Challenging Dogma - Fall 2008

Thursday, December 18, 2008

The Ineffectiveness of Mammography Interventions in Reaching African-American Women in the U.S. – Kate Laporte

The Centers for Disease Control and Prevention (CDC) reported that breast cancer mortality in the U.S. declined by 2.2 percent per year from 1990 until 2004 (1). This is due primarily to widespread use of mammography and early detection of tumors, which leads to a better prognosis than late-presentation cases (2). However, racial disparities in breast cancer mortality have persisted even as the overall incidence rates have dropped. Incidence rates are lower for African-American women compared to white women, but African-American women have higher rates of mortality from the disease (3). African-American women have also been found to have a higher risk of underutilization of mammography, which is a factor in the increased mortality rates (4). Traditional public health approaches to increasing mammography use have focused on raising awareness about the risks of breast cancer and the benefits of mammography through education efforts. The National Breast and Cervical Cancer Early Detection Program of the CDC describes the purpose of their recruitment program as follows: to increase the number of women in priority populations receiving clinical screening services by raising awareness, addressing barriers, and motivating women to use these screening services (5). Intervention strategies to increase the uptake of mammography are focused on three aspects: educational efforts that increase women’s knowledge about breast cancer and mammography, training programs to help physicians increase uptake of mammography by their patients, and increasing access to mammography, typically through mobile mammography clinics and ensuring that women are informed about insurance coverage for mammograms (6). The interventions have failed to reach that goal, especially among African-American women. Researchers at the University of California at San Francisco included over one million women in a recent study of mammography utilization and found that African-American women had a 1.2 odds ratio of not obtaining a mammogram with comparison to white women (4). This study, published in 2006, found that large, advanced-stage tumors and lymph node–involved tumors were more likely to be diagnosed in African-American women than in white women. However, when rates of mammography screening were accounted for, the differences were attenuated or eliminated. Public health interventions to increase mammography use, based largely upon traditional behavioral change models, have had too narrow a focus to be effective in reaching African-American populations. The following arguments, based on the social context theory, the structural influence model and framing theory illuminate the weaknesses of this approach.
Considering the Social Environment
According to social context theory, there are three dimensions of the social environment that should be taken into consideration when modeling social change (7). These are the following: societal structures or shapers (factors external to individuals such as technology, race and social class), social processes (perceptions, attitudes, values) and social realities (common patterns of social behavior). The public health campaigns to promote mammography and evaluations of such programs have been based upon traditional behavioral models that do not include macro-level processes. They have been focused on changing individual women’s attitudes and beliefs toward their susceptibility to breast cancer and the value of mammography without including the larger societal influences that weigh in on behavior. One recent cross-sectional survey of African American women’s knowledge, attitudes and beliefs concerning breast cancer screening was based on the Health Belief Model (8). The Health Belief Model postulates that people will perform a health behavior if the perceived benefits of the behavior outweigh the perceived barriers (9). The researchers found that the women were aware of the value of the screening process and they identified breast cancer as one of the top health concerns for African-American women. This suggests that public health efforts and fund raisers targeted at increasing awareness had been successful. However, actual mammography utilization was drastically lower than recommended standards; less than half of the women reported having had a mammogram in the past year. The solution recommended by the researchers was increased educational campaigns targeted to African-American women. The broader societal forces that impact African American women were not accounted for in this narrow approach. Other evaluations have similar recommendations. According to social context theory, social realities (common patterns of social behavior) impact individual behavior. The folkways and norms that form common patterns of behavior in African-American communities have been reported to contribute to lower rates of mammography (10). For example, African American women have reported that cultural norms prioritize acute care over preventative health care. There is a tendency for women to neglect preventative care when other concerns are pressing. These others concerns were reported to include neighborhood violence, housing issues and substance abuse. None of these concerns are included as barriers to mammography in the traditional public health approach. Social context theory also emphasizes the affects of societal structures (e.g. race, socioeconomic status) on behavior. Since there is a disproportionate amount of African-American women are living in poverty and in unsafe neighborhoods, these societal structures are important determinants of whether a mammography is obtained. Thus, interventions to increase mammography use in this population must take into account the social context in which health behaviors take place. Another example of the impact of community-level beliefs is the integration of other women’s experiences with breast cancer into the folklore and norms of the community. One study found that the shared experiences of the community were found to impact women’s attitudes and beliefs concerning breast cancer screening (10). Negative experiences with mammography or breast cancer were found to impact how women viewed obtaining a mammogram, regardless of whether the information was correct or if it had happened to someone else. Recommendations to obtain mammograms were disregarded due to the integrated beliefs that it was associated with pain and cancer diagnosis. Traditional educational methods do not address community-level norms and beliefs and fail to capture the impact of these negative experiences on other women’s health behavior.
Communication and the Health Care System
The traditional approach has failed to examine the impact of the experiences of African American women with the health care system upon mammography use. Real or perceived negative experiences with the health care system can create a sense of fatalism regarding cancer (10). However, traditional individual-focused approaches to promoting mammography use have not considered the system-level factors of African American women’s experiences with the health care system. For instance, communication difficulties may play a part in discouraging women to obtain a mammography. One study that used focus groups to elicit the ideas and concerns of African-American women regarding mammography reported women’s fears about cancer that arose from a mistrust of the health care system due to negative encounters with health care providers. The negative encounters were largely concerned with communication, such as inadequate explanation of what a mammography entailed or what would be done with the results (10). Another study documented that African-Americans are less likely than whites to have their physicians discuss treatment plans and preventive health care during clinical encounters. This suggests that racial disparities may exist in the amount of information communicated to African-American women about screening mammograms (11). The structural influence model holds that social determinants (e.g. socioeconomic position) and mediating or moderating conditions (socio-demographics of age, gender, and race/ethnicity and social networks of social capital and resources) impact communication outcomes (12). Communication outcomes include information access, information processing and information utilization. A structural level approach to increasing mammography use would recognize that social determinants and mediating conditions influence communication of patients and providers. Race and poverty, in particular, can play mediating roles in the experiences of African-American women with their providers. Since a disproportionate amount of African-Americans are poor, they will encounter the health care system differently. Poverty has a negative impact on the behavior of health care providers and the availability of health services. Those who provide health care for minorities and people in low income areas, for example, are often less informed about preventive care services and are less likely to be board certified (13). This has not been accounted for in traditional provider training programs that have focused largely on increasing physician recommendation for mammography (6). Training that emphasizes increasing recommendations without regard to the other factors that are impacting communication is most likely ineffective. The structural influence model offers a more comprehensive view than the traditional approach of the interaction of African-American women with the health care system and the impact of that communication on mammogram use.
Framing the Issue
Much has been studied about the disparities in breast cancer mortality between African-American women and white women. The disparities have been documented for greater than thirty years (3). Interventions aimed at increasing mammography use have been focused on individual-level behavior and the problem of mammography utilization among African-American women has been thought of as an education and motivation problem. The social determinants of health have not entered the picture of breast cancer health disparities. Framing theory provides the means to readjust the paradigm concerning mammography use as one of a social and systems problem rather than an individual’s failure. A message can frame population health disparities as being caused by internal factors (within control of the individual), external factors (beyond the control of the individual), or some combination of the two (14). Instead of focusing on under utilization of mammography, the shift to a broader perspective of health disparities would take into consideration the underlying social determinants of health. The social determinants of race and the often correlating factor of income level serve little function as descriptions of study group participants. Their impact on women’s experiences with the health care system and the types of barriers faced by women in their everyday lives to preventative health care are critical factors in understanding why disparities have persisted. Investigation into the social determinants of health can lead to policy change that would address the fundamental underlying factors of disparities. Barriers inherent in the health care system such as communication difficulties could then be addressed on a widespread level. The social norms that discourage preventative health care could be addressed within African-American communities. Finally, agencies and organizations that aim to increase mammography uptake could take into account the real-life society-level concerns that African-American women face and design approaches that consider these concerns.
In conclusion, the traditional, educational, public health approach to mammography uptake has failed. Disparities in mammography utilization and related breast cancer mortality in African-American women in the U.S. have persisted despite millions of dollars of educational and awareness programs. Social science theory elucidates the limitations in the traditional approach. A comprehensive picture of the social context of health behavior and the role of the health care system in promoting mammography use offers new perspectives concerning the underlying determinants of health disparities. The new perspective gained can help frame this health disparity in a way that reflects social responsibility.
A novel approach
Through use of the spatial interaction model, Mobley and colleagues describe a comprehensive approach to factors affecting mammography use (15). The model was applied to aggregate pooled information from several heterogeneous states in the U.S. The aim was to demonstrate that pooled data can provide misleading information regarding predictors of health care utilization. The model includes factors that impact mammography use at several different levels, including fundamental/macro factors, intermediate or community factors, interpersonal factors, and individual factors. It draws from different disciplines to create a more comprehensive picture of what impacts health behavior than individual beliefs and perceived risks (Health Belief Model). Each of the levels, from the outermost (fundamental/macro factors) to the innermost (individual factors) impedes on the next level until, ultimately, the cumulative effects weigh in on individual behavior. The model is described below:
Fundamental/Macro factors:
Distribution of wealth, educational opportunities, and political influence; social and economic policies, institutions, regulations, campaigns, topography, climate, water supply
Intermediate or Community:
Social context – neighborhood, workplace, and housing conditions; public infrastructure and investment; police, enforcement services, crime; health care system
Health care system: proximity and density of facilities, physicians; crowding, scheduling and convenience, personal physician, managed care climate, primary care physician shortage; international medical graduate enclave
Physical environment – community capacity and partnership; land use patterns, transportation systems, buildings, public resources, pollution
Interpersonal:
Stressors, social integration and support, psychosocial factors, behavioral settings, social relationships, living conditions, neighborhoods and communities, neighborhood watchfulness, driver courtesy, social or cultural cohesion, population health behaviors or norms
Individual/Population:
enabling/disabling: personal disability, personal resources, type of health coverage, new address, marital status, employment status
predisposing: age, sex, gender; race or ethnicity, educational attainment
need: beliefs, family history, perceived risk, health status
Accounting for social context
The terms that are highlighted were discussed in previous sections as potential mediating factors in African-American women’s mammography rates that were left unaddressed by the traditional approach. The traditional approach does not consider the social context in which African American women live, including social norms particular to their communities. Population health behaviors or norms are integrated into the interpersonal level in the spatial interaction model. Also, concerns that keep African American women from obtaining mammograms such as neighborhood violence and housing conditions are included at both the intermediate and interpersonal levels. The study based on this model found that factors at the intermediate level did, in fact, affect mammography rates differently across states. In particular, the researchers found that in five of the states, the proportion of the workforce who commuted more than sixty minutes each way to work was negatively associated with mammography use. These findings highlight the need to examine specific social contextual factors that traditionally seem unrelated to health care utilization.
Accounting for system level factors
The model also takes into account the characteristics of the health care system that can promote or inhibit mammography utilization. In particular, the availability of primary care physicians may have an impact on mammography use. Physician shortages tend to occur in poorer areas and this factor may have a disparate affect on African American women, since a disproportionate amount of African American women live in poverty. Crowding, scheduling, convenience and the availability of a personal physician may all play a role in determining whether African American women perceive their experiences with the health care system as positive or negative. These types of variables cannot be quantified at the individual level but require a systems perspective. However, valuable qualitative data from women’s experiences can be obtained through focus groups and open-ended questionnaires that can help illuminate the specific areas of concern. Here, the concerns that surfaced through focus groups are included in the model as mediating factors on mammography use.
Reframing the issue
The spatial interaction model includes race/ethnicity as an individual level factor. While this may seem to be akin to the traditional approach, the model accounts for the impact of race/ethnicity as a determinant of health care utilization through multilevel modeling. The macro-level, intermediate and interpersonal factors that are modeled are the same factors that affect people of different racial/ethnic groups differently. Thus, while including race or ethnicity as an individual risk factor, the spatial interaction model also investigates the higher level processes by which people of various races and ethnicities are affected. The authors reframe the issue of disparate mammography use as one of differences in place-specific resources. The issue of place-specific trends is tightly joined to racial health disparities due to the extremely high degree of racial geographic segregation in the US (16). The health disparities that affect one racial or ethnic group, such as African-American women, are reflective of the place-specific resources and conditions acting upon their lives.
Conclusion
The structural and social forces that drive racial inequalities are being recognized gradually in public health research as the underlying, foundational determinants of health disparities. Mammography utilization is particularly important for African American women, whose mortality rates remain high and frequently present with later stage disease than their white counterparts. Education and individual-level interventions have proven ineffective in reducing disparities over the last thirty years. Specific social contextual factors and broader structural determinants must be addressed if this gap in health care utilization is going to be eliminated.

