Measures of psoriatic arthritis: Tender and Swollen Joint Assessment, Psoriasis Area and Severity Index (PASI), Nail Psoriasis Severity Index (NAPSI), Modified Nail Psoriasis Severity Index (mNAPSI), Mander/Newcastle Enthesitis Index (MEI), Leeds Enthesitis Index (LEI), Spondyloarthritis Research Consortium of Canada (SPARCC), Maastricht Ankylosing Spondylitis Enthesis Score (MASES), Leeds Dactylitis Index (LDI), Patient Global for Psoriatic Arthritis, Dermatology Life Quality Index (DLQI), Psoriatic Arthritis Quality of Life (PsAQOL), Functional Assessment of Chronic Illness Therapy–Fatigue (FACIT-F), Psoriatic Arthritis Response Criteria (PsARC), Psoriatic Arthritis Joint Activity Index (PsAJAI), Disease Activity in Psoriatic Arthritis (DAPSA), and Composite Psoriatic Disease Activity Index (CPDAI)

Authors

  • Philip J. Mease

    Corresponding author
    1. Seattle Rheumatology Associates, Swedish Medical Center, and University of Washington School of Medicine, Seattle
    • 1101 Madison Street, Suite 1000, Seattle, WA 98104
    Search for more papers by this author
    • Dr. Mease has received consultant fees, speaking fees, and/or honoraria (less than $10,000 each) from Centocor, BiogenIdec, Roche, UCB, Celgene, and Novartis, and (more than $10,000 each) from Abbott, Amgen, BMS, Lilly, and Pfizer.


INTRODUCTION

The approaches to assessment of psoriatic arthritis (PsA) have matured significantly over the last decade due to the need for reliable measures in clinical trials. Additionally, there is a growing interest in a “treat to target” paradigm in the management of rheumatic diseases, i.e., the goal of achieving minimal disease activity or remission in order to maximize clinical improvement and minimize long-term damage, which requires quantitation of disease activity through validated measures. This paradigm has gained interest because of the increased understanding that has come from trials and clinical registries of patients with rheumatoid arthritis (RA) about the value of inhibiting the impact of disease symptoms and structural damage on function, quality of life, and long-term adverse outcomes related to comorbidities such as cardiovascular disease (1–4). It is becoming apparent that a similar value of tight control and treating to target exists in the management of PsA (5, 6).

Work on outcome measures has been accomplished both in individual centers and the collaborative efforts of these centers through the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) and the Outcome Measures in Rheumatology (OMERACT) associations.

The principle clinical features of PsA to be assessed on physical examination include joint, skin and nail, enthesial, and spine disease, as well as dactylitis, which results from synovitis, tenosynovitis, and enthesitis. Disease activity in these domains may wax and wane in concert, or be divergent, contributing to the somewhat greater complexity of assessment of PsA as compared to RA. Efforts are underway to determine if composite measures of disease activity and response to therapy can be developed that effectively encompass all of these domains. Key domains assessed by patient-reported outcomes include pain, patient global, function, quality of life, and fatigue. Many of the measures of these clinical domains have been successfully adapted from measures used in the assessment of RA, ankylosing spondylitis (AS), and psoriasis. As such, many of these measures will only be touched on briefly in this article, with a focus on their adaptation to PsA, and the reader will be referred to their more extensive description of the basic measure in other articles in this issue. Measures that are not addressed elsewhere in the issue will be described in more detail in this article.

Through analysis of randomized controlled clinical trials, registry data sets, and expert Delphi exercises, a GRAPPA–OMERACT recommendation for a core set of domains to be assessed in PsA clinical trials has been established (7) (Figure 1). The set that should be assessed in all clinical trials (inner ring, Figure 1) includes peripheral joint activity, skin activity, patient global, pain, physical function, and health-related quality of life. Additional domains that ideally should be assessed at some point in a clinical development program include enthesitis, dactylitis, spine disease, nail disease, fatigue, physician global, acute-phase reactants, and structural status by radiograph (second ring, Figure 1). Additional measures still considered in the “research” area of assessments in development (outer ring, Figure 1) are computed tomography, magnetic resonance imaging, and ultrasound imaging scoring systems; tissue analysis (e.g., skin and synovial biopsy); and “participation” (ability to participate in meaningful life activities). The measures used to assess the GRAPPA–OMERACT domain set will be sequentially described in the remainder of this article. References to trials in which the measures have been used are summarized in recent reviews of PsA treatment (5, 8, 9). Many of these measures have not been officially validated in PsA, nor specifics of performance characteristics evaluated, other than the demonstration of their responsiveness and discriminant ability in clinical trials; therefore, psychometric characterization is not available for review for many measures.

