ABSTRACT: Ozuzu-Nwaiwu, J. (2007) Black women’s perceptions of menopause and the use of hormone replacement therapy.www.nursingtimes.net
BACKGROUND:Little research has been conducted into the perceptions and use of hormone replacement therapy among black and minority ethnic (BME) women in the UK. These women have different health and cultural needs to the white majority.
METHOD:A qualitative study using a semi-structured interview was undertaken with 22 BME women aged 45-61.
RESULTS:Participants who had experienced menopausal symptoms managed their symptoms in a range of ways, often using coping methods such as wearing light clothing or having cold showers during hot flushes. It appeared that those who had used HRT had not been given full information on the use of HRT before it was prescribed.
CONCLUSION:BME women have different methods of managing and coping with menopausal symptoms and do not necessarily want HRT as their first line of treatment. They should be assessed adequately before being prescribed HRT, as there could be other underlying factors affecting symptom presentation.
In a multicultural, multiethnic society the NHS must provide ethnically and culturally sensitive care to patients from minority ethnic groups. These patient groups can have different health needs to those of the majority due to social, psychological, biological and genetic factors (Bahl, 1993). Such diversities may affect women’s experience of menopause and their decision-making processes regarding management of menopausal symptoms. This article reports on an investigation of black and minority ethnic (BME) women’s perceptions of menopause and use of hormone replacement therapy (HRT). The term ‘BME’ here is used to describe people of African, African-Caribbean and Asian origin.
The study was undertaken due to the paucity of research on these women in the UK, although such studies have been published elsewhere in the world. The aim was to provide a starting point for other research and increase healthcare professionals’ awareness of how BME women make decisions on and manage menopausal symptoms. It is hoped it will help healthcare professionals to ensure women are enabled to make informed decisions on treatments they are offered.
Menopause - the cessation of menstruation - usually occurs between the ages of 45 and 55 years (Singer and Hunter, 1999), although premature menopause can occur as early as the twenties. While there is a general perception that the menopause is associated with negative physical and psychological symptoms, Dennerstein and Shelly (1998) found that the majority (55-72%) of middle-aged women reported experiencing positive moods most of the time. The study concurred with the findings of Phillips and Rakusen (1999) who found that only a minority of women experience menopause-related problems requiring medical treatment.
Women’s experiences of emotional disturbance such as depression and mood changes varies in the literature, while there is little applying specifically to women from minority ethnic groups. Most of the women studied by Ballard (2003) - who were mainly white - experienced emotional disturbances but did not see them as a problem. However, some of the women experienced loss of self-esteem, suggesting hormonal changes affected their confidence.
Polit and LaRocco (1980) reported that 70% of women experience hot flushes and 40% experience other physical symptoms such as fatigue, irritability and forgetfulness. However, the extent of these symptoms varied from individual to individual, and it was not clear whether the psychological and emotional problems were related to the menopause.
Some effects of menopause can be linked to the fact that it represents a transitional stage in women’s lives and that women are often undergoing a range of life changes at this time, which may affect menopausal symptoms (Rousseau, 1999). Ballard et al (2001) identified seven status passages which can occur for many women around the time of menopause:
Becoming a carer for older relatives (this is common among BME women);
Changes in employment and finances - for example, retirement, redundancy or changing to work part time;
- Changing relationships with children as they grow up and leave home;
- Illness of self or family members;
- Death of family or friends;
- Changes in relationships;
- Changes in perception of age.
Black and minority ethnic women’s health
Healthcare professionals have been known to assume that the health needs of BME women can be met in the same way as those of white women. However, Douglas (1998) argued that this assumption fails to take account of the complex mixture of social and cultural diversity of these women. Where it does focus on their diverse needs healthcare tends to be directed at managing diseases such as diabetes (Okosun et al, 2003), hypertension (Jonas and Landau, 2000), coronary artery disease (Jones et al, 2002) and sickle cell disease (Blythe, 2000). The Kings Fund (2006) pointed out that a major issue affecting BME groups in the UK is access to healthcare itself, which may be hindered by cultural and communication barriers. An NHS Patient Survey on GP services (DH, 2002) also found that BME patients reported more negative experiences than white patients, including in their ability to book GP appointments and problems with receptionists.
