| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Andalusian School of Public Health, Granada, Spain.
Address correspondence and reprint requests to Dra Isabel Ruiz Pérez, Escuela Andaluza de Salud Pública, Campus Universitario de Cartuja, Apartado de Correos 2070, 18080 Granada, Spain. E-mail: isabel.ruiz.easp{at}juntadeandalucia.es
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Methods: The sample included 1402 randomly selected women attending general practice in Spain. Information about IPV, use of psychoactive drugs, psychological distress (measured with the 12-item General Health Questionnaire), and self-perceived health status was obtained through a self-administered structured questionnaire.
Results: Lifetime IPV prevalence was 32%. All types of abuse were significantly associated with psychological distress, tranquilizers, and antidepressants use. Compared with never-abused women, those referring psychological abuse solely were more likely to present negative mental health indicators. Women reporting physical abuse presented worse self-perceived health than never-abused women (odds ratio [OR] = 1.74; 95% confidence interval [CI] = 1.04–2.91 for combined physical and psychological abuse; OR = 2.91; 95% CI = 1.73–4.91 for combined physical, psychological, and sexual abuse). Women referring the 3 types of abuse were the most likely to present impaired mental health. Increased intensity and duration of some abusive experiences also led to increased levels of association with poor mental health.
Conclusions: This study underlines that different types of IPV can affect mental health differently. Results also show that IPV can be present even when there are no physical injuries and highlight the need for clinicians to screen for psychological, as well as physical and sexual, IPV.
Key Words: spouse abuse • domestic violence • Spain • mental health • women
Abbreviations: IPV = intimate partner violence; PTSD = posttraumatic stress disorder; OR = odds ratio; CI = confidence interval; US = United States; GHQ = General Health Questionnaire.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Most of the past studies have only considered physical and sexual violence, probably because these are the most visible and easy to identify forms of IPV. However, psychological abuse is the most frequently reported characteristic of violent relationships, and it has been found to affect women’s health as greatly and significantly as the other types of abuse (14–16).
Another aspect that has rarely been considered is whether the different kinds of abuse could have separate effects on women’s health (17). Literature shows, for example, that sexual assault by an intimate partner could explain the high prevalence of gynecological problems reported by battered women (18). However, when it comes to mental health, this issue has hardly been addressed.
Finally, IPV has been studied in many developed and developing countries. Spain remains one of the few large societies for which the prevalence and correlates of partner abuse are still unknown (19–21), despite the fact that it has become a priority in the political agenda of the country (22). Seventy-two women were killed by their male partners or ex-partners in 2004, and the number of legal demands was 57,527 (23).
This study adds to the existing literature by providing information on the magnitude of IPV in Spain and by describing the separate effects of physical, psychological, and sexual IPV on mental health in a sample of Spanish women attending public family practices. Furthermore, it analyses whether an increase in the intensity and duration of IPV is associated with an increased likelihood of poor mental health. Finally, the study tries to determine if the consequences of IPV on mental health could be long term by differentiating between women reporting current but no past abuse, those referring past but no current abuse, and those referring both.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
Each family practitioner randomly recruited a maximum of 2 women a day. All physicians followed a similar randomization process, based on the time of the patient consultation. If the woman did not meet inclusion criteria, the following one was included. A total of 1631 women were approached.
Survey Instrument and Measures
A self-administered structured questionnaire was specifically developed for this study. It consists of 21 close-ended questions on IPV, demographic variables, and health status that can be answered in less than 15 minutes. The IPV questions were taken from a questionnaire used in previous studies, showing high comprehensibility and acceptability. They were adapted from scales used in other studies, such as the World Health Organization Multi-Country Study on Women’s Health and Life Events (25). Measures considered for this analysis are next described.