Figure 1.

Domains for psoriatic arthritis (7). PGA = physician global assessment; MRI = magnetic resonance imaging; CT = computed tomography; US = ultrasound.

The assessment of PsA has been aided by the development and utilization of formal classification criteria, which are used to select appropriate patients for clinical trials and registries. The criteria of the Classification of Psoriatic Arthritis Study Group were developed from an in-depth clinical, laboratory, and radiographic study of 588 PsA cases and 536 controls with RA, AS, or undifferentiated arthritis (10), using methods of logistic regression analysis, latent class analysis, classification and regression trees methodology, and receiver operating characteristic curve analysis. A patient qualifies for the criteria if they display inflammatory arthritis, enthesitis, and/or spondylitis and 3 points from a list of associated elements (Table 1). The criteria yielded a specificity of 98.7% and sensitivity of 91.4%, superior specificity than previously developed criteria such as those of Moll and Wright or Vasey and Espinoza (10).

Table 1. CASPAR criteria (10)*
  • *

    To meet the criteria of the Classification of Psoriatic Arthritis (CASPAR) Study Group, a patient must have inflammatory articular disease (joint, spine, or enthesial) with ≥3 points from 5 categories. RF = rheumatoid factor.

  • Current psoriasis is assigned a score of 2; all other features are assigned a score of 1.

1. Psoriasis 
 a. CurrentPsoriatic skin or scalp disease present today as judged by a rheumatologist or dermatologist
 b. HistoryA history of psoriasis that may be obtained from patient, family doctor, dermatologist, or rheumatologist
 c. Family historyA history of psoriasis in a first- or second-degree relative according to patient report
2. Psoriatic nail involvementTypical psoriatic nail dystrophy, including onycholysis, pitting, and hyperkeratosis, observed on current physical examination
3. A negative test for RFBy any method except latex, but preferably by enzyme-linked immunosorbent assay or nephelometry, according to the local laboratory reference range
4. Dactylitis 
 a. CurrentSwelling of an entire finger
 b. HistoryA history of dactylitis recorded by a rheumatologist
5. Radiologic evidence of juxtaarticular new bone formationIll-defined ossification near joint margins (but excluding osteophyte formation) on plain radiographs of a hand or foot

GRAPPA is currently initiating an exercise to derive simple clinical definitions for what constitutes “inflammatory” arthritis, enthesitis, and spondylitis in order to aid nonrheumatologists as they attempt to distinguish inflammatory from noninflammatory forms of these conditions (Mease PJ: unpublished observations).

PERIPHERAL JOINT ASSESSMENT

A hallmark feature of psoriatic arthritis (PsA) is the presence of inflammatory arthritis, characterized by tenderness and or swelling due to synovial inflammation. Unlike rheumatoid arthritis (RA), wherein symmetric and polyarticular involvement is frequently seen, PsA may present in an oligoarticular and sometime monarticular pattern, often asymmetric, with a tendency to gradually become more polyarticular and symmetric over time. The pathophysiologic features of joint disease in PsA have recently been reviewed (5).

TENDER AND SWOLLEN JOINT ASSESSMENT

Description

Purpose.

Joints are palpated for the purpose of determining if they are tender and/or swollen, the latter implying the presence of active synovitis, and both implying the presence of inflammation.

Content.

Joints are assessed for tenderness and swelling. The European League Against Rheumatism (EULAR) manual of joint examination in RA (11) demonstrates appropriate examination technique. A rule of thumb is to apply ∼4 kg/cm2 of pressure (enough to blanch the tip of the examiner's fingernail) at the joint line. Joints assessed include the distal interphalangeal (DIP), proximal interphalangeal (PIP), and metacarpophalangeal joints of the hands; the wrist, elbow, shoulder, acromioclavicular, sternoclavicular, temporomandibular, hip, knee, ankle, and midtarsal joints; and the metatarsophalangeal and PIP joints of the feet.

