Young women with chronic disease: A female perspective on the impact and management of rheumatoid arthritis


  • Melanie J. Harrison

    Corresponding author
    1. Weill Medical College of Cornell University, The Hospital for Special Surgery, New York, New York
    • Department of Medicine, Division of Rheumatology, The Hospital for Special Surgery, 535 East 70th Street, New York, NY 10021
    Search for more papers by this author


The incidence and prevalence of autoimmune diseases are dramatically higher in women than in men (Table 1). The factors that predispose women to autoimmune disorders are the subject of ongoing research and probably involve hormonal and immunogenetic elements. Rheumatoid arthritis (RA), like other disorders of this class, is still poorly understood and subject to misunderstanding among patients with the disease (1). As a disease that predominantly affects women, RA bears on a number of sex-related issues, including reproductive physiology, childbearing and childrearing, social behavior, and quality of life.

Table 1. Sex ratios in autoimmune diseases*
  • *

    From the American Autoimmune Related Diseases Association, Inc. Web site at Used with permission.

Hashimoto's disease/hypothyroiditis50:1
Systemic lupus erythematosus9:1
Sjögren's syndrome9:1
Antiphospholipid syndrome9:1
Primary biliary cirrhosis9:1
Mixed connective tissue disease8:1
Chronic active hepatitis8:1
Graves' disease/hyperthyroiditis7:1
Rheumatoid arthritis4:1
Type 1 diabetes mellitus2:1
Myasthenia gravis2:1
Multiple sclerosis2:1
Chronic idiopathic thrombocytopenic purpura2:1

Rheumatoid arthritis is a relatively common autoimmune disease, affecting approximately 1% of the US population (2). The disease is more prevalent in the United States than in some other countries; for example, the overall national prevalence in Sweden is about 0.5% (3, 4). Long-term trends show that for populations over the age of 35 years, the annual incidence rises with age in both men and women (Figure 1) (5). Moreover, age-specific incidence has been shown to vary with birth cohort, suggesting that changing environmental factors may influence the likelihood of triggering the disease (5). Overall, it has been estimated that women are affected up to 4 times more frequently than men (6). This sex bias has implications for the management of RA and for future directions in researching the disease.

Figure 1.

Age-related annual incidence of rheumatoid arthritis in a US population (Rochester, MN) over the age of 35. Data from Gabriel et al (5). Note that other studies have observed different distributions in different birth cohorts, suggesting an influence of environmental factors (74).

In the US today, the typical young woman has her first child in her 20s and may care for children at home into her 40s. These childbearing and childrearing years roughly coincide with the years in which the rate of onset of RA increases markedly. In one study population, the distribution of ages of onset in women appeared to be bimodal, with one peak between the ages of 31 and 35 and another after age 46 (7). In contrast, for men over 35 years of age, the rate of onset of RA was independent of age, and for men under 35 years of age, onset of RA was several-fold lower than that observed in women (7).

Physician attitudes and treatment options are influenced when a disease occurs predominantly in a woman's childbearing years. A disease with this demographic profile may have a different impact on both quality of life and socioeconomic issues from that of a disease that affects, for example, elderly men. This article evaluates RA as an issue of particular concern for women, considers how sex affects disease management, outlines some of the possibilities opened up by new therapies, and concludes with a call for increased public awareness of the importance of early intervention to prevent disease progression.

Clinical differences in RA between men and women

Rheumatoid arthritis is manifested differently in the 2 sexes, not only with respect to age of onset but also with respect to pattern of joint and extraarticular involvement, bone destruction, and behavioral symptoms (Table 2). Such differences may arise from genetic factors, hormonal exposure, structural differences (e.g., bone and muscle composition), and functional differences (i.e., differences in the kind of wear and tear that men and women put on various joints). Thus, the pathogenesis of RA may not be the same in women as it is in men.

Table 2. Sex-based differences in rheumatoid arthritis
• Rheumatoid arthritis prevalence is 2 to 4 times greater in women than men.
• Two peak rheumatoid arthritis onset periods in women (from ages 31 to 35 and after age 46) versus 1 peak in men (after age 35).
• HLA-DR4 and HLA-DR10 associated with rheumatoid arthritis in women; HLA-DR1 associated with rheumatoid arthritis in men.
• Higher risk of articular destruction in women versus men.
• Fewer extraarticular manifestations in women than in men.
• Higher clinical disease severity in women than in men as assessed by Health Assessment Questionnaire.
• Greater rheumatoid arthritis-related pain reported by women than by men.
• More joint surgeries performed for women than for men.
• Higher rates of depressive symptoms in women than in men.