Women find reading materials, friends and relatives the most useful source of information about menopause (Abraham et al, 1995). Omonuwa (2001) pointed out that racial disparity in healthcare between black and white women might be due to lack of health information on subjects such as the menopause and HRT in black popular magazines. This reduces their ability to make informed decisions about their health.
Menopause and cultural differences
Avis and Mckinlay (1993) found that Asian women do not experience menopausal symptoms as severely as Western women, while an Indian study found they experienced few problems associated with menopause compared with Western women (Dennerstein et al, 1999). The women perceived the menopause positively because they could now leave the women’s quarters and had more freedom after a secluded life in purdah. In Japan, menopausal women tended to focus attention on family responsibilities and social duties (Dennerstein et al, 1999) such as caring for older relatives and children.
In North America and Europe there appears to be greater focus on menopausal women’s biology and physical appearance (Coney, 1995). The menopause is regarded as a disease (Coney, 1995), a pathology or hormone imbalance (Hammond, 1999) that requires treatment with medication. Dennerstein and Shelly (1998) showed that in cultures where post-menopausal women gain status, the menopause is not associated with disease.
Pinkowish (1997) found that black women reached menopause about two years earlier than white women, while Moorman et al (2000) found that black women who smoked attained menopause earlier than others. They also found that women who used contraceptives for more than a year and those who were obese had a lower risk of experiencing menopause early. This finding was consistent with the study by Palmer et al (2003) who concluded that earlier arrival of menopause in African women was linked with smoking but reduced with higher BMI and the use of oral contraceptives.
Sommer et al (1999) found that African American women had more positive attitudes towards menopause than Chinese, Japanese and white women. However, the samples were not truly representative of all other groups and therefore cannot be generalised. The study also found that women who had undergone surgically induced menopause were more positive than the older women. Black women have been found to be more likely than other groups to undergo hysterectomy and oopherectomy (Ziegler-Johnson et al, 1998; Haslett, 1996), which bring about early menopause. This may explain their positive outlook towards menopause and its symptoms.
Benefits and risks of HRT
One of the benefits of HRT identified in the literature is the prevention of cardiovascular disease (Paoletti and Wenger, 2002), while the Women’s Health Initiative (2002) study showed a high rate of cardiovascular disease among African American women. Fischman (2003) study found that black women were twice as likely as white women to have heart problems and to die from them and argued that black women do not receive appropriate treatment for their condition.
The other benefit associated with HRT is the prevention of osteoporosis. However, African American women have reduced risk of osteoporosis (Carroll, 2003). Finkelstein et al (2002) stated that bone mineral density is highest among African Americans, lowest in Asians and intermediate in white women.
The most widely known risk associated with HRT is the development of breast cancer. For example, Steinberg et al (1994) estimated that the risk of breast cancer after 10 years of oestrogen use increased by up to 29%. The WHI (2002) study found that breast cancer rates were lower in African American and Hispanic women than in European American women, but that mortality rates were higher among African Americans. A large UK study found a link between breast cancer and HRT (Million Women Study Collaborators, 2003). However, while much of the literature has demonstrated an increased risk associated with the use of HRT it also suggests that short-term use tends to be safer.
Uptake of HRT among black and minority groups
Friedman-Koss et al (2002) examined the relationship between HRT and ethnicity and social class among white, Mexican American and non-Hispanic black women. They found that 40% of white women used HRT, compared with just over 20% of non-Hispanic black women and 24% of Mexican American women, while HRT use was lowest among women with low education and income. However, the study did not examine what factors influenced women’s choice of therapy.
The Research Activities Institute (2000) found that white women in the US were more likely to be prescribed HRT than women who are black, Asian, Hispanic or Soviet immigrants. This is consistent with the study by McNagney et al (1997) who examined the use of HRT among female doctors in the US. Overall, 47% of the sample used HRT but BME physicians were less likely to use HRT than their white colleagues. However, American women’s decisions may have been affected by the fact that they pay for healthcare through health insurance. Zuvekas and Taliaferro (2003) reported that BME women in the US were among the lowest-income earners, so their decisions on HRT uptake may be influenced by lack of adequate health insurance.
Nixon et al (2001) found that BME women have used other strategies than HRT to manage menopausal symptoms such as prayer and complementary and alternative medicines. They also found that rural African American women managed symptoms by enduring them for as long as possible or ‘fighting them’ (relying on will power to overcome the symptoms), only visiting their doctors as a last resort.
