Abstract
Objective This study aimed to explore the experiences of dermatologists and rheumatologists in the early recognition of psoriatic arthritis (PsA) and to identify potential improvements to the current shared-care model.
Methods A 24-question survey addressing referral strategies was constructed by the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) project steering committee and sent to all members (n = 927). Questions addressed the use of screening tools, frequency of PsA in patients with psoriasis, therapeutic decision making, and suggestions for earlier PsA recognition and current unmet needs.
Results There were 149 respondents (16.1% response rate), which included 113 rheumatologists from 37 countries and 26 dermatologists from 16 countries. Of the dermatologists, 81% use PsA-specific screening instruments. Conversely, rheumatologists reported that only 26.8% of patients referred to them from all sources had been assessed with screening tools. Although dermatologists reported that a mean of 67% of suspected PsA cases were confirmed, rheumatologists reported a mean of 47.9% of confirmed cases. Both specialties reported similar views regarding optimization of the diagnostic process and indicated that the best approach involved combining patient-reported (ie, screening tools) and physician-confirmed findings. Moreover, both specialties identified the education of primary care physicians (PCPs) and dermatologists as the greatest priority to improve PsA screening.
Conclusion The survey indicated the current unmet needs in the early recognition of PsA. Important areas to address include improving the use of screening instruments, increasing the education of community-based dermatologists and PCPs, and using a combination of patient-reported and physician-confirmed findings in the screening approach.
Psoriatic arthritis (PsA) is a progressive, chronic, and inflammatory disease that can present in a variety of forms, including peripheral and/or axial joint pain and swelling, enthesitis, and dactylitis. Approximately two-thirds of patients with PsA are thought to experience skin signs and symptoms before any joint involvement.1 The factors behind the diagnostic delay of PsA are multifaceted. Notably, PsA can present in a heterogenous manner, encompassing several domains of musculoskeletal (MSK) symptoms as well as varying extraaxial manifestations.2 Moreover, despite extensive exploration of different biomarkers, including autoantibodies, cytokines, and genetic polymorphisms, a validated diagnostic biomarker for routine clinical use remains elusive.3 Notably, acute-phase reactants (eg, C-reactive protein) do not consistently correlate with PsA disease activity.4
Earlier recognition of joint pain and other MSK symptoms in patients with psoriasis (PsO), and subsequently prompt diagnosis and treatment of PsA, is vital. PsA can develop rapidly, manifesting in destructive radiological changes,5 accelerated joint damage, and reduced quality of life.6 Notably, a 6-month delay between symptom onset and a rheumatology consultation is associated with worse long-term physical function.7
The presence of PsO is a risk factor for developing PsA, although the prevalence rates of PsA in patients with PsO have been shown to vary from 6% to 30%.8,9 Nevertheless, around half of patients with PsO experience joint pain, and a third report dactylitis-like and/or enthesitis-like symptoms. Although these pain symptoms can sometimes be elucidated by alternate diagnoses, such as other arthritic diseases, the fact that skin disease most often precedes the development of arthritis exemplifies the importance of dermatologists and primary care physicians (PCPs) in early PsA recognition. Given the impracticalities of rheumatologists assessing all individuals with PsO for PsA, dermatologists and PCPs in particular represent crucial gatekeepers, with the ability to screen for and refer high-probability PsA cases. Screening tools have been developed attempting to identify PsA among people living with PsO but are not frequently used.
In this survey, we aimed to explore the views and practices of both rheumatologists and dermatologists regarding PsA screening, diagnosis, and care. In particular, we evaluated clinicians’ use of screening tools, their opinions on the optimal method to confirm PsA diagnosis, and the greatest unmet needs that currently impede early PsA recognition.
METHODS
A 24-question survey was constructed by a steering committee of the project consisting of Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) members. The GRAPPA project steering committee, which includes rheumatologists, dermatologists, and patient research partners (PRPs), was also responsible for analyzing the results of the survey. Questions addressed the use of screening tools, frequency of PsA in patients with PsO, therapeutic decision making, and suggestions for earlier PsA recognition and current unmet needs. The survey was sent to all 927 GRAPPA members, which include rheumatologists, dermatologists, and PRPs. Ethics approval was not applicable for this study.
RESULTS
A total of 149 GRAPPA members (16.1% response rate) completed the survey between October 3 and October 16, 2022. Among the respondents, there were 113 rheumatologists from 37 different countries, 26 dermatologists from 16 countries, 7 PRPs, 2 other physicians, and 1 nonphysician. The specialists had a mean of 21.7 (SD 10.7) and 18 (SD 12.3) years of experience, respectively, and the majority (76.9% for rheumatologists and 92% for dermatologists) reported working in a university setting. The responses of the dermatologists and rheumatologists to the survey are summarized in Table 1 and Table 2, respectively.