HLA associations.

Differences have been observed between men and women in the association of HLA-DRB1 with susceptibility to RA (8, 9). A majority of patients with RA carry HLA-DR4, HLA-DR1, or both genotypes (10). One study found that female RA patients were significantly more likely to express the haplotypes DR4 and DR10 than the population at large, whereas in male patients an association was more apparent with DR1 than with DR4 and DR10 (8). However, another study demonstrated only a weak relationship between differential rates of RA occurrence in males versus females and HLA-DR4 phenotype frequency (11). Additional studies are needed to elucidate the association of HLA polymorphism and sex-based differences in RA disease expression.

Symptoms and disease severity.

Women with RA tend to exhibit involvement in the small joints of the hands and wrists, whereas RA in men is more likely to affect the knees and hips. Proximal and/or axial joint involvement is found in 66% of women versus 78% of men with RA (Table 3) (7). In general, women are less likely than men to express extraarticular manifestations of RA. Rheumatoid nodules are found significantly more often in male patients than in female patients, and men with RA are significantly more likely than women to have rheumatoid lung disease (7). Women with RA, however, have an increased frequency of sicca syndrome compared with men (7). Although the ratio of women to men is at least 2:1 in mild-to-moderate RA, the ratio is 1:1 in severe RA marked by vasculitis (7). Women with RA have been shown to rate their symptoms as more severe than men do, but it is not clear whether this is due to more severe disease in women or to subjective differences, as in the perception of RA (12, 13). Rheumatoid arthritis appears to contribute to earlier mortality in both sexes, but studies do not agree as to which sex fares worse in this regard (14).

Table 3. Clinical manifestations of rheumatoid arthritis in women versus men*
Clinical manifestationFemale patients, % (n = 110)Male patients, % (n = 55)P
  • *

    Data from Weyand et al (7).

Rheumatoid synovitis involving proximal or axial joints6678Not reported
Extraarticular involvement   
 Rheumatoid nodules20460.001
 Rheumatoid lung disease426<0.001
 Sicca syndrome1850.05
Frequency of joint surgery50270.01

Disease progression in RA is more aggressive and occurs earlier in men than in women according to such clinical criteria as bone damage and extraarticular involvement, yet demonstrable articular destruction ultimately becomes more pronounced in women (P = 0.02) (7, 15). Although these findings suggest that the window of opportunity to halt disease progression may be wider in women, guidelines have not been developed that translate these attributes into different recommendations for medical therapy for men versus women.

Effects of hormonal fluctuations and pregnancy.

Fluctuations in sex hormones are greater in women than in men, and the influence of these hormones is accordingly more apparent in women with RA. Androgens may delay or prevent the onset of RA; both women and men with RA have lower than average androgen levels (6). Retrospective studies of the use of hormone replacement therapy and oral contraceptives in women with RA have generated conflicting results. Some case-control studies of oral contraceptive use (16–18) and estrogen replacement therapy (19) have indicated that these treatments protect against RA, whereas other studies found no effect (20–22).

More than 75% of women will experience remission during pregnancy; however, approximately 80% of women will experience a disease flare postpartum (6). Additionally, pregnancy has been shown to reduce the risk of RA later in life (23). As a consequence of the joint degradation in RA, natural delivery by pregnant women may be difficult or impossible, particularly when damage has been severe enough to necessitate hip replacement surgery.

Treating RA to slow disease progression.

Both joint damage and functional disability are present during the early period after onset of RA (24–27). Of patients with early RA followed for 1 year, 74% were found to have joint erosions as assessed by magnetic resonance imaging (24). Within the first 2 years of disease, joint space narrowing and joint erosions have been observed in 83% and 67% of patients, respectively (25).

