Effect of immigration on health outcomes among abused Chinese women Agnes Tiwari, PhD, RN, FAAN Professor and Head School of Nursing The University of Hong Kong Member of the Expert Panel on Violence Prevention, American Academy of Nursing 26-27 November 2011 Symposium on Enhancing Quality Primary Care Through Research and Evidence in Hong Kong
Co-investigators Dr. Daniel Fong 1 Dr. Janet Wong 1 Dr. Felix Yuen 1 Mr. Geoffrey Chan 1 Ms. Gloria Lam 1 Dr. Edward Chan 1 Dr. Janice Humphreys 2 Dr. Linda Bullock 3 1 The University of Hong Kong 2 University of California, San Francisco 3 University of Virginia
What is known Western literature: Immigration status has been shown to be a risk factor of intimate partner violence (IPV) Abused immigrant women undertake less formal help-seeking compared to abused non-immigrant women Perpetrators often use the women s immigration status as a form of control
Very little is known Abused Chinese immigrant women: What is the effect of IPV on their health? Would the women s adverse health outcomes improve with time? What would help to improve their health despite the adversity?
Important because > 12% of female population in Hong Kong are new arrivals * from China Immigration status has been found to be a risk factor for IPV IPV is known to impact negatively on women s health * Having lived in Hong Kong for less than seven years
Three studies to be presented 1. Association between immigration and health outcomes in abused Chinese women 2. Differential effects of types of IPV on the health of Chinese immigrant women 3. Effect of an advocacy intervention on depressive symptoms for abused Chinese immigrant women
Study 1 Association between immigration and health outcomes in abused Chinese women
Study design Cross-sectional Participants Study 1 225 Chinese women Aged 18 or above Screened positive for IPV Setting Shelters for abused women A large community centre with 18 outreach sites Duration 2009-2010
Demographic characteristics Abused immigrant women (n=149) Abused nonimmigrant women (n=76) Non-abused women (n=100) Age, Mean ± SD, range 39 ± 7, 23-60 44 ± 9, 21-64 42 ± 8, 22-61 No. of children, Mean ± SD 1.8 ± 0.8 1.9 ± 1.0 2.0 ± 0.8 Divorce or separated or widowed 37 (25%) 29 (38%) 14 (14%) Education ( 9 years or less) 104 (70%) 47 (62%) 64 (65%) Unemployment 117 (79%) 58 (76%) 69 (70%) In debt in the past six months 61 (41%) 33 (43%) 14 (14%) Family monthly income ( 10,000) 105 (72%) 58 (81%) 49 (51%)
Research question 1 Is there any association between immigration and health outcomes in abused Chinese women?
Multiple logistic regression model explaining chronic pain and chronic physical illness by demographic socio-economic, and social support variables (n=145) Chronic pain Chronic physical illness OR 95% CI P OR 95% CI P Demographic factors Age 1.1 1.0 1.1 0.01 1.1 1.0 1.2 0.01 Immigration status 2.4 0.9 6.5 0.09 2.0 0.5 7.1 0.30 Socio-economic factors Financial hardship 3.1 1.1 8.5 0.03 2.0 0.6 7.3 0.28 Perceived social support Low 1.0 -- -- 1.0 -- -- Moderate 0.2 0.1 0.9 0.02 0.3 0.1 1.7 0.18 High 0.2 0.1 0.7 0.01 0.7 0.1 4.2 0.65
Multiple linear regression model explaining depression and PTSD by demographic socio-economic, and social support variables (n=145) Depressive symptoms PTSD symptoms Variable β 95% CI p-value β 95% CI p-value Demographic factors Age 0.01 (-0.11, 0.13) 0.88 0.15 (-0.19, 0.49) 0.38 Immigration status 1.44 (-0.53, 3.41) 0.15 5.50 (0.09, 10.90) 0.05 Socio-economic factors Financial hardship 1.92 (-0.16, 3.87) 0.05 8.20 (2.74, 13.65) 0.01 Perceived social support Low -- -- -- -- -- -- Moderate -1.06 (-4.27, 2.13) 0.51 1.14 (-7.86, 10.13) 0.80 High -3.51 (-6.89, -0.12) 0.04-9.21 (-18.74, 0.31) 0.06
Research question 2 Does the length of residence in Hong Kong have a differential effect on the health of abused Chinese immigrant women?
