic family which
  considers her renunciation to their traditional
  values customs unfilial. Later that week, Jing
  Wang is found dead an apparent suicide.
  Relatives describe suicide as her sole means of
  resistance her ultimate effort to be taken
  seriously.

34
The Suicide-IPV Link

• Up to 80 of female suicide attempters cite an
  abusive relationship as a factor in their
  decision to attempt suicide
• (Stark Flitcraft, 1996) In North America,
  abuse may be the single most important
  precipitant for female suicide attempts yet
  identified.
• Abused women are more likely than nonabused women
  to have a history of suicide attempts
• (Abbott et at., 1995 Bergman Bergman, 1991
  Kaplan et al., 1995 Roberts
• et al.1997 Stark Flitcraft, 1996)
• Between 35-40 attempt suicide at some point
• Battered women are more likely (20) than
  nonbattered women (8) to make multiple attempts
  (Stark Flitcraft, 1996)
• IPV victims who experienced one, two, or three
  forms of abuse were 1.8, 2.3, or 7.8 times more
  likely to attempt suicide

35
The Suicide-IPV Link (cont)

• Increased rates of having attempted suicide were
  evident among women who had been sexually or
  physically abused (Wiederman et al., 1998)
• Bryant Range (1995) found that young women with
  combined sexual physical abuse displayed the
  greatest degree of suicidality.
• Risk factors for suicide on African American
  women IPV victims depressive symptoms,
  hopelessness, drug abuse, child abuse
• Compared to women with no risk factors, women
  with 2, 3, 4-5 risk factors were 10, 25, 107x
  respectively, more likely to commit suicide
  (Thompson, 2002)
• Experiences of more forms of child abuse also
  increased risks of suicide 1.8, 2.3, 7.8x for
  each additional form of child abuse (Anderson,
  2002)
• Only other study outside USA was in Europe
• 82 of Greek women suicide attempters report
  being beaten by male partners
• 32 of Danish women suicide attempters report
  being beaten by male partners

36
Why Study IPV Cross Culturally?

• In US West, gap in suicide completion rates
  between men women is narrowing (Canto
  Lester1995)
• Partner abuse has only recently been investigated
  as a risk factor for suicidal behavior
• 21-34 women are physically assaulted by an
  intimate partner during adult life
• IPV-related depression, anxiety posttraumatic
  stress are major burdens
• Few investigations have examined the effects of
  partner abuse for ethnic Asian/Chinese women
• Both IPV suicidality are associated with
  smoking, drugs, alcohol, other high risk or
  harmful behaviors

37
Mechanisms of Suicidality in IPVIndependent
Correlates

• Traumatic experience (lifetime acute stress)
• Intermittent positive reinforcement punctuated by
  physical abuse (cycle of violence)
• Increase in platelet catechols, fragmented sleep,
  etc.
• Hopelessness (Walker 1984)
• Derived from a learned helplessness, frustration,
  relationship failure, role confusion, etc.
• Cognitive state and thus, modifiable
• Substance Abuse in IPV (Caetano, Larkin,
  others)
• Facilitates attempts lowers inhibitions,
  increases vulnerability to trauma
• Impulsivity/Impulsive Aggressiveness (Mann1997)
• Aggression is a basic factor of human
  personality, but in IPV and suicide, it is often
  turned inward on oneself.

38
IPV Associated Risks Outcomes

• Mental psychological problems
• Depression, anxiety, low self-esteem, learned
  helplessness, PTSD, substance abuse
• Battering common (battered wife syndrome)
• Injuries secondary to battering
• Headache, gastric problems, anxiety insomnia
• Suicidal thoughts, intentions suicide attempts
• IPV Homicide
• African women 6/100,000
• White women 1.4/100,000

39
Emergency Medicine Treatment of Potential Victims

• Screen for suicide IPV both!
• Risk/lethality assessment for both
• ED based treatment
• Brief intervention
• Safety planning
• Drugs
• Motivating follow-up/help-seeking kiosk
• Connect to community resources