Number of items.

For clinical trials, a 68 tender and 66 swollen joint count, including the DIP joints of the hands and excluding hips for swelling, is recommended (7, 12, 13). The DIP joints are included because of their common involvement in PsA, unlike RA. There is also a greater tendency for more asymmetric and oligoarticular joint involvement in PsA than in RA. The DIP joints of the toes are not included because these may be difficult to evaluate reliably, and addition of these has not been demonstrated to improve performance characteristics of joint scoring systems (14). Although the 28 joint count, as used in RA, has been found to have good performance characteristics in a study of PsA phase II anti–tumor necrosis factor randomized controlled trials (RCTs) (14), it has not been recommended as an entry criteria or primary end point for RCTs because of the potential to underassess disease in the lower extremity and DIP joints, which would have resulted in 20% of patients being excluded if used to determine study eligibility (14). However, it should be noted that a retrospective analysis of the phase III infliximab trial in PsA demonstrated good performance characteristics of the simplified joint counts evaluated, including the 28 used in the Disease Activity Score in 28 joints (DAS28), a 32-joint count including the DIPs of the hands and excluding the elbows and shoulders, and a 36-joint count including the DIPs of the hands and ankles and excluding the shoulders, compared to the 68/66-joint count (15). Furthermore, in this study, only 6.5% of patients would have been excluded if these joint counts had been used to determine eligibility for trial enrollment.

Response options/scale.

As in RA, the convention is to count the presence or absence of tenderness and swelling and not grade severity. Unlike RA, involvement of DIP joints is common in PsA.

In RCTs, the tender and swollen joint counts are reported separately and are used to determine the American College of Rheumatology (ACR) response and the DAS and DAS28 disease activity and EULAR response criteria, as described for RA (see article on rheumatoid arthritis disease activity), as well as the Psoriatic Arthritis Response Criteria (PsARC) (12, 13), discussed below.

Recall period for items.

Current presence or absence of tenderness/swelling.

Examples of use.

The tender and swollen joint count is used in all clinical trials of PsA: Mease PJ. Psoriatic arthritis: update on pathophysiology, assessment and management. Ann Rheum Dis 2011;70 Suppl:i77–84 (5).

Mease PJ, Antoni CE. Psoriatic arthritis treatment: biological response modifiers. Ann Rheum Dis 2005;64 Suppl:ii78–82 (8).

Mease PJ. Psoriatic arthritis: pharmacotherapy update. Curr Rheumatol Rep 2010;12:272–80 (9).

Practical Application

Method of administration.

Physical examination of 68 joints. Results are collected on a simple score sheet, on paper, or electronically.

Scoring.

Presence or absence of tenderness and swelling.

Score interpretation.

Used in composite measures of arthritis such as the ACR score, DAS scoring systems, or PsARC, as well as emerging composite scoring systems (see below).

Respondent burden.

Minimal.

Administrative burden.

Minimal; takes ∼2 minutes to complete with assistant to record, or slightly longer if no assistant is available.

Psychometric Information

Method of development.

Historically developed for RA assessment.

Reliability.

Regarding reliability in RA, see article on rheumatoid arthritis disease activity. Regarding PsA, in a reliability exercise involving 20 experts in PsA and ankylosing spondylitis (AS), examining 10 patients with PsA and 10 with AS, the intraclass correlation coefficient (ICC) for tender joint count in the PsA patients was 0.78 (95% confidence interval [95% CI] 0.61, 0.93) and for swollen joint count was 0.50 (95% CI 0.27, 0.78) (16). The greater variability in swollen joint count may have been partly due to minimal prestudy standardization of swollen joint assessment. In a subsequent exercise involving 10 rheumatologists and 9 dermatologists examining 20 PsA patients in a Latin square design, the ICC for tender joint count among the rheumatologists was 0.81 (95% CI 0.68, 0.91) and for swollen joint count was 0.42 (95% CI 0.23, 0.65) (17). The analogous values among the dermatologists were 0.73 (95% CI 0.56, 0.86) and 0.31 (95% CI 0.23, 0.57), respectively.

Ability to detect change.