In the past 10 years, evidence has accumulated to show that early, aggressive medical intervention with traditional disease-modifying antirheumatic drugs (DMARDs), for example methotrexate, hydroxychloroquine, auranofin, and sulfasalazine, may alleviate symptoms, restore function, and slow or prevent progressive articular damage (28, 29). A new cytotoxic agent, leflunomide, has likewise demonstrated an ability to slow joint destruction (30). As a result of these observations, most rheumatologists today favor early, aggressive treatment with DMARDs (29, 31).

Despite the variety of effective DMARDs available, many RA patients still have disappointing responses, and drug toxicity or lack of efficacy may make responses difficult to sustain (32). Most of the traditional DMARDs are antimetabolites; some affect nucleic acid synthesis (33, 34) and some are contraindicated during pregnancy. Perhaps partly for these reasons, a recent study has shown that quality of care for RA falls far short of recommended guidelines (35), with the result that both men and women appear to be undertreated (7).

Two newer agents, etanercept and infliximab, known as biologic response modifiers, block the action of the proinflammatory cytokine tumor necrosis factor (TNF), thereby inhibiting its ability to stimulate the proliferation of synovial fibroblasts and trigger inflammatory events in RA. Clinical studies have shown that the anti-TNF biologic response modifiers rapidly alleviate pain, reduce joint swelling, and restore functionality in patients with RA (36, 37). Widespread interest in these new agents has been triggered by the specificity of their action on the inflammatory cascade and their ability to halt disease progression. In a comparative clinical trial versus methotrexate, etanercept was observed to stop or slow the radiographic progression of joint damage in patients with early RA, showing this effect significantly more rapidly than methotrexate (38, 39). Infliximab has likewise been shown to provide relief of RA symptoms and slow radiographic joint damage (40). Infliximab must be administered with methotrexate to minimize the development of anti-infliximab antibodies.

The anti-TNF biologic response modifiers have been shown to be at least as effective as traditional DMARDs, such as methotrexate and sulfasalazine, in improving both RA symptoms and functional status, and in inhibiting disease progression (33, 38). Although anti-TNF biologic response modifiers are not known to affect nucleic acid synthesis, they have not yet been adequately studied in pregnant women. Together with the relatively early age of onset of RA in women, the potential for abnormal prenatal development (teratogenicity) encountered with the traditional DMARDs warrants such studies; it is possible that anti-TNF biologic response modifiers may not only be more effective but also safer than the traditional DMARDs. From a basic research perspective, investigation into whether biologic response modifiers elicit sex-specific responses in RA patients may represent an important line of inquiry in trying to understand the differences in disease expression between men and women.

Despite evidence favoring early aggressive treatment of RA, study of treatment practices showed that more than one-third of patients were found to never have been treated with a DMARD (receiving only nonsteroidal antiinflammatory drugs or corticosteroids) (7). The undertreatment of women is unfortunate because, according to a meta-analysis of 14 RA clinical trials involving 1,435 patients, both female sex and longer disease duration reduce the likelihood of response to treatment (41). The data from this study indicated that the odds of response to treatment for female RA patients were only 0.72 times those for male patients (P= 0.04), and disease duration had effect on the likelihood of patient response. For example, patient response to treatment decreased from 53% for patients with less than 1 year of disease to 35% for those with a disease duration of more than 10 years (P = 0.001) (41).

Medication issues for women of childbearing age.

Teratogenicity has been a concern with cytotoxic DMARDs used for RA. Methotrexate is contraindicated during pregnancy, and the prescribing information and the American College of Rheumatology Ad Hoc Committee on Clinical Guidelines state that pregnancy should be avoided for at least 1 ovulatory cycle after discontinuing therapy for female patients (42, 43). The authors' policy is to discontinue the drug at least 3 months (i.e., 3 ovulatory cycles) before a patient wishes to become pregnant (44). Aspirin and nonsteroidal antiinflammatory drugs can cause anemia in pregnant women and prolong labor (43). The long half-life of leflunomide mandates caution in prescribing the drug to women of reproductive age and is contraindicated in pregnant women or women of childbearing potential who are not using reliable contraception (45). Leflunomide must be discontinued at least 2 years prior to conception, otherwise, because of its long half-life, the prospective mother must undergo treatment with cholestyramine to bind the drug that is in the body. Although cyclophosphamide is rarely used in patients with RA, it may be used in the treatment of extremely severe cases in women of childbearing age. However, it must be used with caution because normal ovulatory function may not be preserved following its use and discontinuation. During pregnancy, drug regimens in women with RA are generally restricted to corticosteroids. However, corticosteroids, such as prednisone, may exacerbate diabetes and hypertension (43).