40
Why Do Many Abused Women Attempt Suicide?

• Limited choices for coping
• Powerless depressed
• Abusers often control of every aspect of their
  lives
• Limited social networks material resources
• Vestige of power
• Get attention
• African American women who attempted suicide more
  likely than non-attempters to have an IPV history
  (48.8 vs 22.2)
• 2.5 times more likely to report physical abuse
• 2.8 times more likely to report nonphysical
  (emotional) partner abuse (Kaslow 1998, 2000)
• IPV-suicidal behavior link mediated by
  psychological distress, hopelessness, drug abuse
  moderated by social support

41
Risk Factors for Suicide in China
Phillips, et al, Lancet 2002 3601728-36

• BACKGROUND Suicide is the fifth most important
  cause of death in China
• 287,000 deaths per year
• WHO 1999 World Health Report suggests that
  suicide is Chinas fourth most important
  public-health problem
• Rural suicide rate is three-fold the urban rating
• More suicide deaths among women than men
• Different from those reported in other parts of
  the world

42
Risk Factors Suicide in China (cont.)
Phillips, et al, Lancet 2002 360
1728-36

• 8 significant predictors of suicide remained in
  the final unconditional logistic regression
  model.
• High depression symptom score
• Previous suicide attempt
• Acute stress at time of death
• Low quality of life
• High chronic stress
• Severe interpersonal conflict in 2 days before
  death
• A friend or associate with previous suicidal
  behavior
• A blood relative with previous suicidal behavior

43
Risk Factors Suicide in China (cont.)
Phillips, et al, Lancet
20023601728-36

• Risk Increased with exposure to multiple risk
  factors.
• Exposed to one or fewer risk factors died by
  suicide
• 30 with two or three risk factors
• 85 with four or five risk factors
• 96 with six or more risk factors died by suicide
• Risk factors for suicide do not differ greatly
• Preventative efforts should focus on individuals
  exposed to multiple risk factors

44
Suicide in Hong Kong

• Stable crude suicide rate
• High female suicide rates compared with west
• Decline in hanging, jumping most common method
• Charcoal/CO is emerging trends
• Yap(1958), Lo Leung(1985), Hau(1993), Ho
  et al(1995) Yip(1996,1997)
• Hong Kong more YPLLs from suicide than Beijing
  Taiwan
• Yip (1996)

45
Suicide in Hong Kong (cont)

• 496 in 1981 to 718 in 1995
• Proportion of total deaths 2 in 1981 to 2.4 in
  1995
• Crude suicide increased 20, 9.5 to 11.5 deaths
  per 100,000 1981-1995

46
Elderly Suicide in Hong Kong

• One of highest suicide rates in the world.
• 33 of all suicides
• 1 all deaths among persons aged 60 or above
• 12x higher for over 60 yrs highest over 75 yrs
• 4 to 5 times above average
• Male to female ratio 1.31
• (vs 31 in most countries)
• Single males higher than single females
• Married males higher rate than widowed
• Jumping has become increasingly common
• Winter months Chinese New Year lowest rates?

47
Elderly Suicide in Hong Kong (cont)

• Elderly suicide rates/100,000
• Never married 43
• Married 35
• Widowed 11
• Divorced/separated 22
• Rates closer to rural Beijing than urban Beijing
• Urban Beijing 13/100,000
• Rural 38/100,000
• Hong Kong rates higher than Austrailia, New
  Zealand USA

48
Suggestive Evidence of Partner Violence Risk in
Hong Kong Elderly

• Married females suicide rate higher than male,
• 42 vs. 31 per 100,000
• Married females rate gt never married females
• Widowed (both male female) lt than married
• Suicide rate for divorced female elderly less
  than married female
• 1995 four economically active vs. 213 inactive
  elders committed suicide (3 vs 30/100,000)

49
Suicide Risk Factors

• Not present in every case
• Vary from one country to another dependent on
  cultural, political, economic features
• Some genetic and some environmental
• Risk for attempts in the West
• Female
• Lesbian or gay
• Previously married
• Younger age
• Low education
• Unemployed (Kessler et al 1999)

50
Suicide Risk Factors Cultural and
Socio-demographic

• Low socioeconomic status/economic stress
• Poor education
• Unemployment in the family
• Recent change in economic or social status
• Indigenous or immigrants
• Linguistic, emotional, social network skills
• Torture, war injuries, isolation, value conflicts
• especially for girls raised in new/freer country
  with roots in parents strong, conservative
  culture
• Low participation in societal activity