The dermatologists’ perspective on early PsA recognition and care.
The rheumatologists’ perspective on early PsA recognition and care.
Initial referral and assessment. Dermatologists reported that a mean of 24% (SD 11%) of their patients with PsO also have PsA (Table 1A). Almost all dermatologists (96%) indicated they evaluate MSK symptoms in patients with PsO (Table 1B). Although 77% assessed for these symptoms in all patients with PsO, 19% did so only when MSK symptoms were reported by the patient (Table 1B).
Dermatologists used an array of approaches to identify patients with PsO with a high probability of PsA; 89% relied on physical examination (ie, presence of arthritis, enthesitis, dactylitis), whereas 85% used patient-reported MSK symptoms (ie, joint pain, back pain; Table 1C). This was followed by the use of a screening questionnaire (74%), imaging findings (58%), and finally laboratory-reported findings (39%; Table 1C). In general, 81% of dermatologists indicated that they use screening questionnaires for patients with PsO who they believe are likely to have PsA (Table 1D). Out of the dermatologists who used screening instruments, the Psoriasis Epidemiology Screening Tool (PEST) was the most widely used (81%), followed by the Psoriatic Arthritis Screening and Evaluation Tool (PASE; 24%; Table 1E). Dermatologists estimated that they referred a mean of 85% of patients with suspected PsA to rheumatology (Table 1F), whereas 67% (SD 28%) of referrals by dermatology for suspected PsA eventually received diagnostic confirmation (range 15-100%; Table 1G).
Rheumatologists estimated that a mean of 42.9% (SD 24.9%) of patients with PsO presenting to their clinics had seen a dermatologist within a year prior to diagnosis (Table 2A). However, variation was marked, ranging from 0.1% to 99%. Only 26.8% of rheumatologists reported that the patients referred to them have been screened using specific screening instruments (Table 2B). In such cases, PEST was the most widely used (60% of patients; Table 2C). Rheumatologists estimated that 47.9% (SD 24.2%) of patients with suspected PsA referred to them eventually received a diagnosis of PsA (range 0.5-100%; Table 2D).
Therapeutic decision making. We then explored whether there were interspecialty differences regarding which specialty was considered responsible for therapeutic decision making. For patients with PsO and PsA requiring treatment escalation (ie, introduction of conventional synthetic, targeted synthetic, or biologic disease-modifying antirheumatic drugs [DMARDs]), dermatologists most commonly adopted an interdisciplinary approach (54%), discussing treatment plans with rheumatologists (Table 1H). Conversely, such interdisciplinary decision making was the least frequent approach taken by rheumatologists (25.7%; Table 2E). Moreover, 23% of dermatologists stated that the speciality responsible for therapeutic decisions depended on the patient’s clinical manifestations (ie, skin vs MSK involvement), whereas 19% specified dermatologist-led treatment escalation (Table 1H). For rheumatologists, the speciality responsible for therapeutic decisions was most commonly dictated by clinical manifestation (40.4%). This was followed by rheumatologists treating patients without interdisciplinary discussion, regardless of the nature of skin vs MSK involvement (35.8%; Table 2E).
Optimizing the screening process. We asked clinicians how the screening processes for identifying patients with suspected PsA could be optimized. Across all responses, 78.2% believed that a combination of patient-reported (ie, screening questionnaires) and physician-confirmed findings represent the most effective means of screening patients (Figure 1A). This was viewed as the optimal approach by 85% of dermatologists (Figure 1B) and 76.6% of rheumatologists (Figure 1C). The next most commonly suggested approaches across all responses were a lab-based approach (12.9%), followed by sole physician-confirmed findings (6.1%; Figure 1A).
Opinions regarding the optimization of early PsA recognition from the perspective of GRAPPA members (A: all responses, B: dermatologists, C: rheumatologists). GRAPPA: Group for Research and Assessment of Psoriasis and Psoriatic Arthritis; PsA: psoriatic arthritis.
Both specialties were then asked about what they thought was the biggest unmet need in relation to early PsA recognition. Interestingly, there were disparities in the answers; dermatologists most frequently pointed to a lack of education of dermatologists (85%) and PCPs (65%; Figure 2A), whereas rheumatologists most commonly highlighted a lack of education of PCPs (79.8%) and dermatologists (62.4%; Figure 2B). Other unmet needs that were raised by all respondents (including PRPs and other clinicians) were long waiting times (46.3%), lack of rheumatologists/available appointments (42.9%), poor performance of screening instruments (37.4%), and education of rheumatologists (22.4%; Figure 2C).