REFERENCES:
1. http://www.cdc.gov/cancer/breast/statistics/trends.htm; accessed on 12/01/08.
2. Feig SA. Effect of service screening mammography on population mortality from breast carcinoma. Cancer 2002; 95:451–457.
3. Newman LA. Breast Cancer in African-American Women. The Oncologist 2005; 10:1-14.
4. Smith-Bindman R. et al. Does Utilization of Screening Mammography Explain Racial and Ethnic Differences in Breast Cancer? Ann Intern Med. 2006; 18:541-53.
5. http://www.cdc.gov/cancer/nbccedp/ accessed on 11/24/08.
6. Wong FL. The Manual of Intervention Strategies to Increase Mammography Rates. The Centers for Disease Control and Prevention. 1997. http://www.cdc.gov/cancer/nbccedp/publications/; accessed on 12/02/08.
7. Earle L and Earle T. Social Context Theory. South Pacific Journal of Psychology. 1999; 11(2).
8. Sadler GR et al. Breast cancer knowledge, attitudes, and screening behaviors among African American women: the Black cosmetologists promoting health program. BMC Public Health 2007; 7(57).
9. Rosenstock IM. Historical origins of the health belief model. Health Educ Monogr. 1974;2:328-335.
10. Peek ME, Sayad JV and Markwardt R. Fear, Fatalism and Breast Cancer Screening in Low-Income African-American Women: The Role of Clinicians and the Health Care System. J Gen Intern Med 2008; 23(11):1847–53.
11. Oliver MN, Goodwin MA, Gotler RS, Gregory PM, Stange KC. Time use in clinical encounters: are African-American patients treated differently? J Natl Med Assoc 2001; 93:380–85.
12. Taylor-Clark K, Koh H and Viswanath K. Perceptions of Environmental Health Risks and Communication Barriers among Low-SEP and Racial/Ethnic Minority Communities. Journal of Health Care for the Poor and Underserved 2007; 18:165–183.
13. Gerend MA and Pai M. Social Determinants of Black-White Disparities in Breast
Cancer Mortality: A Review. Cancer Epidemiol Biomarkers Prev 2008;17(11).
14. Niederdeppe J, BU QL, Borah P, Kindig DA and Robert SA. Message Design Strategies to Raise Public Awareness of Social Determinants of Health and Population Health Disparities. The Milbank Quarterly 2008; 86(3):481–513.
15. Mobley, LR, Kuo T-M M, Driscoll D, Clayton L and Anselin L. Heterogeneity in mammography use across the nation: separating evidence of disparities from the disproportionate effects of geography. International Journal of Health Geographics 2008; 7(132).
16. Williams DR and Collins C. Racial Residential Segregation: A Fundamental Cause of Racial Disparities in Health. Public Health Reports 2001; 116: 404-416.

Labels: , , ,

Abstinence Education: A Critique and Alternative Approaches for Improvement – Kathy Zheng