Logistic regression models explaining chronic pain and chronic physical illness by length of residence (n=149) Chronic pain Chronic physical illness OR 95% CI P OR 95% CI P HK-born 1.0 -- -- 1.0 -- -- New arrival < 4 years 0.8 0.2 1.9 0.29 0.3 0.1 1.7 0.18 4 years < 7 years 1.2 0.4 3.4 0.84 0.8 0.2 2.6 0.64 7 years < 15 years 2.9 0.7 6.4 0.16 0.6 0.2 2.6 0.54 15 years or above 3.0 1.1 9.4 0.04 2.8 0.9 8.6 0.06
Linear regression models explaining depression and PTSD by length of residence (n=149) Depression* PTSD^ β 95% CI p β 95% CI p HK-born New arrival < 4 years 0.00 -- -- 0.00 -- -- -0.93-3.58, 1.72 0.49-0.50-8.96, 7.95 0.90 4 years < 7 years 1.71-0.94,4.36 0.20 8.33 7 years < 15 years -0.32-3.14, 2.50 0.82-1.59-0.12, 16.79-10.59, 7.40 0.05 0.73 15 years or above 2.77-0.16, 5.69 0.06 10.17 0.83, 19.51 0.03 * R 2 =4.5 %, adjusted R 2 = 1.8% ^ R 2 =5.9%, adjusted R 2 = 3.3%
Study 1: Unexpected findings Association between immigration and health outcomes (chronic pain, chronic physical illness, depressive symptoms and posttraumatic stress disorder symptoms) was statistically not significant. However, abused women who have settled in Hong Kong for 15 years or longer were significantly more likely to report chronic pain and post-traumatic stress disorder symptoms.
Study 2 Differential effects of types of IPV on the health of Chinese immigrant women
Study design Cross-sectional Participants Study 2 208 Chinese women Aged 18 or above Screened positive for IPV Setting Shelters for abused women FCPSUs in the Hong Kong SAR Duration 2010-2011
Characteristics of respondents (n=208) n Percent (%) Age of respondents 20-29 20 9.6 30-39 99 47.4 40-49 48 23.0 Place of Birth HK (local) 33 15.8 Places other than HK 175 84.1 Education level Illiterate 8 3.8 Primary 46 22.1 Secondary School (Junior school) 96 46.2 IPV perpetration Physical violence 208 100 Psychological violence 208 100 Sexual assault 82 39.4 Types of IPV Intimate terrorism (IT) 111 53.4 Situational Couple Violence (SCV) 91 43.8 Violent Resistance (VR) 4 1.9
Characteristics of immigrants (n=175) n Percent (%) Place of recruitment Shelter 114 65.1 Community 61 34.9 Age of respondents 20-29 16 9.1 30-39 87 49.7 40-49 41 23.4 Immigration from China 169 96.6 Other countries 6 3.4 Marriage status Stay in the relationship 68 38.9 Divorce or separate 106 60.6 IPV perpetration Physical violence 175 100 Psychological violence 175 100 Sexual assault 72 41.1 Afraid of partner Yes 133 76.0 No 42 24.0
* Mean possible scores range from 0 to 4 **Independent samples t-test Controlling behaviors used by perpetrators on Chinese immigrant women (n=169) Group A (n=101) Mean* (SD) Group B (n=68) Mean* (SD) Mean difference P value ** Economic control 2.83 (1.16) 1.08 (1.06) 1.75 < 0.001 Use of threats 2.09 (0.97) 0.72 (0.76) 1.37 < 0.001 Intimidation 2.63 (0.70) 0.89 (0.70) 1.74 < 0.001 Emotional abuse 2.74 (1.11) 1.03 (0.82) 1.71 < 0.001 Isolation 2.97 (1.35) 0.68 (1.12) 2.29 < 0.001 Using the children 1.72 (1.20) 0.89 (0.75) 0.83 < 0.001 Minimizing 1.70 (1.20) 0.43 (0.83) 1.27 < 0.001 Total 2.43 (0.69) (high control) 0.83 (0.47) (low control) 1.60 < 0.001
Examples of controlling behaviors Economic control He only gives me a few hundred dollars each month. I only have HK$ 57 in my bank account. Use of threats He tells me that one of these days he will kill our children, then me, then himself. Intimidation When I told him that I was too full to eat anymore, he threw the food at me and started to smash things in the house.
Examples of controlling behaviors Emotional abuse Every day he tells me how useless I am and how grateful I should be to him for having me as his wife. Isolation He checks my mobile phone everyday. If he finds a number that he does not recognize, he would interrogate me. And if I can t give him a satisfactory answer, he would punish me. Use the children He tells the children what a bad mother I am. Minimizing He tells me that I deserve to be punished.
