To the best of our knowledge, this study is the first to use both qualitative and quantitative methods to evaluate impediments to nr-axSpA diagnosis in the United States.
Median time to nr-axSpA diagnosis was 3.25 years (mean, 7.9 years) among surveyed patients. Although half of patients were diagnosed within 2 years of symptom onset, more than 20% lived with undiagnosed or misdiagnosed symptoms for more than a decade. Diagnostic delay was significantly greater among patients who were women, lived in rural areas, were highly educated, and were not working full-time. The delay revealed in this study is consistent with the mean diagnostic delay of 6.7 years for patients with axSpA in a recent systematic review.(27)
This study identified additional factors that contribute to delay. A critical finding is HCPs’ poor awareness and understanding of axSpA. Patients commonly encountered outdated perceptions of axSpA that biased HCPs against this diagnosis in women and people who were HLA-B27 negative. More than a third of surveyed patients also indicated that seemingly normal imaging studies contributed to diagnostic delay. In 2009, the Assessment of SpondyloArthritis International Society (ASAS) developed classification criteria for axSpA that deliberately recognize that axSpA can be present without definitive radiographic sacroiliitis.(1) However, our findings suggest that many frontline HCPs are not yet familiar with nr-axSpA.
Several factors limit the utility of imaging studies in facilitating axSpA diagnosis. Radiologists often lacked skill in evaluating x-ray and MRI images of the SI joints and likely missed subtle abnormalities. These findings support a recent survey of radiologists demonstrating prevalent knowledge gaps related to axSpA imaging.(28) Studies have shown that the bone marrow edema observed in the MRI of the SI joints in axSpA patients is not specific to the disease but can also occur in athletes, postpartum women, and other healthy populations.(29, 30) Furthermore, degenerative spinal disease frequently coexists with axSpA(31, 32) and may diminish attention to less-familiar axSpA-related abnormalities, particularly when axSpA is not suspected. These issues highlight the importance of close collaboration between radiologists and rheumatologists.
Symptoms described by study patients reflect the marked heterogeneity of nr-axSpA. Current efforts to increase awareness of axSpA often emphasize chronic IBP,(19, 33, 34) and for many patients in this study, persistent back pain was an early and dominant symptom. Yet numerous patients did not have this classic symptom pattern (see Additional file 10). Several patients described acute flares of back pain during their teen years and early adulthood. For other patients, the most painful and disabling symptoms occurred at nonaxial sites, consistent with growing evidence that manifestations of axSpA and peripheral spondyloarthropathy have considerable overlap.(35) Given HCPs’ limited understanding of spondyloarthritis, it seems unlikely that they would recognize these diverse presentations of axSpA. Thus, axSpA campaigns that were centered on the timely recognition and referral of the patient with IBP may be overly narrow. Rather, it may be important to bring attention to the broader phenomena associated with spondyloarthritis and the common underlying pathophysiology.(36) Doing so may help HCPs better discern axSpA in patients with varied, asynchronous, and seemingly unrelated manifestations.
Study patients and their physicians often perceived a connection between axSpA symptoms and preceding physical activities or trauma. These perceptions led some patients to initially self manage symptoms in expectation of gradual resolution. In clinical settings, these perceptions likely led patients and physicians to narrowly focus on the chief complaint, overlooking other relevant signs or symptoms.
Patients’ perceptions that their strenuous activities and axSpA symptoms are linked has been reported in a separate study(37) and may have validity. It is increasingly understood that spondyloarthritis arises, at least in part, from an exaggerated inflammatory response to biomechanical stress in predisposed individuals.(38, 39) Several studies have shown a positive association between physically demanding occupations and axSpA progression;(40, 41) however, few studies have explored the temporal relationship between axSpA symptoms and exertional activity or minor trauma. Further research into these associations is warranted and may facilitate early axSpA detection.
Women in this study were significantly more likely than men to experience delayed nr-axSpA diagnosis, supporting findings from a recent systematic review.(42) The present study identified several factors that likely contribute to this disparity. For example, women were more likely to indicate that normal imaging studies, negative HLA-B27 status, and HCP perceptions of axSpA as a “man’s disease” hindered their diagnosis. Women were also more than 3 times as likely to report that their nr-axSpA symptoms was misdiagnosed, and more than 40% of women reported their HCP making disparaging remarks about their symptoms.
These findings are consistent with a study demonstrating striking sex-specific differences in how rheumatologists documented symptoms of patients with axSpA and showing that men were 3 times more likely to receive an axSpA diagnosis at their first rheumatology visit.(24) The authors concluded that HCPs held gender biases that influenced their clinical assessment, thereby affecting time to diagnosis at multiple time points.
Sex-related differences in the experience of pain may likewise contribute to disparate diagnostic delay. A recent study of axSpA pain found that women had significantly higher rates of pain in the cervicothoracic junction and thoracic spine, as well as higher rates of widespread axial and peripheral articular pain.(43) Biological sex differences that affect the brain’s response to pain signals are apparent in infancy and may predispose women to more widespread pain.(44) Psychological and cultural factors also impact how men and women respond to and communicate about pain. These factors collectively influence doctor-patient interactions that center on pain. These disparities were demonstrated by Global Assessment pain and disability ratings that were discordantly higher among nearly 60% of women, but just 40% of men, compared with their physician’s rating.(45) Women’s disparate diagnostic journey is not unique to axSpA but reflects a broader phenomenon seen with medically inexplicable chronic pain.(46)
Finally, patients in this study faced numerous barriers to rheumatology care that hindered timely nr-axSpA diagnosis. Even when patients explicitly requested referral, their HCPs often resisted. These obstacles stem in part from HCPs’ failure to recognize IBP and other indicators of rheumatological disease; however, local scarcity of rheumatology services can also be a factor.(47) In this study, rural location was independently associated with diagnostic delay, and several regions of the United States are underserved by rheumatologists.(48) Patients in this study were further hampered by the lack of rheumatologists with contemporary knowledge of nr-axSpA, indicating a continuing need for education on this topic. Patients and referring clinicians need mechanisms to more easily identify rheumatologists skilled in spondyloarthritis care. Strategies to improve the timeliness of nr-axSpA diagnosis must also anticipate regional variation in rheumatology access and be tailored accordingly.
Our study has some limitations. We used a convenience sampling approach that recruited almost all of the qualitative study participants, and more than half of survey respondents, through SAA outreach. These individuals were predominantly female and may not be representative of the broader population living with nr-axSpA. To diversify the sample, participants were also recruited from patient panels. This strategy was inefficient and required additional quality checks but increased male and nonwhite survey participation. In both study phases, the determination that respondents had rheumatology-diagnosed nr-axSpA and were eligible relied on self-reported survey responses and could not be independently confirmed.
The qualitative study component revealed patterns in early symptoms and care seeking that may significantly influence time to nr-axSpA diagnosis. These findings were not anticipated and thus were not systematically assessed. Further qualitative and quantitative research into patients’ preclinical symptoms and experiences is indicated.