The responsiveness of the 68/66 tender/swollen joint count has been demonstrated quantitatively in clinical trials of infliximab and etanercept in PsA. In a phase II trial of infliximab in PsA, the standardized response mean of 68 tender joint count in the treatment arm was −1.14 and in placebo was 0.07, for a t value of 6.0 (14). In a similar analysis of a phase II trial of etanercept, these values were −1.25, 0.53, and 6.6, respectively (14). In the same trials, the values for 66 tender joint count were −1.15 and −0.11, respectively, with a t value of 5.0 in the infliximab trial, and −1.53 and −0.20, respectively, with a t value of 5.4 (14). By comparison, the t values of the 28 tender/swollen joint counts were 5.4 and 3.8, respectively, in the infliximab trial and 5.8 and 4.1, respectively, in the etanercept trial (14). By further comparison, the t values of the 78/76 tender/swollen joint counts in the etanercept trial were 6.6 and 4.4, respectively (14).

Validity.

Validation of tender and swollen joint count in PsA has not been performed.

Critical Appraisal of Overall Value to the Rheumatology Community

Joint assessment for tenderness and swelling plays a similar role as it does in RA, i.e., as a marker for presence of inflammation in joints. Because of the involvement of DIP joints and the tendency for PsA to be more asymmetric and oligoarticular, an expanded joint count of 68 tender and 66 swollen joints is recommended to more accurately assess the total burden of joint involvement. It appears that although tender joint count appears to be reliably assessed, there are still challenges in interrater reliability in swollen joint counts. In a study using dolorimetry, it has been suggested that patients with PsA display less tenderness with joint pressure than patients with RA (18).

SKIN ASSESSMENT

Psoriasis lesions may occur virtually anywhere on the skin, but are most commonly found on extensor surfaces and in the scalp. In the most common form of psoriasis, plaque psoriasis or psoriasis vulgaris, the lesions have variable degrees of erythema, induration, and scale. Most trials involving psoriasis patients are restricted to patients with this variant. Other less common variants include guttate, erythrodermic, and pustular psoriasis. Separate instruments are used to measure nail changes. The following are commonly-used psoriasis measures in psoriatic arthritis (PsA) randomized controlled trials (RCTs) and registries to assess plaque psoriasis. Although most expertly performed by dermatologists, they can be performed adequately by rheumatology clinicians trained in their use (17).

The measure most commonly used as a primary outcome measure in psoriasis trials, the Psoriasis Area and Severity Index (PASI) (19), is typically a secondary measure in PsA clinical trials. Because its performance characteristics are diminished in patients with low lesional burden, it is not typically calculated in patients with <3% body surface area (BSA) involvement with psoriasis lesions. Therefore, not all patients in an RCT will be PASI measurable. To account for this and insure some psoriasis measurement in all patients, some trials require the presence of at least 1 measurable “target lesion” of at least 2 cm in diameter. Measurable implies that the lesion is not in the scalp or groin. A BSA score (20, 21) of lesional involvement and the Physician Static Global Assessment (PSGA or PGA) (22) are 2 other commonly used measures in PsA trials. Other psoriasis measures, such as the National Psoriasis Foundation Psoriasis Score (23), the Lattice System PGA (22), or the Copenhagen Psoriasis Severity Index (CoPSI) (24) scoring systems, have not been used in PsA trials or registries and therefore are not further commented upon here. These and other instruments that have been developed for psoriasis trials should be considered in the future for use in PsA trials if they show superior psychometric properties to or greater feasibility than the PASI, as has been suggested for the CoPSI (24).

The most basic assessment of psoriasis lesional burden is the BSA score (20, 21). The typical method to assess BSA is to consider the surface area of the patient's handprint (palm and fingers) as representing 1% of the body's surface area. The clinician then estimates how many “handprints” would be filled by the summated lesions on the person's body. Although planimetric study suggests that the area of a flat closed hand is 0.70–0.76% of the BSA, the handprint rule has become accepted as the standard approach to estimation of BSA (25).

PSORIASIS AREA AND SEVERITY INDEX (PASI)

Description

Purpose.

To provide quantitative assessment of psoriasis lesional burden based on the amount of BSA involved and degree of severity of erythema, induration, and scale, weighted by body part.

Content.

The PASI was developed within a clinical trial and measures both surface area and lesional severity of psoriasis (19).

Number of items.

4 items (surface area, severity of erythema [redness], induration [thickness], and desquamation [scale]) evaluated for 4 body areas (head, trunk, and upper and lower extremities).