Disability, pain, and health-related quality of life

Clinicians and researchers have begun to recognize the importance of measuring psychologic and social aspects of health-related quality of life in addition to such purely biologic measures as radiographic change (46). Both the disease itself and the quality of life issues associated with RA may be different in women as compared with men. Women report more disability than men, even after controlling for severity of disease or observed physical function (47), and they have been shown to require more surgical intervention compared with men (7). The latter observation may reflect greater disability from joint destruction as a result of disease progression and may relate to the particular joints likely to be affected in women as opposed to men. Alternatively, female patients may differ from male patients in their motivation to elect surgical treatment. Finally, women report greater pain than men who have RA, and they depend more on emotion-focused coping strategies to deal with the disease (48).

Because no single indicator of disease activity, severity, or therapeutic efficacy suffices to describe RA progression, multivariate evaluations of physical, emotional, and social aspects of the disease (e.g., the Health Assessment Questionnaire [HAQ]) have become important instruments for assessing the effects of various clinical interventions (49). Outcomes in RA may be predicted based on self assessed measures of global severity, pain, and function using the HAQ (26). Women tend to score higher on the HAQ than men, indicating greater functional impairment, perhaps due to greater disease severity in women (13).

Health Assessment Questionnaire assessment of patients treated with the agents proven to slow or halt joint destruction (etanercept, infliximab, leflunomide, and methotrexate) has shown that these therapies have a measurable positive impact on functional outcome and health-related quality of life (33, 36, 37, 49, 50). For example, RA patients treated with etanercept for 6 months showed significant improvements relative to placebo in HAQ scores for general health status (etanercept 33% versus placebo –12%), arthritis-specific health status (44% versus –22%), vitality domain (25% versus 2%), and mental health domain (35% versus 3%) (36). Maintenance therapy of infliximab at the currently recommended dose of 3 mg/kg every 8 weeks (given to patients receiving concomitant methotrexate) resulted in a 13% improvement from baseline in HAQ scores compared with a 3% improvement in patients receiving placebo plus methotrexate. However, this difference was not statistically significant (P = 0.167) (37). The effects of etanercept and infliximab on quality of life indexes occur rapidly (as early as 1 month after initiation of therapy) and may result from reduction of circulating TNF levels (49). Recent evidence, however, suggests that the clinical benefits of infliximab treatment decrease over time (40, 51). Treatment of RA patients with DMARDs may also produce improvements in function and psychosocial domains. In a 1-year study of patients with active RA, both leflunomide and methotrexate resulted in significant improvements in HAQ scores (35% and 20% from baseline, respectively, compared with a 2% deterioration in the placebo group) (50).

Use of the HAQ has revealed possible differences in therapeutic responses between men and women. In the etanercept clinical trials, a subgroup analysis revealed that male and female patients differed with respect to the HAQ disability categories for which significant improvement was shown relative to placebo. Although both sexes showed improvement in several of the same categories (eating, walking, and reach), male patients treated with etanercept showed improvement in grip, while female patients improved in dressing and hygiene (49). This finding deserves further study as it may reflect sex-specific differences in the deterioration of physical functioning or in the therapeutic response to etanercept.

Psychological and psychosocial issues


Rheumatoid arthritis negatively affects such psychological factors as depression, anxiety, coping, helplessness, and pain behaviors (52). The prevalence of depression has been shown to range from 15% to 42% of patients with RA (53–55), and depressive symptoms affect more women than men (56, 57). A longitudinal study of RA patients demonstrated sex differences in the negative, but not the positive, affective component of psychological well-being (57). As might be expected, higher levels of disability correlated strongly with increased depressive symptoms in a study that used the Center for Epidemiologic Studies–Depression scale (58). However, not only does disability lead to depression, but depression is believed to contribute to the patient's perception of disability, as disease factors explain only approximately one-third of the disability observed in RA patients; another 20% is accounted for by psychological status and depression (59). Strategies that mitigate depression or improve patients' perception of their ability to cope (self-efficacy) could thus have a significant impact on reducing rheumatoid arthritis disability (59).