51
Family Pattern Risk Factors

• Destructive family patterns and traumatic events
  in childhood
• Parental psychopathology (affective)
• Alcohol, substance abuse
• Antisocial behavior/poor communication
• Family history of suicide attempts
• Violent and abusive family (physical/sexual)
• Divorce, separation or death of parents/guardians
• Frequent moves to different residential area
• Family rigidity
• Foster/adoptive family
• Very high or very low parental expectations

52
Suicide Physical Disorders

• Chronic physical illness especially if disabled
  or poor prognosis often coexists with depression
• Neurologic diseases
• Epilepsy (impulsivity, aggressivity, chronic
  disabililty)
• Spinal brain injuries eg post stroke 19
  depressed and suicidal
• Neoplasms esp at time of diagnosis or in first
  two yrs. Pain significant association with
  suicide
• HIV/AIDS esp in young in early stages of
  illness

53
Suicide Psychiatric Risk Factors

• Suicide is a multidimensional disorder
• Biological, genetic, psychological, sociological,
  environmental
• 40-60 of completers saw physician in a month
  prior to suicide (usually a GP)
• Not in itself a disease but mental disorders in
  80-100 of completers
• Lifetime risk for mood disorder (mostly
  depression) is 6-15 alcoholism 7-15,
  schizophrenia 4-10
• Hopelessness
• Poor coping skills

54
Risk Situations

• Marked susceptibility to stress (cognitive
  personality traits both learned and genetic)
  leads to difficulty coping with negative life
  events
• Negative life events reactivate helplessness,
  hopelessness despair Beautrais AL,J Amer
  Acad of Child Adol Psych, 1997, 361543-1551.
  DeWilde EJ,Amer J Psych, 14945-51

55
Risk Situations (continued)

• Injuries to self image / wounded personal dignity
• Family disturbances
• Separation from friends, classmates, etc
• Death of a love one or other significant person
• Termination of a love relationship
• Interpersonal conflicts or losses
• Legal or disciplinary problems
• Peer-group pressure or self-destructive peer
  acceptance
• Problems are always multidimensional

56
Risk Situations (continued)

• Bullying victimization
• Disappointment with school results
• High demands at school during examination periods
• Unemployment poor finances
• Unwanted pregnancy, abortion
• Infection with HIV or other sexually transmitted
  diseases
• Serious physical illness
• Natural disasters

57
Health Impact of Intimate Partner Violence on Men
and Women Analysis of the NVAW Survey

58
Significance

• Few population-based studies of IPV frequency by
  type and by gender.
• Few studies assessing longer term physical and
  mental health effects of IPV by type (physical,
  sexual, psychological) in women.
• No such studies among men as victims.
• No population-based studies assessing mental and
  physical health effects of stalking.

59
Significance of Studying IPV in HK/PRC

• Few population-based studies of IPV in ASIA
• No population-based studies assessing mental and
  physical health associations with IPV in China

60
National Violence Against Women Survey

• Nationally representative random digit dial
  survey of 8,000 women 8,000 men, 1995-6.
• Response rate 72 women 70 men.
• Purpose
• Estimate violence victimization by type.
• Measure financial and health effects of IPV.
• Assess demographic correlates of victims and
  perpetrators.

61
IPV A Risk Factor for Suicidal Behavior in
Western Women

• Abbott, Johnson, Koziol-McLain, Lowenstein,
  1995
• Kaplan, Asnis, Lipshitz, Chorney, 1995
• Roberts, Lawrence, OToole, Raphael, 1997
• Kaslow et al., 2000 Kaslow et al., 1998
• Stark Flitcraft, 1996

62
Risk Factors for Suicidal Attempters in IPV
Kaslow et al, 2002

• Numerous /or severe negative life events
• History of child maltreatment
• High levels of psychological distress
  depression
• Hopelessness about the future
• Alcohol drug problems

63
Protective Factors Associated with Non-attempter
Status in IPV (Kaslow, 2002)

• Hopefullness (Learned Optimism Walker,1994)
• Self-efficacy
• Coping skills
• Social support (empowers leaving)
• Browne, 1993 Sullivan, 1999 Jacobs, 1999
• Material resources
• Employment
• Child care
• Safe housing/escape from partner
• May help guide intervention for abused suicidal
  women