Response options/scale.

The head, upper extremities, lower extremities, and trunk are assessed separately and then combined using weighting based on the surface area represented by each area (head = 0.1, upper extremities = 0.2, trunk = 0.3, and lower extremities = 0.4). The degree of erythema, induration, and scale in each area is judged on a 0–4 scale, the sum of which represents disease severity. The area of involvement of each area is graded from 0–6, depending on the estimated percentage of lesional area (0 = 0%, 1 = <10%, 2 = 10–29%, 3 = 30–49%, 4 = 50–69%, 5 = 70–89%, and 6 = 90–100%). These body scores are multiplied by the disease severity score and the weighting for each body area, yielding a score between 0 and 72. In trials, PASI calculators are supplied to facilitate ease of scoring.

Recall period for items.

Current evaluation.

Examples of use.

Widely used in clinical trials of psoriasis and PsA. Not typically used in clinical practice because of complexity.

Practical Application

How to obtain.

The PASI can be obtained online at http://www.dermnetnz.org/scaly/pasi.html, as well as from the primary article on its development (19). Training videos have also been developed. The Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (online at http://www.grappanetwork.org) has developed such a video for the purpose of teaching skin and nail assessment to clinicians and trainees, and for use in the performance of clinical trials.

Method of administration.

Physical examination; can be recorded on paper or entered into a computer or calculator instrument.

Scoring.

The body is divided into 4 regions: head and neck (H; 10% of a person's skin), upper extremities (A; 20%), trunk (T; 30%), and lower extremities (L; 40%). Each of these areas is scored individually, and then the 4 values are combined for the total PASI score. For each body region, the percentage of area of skin involved is estimated and then assigned a grade from 0–6. Grade 0 is assigned for 0% of the area involved, grade 1 for <10%, grade 2 for 10–29%, grade 3 for 30–49%, grade 4 for 50–69%, grade 5 for 70–89%, and grade 6 for 90–100%. Within each region, the severity of psoriasis is estimated by erythema, induration, and desquamation. Severity parameters are measured on a scale of 0–4, from absent to very severe.

The sum of all 3 severity parameters is then calculated for each region of skin, multiplied by the area grade assigned for that body region and multiplied by weight of the respective section (0.1 for head and neck, 0.2 for upper extremities, 0.3 for body, and 0.4 for lower extremities).

equation image

where E = erythema, I = induration, D = desquamation, and A = area.

Score interpretation.

Score range is 0–72. Not reliable in patients who have <3% BSA lesional involvement and since it is rare to have a PASI score >40, nearly one-half of the scale is not used.

Respondent burden.

Minimal.

Administrative burden.

Takes ∼5 minutes to perform with an assistant recording, or slightly longer if not.

Translations/adaptations.

A simplified version has been developed, but not yet used, in PsA trials (26). A patient self-administered PASI has been shown to be reliable and correlates closely with the PASI, suggesting that it can be used when a skilled evaluator is not present (27).

Psychometric Information

Method of development.

The PASI score was developed empirically for a trial of a retinoid therapy of psoriasis in 1978 (19).

Acceptability.

Acceptable for clinical trials but not used in clinical practice.

Reliability.

Subjective scoring of erythema, induration, and scale by a single trained observer has been demonstrated to be reliable compared to objective measures such as laser Doppler flowmeter, spectrodiometer, erythema meter, and chromameter (erythema); ultrasound (induration); and optical profilometry and scanning macrophotographic densitometry (scale) (25). In a study involving experienced and inexperienced clinicians, intrarater reliability was superior in experienced versus inexperienced clinicians (σ = 1.2 versus 3.2) (22). Interrater variation was greater, again superior in experienced compared to inexperienced clinicians (σ = 8.1 versus 9.6). In a study comparing 10 rheumatologists and 9 dermatologists, all of whom had experience with PASI scoring, examining patients with PsA, the intraclass correlation coefficient for the PASI among dermatologists was 0.74 (95% confidence interval [95% CI] 0.58, 0.87) and for rheumatologists was 0.70 (95% CI 0.53, 0.85). This suggested that the instrument could be applied reliably by trained clinicians in both specialties.

Validity.