The United States spent about $1 billion dollars on abstinence education between 1998 and 2007 (1). The goal of this campaign, as promoted by President George W. Bush beginning in 2001, is to decrease the incidence of teenage pregnancy and sexually transmitted diseases through the promotion of abstinence among unmarried individuals. The Department of Health and Human Services established the website “4Parents.gov” as part of the campaign. Its main purpose is to provide information for parents about how to talk to their children about abstinence. However, despite the widespread efforts of the government and the vast funds that have been dedicated, the abstinence campaign has not been successful. Studies show that in areas where abstinence education is explicitly promoted in schools, the rate of sexual activity among teens increased after completion of the education program (2). The number of pregnancies among teens who participated in the abstinence only programs also increased (3). Thus, it is clear that abstinence-only education programs have fundamental flaws. While some of the flaws are rooted in the theoretical basis from which the program is designed, others are in reference to the audience in which the program is directed towards. Several criticisms of the approach, specifically the use of the 4Parents.gov website, as well as suggestions for the improvement of the intervention are presented here.
Criticism 1 – Theoretical Basis
It is evident that the abstinence-only education programs are based on several traditional models of health behavior. These models include the Theory of Reasoned Action (TRA) and the Health Belief Model (HBM). 4Parents.gov emphasizes the importance of parents communicating to their children their desire for them to abstain from sex until marriage (4). The argument is that children who understand their parents’ desire for them to abstain are more likely to do so. One of the facts that the website encourages parents to share is “Do you know that, according to one survey, two-thirds of teens who have had sexual intercourse wish they had waited?”
This approach is clearly modeled after the TRA because it aims to change a person’s attitudes towards a behavior and present abstinence as a socially desirable and acceptable behavior. One major criticism of this approach is that it is an individual level intervention. This is paramountly inappropriate for interventions dealing with teenage pregnancy and STDs because neither is individually acquired. Another argument to this approach is that it assumes that individuals will systematically weigh the benefits (100% prevention of pregnancy and STDs) and public perception and naturally arrive at a mathematical conclusion to abstain. Thus, 4Parents.gov is fundamentally arguing that is it a lack of knowledge and input that contributes to premarital sex. This is exceptionally unreasonable given the major emotional and hormonal input in a teenager’s decision to have sex.
Abstinence-only education is primarily a data based intervention. Proponents often cite statistics about the rate of STD transmission among the sexually active and unmarried (4). They frequently use data about the higher incidence of poverty among unmarried women who get pregnant as a tool to discourage sex before marriage. To its credit, 4Parents.gov presents some information about contraception other than abstinence. However, its merits are quickly abated when upon closer inspection, one notices that the failure rates of each form of contraception is prominently noted. This presentation clearly highlights the viewpoint of the website about abstinence rather than encourage the discussion of its alternatives.
The approach is also based on the HBM. The severity of the consequences of premarital sex and the susceptibility of a person to those consequences are presumed to be the main driving force behind the decision to abstain. Studies have shown that “knowledge of consequences” is the third most often cited reason for the use of abstinence education (5). Interventionists must trust that the teenager will view this information as dangerous enough to abstain from sex. However, it would not be prudent to assume that the teenager would then naturally make the rational decision. Adolescence is characterized by risk taking and rebellion (2). Therefore, a conscious effort by the teenager to act against the rational decision would be more likely. This model also creates an individual level intervention and this is inappropriate because sexual activity is generally not an individual level action. Furthermore, abstinence-only sex education assumes that the decision to have sex is made inside a vacuum, that is, without social or environmental determinants or influences. This is inappropriate because an adolescent’s decision to have sex is well-known to be influenced by the pressures or actions from others. Finally, abstinence-only education based on both the TRA and HBM assume that the adoption of abstinence is planned. This does not take into account the spontaneous nature of teenage activities, especially those sexual in nature (6).
One alternative model of health behavior that may explain the failure of abstinence-only education is Stigma or Labeling Theory. This theory argues that when people are labeled as a particular type of person, there is a self-fulfilling prophecy. If a teenager has already had sex and plans to continue having sex, the only advice the 4Parents.gov website gives to parents is to continue talking to the teenager and have her visit a health care provider. Although, there is information on the website about birth control, as mentioned, there is no information about how to speak to teenagers about using them. Thus, if the teenager is already sexually active, Labeling Theory suggests that teenagers may view themselves as being beyond the scope of abstinence-only education, and therefore may reject sex education altogether. The intervention does not comprehensively address the needs of this population and teenagers may continue with their unsafe sexual practices because there then appears to be no reasonable alternative. In fact, teenagers who find that they are unable to abstain may experience shame (2). This could lead to the teenagers being secretive about their sexual practices and discourage them from seeking out advice about ways to prevent pregnancy and STDs.
Criticism 2 – Contextual/Ecological Factors
As mentioned, abstinence-only education based on the traditional health behavior models is an individual level intervention. Despite the fact that actions to protect oneself from teenage pregnancy or STDs may be individual in nature to a certain extent, interventionists may not be able to act without knowing the context of the problem. There may be higher level factors that abstinence alone cannot reasonably address and they may be the fundamental causes of why a teenager is engaging in risky sexual practices.
One risk factor for such practices may be the community one interacts or lives within (7). For example, if the community is one where most residents are of low socioeconomic status, parents may work more than one job that requires long hours. 4Parents.gov assumes that parents have the time to maintain a continuous dialogue with their teenagers about the risks of sex. Their argument is that this dialogue is most influential for a teenager’s decision to abstain from sex. However, if the parents are unable to devote the time, 4Parents.gov does not offer an alternative solution to the problem.
Areas of low socioeconomic status may also be deprived of recreational activities associated with the community, church, or school for adolescents. Research shows that adolescents who participate in these activities are less likely to engage in risky sexual behavior (8). Abstinence-only education does not address the possibility of this deprivation of social activities and thus ignores the higher order contextual risk factors associated with the initiation of sex among adolescents.
The individual level intervention fails to address the influence of the media in a teenager’s decision to have sex. Images of sex are pervasive in the media and research shows that teenagers exposed to those images on television are more likely to have their first sexual experience sooner (9). However, the study also shows that the effect of sex in the media can be countered by parents watching television with their children and sharing their beliefs about safe sex. This would also assume that the parents can devote the time to partake in such activities. If this assumption is incorrect, then the influence of the media can be substantial.
Criticism 3 – Social and Cultural Factors
Abstinence-only education fails to address the social and cultural differences of teenagers and their families. A lack of understanding of these differences may lead to teenagers feeling marginalized and unworthy. In addition, an intervention that does not reasonably and comprehensively address alternatives to abstinence greatly limits the context in which the intervention can be presented, and context is of primary concern when approaching a subject that is culturally sensitive such as sexual activity.
There are cultures that practice family silence about topics relating to sex. For example, Choi et al states that “Asian families are not able to acknowledge sexual identity” (7). 4Parents.gov suggests that sexual abstinence by teenagers can be achieved mainly through dialogue with parents about the risks of sex. If sex is a taboo subject within a culture, then this dialogue is not possible. As a result, the underlying message propagated by the campaign would be inapplicable to the Asian population. Moreover, since the website does not comprehensively address alternatives to open discussions about abstinence and sex, teenagers may develop repressed sexual urges that may encourage them to seek out sex, safe or not (7).
In addition to racial differences, studies also show that there are gender differences with respect to adolescent attitudes about sex. Girls have more negative perceptions about the benefits of sex and less negative perceptions about pregnancy as compared to boys (10). 4Parents.gov does not address gender differences with their approach. The assumption is that both girls and boys would be equally receptive to the messages presented by the website. Since the research shows that there is a gender difference in sexual attitudes, the approach is unreasonable.
Abstinence-only education also fails to address the sexual practices of those for whom marriage may not be an option. The intervention operates under the assumption that its entire audience will eventually enter into a mutually faithful, monogamous relationship in the context of marriage (11). This approach discriminates against the gay, lesbian, and transgender youth (12). Marriage may not be possible for these individuals due to government policies. Therefore, abstinence-only education interventions either do not condone sexual activity for these people or they feel that this group is unworthy of safe sex intervention. The concept of worthiness is one that is often cited in support of sexual abstinence for teenagers (13). If teenagers who are gay, lesbian, or transgender do not abstain before marriage because marriage is unlikely, then they may develop a feeling of unworthiness if exposed to the abstinence message. In reference to Labeling Theory as discussed, this feeling may lead them to continue with unsafe sexual practices and may even exacerbate the problem.
As shown, there are many fundamental reasons for the failure of the abstinence-only campaign to fight pregnancy and the spread of STDs among adolescents. It is evident that the failure is unrelated to the lack of funding, but rather it can often be attributed to the lack of understanding of its audience. In addition, it can be seen that the traditional health behavior models inadequately address issues with significant social and emotional components that may also happen to be culturally sensitive. As a result, a more comprehensive and less contextually restrictive approach to the problem is warranted.
Alternative 1 – Predictable Irrationality and Alternative Health Theories
The abstinence-only campaign based on the traditional health belief models does not account for the irrationality of teenage behaviors and decisions. Therefore, an alternative intervention is suggested where there is no assumption of a rational decision on the part of the teenager to abstain from sexual activity and the basis for the intervention will be an alternative health behavior model. Studies show that among individuals aged 10 to 24 years old, 47.8% were sexually active in 2007 (14). About half of teenagers do not remain abstinent despite the 4Parents.gov campaign. Studies show that comprehensive sex education programs that include promoting methods of safe sex do better to reduce the number of partners, improve condom use, and reduce pregnancy among teenagers compared to abstinence-only education programs (15-16). These comprehensive programs are critical because they address the needs of teenagers who are already sexually active.
One suggested alternative health theory to use as a basis for an intervention is Marketing Theory or the Social Marketing approach (17-18). Unlike the TRA and HBM, Marketing Theory is a group level intervention because it does not seek to target each individual separately. The issue of preventing teenage pregnancy and STDs lends itself better to a group level intervention because of the social and peer influences associated with sexual decisions. Also unlike the TRA and HBM, this theory does not present the negative consequences of sex, but rather the positive outcomes of safe sex. Advertising Theory, which is a subset of Marketing Theory, argues that interventionists should first identify teenagers’ aspirations. The promise of a teenager’s fulfillment of those aspirations if they adopt safer sexual behaviors can be made through visual representations in the media.
Interventionists should look to the success of popular clothing brand campaigns for ways to market aspirations to teenagers. The television, billboard, or magazine advertisements for these clothing brands often do not even need to rely on a presentation of the product to be effective. Rather, the aspirations being presented may include teenagers being perceived as physically attractive, popular, independent, etc. Therefore, it is reasonable to assume that the same formula can be effective for a safe sex campaign. Commercials and print advertisements portraying attractive and popular teenagers in relatable yet glamorized settings could be presented along side messages promoting condom or contraception usage.
Alternative 2 - Addressing Higher Level Factors
As mentioned, the United States spent about $1 billion between 1998 and 2007 on abstinence education (1). There is clearly no lack of available funding for sex education. The issue is then to find better a way to utilize those resources to produce lower incidences of teenage pregnancy and STD infections. A criticism of the 4Parents.gov initiative is that it does not address the possibility of the deprivation of social activities within neighborhoods of low socioeconomic status. A lack of participation in recreational activities associated with the community, church, or school is a risk factor for engaging in risky sexual behavior among teenagers (8). Therefore, a successful intervention must do more than address the risk factors association with teenage pregnancy and STD transmission; it must also address the risk factors associated with risky sexual behaviors. The recommendation is to allocate some of the resources reserved for sex education towards funding for increased after-school activities or community recreational activities geared towards teenagers.
4Parents.gov also fails to address the influence of the media. However, an intervention based on Marketing Theory utilizes the influence to its advantage. If studies show that teenagers exposed to sex in the media are more likely to have their first sexual experience sooner, or media campaigns such as Florida’s “truth” initiative are able to decrease the incidence of teenage smoking by 7.4% in 30 days, then it is reasonable to believe that the media can also be a powerful tool to influence teenagers to adopt safer sexual practices (9, 19). Therefore, increased funding for recreational activities for teenagers in conjunction with the use of a Marketing Theory-based media intervention would have improved success over the 4Parents.gov campaign.
Alternative 3 – Cultural Competence and Social Awareness
In addition to community based programs to intervene where parental involvement may be limited, programs should also be established to provide support to parents who are able to take an active interest in their children’s sexual wellbeing. Addressing the social and cultural factors that limit the success of the 4Parents.gov campaign requires the establishment of such programs.
It is important to recognize culture as an important predictor of sexual initiation among teenagers. Although the relationship between culture and attitudes towards sex in the United States has not been fully explored in the literature, a study performed in the United Kingdom shows that an open dialogue about sex within Chinese families was uncommon often due to the language and cultural barriers between the UK-born teenagers and the China-born parents (20). Although abstinence-only education may seem inappropriate given that half of all United States teenagers have sex, it is important to note that the a generalized comprehensive sex education program may also be inappropriate for families of cultural backgrounds with conservative views about sex.
It is important for interventionists to recognize that some cultural views about sex may be more conservative than mainstream views (20). One approach would be to dedicate funds to establish community programs to teach parents how to overcome language, cultural, and generational barriers to effectively convey their own attitudes about sex. These programs should be led by individuals who have the same cultural backgrounds as those being served because there will be an intimate awareness of what is and is not appropriate. Another approach would be to produce television and print campaigns to air on foreign language channels and publications. These foreign language campaigns could show images of happy families and healthy teenagers with the message that effective culturally appropriate communication about sex could lead to these aspirations. In addition to comprehensive sex education programs directed towards teenagers who choose not to abstain, these approaches could show cultural groups how to give parental support to those teenagers who do abstain because of their own beliefs without feeling alienated by mainstream views.
The lesbian, gay, bisexual, and transgender (LGBT) community is one where abstinence education is inappropriate, for reasons described above. A feeling of worthiness should be reinforced within this community despite the fact that sexual activity occurs outside of marriage. Television and print campaigns, such as those previously suggested, can be tailored to present positive images of LGBT teenagers more prominently. Another direction campaigns geared toward LGBT’s can adopt is to make methods of safe sex provocative and fun. For example, there are condom commercials which show the sexual attractiveness of a man who has a condom to women. The same types of commercials can be produced showing LGBT couples.
Lastly, an important consideration for both the use of media and community programs for an intervention is gender differences. Studies have shown that girls and boys have differences in their perceptions about sex (10). As a result, interventionists should produce one set of media campaigns geared toward girls, such as portraying aspirations of being respected by peers for being sexually responsible, and another gears towards boys, such as portraying the independence one could have if one was not burdened by the responsibility of early fatherhood.
Ultimately, the use of the media as well as establishing community programs to address higher order societal risk factors for teenage pregnancy and STD transmission are improvements upon the current interventions adopted by the government. A government campaign to battle teenage pregnancy and STD transmission has the unique advantage of not being limited by funding, as public health initiatives tend to be. Public health professionals should look to successful marketing campaigns for commercial goods aimed at teenagers for inspiration on how to design a successful safe sex campaign. The funding is available for interventionists to purchase the same expensive air time slots as consumer product companies buy. Perhaps the unorthodox approach of marketing safe sex as a product rather than as a behavior is one that teenagers will finally respond to.
REFERENCES
1. Pilkington E. $1bn ‘don’t have sex’ campaign a flop as research shows teenagers ignore lessons. Guardian News and Media. http://www.guardian.co.uk/world/2007/apr/16/schoolsworldwide.usa
2. Katz A. Abstinence-only education programs. AWHONN 2006; 10:30-32.
3. DiCenso A. et al. Interventions to reduce unintended pregnancies among adolescents: systematic review of randomized controlled trials. BMJ 2002; 324:1426-1430.
4. Department of Health and Human Services. Washington, DC. http://4parents.gov/index.html
5. Goodson P. et al. Is abstinence education theory based? The underlying logic of abstinence education programs in Texas. Health Education and Behavior 2006; 33:252-271.
6. Salazar MK. Comparison of four behavioral theories: a literature review. AAOHN Journal 1991; 39:128-135.
7. Choi KH, Yep GA, and Kumekawa E. HIV prevention among Asian and Pacific-Islander American med who have sex with men: a critical review of theoretical models and directions for future research. Aids Education and Prevention 1998; 10:19-30.
8. Ramirez-Valles J, Zimmerman MA, and Newcomb MD. Sexual risk behavior among youth: modeling the influence of prosocial activities and socioeconomic factors. Journal of Health and Social Behavior 1998; 39:237-253.
9. Collins RL et al. Watching sex on television predicts adolescent initiation of sexual behavior. Pediatrics 2004; 114:e280-e289.
10. Cuffee JJ, Hallfors DD, and Waller MW. Racial and gender differences in adolescent sexual attitudes and longitudinal associations with coital debut. Journal of Adolescent Health 2007; 41:19-26.
11. Hampton T. Abstinence-only programs under fire. JAMA 2008; 299:2013-2015.
12. Santelli J. et al. Abstinence-only education policies and programs: a position paper of the Society for Adolescent Medicine. Journal of Adolescent Health 2006; 38:83-87.
13. Morrison-Beedy D. et al. Understanding sexual abstinence in urban adolescent girls. J Obstet Gynecol Neonatal Nurs. 2008; 37:185-195.
14. Eaton DK. Et al. Youth risk behavior surveillance – United States 2007. MMWR Surveill Summ. 2008; 57:1-131.
15. Kirby D. Effective approaches to reducing adolescent unprotected sex, pregnancy, and childbearing. J Sex Res. 2002; 39:51-57.
16. Kohler PK et al. Abstinence-only and comprehensive sex education and the initiation of sexual activity and teen pregnancy. J Adolesc Health. 2008; 42:344-351.
17. Kotler P and Zaltman G. Social marketing: an approach to planned social change. J Market. 1971; 35:3-12.
18. Kotler P. Marketing for non-profit organizations. Englewood Cliffs, NJ: Prentice-Hall, 1975.
19. Hicks JJ. The strategy behind Florida’s “truth” campaign. Tobacco Control. 2001; 10:3-5.
20. Yu J. British-born Chinese teenagers: The influence of Chinese ethnicity on their attitudes towards sexual behavior. Nursing and Health Services. 2007; 9:69-75.
21. Garofolo R, Mustanski B, and Donenberg G. Parents know and parents matter; is it time to develop family-based HIV prevention programs for young men who have sex with men? J Adolesc Health. 2008; 43:201-204.
22. Waldner LK and Magruder B. Coming out to parents: perceptions of family relations, perceived resources, and identity expression as predictors of identity disclosure for gay and lesbian adolescents. J Homosex. 1999; 37:83-100.