Measures of IPV
Each participant woman was first asked if any intimate partner in the previous year had abused her physically (hit, slapped, kicked, pushed . . .), psychologically (threatened, insulted, humiliated, been extremely jealous, scared her . . .), and/or sexually (forced her to have sexual activities against her will). These questions had 3 possible responses: "many times," "sometimes," and "never." A woman was considered to experience "current abuse" if she answered "many times" or "sometimes" to any of these 3 questions. Women who had not been in an intimate relationship in the previous year were asked to leave these 3 questions blank, and they were considered negative for current abuse. Second, each participant was asked if any other partner in the past had abused her physically, psychologically, and/or sexually. If she answered positively to any of these 3 questions she was considered to have experienced "past abuse." Again, women who had not been in a relationship in the past (different for the current one) were asked to leave these questions blank and were considered negative for past abuse. Women who had no current or past intimate relationships were excluded from the analysis.
A woman was considered to have experienced lifetime physical IPV if she answered "many times" or "sometimes" to any of the 2 questions about physical IPV, and the same applied to psychological and sexual abuse. Because there is usually considerable overlap between IPV types, we created 4 mutually exclusive hierarchical categories of lifetime abuse for the analysis of IPV and mental health. The first group included respondents ever experiencing psychological IPV solely. The second group included those ever experiencing physical and psychological IPV yet not sexual. The third group experienced lifetime psychological and sexual IPV yet not physical. The last one included women ever experiencing the 3 types of abuse. We did not include other additional groups as categories of abuse (physical solely, sexual solely, and physical and sexual) because the number of women in these categories did not allow us to conduct the analysis on mental health.
The duration of the abusive experiences was also assessed with the following question: "For how long did you suffer the abusive experiences (any form of abuse)?" The response options were "<1 month", "1 month–6 months," "6 months–1 year," "1 year–5 years," and ">5 years." This variable was recoded for analysis purposes.
Mental Health Indicators
We included 3 mental health indicators: substance use, psychological distress, and self-perceived health status.
Women were asked to check among tranquilizers, antidepressants, painkillers, alcohol, and illicit recreational drugs which substances they had used within the preceding 3 months.
Psychological distress was assessed with the 12-item General Health Questionnaire (GHQ-12) (26). This short-form screening instrument has been validated in different countries and contexts (27,28), including Spain (29), and it has also been used in studies on the impact of IPV on women’s health (30). It covers anxiety, depression, and self-esteem, as experienced in the last month. This instrument is widely used to screen for mental health problems in nonpsychiatric services, and its application in research settings as a screening tool is well documented. There is evidence that the GHQ-12 is a consistent and reliable instrument when used in general health care samples (31).
Finally, self-perceived health was assessed with the following question: "How would you say your current health status is, compared with your health status last year?" The response options were "better," "the same," and "worse." This measure that reflects both physical and mental health has widely accepted validity and a demonstrated association with morbidity and mortality (32).
Demographics
Information on age, number of children, current marital status, employment status, education, and monthly family income was obtained.
Procedure
Data were collected by the family practitioner at the end of the woman’s consultation. If the patient met eligibility criteria, the physician invited her to participate in a study about women and health. The doctor explained that the questionnaire was anonymous and confidential and that she/he could help in filling it out if necessary. After giving consent for participation, the woman was handed the questionnaire that was self-completed either in the waiting area or in a private space specifically provided for the study. Once the questionnaire was completed, the woman put it in an envelope, closed it, and deposited it in a "questionnaire box" specifically provided for this purpose. Within the envelope, the woman was given information on available community resources for battered women in the area.
Statistical Analysis
All univariate, bivariate, and multivariate analysis procedures were performed with the statistical software SPSS, version 11.5.
Prevalence for the different lifetime IPV categories and for current and past IPV in our sample was first calculated. The association between IPV and mental health indicators was calculated with the 
2 test. Statistical significance was set at p < .05. Ever experiencing any kind of IPV was considered another category for analysis purposes, and the comparison group in all analysis was comprised of never-abused women. A logistic regression analysis was also conducted to control the effect of possible confounding variables (age, employment status, and monthly family income). Adjusted odds ratios (OR) and 95% confidence intervals (CI) are reported.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
The mean age of the sample was 38.83 years (SD = 11.15 years), and 33.3% had 2 children. Almost 63% were married, and more than half of the sample were employed (51%). Most of the women did not have a university degree (65.3%). Thirty-six percent of the sample had a monthly family income over 1200
, and 15% reported an income below 600 (1 = $1.32).