64
Limitations of Previous Work

• Kaslow et al focused on African Americans
• Largely self report and retrospective data calls
  into question response validity
• Cross sectional studies cannot show causation
• Unclear which risk protective factors are
  redundant both within between risk protective
  factor constellations
• Future work must include diverse ethnic groups in
  multiple settings longitudinally
• Structural equation modeling may help clarify
  distinct contributions of each risk protective
  factor

65
Intimate Partner Violence Measures

• Lifetime and current IPV
• Intimate spouse, live in boy/girl friend
• Physical (12 item Conflict Tactics Scale)
• Forced sex (3 items, completed sexual assault)
• Psychological IPV (12 item Power Control scale)
• Stalking (20 items occurred gt once, victim
  afraid)

66
Suicide Attempts by Battered Wives

• 22 (19) of battered wives mad 82 suicide
  attempts (16 yrs)
• More than 8 times of women being treated
  attempted suicide
• Predominately passive methods
• Conflict with husband was most common
• Mental disorder another main cause
• Doctor should consider possibility ongoing
  physical abuse in suicide patients

67
Suicide Attempts by Battered Wives

• 117 battered women, 22 (19) made one suicide
  attempt resulting in inpatient care during the
  16-yr study period
• Every year 1 of battered women attempted suicide
• 1,175 per 100,000

68
Health Outcomes Mental Health

• Current depression (SF-36 health survey)
• Chronic mental schizophrenia illness
  (depression, bipolar disorder,) age at diagnosis
• Current drug use (antidepressants,
  pain killers, tranquilizers, recreational
  drugs, heavy alcohol use)

69
Health Outcomes Physical Health

• Current self-perceived health status excellent
  (1) to poor (5)
• Developed a chronic disease (hypertension,
  diabetes, arthritis, asthma or emphysema,
  cancer) age at diagnosis
• Developed an injury (head, back, neck, spinal
  cord) age at injury

70
Developed an Injuryand IPV by Type and Sex
Adjusted RR ( 95 CI) Adjusted RR ( 95 CI)
Women N6790 Men N7122
Physical IPV 2.9 (2.2, 3.7) 2.3 (1.6, 3.3)
Psychological 1.9 (1.4, 2.6) 1.4 (1.1, 1.9)
Stalking 2.7 (1.9, 3.7) 3.1 (1.7, 5.6)
RR Adjusted for age, race, health insurance
status, and childhood abuse
71
Statistical Analysis

• Hierarchical categories of IPV Physical IPV
  (Sexual or physical)
• Psychological IPV without physical IPV
• Cohort analysis Use age at IPV and health
  outcome to ensure IPV precedes outcome.
• Adjusted RR Mantel Haenszel, test based CI.
• 10 change rule for confounding decision.