Labels: , , ,

Wednesday, December 17, 2008

Prevention of Eating Disorders: Failing to Protect Young Women from the Thin-ideal. – Marissa L. Garofano

The prevalence of eating disorders is growing problem, and over the past 50 years the number of cases has steadily increased. Eating disorders; anorexia and bulimia, affect a small proportion of people in the United States, 0.5% - 1% and 1 – 2% respectively and possibly as much as 10% (1, 2) However, they are costly to treat and effective prevention methods are necessary. At least an additional 10% of women and girls in late adolescents have eating disorder symptoms at any given time (2). Health care professionals recognize the importance of primary prevention due to the difficult nature of treating eating disorders (3). In addition, only the small proportion of people who have a clinically diagnosable eating disorder will receive treatment, an unknown amount of others who have some symptoms receive no treatment.

There are a multitude of risk factors that may lead to the development of an eating disorder. The factors that receive the most attention in prior prevention efforts are individual factors which include: child abuse, having been teased, perfectionism, obsessive thoughts, and body dissatisfaction among others (1). In addition to individual risk factors, sociocultural influences including parents, peers and the media are being shown to have an effect on the development of eating disorder (1,4). There is a relationship between the two but there has been a greater emphasis on individual risk factors rather than social ones.

There has been an inverse relationship to the amount of available food and ideal body of certain groups of people; as there begins to be a surplus of food, the ideal body size tends to become smaller (1). The media norms, on television and in magazines, have helped to accelerate this process. American women depicted in the media have become steadily thinner in the past 50 years, and more than half meet the criteria for anorexia nervosa (5). The media sends constant messages to children and adolescents about what it means to be beautiful, successful, feminine, attractive and desirable: for women this comes with slenderness (6).

The Thin-Ideal and Drive for Thinness

The media is the greatest contributor to a culture’s thin ideal. The media creates a social norm that girls feel they need to live up to. Many young girls aspire to be actresses or models that are greatly admired by Western cultures and have a high status and a lot of money. Currently most every show on television and fashion magazines depicts thinness as the norm (4, 6). Slender, airbrushed women with very little fat, and no wrinkles or pimples are shown on television or fashion magazines. These images set the standards for women so high that failure is inevitable.

In addition to the glorification of slenderness in the media being fat is also looked down upon (6). Because of this and due to the amount of exposure to these ideals on the media, girls may begin to compare themselves to the slender actresses and develop a negative view of their own bodies (6). Today, children under the under the age of six spend around two hours a day watching television, or movies (7). Between the ages of eight and eighteen children spend around four hours watching television and an additional two in front of a computer (7). The amount of time older children and teens spend in front of a screen is nearly one fourth of their day. From this, it is evident how the media is able to instill these ideals in youth. It is the internalization of the thin-ideal that has been associated with weight concerns and disorder eating habits (8)

Current prevention methods have failed and this is demonstrated by the growing number of cases of eating disorders and increasing prevalence of dieting and unhealthy eating behaviors. The failure of current eating disorder prevention methods is due to the ineffectiveness of current education programs, the failure to alter social norms, and the media’s promotion of the thin ideal.

Current Education Methods are Ineffective

It has been shown that preventing eating disorders through changes in knowledge, attitude, and behavior via education is difficult and often unsuccessful (9). The majority of previous educational interventions have included information on the nature and consequences of eating disorders and dieting (10, 11). Many of these studies have shown an increase in knowledge but no change in eating behavior (11). Interventions of this nature use the health belief model (HBM) for behavior change. This model assumes that increasing the knowledge about the benefits of eating healthy and consequences of dieting and eating disorders will lead a person to weigh the risks and benefits of change against what they perceive to be the severity of eating disorders and how susceptible they feel they are to them. If the benefits of change are high along with the severity and susceptibility, this will lead to the intention to change, and then the adoption of a healthy behavior change (12). Prevention methods based upon the HBM address individual factors and does not account for outside social factors that influence a behavior. Also, the HBM is most effective when used for short term change or one time decision, as demonstrated by how soon after an intervention their was a greater change in attitudes but after time had passed the strength of theses attitudes decreased (10). Maybe the first night when these facts are still fresh in her memory, a girl may not eat a large tub of ice cream and then purge or fast afterwards. However, three months later, during a stressful time period, this may not be the case. The HBM suggests that teenage girls make rational decisions, and highly value their health over being accepted by their peers.