Ability to work.

The disability caused by RA often prevents patients from working (60). Several studies have found that the inability to work outside the home may increase social passivity and depression in women with RA, negatively affecting their psychological and physical health (58, 61, 62). The authors concluded that patients should strive to maintain their work habits and social activities to avoid depression and other psychological sequelae. Although the dual pressure of work and household demands might be expected to increase depression, social support and autonomy (e.g., freedom to adjust work schedules) have been shown to reduce the effects of work demands on depressive symptoms among women with RA (58). Accordingly, a recent study has shown that women with RA were dissatisfied with those areas of their lives that threatened their independence (63), underscoring the importance of therapies that enable such individuals to maintain their independence.

Ability to manage a household.

Whether employed or not, women with RA continue to assume major homemaking responsibilities, but their ability to accomplish household tasks and provide emotional family support can be negatively affected by their disease (58, 64, 65). In one study, family members of households in which the wife or mother had moderate-to-severe RA spent 7 more hours each week performing household chores than did household members in which the wife or mother had either mild or no RA (66). Clearly, treatments that relieve symptoms and improve functionality can help women manage their dual responsibilities of work and household, and thereby improve their psychological status and general quality of life.


Given the high prevalence of RA in women and the significant effects that this disease exerts on function and psychosocial parameters, significant gaps still remain in our knowledge of how best to manage RA in women. Studies that focus on sex differences in RA may contribute to better basic understanding of the disease. Local registries provide one way in which to track women with RA. Dr. Cornelia M. Weyand and colleagues at the Mayo Clinic in Rochester, Minnesota, have extracted observational data about female RA patients from a general medical registry maintained by Olmsted County, Minnesota (7). Implementation of such programs nationwide would provide a useful adjunct to ongoing clinical studies of RA. As a specific example, this author's institute has recently begun a registry of pregnant women who have undergone RA–related hip replacement surgery. The ability of each female patient with RA to deliver naturally is recorded. The hope is that analysis of these accumulated data will aid pregnant patients in making wise, well-informed decisions about safe delivery methods. Other areas of study that would help clinicians better address the needs of affected women include sex differences in diagnosis (12), effect of RA on psychological status (67), effect of environmental influences (68), family history as a risk factor (69, 70), and role of ethnicity (71, 72). A particularly important topic for further study is sex differences in treatments and potential benefits that the newer therapies might have for women, particularly those in their childbearing years. Further development of DMARDs and biologic response modifiers that can be administered safely to women throughout their childbearing and childrearing years remains a meaningful goal. Clinical evaluation of the genotoxic potential of TNF-neutralizing agents may be a significant step toward the development of DMARDs that can be used more safely through the childbearing years.

In addition to studies focused on women's issues in RA, educational programs should alert women and their primary care providers to symptoms suggestive of RA and prompt rapid pursuit of effective therapeutic regimens. Primary care providers need to be educated so that they can recognize early symptoms in women and, when appropriate, recommend early referral to a rheumatologist. Because RA is a chronic disease that may affect the patient's quality of life for several decades, it is particularly important that primary care providers be made aware of findings concerning the beneficial effects of early therapy with DMARDs or biologic agents. The improved quality of life that can be achieved with the currently available RA agents clearly justifies their use. Because women seem to experience more pain and disability than men, early treatment with agents like etanercept, infliximab, leflunomide, and methotrexate may be able to produce a satisfying outcome in quality of life. Insurers also need to be educated about these issues and should be encouraged to allow early referral to rheumatologists and the use of effective treatments without the roadblocks that can impede these beneficial treatment options.

Finally, women themselves need to be educated about this disease. The public is largely uninformed about RA, partly because of confusion about nomenclature. Older women tend to be vague about what type of arthritis they have—a survey of 2,424 women over age 65 found that only 20.5% of women who self-reported having RA actually had the disease (73). Yet public recognition of this disease is important because appropriate medication can make a difference in outcomes. Thanks largely to recent pharmacologic advances, the advantages of early detection and treatment of RA in women are greater than ever. Only by means of widespread public education about RA and its effective treatment can women learn to recognize symptoms and seek care early enough to make the crucial difference between disease amelioration and symptom palliation.