Gender and the treatment of immune-mediated chronic inflammatory diseases: rheumatoid arthritis, inflammatory bowel disease and psoriasis: an observational study

Rheumatoid arthritis (RA) is a chronic systemic disease that affects the synovial joints. Joint inflammation may lead to progressive destruction of bone and cartilage, which eventually causes loss of function and disability [1]. To control symptoms and prevent joint damage, treatment with disease-modifying antirheumatic drugs (DMARDs) such as methotrexate and sulfasalazine, is the cornerstone of RA management. More recent treatments include biologic drugs of which anti-tumor necrosis factor (anti-TNF) agents are the most frequently used [1, 2].

The expression and clinical course of RA are influenced by gender. In developed countries the prevalence of RA is 0,5 to 1.0%, with a male:female ratio of 1:3. The reason for this imbalance is not clear, but both genetic and hormonal factors are thought to be involved. In most studies focusing on the relationship between gender and RA, women were found to have higher disease activity scores, more pain and greater loss of function, both in early and established disease [3‐6]. Other studies have reported faster progression of disability in women as measured by the Health Assessment Questionnaire (HAQ), and a lower remission rate in women with early RA [3, 7]. Although these studies suggest a less favorable course in women, there is also evidence to the contrary. One study found that erosive disease (that is, severe joint destruction) was more common in men, and also occurred earlier in the disease course. However, women underwent more surgical operations to correct the consequences of joint destruction [8].

Crohn's disease (CD) and ulcerative colitis (UC) are two of the most common forms of inflammatory bowel disease (IBD), a group of chronic systemic diseases causing inflammation of the gastrointestinal tract. UC affects only the large bowel, whereas CD can involve the whole gastrointestinal tract. In the treatment of IBD, corticosteroids, aminosalicylates, and immunosuppressive agents are used to induce and remain long-term remission. When these fail to achieve sufficient disease control, biologic agents can be used, with infliximab being the most widely used biologic in IBD treatment [9]. The literature concerning gender and IBD is relatively limited and to some extent contradictory. The male:female incidence ratio is estimated to be around 1:1.5 for IBD [10]. Female gender was reported to be a risk factor for earlier recurrence of CD after surgery in two studies [10, 11], but male gender was found to be a greater risk in another study [12]. Gender was not a prognostic factor for disease course [13]. IBD is associated with an increased risk of colorectal cancer, but this risk seems to be lower in females [14].

Psoriasis encompasses a group of chronic, immune-mediated inflammatory skin and joint diseases, of which chronic plaque psoriasis is the most common form. In the treatment of psoriasis, topical agents are used first, followed by or in combination with phototherapy and/or finally systemic therapy [15]. Both non-biologic (methotrexate, acitretin, cyclosporine and UV therapy) and biologic agents (etanercept, adalimumab, infliximab and ustekinumab) can be used as systemic therapy [16]. There is no difference in the male:female ratio, with an estimated prevalence of 2% for both sexes [17]. A few studies on health -related quality of life (HRQOL) have suggested that with the same degree of skin disease, women experience more stigmatization and worse HRQOL [18, 19]; however, a systemic review did not find any relationship between gender and quality of life [20].

Thus, there is some evidence for gender differences in all three of these immune-mediated inflammatory diseases. In previous studies, a number of explanations have been given for these gender differences, including that they are caused by differences in the treatment given to men and women. Therefore, we investigated gender differences in treatment by analyzing disease characteristics just before initiation of biologic therapy in patients with RA, IBD, or psoriasis.

Patient consent for data registry and subsequent use in research was obtained from patients at the time of inclusion in all three registries, and local ethics committees approved the use of the data for these studies.

In this study, we investigated gender differences in biologic treatment by examining disease characteristics at the time of biologic initiation in RA, IBD, and psoriasis. We found that there was a treatment difference in both IBD and psoriasis, with men receiving biologic medication more often than women. With regard to the disease characteristics, we found that women with RA or IBD had higher scores than men on subjective but not on objective measures. Only for psoriasis did men score higher on objective disease measurements.