While education has very little effect on positive behavior change it may also have a negative effect on behavior and will lead to more unhealthy eating behaviors. It has been shown that providing detailed information about the symptoms of eating disorders to children and adolescents can increase the amount of disordered eating habits that occur (10, 13). While behavior and attitudes may change positively right after the implementation of an intervention, months later the girls may be back to their original views and behaviors, or they may have developed more negative views and have more disordered eating behaviors (10). The idea of providing teachers and faculty at schools with a detailed overview of the causes and symptoms of eating disorders has also been suggested (14). However this knowledge may create false accusations of students and stigmatization of them if not enough information is provided. If a stigma is created the student may fall into a self-fulfilling prophecy where she engages in disordered eating behaviors because others already accept it from her (15).

Current interventions for prevention are typically being implemented in adolescence or young adulthood. The time at which most educational interventions have been implemented are at or around the onset of puberty (11). By this time many disorders attitudes towards eating habits my have developed as by the time girls are 10, 81% are afraid of being fat (16). Also, by adolescence one’s views are already set and it is more difficult to alter these opinions. This suggests that intervention should be implemented at a younger age when children are more malleable and absorb the most information.

In addition current educational interventions fail to account for important influences on behavior; social and contextual factors (12). Education assumes that people want to be healthy rather than fit in with social norms. Also it fails to address the reasons for why these behaviors were adopted in the first place.

Failure to Alter Social Norms

The occurrence of dieting and the desire to be thin are extremely common among women and girls and begin at a very young age, as young as six. Forty-two percent of 1st through 3rd graders want to be thinner (16). It is common that when with a group of young women you will hear talk about current diets, being too fat, and needing to lose their excess weight. Trends in dieting are discussed as frequently as trends in fashion. Women will attempt to conform to the current definitions of what is beautiful and attractive in order to fit in with their social group (15). However, what women perceive as normal may not coincide with actual societal norms. They perceive that the most attractive figure to men is one that is very slender (18) when men actually prefer a more curvaceous woman with muscle tone to one who is super skinny and mostly bone. Women envision an ideal body shape that is smaller than their current one, feeling that they need to lose weight, to achieve that of what is perceived as attractive and what their peers hold as ideal (17). An attempt to achieve this ideal and fit in with their peer’s norm, may lead girls to begin methods of unhealthy methods to achieve the perceived ideal. The CDC found that, in 2000, 37.5% of girls trying to lose weight did so in an unhealthy manner (18). A study of high school girls showed that an increase in an individual’s unhealthy eating habits was correlated with perception of her friends dieting habits (19) which suggests that eating behaviors are not solely based on individual factors and the weighing of risks and benefits but group level factors as well.

Prevention methods that stress education do not account for perceived social norms or group level factors. Facts may be provided about the average size of a woman and comparing her to the average model but there is a lack of discussion regarding the perceived norms the girls have and how they feel about them. Marketing theory suggests that focus groups need to be created to determine what the girls think would be effective and what their current ideals are (20). After creating an intervention with the girls concerns in mind the outcomes will be more successful, as education does address these concerns and the programs are not created with the help of the girls who are affected by the norms.

The Social Networking theory gives insight as to how these ideals and practices have become normal. The theory relies on the relationships and interactions of the individuals within a group (21). If you have a group that consists of high school girls and analyze the relationships of them you will most likely see the thin girls who comply with society’s norm in the middle with the most friends and the heavier girls who are perceived as unattractive will be marginalized (21). Body shape will also dictate who will be in the subgroups or cliques in the network. The heavier girls will have less interpersonal contacts within the group. Social norms provided by the media suggest that people who are heavy are outsiders, and possess undesirable qualities and therefore it is acceptable to not interact with them and avoid contact (5). In order to change this social network one of two things would have to occur, either a change in norms to accept heavier people as attractive or, more likely, the heavier girls will adopt disordered eating behaviors to try to move to the center of the network and gain acceptance from peers.

Also in the social network would be parents and family members, neighbors, teachers and all other people the girls interact with (22). Parents and teachers play an important role in teaching the children to accept everyone for who they are and to help develop a positive self image (3, 21).

The Media’s promotion of the Thin Ideal

While there is a strong correlation between the idealization of thinness and occurrence of eating disorders, little has been done to protect children, especially girls, from this ideal. With nearly ¼ of a child’s or teen’s day spent in front of a computer or television, then combined with the amount of time spent reading fashion magazines, it is evident how the media can impact a person’s views on what it normal (7). Media awareness interventions that have been tested have not demonstrated much success at changing attitudes about eating and body image. Suggesting that the length of previous interventions were inadequate and long term programs may be more effective.

The social networking theory described above is dictated by societal norms, and these norms are greatly influenced by the media. The media depicts thin as being the right way and fat people, when shown in the media, are often shown as undesirable and lazy (5). The desire to be considered as beautiful and attractive is one that is highly sought out by girls and women young and old. With the influence of the media the attempt to achieve the ideal and avoid being considered lazy and undesirable may lead to an increase disordered methods of obtaining them. Becker, et al. (2002) conducted a study in Fiji where girls had not been exposed to the thin ideal media and where a larger figure was considered desirable and the norm (23). The study showed an increase in disordered eating habits and a shift in the ideal body over a three year period of time after the introduction and exposure to thin ideal media (23). These findings can be generalized to give insight as to how the United States and westernized cultures may have adopted this view.

The media may counteract any improvements made through education because it dictates what is perceived as both ideal and normal (2, 15, 23). As time passes there is less focus on being healthy and more on social ideals and now girls have the information on how to achieve this ideal because of an increased knowledge symptoms and unhealthy eating habits. The media has the ability to create a norm in which all sizes are accepted, and because of this there is a need to regulate what is shown on the media and diversify the images of women seen (6).

Most of the studies previously conducted on the relationship between the media and eating behaviors have been quantitative in nature and have not asked the how and why questions (23). Marketing theory would suggest that for an intervention to be successful one needs to determine the needs of the consumers (21). In this circumstance to discuss with adolescent girls how they are affected by the media, why they think this way, the changes they would like to see, etc. From the answers received by the girls a program can be developed that is specifically designed to influence the needs of this target population, something has not been considered when using educational methods.

However a reduction in the internalization of the thin ideal has been associated with positive outcomes suggesting that it is possible to significantly reduce these ideals if an effective method is discovered. (8).

There is a need to prevent eating disorders since they are costly to treat and have poor health outcomes. Also, as the prevalence of anorexia and bulimia increases there is an increased need for an effective public program. Current eating disorder prevention interventions have failed due to the overemphasis on education, lack of attention to the influence of social norms, and the influence of thin ideal media. The effects of education do not last over an extended period of time and may actually create more disordered eating habits than they prevent. Current social norms support these disordered behaviors in order to achieve the ideal body type valued by society. The media dictates these social norms and is a responsible for interacting with other risk factors to increase the prevalence of disordered eating behaviors.

These reasons for failure suggest that there is a need to reform the prevention methods that are used. New prevention methods should include the voice of girls to determine what they want and there opinions on these topics. New prevention measures should attempt to change the societal norm to one that does not focus on one ideal body but rather the acceptance of all healthy bodies.

Implications for future prevention methods

To improve upon the flaws of prior eating disorder prevention methods, new approaches must be implemented to improve upon the failures of past intervention because of their focus on education, lack of acknowledgment of social norms, and the media’s influence on the ideal. The new intervention would take place over a long time period and would use theories not typically used in the public health field, using mostly theories about marketing, social influence, and social networks. The basic structure of the new prevention program would consist of mapping a social network of teen girls in a representative area to determine the following: who is marginalized, who is heavy or thin, and who engages in disordered eating behaviors.

Next, girls who would fall into one or more of the above categories would be interviewed in focus groups with other girls similar to them to determine why they engage in certain behaviors, how other girls who are in their peer group influence them, and how the media influences them (24). Using the information collected, healthy eating behaviors can be marketed to this population in a way that shows this as the norm. The girls’ feedback on the media can be used to create media awareness campaigns, and regulations of the media that will create size acceptance and minimize the thin ideal.

Moving beyond education

New interventions would address education’s inability to produce long term change in the eating behaviors of teen girls. The use of marketing theory will take into consideration the needs and wants of the target population instead of coming from the point of view of health care professionals (24). As health professionals will focus more on the clinical aspects that are important to them rather than social ones which are more important to adolescent girls. The idea of gaining insight, through focus groups, as to why a consumer engages in a certain behavior is beneficial when determining the best way to “sell” healthful eating (24). Questions need to be asked to find out why the girls engage in a certain behavior at surface level and then continue to question them to discover the true motives for this behavior that are not at the surface level (24). Questions asked may include: how it makes them feel afterwards? How they usually feel before? Are there any external pressures they feel to engage in this behavior? Do the girls engage in healthy eating behaviors and if so why, what to they gain or lose through this action?