72
Frequency of Lifetime IPV by type among Women and
Men
Women N6790 Men N7122 Adjusted RR for Gender
Sexual IPV 4.3 0.2 33.3 (16.7, 50.0)
Physical, no sex. 10.8 4.8 2.4 (2.1, 2.8)
Psychological 12.2 15.2 0.9 (0.8, 1.0)
Stalking 5.7 1.2 5.3 (4.0, 6.3)
RR adjusted for age, race, insurance status
73
Lifetime Hierarchical IPV by type Among Women
Sexual IPV
Psychological Alone
Physical not sexual
74
Lifetime IPV Victimization by Hierarchical Type
Among Men
Sexual IPV
Physical not sexual
Psychological IPV
75
Demographic Correlates of Lifetime IPV
Victimization
Women Men
Increasing age ? ?
Race (Native Amer. / Mixed Race) ? ?
Lower education ? ?
Lower income ? ?
Currently Divorced or Separated ? ?
Childhood physical sexual abuse ? ?
Alcohol abuse by current partner ? ?
Current unemployment ?
76
Current Poor Health IPV by Type
Adjusted RR ( 95 CI) Adjusted RR ( 95 CI)
Women N6790 Men N7122
Physical IPV 2.0 (1.4, 2.8) 1.9 (1.0, 3.6)
Psychological 2.0 (1.4, 3.0) 1.6 (1.1, 2.3)
Stalking 1.1 (0.6, 1.9) 2.3 (0.8, 6.4)
RR Adjusted for age, race, health insurance
status, and childhood abuse
77
Developed Chronic Disease and IPV by Type
Adjusted RR ( 95 CI) Adjusted RR ( 95 CI)
Women N6790 Men N7122
Physical IPV 1.7 (1.3, 2.2) 1.4 (0.9, 2.2)
Psychological 1.0 (0.8, 1.4) 0.8 (0.6, 1.1)
Stalking 1.4 (1.1, 2.2) 1.5 (0.6, 3.6)
RR Adjusted for age, race, health insurance
status, and childhood abuse
78
Current Depressionand IPV by Type and Sex
Adjusted RR ( 95 CI) Adjusted RR ( 95 CI)
Women N6790 Men N7122
Physical IPV 1.9 (1.6, 2.2) 1.7 (1.3, 2.3)
Psychological 1.7 (1.4, 1.9) 1.6 (1.3, 1.8)
Stalking 2.0 (1.6, 2.5) 2.3 (1.4, 3.7)
RR Adjusted for age, race, health insurance
status, and childhood abuse
79
Developed Chronic Mental Illnessand IPV by Type
and Sex
Adjusted RR ( 95 CI) Adjusted RR ( 95 CI)
Women N6790 Men N7122
Physical IPV 3.5 (2.2, 5.6) 1.4 (0.5, 4.1)
Psychological 1.1 (0.5, 2.3) 2.4 (1.4, 4.2)
Stalking 2.4 (1.3, 4.3) 2.6 (0.6, 11.1)
RR Adjusted for age, race, health insurance
status, and childhood abuse
80
Current Drug Use Any IPV By Sex
Adjusted RR ( 95 CI) Adjusted RR ( 95 CI)
Women Men
Heavy alcohol use 1.6 (1.2, 2.1) 1.4 (1.2, 1.6)
Tranquilizers 2.0 (1.6, 2.6) 2.0 (1.4, 2.8)
Pain killers 1.9 (1.6, 2.1) 1.7 (1.4, 2.2)
Antidepressants 2.1 (1.6, 2.6) 2.1 (1.4, 3.2)
Recreational drugs 1.5 (0.9, 2.2) 1.7 (1.3, 2.2)
RR Adjusted for age, race, health insurance
status, and childhood abuse
81
Strengths Limitations

• Limitations
• Chronic illness self reported.
• Misclassification in age at IPV and age
  developing chronic mental or physical illness.
• Limited power to detect assoc. for rare outcomes.

• Strengths
• Largest study of men.
• 1st pop-based study to address physical and
  mental health effects of all IPV types, both
  genders.
• Correct temporal sequence.

82
Conclusions

• Physical psychological IPV associated with same
  health outcomes for men as for women.
• Screen for physical, sexual, asychological IPV
  for women men.
• Clinician and community support needed to
  identify intervene to prevent IPV.

83
Role of Public Health Professionals in Partner
Violence Prevention

• Validate the experiences of victims by
• Studying the phenomena
• Collaborate with other health professionals to
  design and implement interventions

84
Conclusions

• Sexual and physical IPV stalking by an intimate
  much higher among women.
• Men as likely as women to report psychological
  IPV only.
• 56 of IPV in women and 83 in men is
  psychologic.
• Same correlates of IPV victimization for men as
  for women.

85
QUESTIONS?
86
IPV Epidemiology

• 1.9 million women physically assaulted (Tjaden
  Theonnes, 1998)
• 12 - 29 of female ED patients report physical
  or sexual abuse (Dearwater
  et al., 1998 Ernst et al., 1997)
• gt250,000 ED visits for direct injuries
  (Rand Strom, 1997)

87
IPV --- Mental Health Effects

• High incidence of STRESS
• Depressive Anxiety symptoms
• PTSD
• Substance Abuse / Dependence
• Suicidal ideation
• Elevated general distress (SCL-90)
• (Bohn Holz, 1996 Campbell Lewandowski,
  1997 Geffner Pagelow, 1990 Koss,
  1990 Vitanza 1995)

88
STRESS

• Physiological Psychological precursor to
  illness
• Complex, multifaceted phenomenon
• Endocrine activity key marker of stress
• Biomarkers
• catecholamines
• corticosteroids
• superior to self-report psycho- physiological
  measures (BP,GSR, HR)

89
IPV STRESS

• Associations between IPV and physical mental
  health may be mediated by stress.
• Mechanism of stress/illness victimization
  unstudied
• The extent to which stress is related to the
  level of DV exposure and associated
  neuroendocrine changes is unknown.