Current unmet needs in relation to early PsA recognition in the opinion of GRAPPA members (A: all responses, B: dermatologists, C: rheumatologists). GRAPPA: Group for Research and Assessment of Psoriasis and Psoriatic Arthritis; PsA: psoriatic arthritis.
Qualitative comments. At the end of our survey, we asked respondents if they had any additional comments regarding current challenges to PsA screening and early recognition. The most commonly raised areas included patient education, biomarkers, imaging, and interspeciality collaboration. Here we explore the key areas raised in greater detail.
Eight answers referred to the lack of specific, cost-effective biomarkers and/or screening tools for diagnosing PsA. This is summarized by one rheumatologist who said, “It would be great if we could come up with a diagnostic screening tool for rheumatology/dermatology with good specificity that would help clinicians in the referral process and us in [achieving] early diagnosis.” Interestingly, several respondents also highlighted that imaging, particularly ultrasound, could be an important facet of screening, although one clinician noted that, “imaging including MRI [magnetic resonance imaging] may be falsely negative in patients with active enthesitis or spondyloarthritis.”
Three clinicians and 2 PRPs highlighted a need for patient education, particularly for patients with PsO. Indeed, one PRP highlighted, “When a patient is initially diagnosed with PsO, they should be given signs to look out for PsA.” Moreover, another PRP cites the responsibility of the clinician when diagnosing PsA or PsO, “There is also a lack of knowledge of most patients when confronted for the first time with any information about the conditions. The first physician to diagnose PsO or PsA should be aware of such a gap.”
Three respondents emphasized the importance of maintaining interspecialty approaches to practice. One answer noted that this approach extends beyond just rheumatology and dermatology: “In addition to rheumatologists, dermatologists and family physicians, immunologists and physical medicine and rehabilitation specialists in some countries diagnose, treat and monitor PsA.”
Finally, although the need for further clinician education was conveyed in the quantitative data, some key educational points were raised. First, one rheumatologist stated, “There should be an extensive education among physicians (primary care physicians, dermatologists and rheumatologists) concerning a specific point: a normal blood test does not exclude PsA and the absence of arthritis and enthesis swelling does not exclude PsA.” Another rheumatologist expressed, “Rheumatologists need to be educated that early PsA is not as easy to diagnose and not the same as early RA [rheumatoid arthritis] (ie, majority get erosions), so managing and expressing uncertainty to patients is required, rather than making an incorrect diagnosis.”
DISCUSSION
In this global survey, we explore the clinician-based factors that may contribute to delayed recognition of PsA. In particular, the use of screening tools was widely discussed and remains an unsolved issue in the field. Notably, screening tools were reported to be used by three-quarters of responding dermatologists; nevertheless, they were only used in approximately a quarter of all rheumatology referrals for suspected PsA. This finding most likely reflects the disparities in practice between community-based dermatologists and GRAPPA dermatologists. Indeed, the latter are commonly expert specialists based in university care centers, with a special interest in PsO and PsA, and therefore may be more likely to employ PsA screening tools. The disparity may also be elucidated by the fact that a substantial share of patients were referred to rheumatology by alternate routes, such as primary care and self-referral.