Conducting qualitative research instead of generating quantitative data is useful especially for studying a behavior and determining the reasons for a particular behavior. Quantitative research just discusses how common an event is (25). Discussing in small groups of girls who are either centralized or marginalized within a social network will provide a great deal of insight into the reasons for such behaviors prior quantitative research was unable to do this, however it is necessary to know the reasons for why one engages in behaviors to know how to intervene and prevent them. Education about the consequences of eating disorders, and their symptoms, as well as what foods are healthy and which ones are not should be discarded as it has not been proven effective in the past and may be causing more harm than good (10).

Additionally, another way marketing and social network theory improve upon education through the Health Belief Model, is how they do not assume that behaviors are made through a rational weighing of the risk and benefits (12). Unlike education methods based upon the Health Belief Model, they take into consideration the influence of outside forces through marketing theory and social networking theory in this new intervention would able to create effective programs to influence the healthy and unhealthy eating behaviors of teenage girls (19, 24).

Altering social norms

Currently society places normality on being thin and because of this there has been a marginalization of overweight and obese people (19). Social norms are extremely influential on adolescents, as they are still gaining a sense of self and have a strong desire to fit in with their peers. The current social network needs to be altered so that there is a mixture of thin and heavy people through out the group and there is not one group who is favored above the other. New prevention methods should focus on altering social norms by creating size acceptance and promoting the importance of healthy eating and how to do this. Decreasing the pressures to be thin will have a great impact on the amount of disordered eating behaviors in which a girl engages. If the social norms that influence whether or not a girl will engage in unhealthy eating behaviors are altered or reduced than the eating behaviors that follow will also be reduced through the role of social norms on how they depict what peer influences will impact other girls (19).

By changing the social influences on adolescent girls through the creation of size accepting norms there will be less marginalization of overweight individuals who would not fit in with the ideal body that is currently portrayed. If a second network was mapped of a similar group of teen girls after the alteration of social norms occurred, its structure would show a mixture of heavy and thin people centralized and those who are extremely obese or engage in disordered eating habits would be marginalized and seen on the outskirts of the social network. Social network analysis believes that actions are interdependent on others in an environment and not based upon an individual behaving in an independent manner (21). This analysis would demonstrate a norm for being healthy and not approving of unhealthy eating behaviors, such as dieting, purging, excessive exercise, fasting, and others. The media can be used to alter the social norms regarding ideal body.

Combating the media’s influence

The media is one of the most powerful tools in creating social norms and has the great impact on youth. Currently the media helps to create unrealistic and unhealthy standards for female physical appearance which American culture has come to accept, support, and strive for. Social norms about weight have fluctuated over time suggesting that it may be possible to shift the desired body to one of a healthier curvy figure. Americans need to become educated consumers about what is depicted on television and in fashion magazines, and how it is neither normal, healthy, or real (6). It is possible that media education can be integrated in to the classroom setting and be taught similar to that about alcohol use and violence (6, 16).

Through use of the focus groups used to collect data on the eating behaviors of teen girls can also be used to collect opinions about the media and the influence it has on the decision making of these girls (24). For instance what types of shows are currently watched and which ones were watched when the girls were younger. After learning what aspects of shows have had the greatest impact on the girls, it may be possible to implement some regulations in the media to alter this affect; such as the times when certain body types can be portrayed as ideal, the body types that are portrayed in children’s shows. Currently on television shows, overweight individuals are rarely portrayed and when they are it is usually in a negative manner where they are being depicted as clumsy, lazy, stupid, and unsuccessful (5, 6). If social networks of shows in the media do not marginalize heavier people than this will influence the setup of teen social networks. Becker, et.al (2002) demonstrated how the introduction of thin ideal media can change a society’s ideal from a heavier set on to an extremely thin one and how the pressures from the thin ideal lead to an increase in disordered eating behaviors (23). This also suggests that if the media were to introduce shows in which different sizes of people are shown, the thin ideal could be reduced and one regarding a healthy body can be created.

Using the media to alter social norms in a way that would suggest eating healthy and being of a healthier weight is seen in a positive manner and as more attractive to men. The qualitative data collected through use of marketing theory could also be used to create effective campaigns by addressing what is important to adolescent girls and portray these messages in a way that will actually impact them more than someone telling them what they should do.

Conclusions

In the past the ideal shape for a woman’s body has been curvy and heavier than the current societal ideal suggesting that it is possible to alter current norms (16). If it is possible to gain insight as to why girls begin to engage in disordered eating behaviors, then it will be more likely that they can be prevented. By focusing on external factors rather than individual ones eating disorder prevention is more likely to be successful. New prevention measures need to improve upon the failures in past approach in that they should not focus on education but rather the impact of social norms and the media on disordered eating behaviors.

References

1. Polivy, J., Herman, P. (2002). Cause of eating disorders. Annual review of psychology 53: 187 – 213.

2. Academy for Eating Disorders. (2005). Prevalence of eating disorders. Aedweb.org/eating_disorders/prevalence.cfm

3. NEDIC. Prevention of Eating Disorders

4. Smolak, L., Levine, M.P. (2001). Body image in children in Thompson and Smolak Body Image, Eating Disorders and Obesity in Youth.

5. Harrison, K. (2000) The Body Electric: Thin-Ideal Media and Eating Disorders in Adolescents. Journal of Communication: 119 – 143.

6. Levine, M.P., and Smolak, L. “The mass media and disordered eating: Implications for primary prevention” in The Prevention of eating disorders.

7. Gavin, M.L. (October 2008). How TV effects your child. http://kidshealth.org/parent/positive/family/tv_affects_child.html

8. Low, K., Charanasomboon, S., Brown, C., Hiltunen, G., Et al. (2003). Internalization of the thin ideal, weight and body image concerns.

9. Scime, M.S. and Cook-Cottone, C. (2008). Primary prevention of eating disorders: A Constructivist integration of mind and body strategies.

10. Carter J.C., Stewart, A., Dunn, V.J., Fairburn, C.G. (1996) Primary prevention of eating disorders: Might it do more harm than good.

11. Stewart, A. (1998). Experience with a school-based eating disorders prevention programme. In The Prevention of Eating Disorders

12. Becker, M.H. (1974). The health belief model and personal health behavior. Health Education Monogr (2).

13. Keca, J, and Cook-Cottone, C. (2005). Eating Disorders: Prevention is worth every ounce.

14. Bear, M. (2003). Prevention of Eating Disorders. NEDIC

15. Healthy Within. Statistics of Eating Disorders. http://www.healthywithin.com/stats.htm

16. Aronson, E., Wilson, T.D., and Akert R.M. (2007). Social Psychology (pp. 246 – 249).

17. Cohn, L.D., and Adler, N.E., Female and male perceptions of ideal body shape.

18. Centers for Disease Control and Prevention. (2000). CDC surveillance summaries. MMWR, 49, 1–5. http://www.cdc.gov/mmwr/preview/mmwrhtml/ss4905a1.htm.

19. Eisenber, M., Neumark-Sztainer, D., Story, M., and Perry, C. (2004). The role of social norms and friends’ influences on unhealthy weight-control behaviors among adolescent girls. Social Science and Medicine 60: 1165 – 1173.

20. Kotler, P. (1989). Social Marketing: an approach to planned social change. Free Press: New York, NY.

21. Wasserman, S., Faust, K. (1994) Social Network Analysis. Cambridge University Press: Cambridge.

22. Story, M., Neumark-Stzainer, D., French, S. (2002). Individual and environmental influences on adolescent eating behaviors. The American Dietetic Association 102(3): s40 – s51.

23. Becker, A., Burwell, R., Gilman, S., Herzog, D., Hamburg, P. (2002). Eating behaviours and attitudes following prolonged exposure to television among ethnic Fijian adolescent girls. British Journal of Psychiatry 180: 509 – 514.

24. Wansink, B. (2005). Marketing Nutrition. University of Illinois Press: Chicago.

25. Marshall, Catherine & Rossman, Gretchen B. (1998). Designing Qualitative Research. Thousand Oaks, CA: Sage.

26. Irving, L.M. & Neumark-Stzainer, D. (2002). Integrating the Prevention of Eating Disorders and Obesity: Feasible or Futile? Preventative Medicine 34; 299 – 309.

Labels: , ,

Tuesday, December 16, 2008

Improving Interventions for Domestic Violence: A Community-Based Ethnographic Approach – Rachel Stein

Current Interventions for Domestic Violence

Nearly one third of American women become victims of domestic violence at some point during their lifetimes (1). The true prevalence of domestic violence is likely higher than the percentage reported, as many women are reluctant to admit to being victimized and abused (2). Their reluctance demands an explanation. What is it about how domestic violence is screened that prevents women from disclosing their experiences? Current approaches to screening have multiple limitations, and their weaknesses are evident in three key areas: the setting of the screening, the questions asked, and the recommended interventions for a positive screen. These approaches to domestic violence are deeply rooted in the individualistic notions of western biomedicine, and it is this paradigm that renders these approaches inadequate to address this pervasive public health problem.

Setting

Most screening for domestic violence occurs in a clinical setting, usually between a physician and a patient. Doctors are trained to ask questions about domestic violence as part of taking a patient’s social history. As the average time allotted for primary care appointments is fifteen minutes, this screening must elicit the necessary information in the shortest amount of time possible. Moreover, the social history tends to be the last element of a patient’s history that is taken, and many physicians, pressured by their chronically overbooked schedules, are not able to find time to ask these questions. Consequently, only one to fifteen percent of all women are asked about domestic violence by their primary care doctors (3).