90
Purpose

• To examine differences in platelet catecholamine
  levels (a biomarker of chronic strees) in three
  groups of female ED patients
• currently exposed to IPV
• those exposed to IPV in the remote past
• those never exposed to IPV

91
Cultural and Sociodemographic Risk Factors
(continued)

• Children who lack cultural roots
• Identity problems create challenges for conflict
  resolution promoting self destructive behavior
• Gender nonconformity/identity issues
• sexual orientation
• Rejection by families, peers, schools

92
Intimate Partner Violence Platelet
Catecholamine LevelsAll Stressed-up and Nowhere
to Go
Gregory L. Larkin, MD, MSPH Kelly B. Hyman,
PhD Candidate Sabra S. Inslicht, MS Andrew
Baum, PhD

• The Mercy Hospital of Pittsburgh, Pittsburgh,
  PA
• Emergency Medicine Association of Pittsburgh,
  Pittsburgh, PA
• University of Pittsburgh, Pittsburgh, PA

93
Setting

• Urban, Level I Trauma and Burn Center with an
  annual emergency department (ED) census of 43,000
  patients

Design

• Prospective, random sample of female ED patients

94
Population

• Included
• Female
• 18 - 65 years old
• Likely to be discharged
• Excluded
• Unable to read or answer questions in ED
• Intoxicated or couldnt provide written consent
• Experienced non-IPV trauma in last year

95
3 Groups ---Domestic Safety Assessment ISA

• Current IPV lt 3 months
• Past IPV gt12 months
• Control No IPV ever

96
Measures

• Demographics (age, ethnicity, income, education,
  marital status)
• Background Stress
• Traumatic Stress Schedule (Norris 1990)
• SCL-90
• Health Questionnaire (SF-12)
• Biological markers of stress
• Platelet Catecholamines
• EPINEPHRINE
• NOREPINEPHRINE

97
Measures (continued)

• Findings are part of a larger study that included
  assessment of
• Physical health
• Psychosocial factors
• Structured Clinical Interview for DSM-IV

98
Platelet Catecholamines

• Index of SNS arousal
• Stable (vs plasma) for 10-12 days
• Efficient (vs 24 hr urinary cat)
• Plts absorb cats from plasma thru
  diffusion/uptake thru serotonin transport
• Assay HPLC for epi norepi

99
Statistical Analyses

• ANOVA ---
• (platelet cats by IPV status, controlling for
  demographics, income, background stress, life
  events, diet)
• ?2 and Multivariate Linear Regression

100
Results

• N136 (38control/41 past/57 current IPV)
• Mean age 31.5 10.2
• 51.9 Black, nonHispanic 46.9
  Caucasian, nonHispanic
• 45.7 Single (but involved) 23.8
  Married or cohabitating
• --Not significantly different by group status --

101
Significant Demographic Differences by Group
Status --- Income
39.6
Control n38
32.1
6.4
26.4
15.5
12.7
Past n41
11.1
57.8
Current n57
19.0
15.6
61.9
F 4.99, p .009 r .384 (income education)
102
Current Major Depression
IPV Status
OR (Control Past vs Current) 4.65
(2.02-10.69)
103
Platelet Epinephrine (pmol/mg) by IPV Status
F5.27 (p.006)
104
Platelet Norepinephrine (pmol/mg protein) by IPV
Status
F3.44 (p.035)
105
Current PTSD (IPV Non-IPV)
IPV Status
OR (Control vs Past Current) .03 (.00 -
.22)
OR (Past vs Current) 3.16 (1.24 - 8.04)
106
Limitations

• Generalizability of Urban ED pop
• Current IPV victims over-sampled
• Women excluded if experienced other trauma in
  last year
• Cross-sectional (assoc v. causal)
• Biomarkers dont define total stress
• Other measures of stress not included