Mental Health Status of the Sample
More than 19% of the women reported using tranquilizers, 15.4% antidepressants, and 14% alcohol. Painkillers were the most frequently used substances (36.8%) and recreational illicit drugs the least (2.8%).
Almost 40% of the women presented psychological distress. More than half of the sample (59.5%) considered their current health status as good (or as bad) as it was the previous year, whereas 20.4% reported it was better and 20.1% worse.
IPV Prevalence
Thirty-two percent of the women (n = 445) had experienced any type of lifetime IPV; 14.4% (n = 198) reported lifetime psychological abuse solely, 7.2% (n = 99) lifetime physical and psychological abuse (no sexual), 2.5% (n = 35) lifetime psychological and sexual IPV (no physical), and 6% (n = 83) lifetime physical, psychological, and sexual violence. Only 3 women reported sexual abuse and 9 physical abuse solely. It should be noted that 18 women who reported lifetime IPV did not respond to all IPV questions, and, therefore, they could not be placed under any IPV category.
IPV and Mental Health
Table 1 shows the adjusted association between having ever experienced IPV and the different mental health indicators. Abused women were more likely to use tranquilizers, antidepressants, and illicit recreational drugs than never-abused women. They were also more likely to present psychological distress and to refer a worse self-perceived health compared with previous year.
|
The adjusted association between the different IPV categories and mental health indicators is shown on Table 2. Tranquilizers and antidepressants use was significantly associated with all IPV categories. This association was stronger for the categories that included sexual violence. In this way, women who reported lifetime psychological and sexual abuse were 5.12 times more likely to use tranquilizers (95% CI = 2.44–10.73) and 3.39 times more likely to use antidepressants (95% CI = 1.56–7.37) than never-abused women.
|
Regarding alcohol use, this was only significantly associated with the IPV categories including sexual violence. Women who experienced any of these categories of abuse were also more likely to report illicit recreational drug use than never-abused women, as well as females referring psychological abuse solely (OR = 4.18; 95% CI = 1.67–10.44). This last category of IPV also reduced the probability of using painkillers.
Psychological distress was associated with all IPV categories, and this association was again stronger for the categories that included sexual abuse. However, only women referring physical abuse were significantly more likely to report a worse self-perceived health than never-abused women (OR = 1.74; 95% CI = 1.04–2.91 for physical and psychological; and OR = 2.91; 95% CI = 1.73–4.91 for physical, psychological, and sexual IPV).
It should be noted that women who referred having experienced the three types of abuse (physical, psychological and sexual) were the most likely to present negative mental health indicators.
When analyzing mental health indicators by the intensity of IPV, a dose-response gradient was observed in some of the categories of abuse, so that increasing IPV intensity scores were associated with an increased likelihood of psychological distress and worse self-perceived health than the previous year. This gradient was observed only in some of the IPV categories and was more evident in women experiencing the 3 types of abuse (Table 3).
|
Higher IPV duration scores were also associated with an increased likelihood of worse self-perceived health in some of the IPV categories. However, and contrary to what occurred with IPV intensity, compared with never-abused women, those referring the 3 types of abuse for a shorter period of time (1 month to 1 year) were more likely to present psychological distress (OR = 3.66; 95% CI = 2.23–10.87) than women who experienced the abuse for a longer period (OR = 2.86; 95% CI = 1.59–5.15). This was also observed in women experiencing psychological and sexual IPV (Table 4).