When a physician is able to make the time to screen for domestic violence, women may not feel comfortable sharing this intimate aspect of their lives. It has been reported that over 90 percent of women screened for domestic violence feel comfortable answering the questions that clinicians ask them (4). Public health practitioners subscribing to the western biomedical paradigm would find this screening intervention an enormous success. These practitioners, however, may not take into account that most of the women in the sample studied were low-income African American women who were employed and unmarried, as the biomedical paradigm neglects the social factors that contribute to an individual’s health behavior. Given this sociocultural context, it may not be accurate to assume that 90 percent of all women feel comfortable answering screening questions about domestic violence.

Domestic violence transcends race, ethnicity, class, and other sociocultural factors, and women that belong to different subtypes of these categories have different conceptions of and levels of comfort with disclosing information about their intimate relationships. For example, although African American women may confide in their physicians readily, as in the example given above, Orthodox Jewish women may not. Orthodox Jewish culture discourages speaking about private issues and life experiences, including marriage and intimacy, outside of one’s immediate family. In this culture, it is also shameful to admit to having marital problems. An Orthodox Jewish woman is therefore less likely to report domestic violence and to seek help (5).

In addition to considering cultural differentials in privacy, it is important to be aware of how a woman would feel answering questions about domestic violence when her children are with her, which is often the case in a clinical setting. Overall, women are more comfortable answering questions about domestic violence when they are alone than when they are with their children, particularly when the questions refer to feeling unsafe. Latina women tend to be less comfortable disclosing incidences of domestic violence when their children are present than White and Black women. Their higher level of discomfort may be explained the desire to adhere to the “machismo” culture, to be loyal to their families, and to maintain their privacy in front of their children (6).

These findings suggest that the clinical setting may not be the most appropriate place to screen for domestic violence for all women. Practitioners who subscribe to the biomedical approach may not come to this realization, and many women and children may wrongfully be assumed to be safe in their home environments.

Questions

Domestic violence screening based on the western biomedical paradigm not only occurs in an inopportune setting but also may ask inappropriate questions. Two screening tools that are recommended in the clinical setting, likely because of their brevity, are the HITS (Hurt-Insult-Threaten-Scream) and the WAST (Woman Abuse Screening Tool). The HITS asks women about the frequency of physical violence, insults, threats, and screaming or cursing in their relationships. The WAST inquires about the amount of tension in a relationship, degree of difficulty in resolving arguments, feelings of self worth, fear, and physical, emotional, and sexual abuse (7). These questions are problematic on multiple levels: the various ways that they can be interpreted, their potential to be considered disrespectful in certain cultures, and their disregard for context.

People of different backgrounds have different understandings of what domestic violence entails and have different expectations for what they will be asked to disclose to a stranger, including to a physician. Latina women, who report greater discomfort than White and Black women when answering questions about domestic violence in the presence of their children, have different conceptions of domestic violence than White and Black women. They perceive it as physical and emotional abuse more than sexual and financial abuse, all of which health care providers define as domestic violence. Latina women also recognize men’s expectations of women to shoulder a greater burden of household work as a form of abuse, while most providers do not (8). The questions asked in the HITS and WAST may therefore be confusing and disconcerting for Latina women because they ask about experiences that these women may not consider domestic violence. The screening questions may imply that certain experiences that Latina women consider normal are pathological and may suggest that certain experiences that these women find problematic are of no concern.

Latina women may also interpret the precise words used in the questions differently from their physicians. For example, the HITS asks how often a woman’s partner curses at her. This use of the word “curse” refers to swearing or using insulting language. In Latino culture, however, “curse” refers to the invocation of evil spirits against another person (9). This question would likely not elicit the information about which a health care provider had intended to inquire.

In addition to understanding cultural differences in conceptions of domestic violence and in interpretations of the screening questions, the structure of and manner of asking these questions is also important and may present a barrier to screening. Modes of communication are highly culturally specific, and disregarding these differences may pose severe problems in addressing domestic violence. This inadequacy is clearly delineated in the experience of American Indian women. The prevalence of domestic violence is higher among American Indian/Alaska Native women than among White, Black, and Asian/Pacific Islander women (10). Screening American Indian women for domestic violence, however, may be more harmful than helpful to them. According to Mescalero Apache culture, it is impolite to ask direct questions because they force an individual to give a forthright and potentially embarrassing answer (11). The HITS and WAST, which are designed to be concise and direct to accommodate the short amount of time allotted for primary care appointments, would be inappropriate and disrespectful to ask to a Mescalero Apache woman and may prevent her from seeking medical care in the future.

The questions asked in the HITS and WAST reflect the western biomedical paradigm in their disregard for cultural differences and decontextualization of domestic violence. It is important for clinicians to be aware that the questions they ask and the manner in which they ask those questions may be interpreted differently by women from various backgrounds. Additionally, the questions do not enable clinicians and patients to understand what the other means by “domestic violence.” These discrepancies limit the ability of domestic violence screening to accurately assess the prevalence of domestic violence and prevent clinicians from helping and empowering women to escape abuse and victimization.

Interventions

The lack of cultural competency in screening for domestic violence, both in the setting of the screening and in the questions asked, deters women from disclosing their true life experiences. Even if these limitations are overcome and a woman admits to being victimized, a clinician may not be equipped with appropriate resources to help her. Many clinicians do not screen for domestic violence because they would not know what to do if the screen was positive. Furthermore, the interventions that are recommended are directed towards individual women in isolated circumstances, failing to address the context of domestic violence and to approach this issue at the group or societal level.

The United States Preventive Task Force has identified certain criteria to evaluate the effectiveness of a screening tool. A test is required to have a certain degree of accuracy, measured by sensitivity and specificity, and the screening as well as the subsequent action taken must lead to improved outcomes. Regarding the second criteria for follow-up care, unlike a positive screening for cervical cancer, which is followed by a standard protocol consisting of repeat pap smears and possibly a colposcopy, the follow-up for a positive screening for domestic violence is an intimidating enigma for many physicians (12). In response to this uncertainty, a group of clinicians in California developed the AVDR – Asking, Validation, Documentation, and Referring – model for physicians to follow when confronted with a case of domestic violence. In the first step, physicians are encouraged to ask women about safety in general and specifically in their relationships. If they disclose that they are not safe, the model then suggests that the physician state clearly that abuse is not acceptable, express concern for the patient, and explain to her that she is not to blame. In the third step, a clinician is to document signs and symptoms of abuse and record the patient’s story. Finally, the model provides physicians with a list of resources where they may refer a patient, including local agencies, hotlines, and shelters, and suggests that they create a plan to follow up with her (13).

Although the AVDR model demonstrates compassion and encourages a doctor-patient relationship based on support and trust, it fails to contextualize domestic violence. Women do not exist in a vacuum, and without taking into account sociocultural determinants of behavior, physicians cannot fully understand why a woman would not disclose being victimized or would not leave an abusive relationship. The model does not suggest that providers inquire about cultural worldviews or social policies that may unintentionally permit domestic violence, such as the emphasis of gender role separation in Orthodox Jewish culture (14). The AVDR model also holds individual women responsible for taking action, while in many cultures, change at a group or societal level would be more beneficial. The biomedical approach to societal problems views society as “individuals en masse” rather than as a whole being, and this conception is not effective “when the target [domestic violence] is a social entity with its own laws and dynamics” (15).

The current approach to screening for domestic violence is inadequate in its setting, in the content and manner of the questions asked, and in the recommended interventions for a positive screen, as is evident by the reluctance of many women to disclose this information even when their own and their children’s well-being are at risk. The limitations of this screening are rooted in its foundation in the western biomedical paradigm, which neglects the sociocultural determinants of behavior and which approaches behavior change at the individual level rather than at the group or societal level. Only by considering these crucial factors and by confronting domestic violence at these multiple levels can effective screening tools and interventions be designed to empower women to ensure the safety of themselves and their children.

An Alternative Intervention for Domestic Violence

Given the limitations of current approaches to domestic violence based on the western biomedical paradigm, alternative screening interventions founded upon theories of the social sciences may be more effective. These interventions would directly address the context of domestic violence and would approach this issue at the individual and group levels. In doing so, they would construct comfortable settings for screening, develop appropriate screening questions, and recommend proper interventions.

Setting

Screening for domestic violence in a medical setting is often not conducive to eliciting disclosures of domestic violence. To locate more appropriate settings for screening women, it may be helpful for public health practitioners to conduct field work in the targeted communities, as recommended by the Cultural Theory. This theory emphasizes the influence of society and culture on human behavior, and fieldwork is one method of data collection that enables investigators to discover the social and cultural nuances of specific communities.

By immersing themselves in the community, observing people and events, and interviewing community members, public health practitioners may find that rather than screening women in primary care clinics and health centers, screening women in settings where they spend time on a regular basis and where they feel most comfortable may be more effective. Screening could take place in schools, community centers, and religious institutions. In Early Intervention programs, which seek to help children at risk for developmental, emotional, social, behavioral, and school problems, ongoing service coordinators meet regularly with the children’s caretakers in their homes and at the children’s day care centers to ensure that they are receiving the services that they need (16,17). Head Start programs, which help prepare low-income pre-school children for school and provide social services to their families, have a similar design. Each family is assigned a family case manager, who supports families with issues of employment, housing, immigration, health care, education, finances, and family communication and relations. Family case managers meet with the children’s caretakers regularly at the Head Start sites (18). Screening for domestic violence in Early Intervention sites, Head Start preschools, and other locations where women regularly spend time would not place a great burden on them in terms of travel, and these locations are familiar places where women may feel more at ease.