Two-thirds of patients with suspected PsA referred by dermatologists received diagnostic confirmation; conversely, less than half of all the referrals received by rheumatologists were confirmed as PsA. This disparity may reflect both the different clinical manifestations that present to each specialty as well as the effective use of screening tools (and use of different tools) by dermatologists. Again, this finding may also reflect differences between dermatologists who are GRAPPA members and those who are community-based. GRAPPA dermatologists reported use of validated screening instruments and evaluation of MSK manifestations by themselves, whereas the majority of patients referred to rheumatologists received no dedicated screening for PsA at all. To date, the literature regarding the real-world practice of PsA screening among dermatologists, PCPs, and other clinicians is sparse. Research around screening instruments has mainly explored the implementation and validation of these tools in secondary care settings,10 and only a handful of studies have explored their use in primary care.11
Although we did not explore the breakdown of the different specialties that make referrals to rheumatology, the overall low rates of screening instrument usage reported by rheumatologists are striking. Guidance regarding PsA screening in patients with PsO varies across different countries. Notably, in the United Kingdom, the National Institute for Health and Care Excellence (NICE) guidance stipulates that patients with PsO should be screened for PsA annually.12 This policy is supported by evidence from a systematic review that predominantly focused on North American and European data and showed the prevalence of undiagnosed PsA in patients with PsO to be 15.5%.13
Effective use of screening tools represents a prerequisite for early PsA recognition. Current evidence highlights that such instruments can play an important role in PsA screening across both primary care and dermatology clinic settings. In the Prevalence of Psoriatic Arthritis in Adults With Psoriasis: An Estimate From Dermatology Practice (PREPARE) study of 949 patients with PsO across Europe and North America, comparative analysis of 3 different screening tools revealed high negative predictive values (≥ 0.83) and sensitivity values of 0.67 (Psoriasis and Arthritis Screening Questionnaire [PASQ]), 0.77 (Toronto Psoriatic Arthritis Screen [ToPAS]), and 0.84 (PEST).14 These are consistent with previous studies based on dermatology-centered screening, which showed sensitivity ranging from 0.68 to 0.85.10,15 In a primary care setting, PEST was shown to be the most favorable screening instrument, based on its sensitivity (0.68) and specificity (0.71).16
A body of early evidence suggested that the specificity of screening tools tends to be notably lower than sensitivity.10,15 However, a subsequent systematic review from 2019, which conducted pooled specificity analysis for 4 commonly used screening tools (ToPAS, PASE, PEST, and Early Psoriatic Arthritis Screening [EARP]), showed encouraging specificities (ranging from 0.68 to 0.85).17 This discrepancy is potentially elucidated by the heterogenous nature of PsA, as well as variations in the study design and population. Nevertheless, altogether these findings reinforce the notion that screening instruments are useful, but alone they are insufficient; namely, diagnosis also requires subsequent clinical evaluation.
To that end, approximately 80% of clinicians expressed the view that a combined approach using both patient-reported (ie, screening tools) and physician-confirmed findings represents the optimal means of identifying PsA in patients with PsO. Indeed, screening instruments could initially be used to accurately and quickly exclude patients without PsA, thus reducing rates of inappropriate referral to rheumatologists. Subsequently, physician-confirmed findings could help facilitate differentiation of PsA from other arthritic diseases; such an approach could circumvent some of the aforementioned limitations of screening tools. This strategy would likely manifest in varying ways across different specialties. For example, in addition to the screening tool results, physiotherapists could assess the range of movement around key joints. Elsewhere, dermatologists and PCPs could perform targeted MSK examinations. Ultimately, there is a need for a future study that considers our findings in the context of different referral practices to find an optimal referral strategy that is highly specific, sensitive, and cost-effective. Such work can help streamline the referral process to rheumatology and facilitate earlier recognition of PsA in patients with PsO.
Aside from the clinician-centered factors that may underpin undiagnosed/delayed PsA diagnosis, an important alternative factor to consider is patient education. Here, in our qualitative data, both clinicians and PRPs referenced the need to address awareness of the increased risk of PsA and the associated symptoms in patients with PsO. Currently, understanding of the skin symptoms, triggers, and management by patients with PsO have been well explored.18 However, aside from 1 survey that showed that 91% of patients with PsO know that PsA can occur on a background of PsO, there is a paucity of data that addresses either their awareness of the increased risk of developing PsA or knowledge of PsA symptoms. Efforts to educate patients with PsO about PsA are vital and could help empower them to seek earlier clinical advice relating to potential PsA symptoms. Indeed, previous research has shown that patients with PsO want to be actively involved in decision-making processes relating to their care, however insufficient patient knowledge is a barrier to this and is associated with poorer patient satisfaction.19
Our data suggest that dermatologists tend to adopt an interdisciplinary approach to treatment. This reinforces findings from the Multinational Assessment of Psoriasis and Psoriatic Arthritis (MAPP) survey, in which dermatologists were predominantly responsible for managing skin symptoms and rheumatologists were responsible for overall prescribing decisions.20 This finding may be reflective of the nature of patients presenting to each specialty; namely, rheumatologists may see patients with more mild PsO who do not require input from dermatology. Nevertheless, it is notable that in the MAPP survey, 73.2% of rheumatologists were solely responsible for prescribing decisions for patients with PsA, compared to 36% of rheumatologists in our study. This disparity may be because since the MAPP survey, there has been a growing trend to promote collaboration between different specialties in PsA management. This is best exemplified by the emergence of combined dermatology-rheumatology clinics, which facilitate integrated care of patients with PsO and PsA.21
Overall, education of dermatologists and PCPs in relation to early PsA recognition were considered as the largest unmet needs. Notably, twice as many respondents identified education of PCPs as an unmet need compared to the poor performance of screening tools. This is consistent with a study from Wade et al in 2016 showing that 28% of patients with PsO had PsA that had previously been undiagnosed/not recorded in their primary care medical records.22 Despite this, there is evidence that the referral process from primary care for suspected early inflammatory arthritis is improving. Notably, in the UK, data from the National Early Inflammatory Arthritis Audit (2021-2022) showed that the speed of referrals is improving: 54% of patients are now referred to secondary care within 3 days of initial presentation, compared to 47% 2 years prior.23 Thus, the next steps will be to survey PCPs directly and gauge whether PsO and PsA education is seen as the most significant unmet need regarding early PsA recognition, or whether other factors (eg, lack of patient knowledge, delayed initial presentation) represent greater barriers in primary care.