The WHI (2002) study highlighted several issues regarding HRT and BME women, including the fact that they have higher incidence of chronic disorders (Carroll, 2003) such as cardiovascular diseases and cancers during and after menopause than other groups.
Long-term use of HRT has been reported to be common among white women (Moorman et al, 2000), who use HRT for an average of 90 months compared with 63 months among black women. Moorman et al (2000) also found that black women who used oral contraceptive, who had been in higher education and those who smoked used HRT more than those who did not.
The approach adopted for this study was embedded in the phenomenological framework, which seeks to describe participants’ lived experiences. The study was carried out in a natural setting, trying to make sense and understand black women’s perceptions and the meanings they bring to menopause, its symptoms and HRT usage.
Ethical approval was gained from Northampton Ethics Committee, while the Mary Seacole Development Award Committee also offered advice and support. Participants remained anonymous and could withdraw from the study at any time.
Sample and settings
The sample consisted of 22 BME women aged 45-61 years from a range of ethnic groups (Table 1), who were recruited from a number of BME organisations in Northampton (Box 1). Letters explaining the purpose of the research were distributed to 30 women from the target age group two weeks before recruitment. Those agreeing to participate were visited in the centres to establish rapport and contact; the sample size was sufficient to generate adequate data for analysis.
Table 1. Ethnic groups represented in the study.
The Afro Caribbean Association
The Black History Association
Dostiyo Asian Women Group
Somalian Women’s Organisation
Box1. Organisations visited for recruitment
Data was collected by one-to-one semi-structured interviews. The interview took approximately one hour to conduct and included the following questions:
- What do you understand by menopause?
- Do you experience any symptoms?
- How do you manage your symptoms if any?
- Do you use HRT or intend to use it?
- What are your views regarding HRT?
- Do you use any other alternative or coping methods to manage the symptoms?
- Did your mother experience any symptoms?
Interviews were tape-recorded to enable me to concentrate on the participant and maintain eye contact and communication, which encouraged the women to talk. Interviews were conducted in the women’s homes or at their cultural centres, depending on their preference, and at times convenient to them. At the end of each interview, the tape was replayed to ensure the women were happy with what they had said. The recordings were transcribed and again participants were given copies of the transcripts to ensure they recognised and agreed with the wording. Minor changes were made as a result.
The data collected was transcribed verbatim and emerging themes were arranged to form the discussion points. The study revealed that educated women knew and understood more than uneducated women about menopause and HRT.
The most common symptoms reported were hot flushes, night sweats, headaches, high blood pressure, tiredness, irritability and forgetfulness. Some women also experienced cramps on their hands or fingers. The other main themes that emerged during the analysis are outlined below.
Meaning and understanding of menopause
Only four participants clearly understood the term ‘menopause’, although seven partially understood it, for example referring to it as old age or no longer being able to bear children, while another seven described the process in terms of the symptoms they experienced it. The remaining four women did not understand the term.Some of the participants who were going through the menopause at the time of interview were experiencing social and family problems in addition to menopausal symptoms. They found it difficult to differentiate their symptoms from the stress of the family and work, and even other medical problems such as high blood pressure or diabetes.
The participants managed their symptoms in a range of ways (Table 2). Some used more than one method, particularly those who used ‘alternative’ methods such as evening primrose, cod liver oil, lavender oil and black cohosh (a plant-derived natural remedy). These women also tended to use coping methods such as eating healthy diets, exercising, wearing light clothing, having cold showers or drinks and opening doors or windows during hot flushes. One felt that making others laugh helped her to deal with depression. Only one participant was taking HRT at the time of interview, while six had done so in the past. One of these women would have preferred to remain on HRT but her GP believed she had remained on it for too long and refused to continue prescribing it. The participant taking HRT described it as the best thing that had ever happened to her and encouraged other women to consider it. The eight using no management methods felt their symptoms were not severe enough, believed that HRT would encourage thereturn of menstruation or cause cancer, preferred using alternatives or felt they could cope without it.
Management/No of women
|Took HRT in the past/6|
Table 2. Methods of symptom management.