107
Drug or Alcohol Abuse or Dependence
IPV Status
OR (Control Past vs Current) 2.18 (.6249 -
7.62)ns
108
Current Drug Abuse or Dependence
109
Any DSM - IV Axis I Diagnosis
IPV Status
OR (Control vs Past Current) .10 (.04 -
.27)
OR (Past vs Current) 3.17 (1.39 - 7.23)
110
Comparison of Current DSM - IV Diagnosis
Current
A. MDE B. MDE other C. Primary Psychotic D.
PTSD E. Anxiety Other F. Drug or ETOH G.
Somatization H. Eating
Past
111
Conclusions

• IPV exposure is associated with elevated platelet
  catecholamine regardless of income, life events,
  or diet.
• Platelet catecholamine levels are proportional to
  IPV exposure in a graded (dose-response) fashion.

112
Discussion

• PTSD symptoms correlated w/length of relationship
  recency of abusive episode (Kemp et al., 1991
  Astin et al., 1993)
• 2.01 times more likely to have DSM-III-R disorder
  (Danielson et al., 1998)
• Related to increased PTSD but not depression
  (Cascardi, 1994)

113
Future Directions

• Conduct longitudinal, cohort studies to examine
  causality
• Explore the comorbidity of MDE and PTSD in IPV
  victims
• Further explore the links between stress and
  other illness mechanistically.
• Use platelet cat assay as indicator of
  improvement after novel interventions

114
EFFICACY OF UNIVERSAL SCREENING FOR INTIMATE
PARTNER VIOLENCE (IPV) THE IMPORTANCE OF
PATIENT PROVIDER FACTORS

• GREGORY L. LARKIN, MD MSPH FACEP
• KELLY B. HYMAN, MS

THE EMERGENCY MEDICINE ASSOCIATION OF
PITTSBURGH THE MERCY HOSPITAL OF PITTSBURGH THE
UNIVERSITY OF PITTSBURGH SCHOOL OF MEDICINE
115
BACKGROUND

• IPV INTIMATE PARTNER VIOLENCE
• LARGEST CAUSE OF INJURY TO WOMEN
• OVER 1 MILLION ED VISITS ANNUALLY
• 22-35 ALL FEMALE ED PATIENTS
• 5-10 BILLION
• ONLY 5 APPROPRIATELY IDENTIFIED

116
OBJECTIVE
EFFICACY OF UNIVERSAL IPV SCREENING

• FEASABILITY COMPLIANCE
• ROLE OF PATIENT PROVIDER FACTORS
• LIFETIME PREVALENCE OF IPV
• HØ RATES OF SCREENING AND IDENTIFICATION DO NOT
  VARY WITH PATIENT FACTORS (DEMOGRAPHICS, CC,
  VISITS, ACUITY) OR PROVIDER FACTORS (SHIFT
  GENDER)

117
METHODS
SETTING

• URBAN LEVEL 1 TRAUMA CENTER
• 45,000 ED VISITS
• ON-SITE VICTIM ADVOCACY PROGRAM
• UNIVERSAL SCREENING POLICY SINCE 2/94
• FAMILY VIOLENCE PREVENTION FUND MODEL PROGRAM

118
METHODS
DESIGN

• RETROSPECTIVE RECORD REVIEW
• STRATIFIED RANDOM SAMPLE
• FEMALES gt18 YRS ( N 14,200)
• DISCHARGED FROM ED BETWEEN 7/94-6/95
• n 1638 (12 SAMPLING FRACTION)

119
METHODS
MEASURES

• PATIENT FACTORS
• AGE
• RACE (Minority vs Caucasian)
• MARITAL STATUS (Single vs Married)
• EMPLOYMENT STATUS (Yes vs No)
• RELIGION (Judeochristian vs Other)
• INSURANCE STATUS (DPA vs Other)
• ACUITY (Fast Track vs Acute ED)
• CHIEF COMPLAINT (Medical, Trauma, or Psych)
• PROVIDER FACTORS
• NURSE GENDER
• SHIFT WORKED (7AM-1059PM vs 11PM-659AM)