|
Regarding the timing of the violence, 12.1% (n = 168) of the women reported any form of current IPV (experienced within the previous year), 16.2% (n = 225) referred past IPV, and 3.7% (n = 51) experienced partner abuse both in the past and in the previous year. Table 5 shows that the adjusted association between IPV and mental health indicators was stronger for current IPV than for past IPV. It should be noted, nevertheless, that women referring past IPV were 1.74 times more likely to present psychological distress (95% CI = 1.26–2.40) than never-abused women. Finally, females experiencing abuse both in the past and in the previous year were the most likely to present psychological distress and worse self-perceived health.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
There are also other limitations related to the exclusion criteria. On the one hand, women who attended the practice with their partners were excluded, and it could be that these women were more likely to be controlled or abused by their partners than women who attended the practice alone. However, the Ethical and Safety Recommendations for Research on Domestic Violence Against Women suggest that these women should not be included in research (24). On the other hand, despite the association found in previous studies between IPV and low socioeconomic levels, given that participants in our study were asked to complete a written questionnaire, illiterate women were excluded. All these limitations, however, could have only led to an underestimation of the real IPV associations.
Additionally, this is a cross-sectional survey and inferences regarding causality cannot be made. According to the results of this study, it could be plausible both that the mental health problems are a consequence of IPV or that women with poorer mental health are more likely to be abused. Until very recently, abused women were considered in some way "responsible" for the abuse they experienced from their partner because of neurotic disorders or personality traits that made them more vulnerable and more likely to be abused. However, in recent years, this hypothesis has been widely questioned. There is currently strong evidence that suggests that IPV can have serious mental health consequences. For example, a meta-analysis by Golding (6) used a subset of the causality criteria proposed by Hill (36) (strength, consistency, temporality, biological gradient, and experiment) to evaluate whether IPV is likely to be causally related to mental health problems. This author concluded that "the existing research is consistent with the hypothesis that IPV increases risk for mental health problems." Furthermore, Campbell and Soeken (37) found that the first episode of depression in abused women often coincides with the beginning of an abusive relationship. It was also found that the frequency of depression among abused women who had just left a shelter for battered women was 83%, and it decreased to 49% 6 months after abandoning the abusive relationship (38). Therefore, although results of this study should be interpreted with caution, all these data come to support the hypothesis that mental health problems are more likely to be the consequence of IPV than its cause. Moreover, we found strong associations between IPV and mental health problems in this study, and, in addition, intensity and duration of some abusive experiences also led to increased levels of association with impaired mental health.
The IPV prevalence data in our study are consistent with those of other studies (33–35,39–43). Hegarty and Bush (42) found a lifetime IPV prevalence of 37% in a sample of women attending general practice in Australia, and Richardson et al. (39) found that psychological abuse, measured by controlling and threatening behavior by partner, was more frequent than physical violence in women attending family practices in London. It should be noted that Spain has a public National Health System that aims to provide universal medical care. Moreover, primary care is the first contact people usually have with the health system, regardless of their age or gender. For this reason, although women in this study are not by any means representative of all Spanish women, we should acknowledge that findings from studies in the general practice are probably the closest data we can have to the real magnitude of the problem.
The prevalence of psychological distress in our sample (40%) is similar to that found in other Spanish studies conducted in the primary health care setting (44,45). In a study conducted in a sample of patients attending general practice in the north of the country, 38% of the women included presented psychological morbidity according to the 28-item GHQ (45).
Also of interest is the finding that, compared with women who reported no IPV, those referring lifetime psychological abuse solely were more likely to present mental ill health. In a population-based study conducted in the United States, Coker et al. (5) found that higher psychological IPV scores were more strongly associated with current depressive symptoms than were physical IPV scores. Furthermore, the psychological component of IPV has also been found to be the strongest predictor of PTSD (46). Our findings corroborate both that psychological violence is frequent in abusive relationships and that it is strongly associated with impaired mental health.
Women experiencing any kind of IPV also make a greater use of antidepressants and tranquilizers than women reporting no violence. This has also been found in previous studies (9,30,47). Recent research showed that the mean total expenditure difference for physician, drug, and hospital utilization between women who were currently experiencing severe IPV and those who were not abused by their partners was $1064 (95% CI $623–$1506) in a 2-year period (48). Early identification of IPV would greatly contribute to reduce the health impact and costs of IPV.