Deciding to use HRT
The participants who had used HRT in the past did so because their symptoms became difficult to manage. Due to language and communication problems it was not always clear how they came to conclude that HRT was best suited for them, although it appeared that they were not given full information on the use of HRT before it was prescribed. Most were persuaded to take HRT because blood tests confirmed the arrival of the menopause. The women also said no alternative treatment was suggested to them before HRT was prescribed.
How they received information
The majority of the women obtained information on the menopause and HRT from friends, colleagues and relatives, media, books, the internet and their doctors. Those who reported receiving their information from their doctors, however, said their GP surgeries had information leaflets on menopause and the use of HRT but none on alternatives to HRT. Three of the women had made no attempt to obtain information but it is not clear why. This raises several questions:Were they unaware of the local NHS clinic?If they were aware of the clinic why did they choose not to access it for information?Could fear or lack of trust in the information that they might be given have prevented them from using their NHS clinic?
The six women who had used HRT in the past said they had done so because their doctors wanted them to and that they felt they had no choice. There was a general feeling that women did not receive adequate information from their GPs and would prefer GPs from their own ethnic group. This supports the findings of the NHS patient survey of GP services (DH, 2002), which also found that BME patients were least likely to have access to a GP from their own ethnic group. One participant said she felt BME women would prefer GPs who spent time explaining issues listening to them, and would also prefer to see female doctors.
Experiencing parental symptoms
Eight women did not witness their mother go through the menopause, and most of these said the subject was never discussed. Others reported that their mother died before they could witness the change, while some had left home before their mother experienced menopause. Seven women were unsure whether they had witnessed their mother experience the menopause while the remaining seven had witnessed it. Of these, it appeared that the ones whose mothers did not use anything for their symptoms did not tend to use anything either. It also appeared that their mothers’ coping methods did not always help. One participant explained that her mother could not separate her hot flushes from the weather conditions of the hot country where they lived.
There was a range of reasons why the women did not opt for HRT. Some felt their symptoms were not severe enough to justify it or that they could overcome them through willpower. Cancer scares related to HRT had deterred some, while others did not want to interfere with a natural process and felt that coping and alternative methods were more effective and safer.
Of the 22 women interviewed, six had used HRT in the past. This is consistent with McNagney et al (1997) who found that HRT use was lower in BME women than in white women. They suggested this was due to the fear of breast cancer and the belief that HRT caused the resumption of menstruation. Scheid et al (2003) highlighted the fact that women who used HRT perceived it to reduce their risk of ill health, while non-users perceived it to increase the risk of breast and uterine cancer. Some participants in this study also believed that taking HRT involved taking tablets daily, showing a lack of understanding that HRT can be used in different ways, such as patches and creams. The only participant taking HRT at the time of the study felt positive about it and claimed that she could not manage without it. She explained that she felt vibrant, very fit and had been training to run a marathon for a cancer charity.
Two participants believed that God would take control of the symptoms and heal them. This appears to be consistent with Nixon et al (2001) who found that African American women managed their symptoms through the use of prayer and Holland and Hogg (2001) who stated that people of African and Caribbean origin perceived illness to be caused by or treated with a supernatural force.
The information available to participants had often been insufficient to enable them to make informed decisions about whether or not to take HRT and most of that available was not properly geared towards BME women. Their GPs did not take time to explain the risks and benefits associated with HRT. It also appeared that their doctors did not understand all the symptoms that could present in black women. Some of the women said their doctors misinterpreted their symptoms and made incorrect diagnoses and sometimes merely told them to take paracetamol. They felt that their doctors did not take their complaints seriously and usually dismissed them as psychosomatic. This left the women distressed and caused loss of faith in their doctors and so they no longer visited them.
Some participants used ‘willpower’ to cope with their symptoms, and believed it was unnecessary to interfere with nature and therefore to use HRT. Their mothers’ attitudes towards menopause itself and its symptoms also influenced the women’s choices. Those whose mothers did not regard menopause as a problem or use products to manage their symptoms were less likely to use HRT or any other product. They felt that if their mothers could cope without medication or therapy, they should also be able to cope.
Although some women understood the meaning of menopause, most did not understand the health implications of some of the symptoms. Some chose to ignore their symptoms due to other social factors happening at the same time as their menopause - such as family commitments and changes, and financial issues - which were competing with menopause in their lives.
It appeared that women who had emigrated from hot climates coped better with night sweats and hot flushes. They were accustomed to coping with such symptoms caused by extreme weather.