Thus, the greater proportion of men receiving biologic treatment for IBD or psoriasis compared with women was not reflected in the population prevalence rates, which were similar for both sexes in both diseases. This in contrast to RA, for which treatment and population male:female ratios matched each other. For RA, our data confirm earlier studies that did not note any gender differences in the overall use of biologic medications [5, 7, 30, 31]. For both psoriasis and IBD, it had previously been suggested that males are more likely than women to receive systemic treatment (biologic or non-biologic) [32, 33], and in the case of psoriasis that female patients are more likely to receive topical treatment whereas men are more likely to receive phototherapy [34].

The gender difference we found in treatment proportions for men and women with IBD or psoriasis gives rise to a number of hypotheses that could explain these findings. Possible reasons that more men than women are treated with biologics include the following: men have more severe disease activity; men may be more likely than women to express treatment preferences for biologics/systemic agents; there may be gender differences in treatment access; the risk:benefit ratio as determined by the physician may be modified by the patient's gender; or there may be a gender difference in the effectiveness of biologicals and of treatment alternatives. With respect to the last point, an interesting point was made by Nyberg et al., who suggested that females are more likely to believe that they can influence their disease themselves and therefore tend to assume or be given greater responsibility for their disease (that is, are prescribed more self-care topical treatment) [34]. Later in-depth interviews showed the same tendency [35]. This could mean that through more and better use of an alternative treatment, women with psoriasis need biologic treatment less often than men. Because some patients in this psoriasis cohort were already using biologics at inclusion, similar effects with better adherence to oral medication by women might have influenced PASI scores. However, to our knowledge no reports exist to support this hypothesis. With regard to the risk:benefit ratio, biologic therapies might be considered more dangerous for young women of childbearing age than for men because of possible teratogenic effects, although for biologics this is not yet proven [36, 37]. This concern, in combination with age differences, might have been an influence on the discrepant gender ratio for biologic treatment in IBD and psoriasis, as the female patients with RA were on average 20 years older than their IBD and psoriasis counterparts when started on biologics (54.6 for RA versus 35.4 years for IBD and between 20 and 40 years for psoriasis [13]).

With regard to the disease characteristics pertaining before biologic therapy initiation, significant gender differences were found for both RA and psoriasis, and similar numerical differences were found for IBD. At treatment start, female patients with RA had higher values for ESR, patient global assessment, TJC, HAQ, DAS28, and DAS28CRP compared with men. One study on anti-TNF therapy [38] found a similar gender imbalance in objective and subjective measurements as we found in this study. Furthermore, recent data from Arkema et al, from the Anti-Rheumatic Therapies in Sweden (ARTIS) registry also support a difference in subjective disease parameters for patients with RA starting biologic therapy.[39] Most studies on gender in RA used time points other than biologic initiation, but the results parallel ours [3‐6, 40].

Because biologic medications are now used earlier in the disease course than at the time these therapies were first introduced, this could possibly induce gender differences. However, we did not find such evidence, and this is in line with similar studies [39]. The differences in TJC and SJC are not large, they could nonetheless be important when studied at group level. Furthermore, measurements of TJC and SJC in rheumatology practice in the hospital system in which these patients were seen are highly standardized (based on the European League Against Rheumatism handbook) so that interobserver variability is minimized. Another confounder could be a gender difference in the specific joints that are inflamed, but as the individual joints are not recorded in the STURE database, this question remains open.

In psoriasis, a gender difference was found for both the PASI and the DLQI, with men more often having high PASI scores and women more often having high DLQI scores. Our findings for the DLQI parallel a few, but not all, studies on QOL in psoriasis [18‐20, 41]. For IBD, the SHS, which can be regarded as a short QOL assessment, tended to be higher in women. This matches earlier research in which female gender was found to be a predictor of impaired HRQOL and level of concern [42‐44].