Our analysis also indicates that illicit drug use was positively correlated with most abusive experiences. This result supports the theory that drugs and alcohol are often used as a coping strategy to alleviate the stress associated with IPV (9). Coker et al. (5), for example, found that women reporting physical and sexual IPV reported using more alcohol, recreational drugs, and tobacco than nonabused women. In our study, however, only women who referred sexual abuse used significantly more alcohol than nonabused women. We should take into account that in Spain alcohol is used by both men and women across all generations and that this substance is increasingly being integrated into the norms of social interaction, being hence no longer exclusive of vulnerable populations. Moreover, it should also be considered that the nonexhaustive and crude measurement of alcohol use in this study could have led to a misclassification in this variable that may have accounted for its lack of association with IPV.
This research adds to the existing literature by strongly suggesting that the different kinds of IPV may have separate effects on women’s mental health. Despite the fact that women experiencing any category of abuse were more likely to present psychological distress than never-abused women, only those experiencing physical violence were more likely to report a poor self-perceived health. Given that self-perceived health status has demonstrated to be a very good predictor of morbidity and mortality (32) and that it reflects both physical and psychological health, one would expect a high positive correlation between psychological distress and self-perceived health in all abused women and not only in those experiencing physical violence. This finding could suggest that women tend to underestimate the severity of the situation by identifying only physical abuse as IPV but not other forms of abuse, such as psychological. In the same line, IPV categories including sexual abuse showed the strongest association with mental health indicators, even when the sexual abusive experiences occurred for less than a 1-year period. Again, this form of abuse also seems more visible and easy to identify than psychological abuse solely.
It also deserves mention that women experiencing the 3 types of abuse were the most likely to present negative mental health indicators. This was also observed when analyzing the mental health of abused women by the IPV intensity scores. This result seems to support the hypothesis of an accumulative effect of the diverse kinds of IPV on women’s health. Wingood et al. (12), for example, found that women experiencing both sexual and physical abuse were more likely to report adverse mental health consequences than women who had a history of physical but not sexual abuse.
Finally, the results of this study show that current abuse was strongly associated with poor mental health and, more interestingly, that women who experienced abuse in the past were also more likely to present current psychological distress than never-abused women. This finding suggests that the effects of IPV on mental health may not only be immediate and acute but could be long term or even become chronic. This has also been demonstrated by Romito et al. (30).
The results of this study highlight that IPV can be present even when there are no physical injuries. Furthermore, these findings underline the need for clinicians to screen not only for physical and sexual assault in intimate relationships but also for psychological forms of IPV (49,50). Informed, trained health professionals could have great potential for helping victims of IPV by providing them with the protection, resources, and support they need to make the changes necessary to improve their lives.
We thank the 23 general practitioners that recruited the sample. We also acknowledge the careful reading and thoughtful suggestions on this article from Antonio Daponte, PhD, Antonio Olry de Labry, MPH, Gastón Babio, MPH, and Ricardo Ocaña, PhD.
|
NOTES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This study was supported by a research grant from the Spanish Network for Research on Health and Gender (Carlos III Health Institute) (G03/042) and by the Epidemiology and Public Health Network (Carlos III Health Institute) (C03/09).
DOI:10.1097/01.psy.0000181269.11979.cd
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTESREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. Reed INTIMATE PARTNER VIOLENCE: A GENDER-BASED ISSUE? Am J Public Health, February 1, 2008; 98(2): 197 - 198. [Full Text] [PDF]
|
|
|
|
 |
|
 I. Ruiz-Perez, J. Plazaola-Castano, and C. Vives-Cases Methodological issues in the study of violence against women J. Epidemiol. Community Health, December 1, 2007; 61(Suppl_2): ii26 - ii31. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Ruiz-Perez, J. Plazaola-Castano, and M. del Rio-Lozano Physical health consequences of intimate partner violence in Spanish women Eur J Public Health, October 1, 2007; 17(5): 437 - 443. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Ruiz-Perez, J. Plazaola-Castano, M. del Rio-Lozano, and and the Gender Violence Study Group How do women in Spain deal with an abusive relationship? J. Epidemiol. Community Health, August 1, 2006; 60(8): 706 - 711. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