Some participants dealt with their symptoms in the most natural way possible. They tended to drink plenty of water and fluids during hot flushes, wore light and cotton clothing at night, opened their windows and avoided using heavy bedclothes. They exercised, ate fresh vegetables, and took paracetamol for headaches. Some supplemented their diets with multivitamins.
Women who chose alternative methods used products such as vitamin E, evening primrose, sage leaves, cod liver oil, black cohosh, lavender bath oils and thyme. However, most did not adhere strictly to taking these products, except for one woman who said she was ‘addicted’ to evening primrose and ‘could not do without it’.
Doctors’ attitudes also contributed to the women’s behaviour. Some felt their doctors simply put them on HRT without explaining what they were being given or why. When the women realised what they had been given and their side-effects, they opted to discontinue with the drug.
Most of the women in this small study sample did not choose HRT to treat their menopausal symptoms. Of the 22 interviewed, only six had used HRT in the past while one woman was still using it. They generally adopted a rather positive attitude towards menopause and ignored their symptoms or used natural and coping methods to manage them.
The research was an exploratory study of the perceptions of BME women on the use of HRT to treat their menopausal symptoms, and I believe its aims and objectives have been achieved. The responses from the women showed consistency in thoughts and reflections and also agreed with some of the literature findings. The use of an unstructured interview allowed the women to use their own words instead of those of the researcher and gave them a voice. The qualitative phenomenological research approach has generated interesting findings that, although exploratory in nature, will provide a basis for future research.
The study had some limitations. Language and accent were often problematic during interviews and transcription of tape-recordings, while it was undertaken by a novice researcher.
Healthcare professionals should understand that BME women have different methods of managing and coping with menopausal symptoms and do not necessarily want HRT as their first line of treatment. They should be assessed adequately before being prescribed HRT, as there could be other underlying factors affecting symptom presentation.
This study should be replicated with a larger sample and should be conducted in two localities so that comparisons can be made to inform professional practice.
Healthcare professionals should be aware of the cultural, social and economic differences that affect the experience and management of symptoms among BME women. Proper communication systems should be put in place to ensure that information about the menopause, the risks and benefits of HRT and the health problems women often experience during the menopause. The information should be appropriately and sensitively delivered to enable the women to make informed decisions and take control of their lives and health.
Abraham, S. et al(1995) Australian women’s perceptions of hormone replacement therapy over 10 years. Maturitas; 21: 2, 91-96.
Avis, N., Mckinlay, S.(1993) A longitudinal analysis of women’s attitudes toward the menopause: result from the Massachusetts women’s health study. Maturitas; 13: 2, 65-79.
Bahl, V.(1993) Development of a black and ethnic minority health policy at the department of health. In: Hopkins, A., Bahl, V. (eds) Access to Health Care for People From Black and Ethnic Minorities. London: RoyalCollege of Physicians of London.
Ballard, K.D. et al(2001) The role of the change of life. Sociology of Health and illness; 23: 4, 397-424.
Ballard, K.(2003) Understanding Menopause. Chichester: Wiley.
Blythe, J.(2000) Better Screening needed for sickle cell and thalassaemia. British Medical Journal; 321: 7256, 260B.
Carroll, L.N.(2003) Implications of menopause hormone therapy for African American and Hispanic women. Journal of the American Medical Association; 95; 4: 253-256.
Coney, S.(1995) The Menopause Industry: a Guide to Medicine’s Discovery of the Midlife Woman. London: Women’s Press.
Dennerstein, L. et al(1999) Mood and the menopausal transition. The Journal of Nervous and Mental Disease; 187: 11, 685-691.
Dennerstein, L., Shelly, J.(1998) A woman’s guide to menopause and hormone replacement therapy. Washington, DC: Psychiatric Press.
Department of Health(2002) National Surveys of NHS Patients 2002: GP Practices. London: DH.
Douglas, J.(1998) Meeting the health needs of women from black and minority group ethnic communities. In: Doyal, L. (ed) Women and Health Studies. Buckingham: Open University Press.
Finkelstein, J.S. et al(2002) Ethnic variation in bone density in premenopausal and early perimenopausal women: effects of arthropometric and lifestyle factors. Journal of Clinical Endocrinology and Metabolism; 87: 7, 3057-3067.