Interestingly, out of the dermatologists surveyed, the greatest unmet need identified was education of dermatologists. This is consistent with previous data, which has shown that over a third of dermatologists have difficulty differentiating PsA from other arthritic disease.20 Moreover, 87.1% of dermatologists stated that PsA is likely underdiagnosed due to a failure to recognize the connection between the skin and joint manifestations.20 These data highlight the importance of educating clinicians about the early signs and symptoms of PsA. In particular, the features that distinguish PsA from other rheumatic diseases (eg, RA and osteoarthritis), as summarized in a recent review by Saalfeld et al,24 should be emphasized.
There are some limitations of this study that warrant discussion. First, the survey was solely distributed among GRAPPA members, with only 16% of members responding to the survey. Although low, this is likely an underestimate, as many GRAPPA members are not practicing clinicians and thus would not have responded. GRAPPA members tend to be expert clinicians based in university settings with an inherent interest in PsO and PsA. Therefore, an important next step is to explore whether the results from our survey are representative across non-GRAPPA clinicians, especially those based in nonuniversity hospitals. For instance, it is likely that the reported use of screening questionnaires (especially in dermatologists) is higher in this group than it would be by other clinicians. Moreover, although the dermatologists surveyed highlighted a need for more PsA education, only 26 dermatologists responded. The opinions expressed here need to be verified across more dermatologists, including non-GRAPPA clinicians. Second, given that our study is based on clinician-reported data, certain findings (eg, the proportion of patients who receive diagnostic confirmation of PsA) may be prone to subjectivity and recall bias.
Finally, the education of PCPs was identified as a major unmet need by rheumatologists, dermatologists, and PRPs. On one hand, identifying the views of dermatologists and rheumatologists is valuable as they are likely to see a higher proportion of patients with PsO (and those at greater risk of developing PsA) than PCPs. Moreover, through encountering patients that have been referred by PCPs, secondary care clinicians are likely to have significant insight into PsA screening processes across primary care. Nevertheless, the next steps will involve surveying PCPs directly to elucidate their current practices, their perceived barriers to early PsA recognition, as well as their amenability to PsA and PsO education.
In conclusion, according to the results of the survey, rheumatologists mostly see patients with PsO with suspected PsA who are referred without any standardized referral instruments, which leads to PsA confirmation in less than half the referred patients. In accordance with dermatologists and rheumatologists, future work to identify the optimal screening strategy for PsA in individuals with PsO should include a combination of patient-reported (ie, screening tools) and physician-confirmed findings. The education of PCPs and dermatologists regarding early PsA recognition in patients with PsO was identified as the biggest current unmet need, above long waiting times and poor performance of screening tools. We suggest that such education should focus on identifying the early signs and symptoms, especially clinical manifestations that differentiate PsA from osteoarthritis and RA.
ACKNOWLEDGMENT
Some of the data from this study have been used in an abstract that has been accepted for the European Alliance of Associations for Rheumatology (EULAR) Congress 2023.
Footnotes
L.C. Coates and D. Poddubnyy contributed equally to this work.
DP receives research support from AbbVie, Eli Lilly, MSD, Novartis, Pfizer; consulting fees from AbbVie, BIOCAD, Eli Lilly, Galapagos, Gilead, GSK, Janssen, MSD, Moonlake, Novartis, Pfizer, Samsung Bioepis, and UCB; speaker fees from AbbVie, BMS, Eli Lilly, Janssen, MSD, Medscape, Novartis, Peervoice, Pfizer, and UCB; and is a member of the executive committee of ASAS and a member of the GRAPPA steering committee. JFM is a consultant and/or investigator for Amgen, BMS, AbbVie, Dermavant, Eli Lilly, Incyte, Novartis, Janssen, UCB, Sanofi-Regeneron, Sun Pharma, Biogen, Pfizer, and Leo Pharma. NG owns stock in UCB and Abcuro. PSH, LCC, and LE are on the editorial board of The Journal of Rheumatology. The remaining authors declare no conflicts of interest relevant to this article.
- Accepted for publication July 13, 2023.
- Copyright © 2023 by the Journal of Rheumatology