Although different assessments were used for the three diseases, a general comparison can be made in that in all three diseases the gender imbalance seems to occur only in subjective measurements such as pain, functional status, and QOL. These results support the hypothesis that the inherent biology of the diseases is similar for both genders, but that females experience their illness as worse, and consequently have a higher symptomatic disease burden. For RA, this is further supported by a study in which women were shown to have more symptoms than men despite similar radiographic joint damage [30]. The higher ESR in female patients with RA could be an exception, but women are known to have higher ESR levels than men, which is most likely due to hormonal factors and a difference in hematocrit concentration [39, 45].

The literature regarding gender differences in RA has focused on the possible non-sex neutrality of the disease activity measurements used. For psoriasis and IBD, the same questions may apply. Pain and related measurements are often discussed as being non-sex-neutral. Women are more likely than men to experience different kinds of recurrent pains and also to report more severe levels of pain [46, 47]. Moreover, with the same noxious stimuli, women experience more pain than men [48]. These factors can consequently influence pain measurements (TJC, global pain), assessments in which pain is included (DAS28, SHS symptoms, HBI) or assessments that are affected by pain (HAQ, DLQI, SHS). In the case of RA, concomitant fibromyalgia could also influence pain measurements [49].

Another example is the HAQ score, which is known to be higher in females with RA [5, 6, 30, 50]; and this could be attributable to the fact that women have less muscle strength, but it is also possible that men overestimate their functional capacity or that women have higher pain scores [30, 48]. Therefore the higher HAQ scores might be caused by the properties of the HAQ rather than by RA; however, it is interesting to note that in the healthy population, no sex differences in HAQ scores are seen [51, 52].

Although not formally studied, the PASI might also be influenced by gender. The component of 'scaling' is immediately improved after the use of emollients, and even experienced scorers would probably give a lower score for a skin that has just been treated with emollients. As stated earlier, women are more likely than men to adhere to topical treatment, and consequently, this could mean systematic underscoring of the PASI in female patients.

Thus, it is not clear whether standard disease assessments are sex-neutral, but there is also another way of looking at this question. For example, if women do feel more pain with the same stimulus can it then really be said that the disease is the same? Are fewer compensation resources sufficient reason to disregard a gender difference? It would seem that the subjective disease experience may to some extent be discounted by physicians, whereas for the patient this may be the most important part of their disease.

In summary, we found gender differences in disease characteristics at the start of biologic or systemic treatment in RA, IBD, and psoriasis, with an over-representation of men in the latter two diseases, and greater effect of the disease and worse QOL scores in women with all three diseases. This suggests a potential subtle undertreatment of women with these diseases. We note that undertreatment in RA might then also partly explain the worse outcomes previously reported in longitudinal studies for women with RA.

This study has several limitations, as it was a retrospective observational study with all data having been collected at clinical visits. In addition, it must be noted that the psoriasis results came from an annual report, and were not computed using the raw data. Furthermore, some of the measurements in psoriasis were performed after treatment initiation, and thus the real gender difference might therefore be even bigger. The PASI and DLQI scores were given only for the total group, and not separately for each treatment, therefore subgroup analysis was not possible, thus making comparison with RA and IBD more difficult.

Women with RA or psoriasis scored significantly higher than men on subjective, but not objective, disease activity measures. A similar trend was seen for IBD. This indicates that all three diseases have a greater effect on women than on men at the same level of treatment. These findings might suggest that in all three diseases, subjective measures are discounted to some extent in the therapeutic decision-making process, which could indicate undertreatment in female patients.

NL and RFvV were the major contributors responsible for study design, data analysis, and manuscript writing. RB and FN provided patient data and participated in manuscript writing. All authors had full access to all of the data in the study and can take responsibility for the integrity of the data and the accuracy of the data analysis. All authors have read and approved the final manuscript.