Depressive symptoms by type of IPV BDI-II Score IT (n=101) SCV (n=68) P value * Minimal depression 0-13 6 40 <0.001 Mild depression 14-19 8 9 Moderate depression 20-28 22 10 Severe depression 29-63 65 9 * Pearson Chi-Square test used IT (n=101) Mean (SD) SCV (n=68) Mean (SD) Mean difference P value * BDI II Total Score 35.02 (15.51) 14.31 (14.69) 20.71 < 0.001 * Independent samples t-test used
Post Traumatic Stress Disorder (PTSD) Symptoms by type of IPV IT (n=101) Mean (SD) SCV (n=68) Mean (SD) Mean difference P value* Re-experiencing 1 3.96 (1.11) 2.30 (1.17) 1.66 <0.001 Avoidance 1 3.82 (1.23) 2.15 (1.17) 1.67 <0.001 Hyper-arousal 1 4.10 (1.03) 2.39 (1.24) 1.71 <0.001 PTSD Total score 2 67.05 (17.79) 38.51 (19.26) 28.54 <0.001 1. Mean possible scores range from 1 to 5 2. Total scores range from 17 to 85 * Independent samples t-test used
Study 2: first to report For abused Chinese immigrant women experiencing high coercive control by their partners, reports of depressive symptoms and post-traumatic stress disorder symptoms were significantly higher compared to those Chinese immigrant women experiencing low control
Study 3 Effect of an advocacy intervention on depressive symptoms for abused Chinese immigrant women
Study design Study 3 Randomized controlled trial Participants 200 Chinese women Aged 18 or above Screened positive for IPV Setting A community centre with a large catchment area Duration 2007-2009
A 12-week advocacy intervention for community-dwelling abused Chinese women
Based on Dutton s empowerment model Cohen s social support theory
Dutton s empowerment model Modified from McFarlane and Parker s Abuse Prevention Protocol Efficacy for use with Chinese abused women was previously confirmed Components: safety, enhanced choice-making and problem-solving Empowerment - the Chinese way : work through their ambivalence, explore informal and formal resources, respect their values and beliefs, and support their decisions
Cohen s social support theory Modified from Bullock et al. s Baby-Beep intervention Incorporates: non-judgmental listening ear, women s self-identified needs, give information when asked to, and carry out referral if appropriate Social support the Chinese way : the woman knows the provider, same person providing help as far as possible, and provider skilled at handling crises
Comparing the demographics of non-immigrants and immigrants Demographics Non-immigrants Number (%) (n=137) Immigrants Number (%) (n=60) p-value Age (Mean ± SD) 39.2 ± 7.8 35.6 ± 4.8 0.001* Education: Grade 6 Grade 7 or above 40 (29.2%) 97 (70.8%) 14 (23.3%) 46 (76.7%) 0.56 Employed 50 (36.5%) 11 (18.3%) 0.012* Number of children 3 27 (19.7%) 12 (10%) 0.003* Married or cohabited 118 (86.1) 58 (97.7%) 0.001* Receiving financial support from friends or relatives 86 (62.8%) 34 (56.7%) 0.432 Receiving social security support 28 (20.4%) 13 (21.7%) 0.85 ISEL score 8.2 ± 8.9 4.3 ± 5.5 0.006* Psychological aggression (Mean ± SD) Physical assault (Mean ± SD) Sexual coercion (Mean ± SD) 18 ± 10.5 1.4 ± 3.9 0.5 ± 2.8 20.4 ± 9.8 2.1 ± 4.7 0.2 ± 0.9 0.059 0.215 0.551 Study group: Intervention group Control group 65 (32.5%) 72 (36%) 34 (17%) 26 (13%) 0.279
Impact of immigration status on intervention effect Study Outcomes Non-immigrants Intervention control Immigrants Intervention control Estimate 95% CI p-value Estimate 95% CI p-value Group by immigration interaction p-value Depressive symptoms 3 months -4.9 (-8.38, -1.41) 0.006* 2.65 (-2.67, 7.97) 0.328 0.02* 9 months -2.34 (-5.83, 1.14) 0.187-3.42 (-8.73, 1.90) 0.740 0.74 Note: Data analysis was done by mixed effects method. Results were adjusted for age, marital status, no of children, employment status and ISEL score
Study 3: Unexpected findings The advocacy intervention was effective in reducing depressive symptoms of abused Chinese non-immigrant women but not for abused Chinese immigrant women on completion of the intervention However, the impact of immigration status on intervention effect was not sustained at 6 months post-intervention suggesting that immigrant women may need longer time in order to benefit from the intervention
Conclusion Abused Chinese immigrant women have health needs that are related to IPV as well as to their vulnerability as immigrants IPV among Chinese immigrant women is not a single phenomenon but differentiated by coercive control with differential effects on health Time should be taken to establish rapport before providing interventions for abused Chinese immigrant women
Acknowledgements The University of Hong Kong Small Project Fund (Study 1) The General Research Fund (Study 2) The Health and Health Services Research Fund (Study 3)