Screening interventions for domestic violence may also benefit from modeling the Early Intervention and Head Start programs’ employment of community members as ongoing service coordinators and family case managers. One third of Head Start staff members were initially involved as parents of children in Head Start programs (19), which places them in a unique position to help other parents in their communities. Women may feel more comfortable sharing personal information with their peers than with their doctors. There are also cultural differences in notions of privacy that impose barriers to the disclosure of domestic violence. While African American women may feel more comfortable sharing intimate aspects of their lives with their physicians (20), Orthodox Jewish women may not. The pressure that the Orthodox Jewish community places on avoiding a shanda, a shame that brings disgrace to all Jews by revealing the imperfections of the community, discourages women from disclosing incidences of domestic violence to their health care providers. Providing these women with peer counselors and advocates may empower them to speak freely about the more intimate aspects of their lives without causing a shanda (21). Domestic violence screening conducted in settings such as Early Intervention and Head Start sites and implemented by service coordinators and case managers who are members of the target community would thus likely improve the accuracy of the screening and ensure that women who are victims of domestic violence are receiving the services that they need.

Questions

The ethnographic focus of the Cultural Theory could also be used to design more culturally appropriate and respectful domestic violence screening questions. Conducting fieldwork among the Mescalero Apache has taught anthropologists that members of this community find direct inquiries disrespectful and discourteous because they force the individuals being questioned to reveal personal information that they may not feel comfortable sharing (22). Asking less direct and more open-ended questions may not only help women feel more at ease but also elicit more complete and helpful responses. These types of questions are those for which physicians strive yet often fail to ask due to time constraints, but they are questions that anthropologists are experts in asking. They directly address social context and past experiences, and they enable domestic violence screening to unravel a woman’s true story.

Pediatricians, child psychiatrists, and early childhood experts have suggested taking a similar ethnographic approach to parenting. They encourage parents to reflect back on their own childhoods, as unresolved fears and experiences from the past may compromise their relationships with their children (23). The questions they recommend that parents ask themselves and that pediatricians address inquire about their most vivid memories, their relationships with their parents, how they felt when they were separated from their parents, how they were disciplined, their experiences of trauma and loss, their meaningful relationships with people other than their parents, and how they see their own childhoods influencing how they interact with their children (24).

Applying this model to domestic violence, screening questions could be designed to provide greater context of women’s past and current situations. These questions would be asked in a sequence from least to most intrusive, allowing women to feel more comfortable with and to begin to trust their peers, service coordinators, or case managers implementing the screening. First, women may be asked whether they feel safe in their current relationships, and, as the Cultural Theory would recommend, what they mean by “safe.” Subsequent questions may ask women whether they experienced domestic violence as a child, whether they know others who are or were in abusive relationships (although perhaps using a more mild term), how they feel that their relationships with their partners affect their children, and other open-ended questions. Such questions would promote a strong relationship between women and their peers conducting the screening, and the stories that these questions elicit will enable their peers to connect them to appropriate and effective resources.

Interventions

Many physicians do not screen for domestic violence because they would not know what to do if confronted with a positive result. Additionally, the resources that they are able to offer women are often limited. Problem Solving Education (PSE) is an alternative intervention, currently used to treat depression, that addresses many of the inadequacies of current interventions. PSE draws from both the Cultural Theory and Maslow’s Hierarchy of Needs and can be effectively implemented by peer counselors in informal settings.

PSE consists of seven steps that an individual and counselor work through together to help that individual overcome the daily challenges contributing to his or her depressed mood. These steps include defining the problem, setting realistic and tangible goals for problem resolution, brainstorming multiple solutions to achieve those goals, creating guidelines for decision-making, using those guidelines to evaluate potential solutions, enacting the solution chosen, and evaluating the outcome. PSE is conducted over four to six 30-minute sessions, and individuals and their counselors focus on a different problem or challenge during each session. Reflecting principles of Maslow’s Hierarchy of Needs, these problems must be simple, specific, and describable in objective terms, such that lower order needs are met before higher order needs are addressed (25). PSE has been shown to be successful in treating depression in the primary care setting (26) and is currently being evaluated as an intervention for depression among mothers with infants in neonatal intensive care units (I acquired most of the information about PSE discussed in this paper through my experience working with the pediatrician conducting this clinical trial). Applying this model to interventions for domestic violence, women and their peer counselors, or problem solving educators, could take small steps to ultimately work towards the goal of leaving an abusive relationship.

As suggested by the Cultural Theory, culture is a crucial factor motivating behavior change, and in designing PSE as an intervention for domestic violence, it is important to take culture into account. Problem solving educators would ideally be members of their clients’ communities. They may be familiar with the challenges that these women face, be well-informed about available resources, and be able to help these women devise creative and culturally acceptable solutions to their problems. Problem solving educators from the women’s communities would also be able to meet them in accessible and familiar places. Furthermore, because PSE requires only a few hours to complete, service coordinators and case managers from Early Intervention, Head Start, and similar programs could serve as problem solving educators and, after having screened women for domestic violence, could integrate PSE into their existing counseling sessions.

To improve upon the inadequacies of current domestic violence screening and interventions, which are rooted in the individualistic principles of western biomedical theory, public health practitioners may employ social science theories to design more culturally acceptable and effective interventions. These alternative interventions would be implemented in appropriate and comfortable settings for women, would ask culturally sensitive and open-ended questions based on ethnographic techniques, and would promote a community-based team approach to overcoming the daily challenges that women face. These interventions would address the larger context underlying cases of domestic violence and would approach the issue at both the individual and group levels. In doing so, these alternative interventions would be able to more accurately screen women for domestic violence and more successfully meet the needs of women who screen positively.

References

  1. Family Violence Prevention Fund Get the Facts. Domestic Violence Is a Serious, Widespread Social Problem in America: The Facts. San Francisco, CA: Family Violence Prevention Fund, 2008. http://endabuse.org/resources/facts/.
  2. Felson R, Pare, P-P. The Reporting of Domestic Violence and Sexual Assault by Nonstrangers to the Police. University Park, PA: Pennsylvania State University, 2005.
  3. Chen P et al. Randomized Comparison of Three Methods to Screen for Domestic Violence in Family Practice. Annals of Family Medicine 2007; 5:430.

4. Chen 433.

5. Horsburgh B. Lifting the Veil of Secrecy: Domestic Violence in the Jewish Community (pp. 207, 220). In: Sokoloff NJ, ed. Domestic Violence at the Margins: Readings on Race, Class, Gender, and Culture. Piscataway, NJ: Rutgers University Press, 2005.

6. Zink T et al. Mothers’ Comfort with Screening Questions about Sensitive Issues, Including Domestic Violence. Journal of the American Board of Family Medicine 2006; 19:363-364.

7. Chen appendix.

  1. Lewis MJ et al. Perceptions of Service Providers and Community Members on Intimate Partner Violence within a Latino Community. Health Education and Behavior 2005; 32:75.

9. Curse (p.661). In: Langenscheidt Editorial Staff, ed. Langenscheidt Compact Spanish Dictionary. New York, NY: Langenscheidt, 2004.

  1. Grossman SF, Lundy M. Domestic Violence across Race and Ethnicity: Implications for Social Work Practice and Policy. Violence against Women 2007; 13:1032.

11. Farrer CR. Play and Inter-Ethnic Communication: A Practical Ethnography of the Mescalero Apache. New York: Garland Publishing, 1990.

  1. Cole TB. Is Domestic Violence Screening Helpful? Journal of the American Medical Association 2000; 284: 552.
  2. Gerbert B et al. Simplifying Physicians' Response to Domestic Violence. Western Journal of Medicine 2000; 172:329-330.

14. Horsburgh 207.

15. Susser M, Susser E. Choosing a Future for Epidemiology: I. Eras and Paradigms. American Journal of Public Health 1996; 86: 671.

16. Feldman MA. Introduction: What is Early Intervention? (pp. 1-2). In: Feldman MA, ed. Early Intervention: The Essential Readings. United Kingdom: Blackwell Publishing, 2004.

17. New York State Department of Health. The Early Intervention Program: A Parent's Guide for Children with Special Needs - Birth to Age Three. Albany, NY: New York State Department of Health. http://www.health.state.ny.us/community/infants_children/early_intervention/parents_guide/step4.htm.

18. Boston Head Start and Children’s Services. Head Start. Boston, MA: Action for Boston Community Development, Inc. http://www.bostonheadstart.org/our-services/head-start/.

19. Boston Head Start and Children’s Services. Head Start Families. Boston, MA: Action for Boston Community Development, Inc. http://www.bostonheadstart.org/families/.

20. Chen P 433.

21. Horsburgh 207.

22. Farrer C.

23. Zuckerman B, Zuckerman P, Siegel D. Promoting Self-Understanding in Parents - for the Great Good of Your Patients. Contemporary Pediatrics 2005; 22: 78.

24. Contemporary Pediatrics Guide for Parents. Ask Yourself About Your Childhood - and Make Yourself a Stronger Parent. Woodcliff Lake, NJ: Contemporary Pediatrics. http://contemporarypediatrics.modernmedicine.com/contpeds/data/articlestandard//contpeds/162005/156681/article.pdf.

  1. Hegel MT et al. Training Residents in Problem-Solving Treatment of Depression: A Pilot Feasibility and Impact Study. Family Medicine 2004; 36: 204-208.
  2. Mynors-Wallis LM, Gath DH, Lloyd-Thomas AR. Randomised Controlled Trial Comparing Problem Solving Treatment with Amitriptyline and Placebo for Major Depression in Primary Care. BMJ 1995; 310: 441-445.

Labels: , , , , , ,