Fischman, J.(2003) Blacks have more heart risk but less care. US News and World Report. Online access: http://goliath.ecnext.com/coms2/gi_0199-3241859/Blacks-have-more-heart-risk.html#abstract.
Friedman-Koss, D. et al(2002) The relationship of race/ethnicity and social class to hormone replacement therapy: results from the third National Health and Nutrition Examination Survey 1988-1994. Journal of American Menopause; 9: 4, 264-272.
Hammond, C.(1999) A new look at menopausal hormone replacement therapy: evidence-based observations. Journal of British Menopausal Society; 15: (Supplement S1, 4-10.
Haslett, S.(1996) Hysterectomy. Nursing Standard; 10: 38, 49-55.
Holland, K., Hogg, C.(2001) Cultural Awareness in Nursing and Healthcare: An Introductory Text. London: Edward Arnold.
Jonas, B.S., Landau, J.F.(2000) Negative affect as a prospective risk factor for hypertension. Psychosomatic Medicine; 62: 2, 188-196.
Jones, D.W. et al(2002) Risk factors for coronary heart disease in African Americans: the atherosclerosis risk in communities study, 1987-1997. Archives of Internal Medicine; 162: 22, 2565-2571.
Kings Fund(2006) Access to Health Care and Minority Ethnic Groups. London: Kings Fund.
McNagney, S.E. et al(1997) Personal use of post menopausal hormone replacement therapy by women physicians in the United States. Annals of Internal Medicine; 127: 1093-1096.
Million Women Study Collaborators(2003) Breast cancer and hormone replacement therapy in the million women study. Lancet; 362: 419-427.
Moorman, P.G. et al(2000) Menopausal hormones and breast cancer in a biracial population. American Journal of Public Health; 90: 6, 966-972.
Nixon, E. et al(2001) Staying strong: how low-income rural African American women manage their menopausal changes. Women and Health; 34: 2, 81-95.
Okosun, I.S. et al(2003) Diagnosed diabetes and ethnic disparities in adverse health behaviours of American women. Journal of the National Medical Association; 95: 7, 523-532.
Omonuwa,S.C.(2001) Health disparity in black women: lack of pharmaceutical advertising in black vs. white-oriented magazines.Journal of the National Medical Association; 93: 7/8, 263-267.
Palmer, J.R. et al(2003) Onset of natural menopause in African American women. American Journal of Public Health; 93: 2, 299-306.
Paoletti, R., Wenger, N.K.(2002) Review of the international position paper on women’s health and menopause: a comprehensive approach. Circulation; 107: 9, 1336-1339.
Phillips, A., Rakusen, J.(eds) (1999) The New Our Bodies, Ourselves. London: Penguin.
Pinkowish, D.M.(1997) Racial differences in age at menopause. Patient Care; 31: 5, 196.
Polit, L., LaRocco, M(1980) Social and psychological correlates of menopausal symptoms. Psychosomatic Medicine; 4: 3, 335-345.
Research Activities Institute(2000) White women are more likely than minority women to be prescribed hormone replacement therapy. Available at: www.datastarweb.com; accessed 9 March 2004.
Rousseau, M.(1999) Clinical commentary. Journal of Nursing Scholarship; 31: 4, 338.
Schied, D.C. et al(2003) Do perceptions of risk and quality of life affect us of hormone replacement therapy by post menopausal women? Journal of the American Board of Family Practitioners 16; 4: 270-277.
Singer, D., Hunter, M.(1999) The experience of premature menopause: a thematic discourse analysis. Journal of Reproductive and Infant Psychology; 17: 11, 63-68.
Sommer, B. et al(1999) Attitudes towards menopause and aging across ethnic/racial groups. American Psychosomatic Society; 61: 6, 868-882.
Steinberg, T. et al(1994) Breast cancer risk and duration of estrogen use: the role of study design in meta-analysis. Epidemiology; 5: 4, 415-421.
Women’s Health Initiative(2002) Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. Journal of the American Medical Association; 288: 321-333.
Ziegler-Johnson, C.M. et al(1998) Subclinical atherosclerosis in relation to hysterectomy status in black women. Stroke; 29: 4, 759-765.
Zuvekas, S.H., Taliaferro, G.S.(2003) Pathways to access: Health insurance, the health care delivery system, and racial/ethnic disparities, 1996-1999. Health Affairs; 22: 2, 139-145.