120
DATA ANALYSIS

• DATA CLEANING RANDOM AUDIT (20)
• STATISTICS
• DEPENDENT VARIABLES
• SCREENED vs NOT SCREENED
• LIFETIME IPV STATUS ( vs -)
• CHI-SQUARE LOGISTIC REGRESSION
• UNIVARIATE MULTIVARIATE LOGIT
• STEP DOWN PROCEDURE
• alpha .05

121
RESULTS
COMPLIANCE WITH SCREENING

• UNIVARIATE ANALYSES
• SIGNIFICANT PATIENT FACTORS
• LOW ACUITY
• MINORITY RACE
• DPA /MEDICAID
• UNEMPLOYED
• NONPSYCHIATRIC COMPLAINT
• SIGNIFICANT PROVIDER FACTORS
• DAYLIGHT AFTERNOON SHIFT
• NON SIGNIFICANT FACTORS
• NURSE GENDER, PT AGE, RECIDIVISM,MARITAL STATUS,
  RELIGION

122
RESULTS
COMPLIANCE WITH SCREENING (cont.)

• MULTIVARIATE ANALYSES
• SIGNIFICANT PROVIDER FACTORS
• NIGHTSHIFT OR 0.59 (.39-.89)
• SIGNIFICANT PATIENT FACTORS
• LOW ACUITY OR 2.12 (1.66-2.71)
• MED vs PSYCH OR 2.76 (.98-7.83)
• TRAUMA vs PSYCH OR 3.17 (1.11-9.12)
• NONSIGNIFICANT FACTORS
• INSURANCE STATUS, RACE, EMPLOYMENT STATUS FELL OUT

123
RESULTS
LIFETIME PREVALENCE OF IPV

• UNIVARIATE ANALYSES
• SIGNIFICANT PROVIDER FACTORS
• FEMALE NURSES MORE LIKELY TO IDENTIFY
• SIGNIFICANT PATIENT FACTORS
• AGE
• HIGHER ACUITY
• PSYCHIATRIC ILLNESS
• MINORITY RACE
• SINGLE vs MARRIED
• UNEMPLOYED
• DPA/MEDICAID
• NONSIGNIFICANT FACTORS
• TIME OF DAY PATIENT RELIGION

124
RESULTS
LIFETIME PREVALENCE OF IPV (cont.)

• MULTIVARIATE ANALYSES
• SIGNIFICANT PROVIDER FACTORS
• MALE NURSES OR 0.20 (.05-.87)
• SIGNIFICANT PATIENT FACTORS
• LOWER ACUITY OR 0.39 (0.22-0.67)
• MINORITY RACE OR 1.96 (1.15-3.34)
• DPA/MEDICAID OR 2.56 (1.41-4.63)
• NONSIGNIFICANT FACTORS
• TIME OF DAY PATIENT RELIGION, AGE, RECIDIVISM
  MARITAL STATUS EMPLOYMENT STATUS TYPE OF
  ILLNESS OR INJURY
• PSYCH vs MEDICAL OR.65 (.38-1.12)

125
LIMITATIONS

• RETROSPECTIVE STUDY
• SMALL SAMPLE SIZE
• MECHANISM OF INJURY
• SCREENING COMPLIANCE POOR
• RANDOMIZED TO VISITS
• 1638 VISITS 1509 PATIENTS
• EXTERNAL VALIDITY/GENERALIZABILITY
• ED PATIENTS
• ONE CENTER
• URBAN
• PITTSBURGH DEMOGRAPHICS

126
CONCLUSIONS
UNIVERSAL SCREENING

• INCREASES THE IDENTIFICATION RATES OF IPV-EVEN
  WITH POOR COMPLIANCE
• PSYCH PATIENTS MAY BE AT RISK OF INADEQUATE
  SCREENING
• OTHERWISE, CAN BE CONDUCTED WITHOUT PROVIDER BIAS
  
• WORSE ON NIGHTSHIFT AND WHEN WORKING WITH MORE
  ACUTE PATIENTS

127
CONCLUSIONS
LIFETIME PREVALENCE OF IPV

• MALE NURSES LESS LIKELY TO IDENTIFY
• IN THIS ED POPULATION, PATIENTS WITH MORE ACUTE
  ILLNESS OR INJURY, ON PUBLIC ASSISTANCE,
  MINORITY RACE WERE AT HIGHER RISK OF IPV