Content validity is based on the fact that the PASI measures objective skin lesion parameters of severity. However, the PASI does not include a fuller set of elements considered important to patient's assessment of severity such as “embarrassment over appearance” and itching (25). The lack of a gold standard measure of psoriasis severity limits the ability to establish criterion validity (25).

Ability to detect change.

The PASI score has been used widely in clinical trials of psoriasis and PsA and demonstrates excellent ability to detect change and discriminate from placebo.

Critical Appraisal of Overall Value to the Rheumatology Community

There is poor sensitivity to change and responsiveness in mild psoriasis, so application is restricted to those with ≥3% BSA, and since it is rare to encounter a patient with a PASI score >40, nearly one-half of the range of the scale is unused. The features of erythema and scale may vary with changes in temperature and humidity and the use of emollients, instituting variables unrelated to an interventional therapy. The PASI instrument is also impractical to use in clinical practice (25). Despite these limitations, the PASI remains the most commonly used quantitative instrument to assess psoriasis in clinical trials. A standard threshold to report efficacy in a clinical trial is the PASI75 response, i.e., the percentage of patients achieving at least 75% improvement in the PASI score. This benchmark was established in a meeting between the Food and Drug Administration and the Dermatology Advisory Council in 1998. Some have criticized that this is too stringent a benchmark, in that many patients do not seek to change therapy until below a PASI50 response; the PASI50 is also associated with significant improvement in quality of life and is discriminant in clinical trials (28). However, despite these points, the PASI75 has remained the benchmark. As more effective therapies have emerged recently, PASI90 response is also measured.

The target lesion score has been used in PsA clinical trials to allow assessment of at least 1 psoriatic lesion in patients, since patients with low BSA involvement with psoriasis are not reliably measured by the PASI score. An evaluable lesion (not in the scalp, groin, or axilla) of at least 2 cm in diameter is serially evaluated for change in size (diameter) and degree of erythema, induration, and scale on a 0–3 scale of severity.

PSGA or PGA

The PGA is an overall assessment of the patient's skin lesions, on a scale of 7 descriptors, in which 0 = clear and 6 = very severe (22). It is less quantitative than the PASI but is simpler to use and is widely recognized and accepted by dermatologists.

NAIL ASSESSMENT

There is evidence of nail disease in up to 50% of patients with psoriasis and up to 80% of patients with psoriatic arthritis (PsA) (17). Characteristic nail changes involving the nail matrix include pitting, leuconychia, lunular red spots, and nail plate crumbling, whereas changes in the nail bed yield onycholysis and subungual hyperkeratosis, “oil-drop change,” salmon spots, or splinter hemorrhages. The Nail Psoriasis Severity Index (NAPSI) is the most comprehensive assessment of nail disease used in psoriasis clinical trials (29). In this system, the nail is divided into 4 quadrants and 1 point is awarded if there is any finding of nail matrix and 1 point is awarded for nail bed change that is seen, per quadrant, or 0–8 per nail. This yields a potential total score of 80 if just the fingers are used and 160 if the toes are included. The original study describing the instrument showed good reproducibility (21) and a subsequent study showed good interrater reliability (30). This instrument is routinely used in psoriasis clinical trials. A modification of this system (mNAPSI) is a shorter and more feasible scoring system that has demonstrated excellent interrater reliability (31) and has been used in PsA clinical trials.

NAIL PSORIASIS SEVERITY INDEX (NAPSI)

Description

Purpose.

To develop an objective reproducible tool for scoring nail psoriasis.

Content.

Each nail is scored by the presence or absence of nail bed psoriasis and nail matrix psoriasis. Nail bed psoriasis includes onycholysis (separation of the nail bed), splinter hemorrhages (small, dark brown, linear marks under the nail), hyperkeratosis (thickened nail keratin), and oil-drop dyschromia (reddish-brown discoloration under the nail plate), while nail matrix psoriasis includes pitting (sharply defined depressions in the nail surface), leukonychia (white spots in the nail plate), crumbling, and red spots in the lunula.

Number of items.

Evaluation of the nail bed and nail matrix are performed for each nail.

Response options/scale.

Each fingernail is divided into 4 quadrants. For nail bed psoriasis, if no nail bed features are present, a score of 0 is assigned. A score of 1 is assigned if nail bed features are present in 1 quadrant of the nail, 2 if present in 2 quadrants, 3 if present in 3 quadrants, and 4 if present in 4 quadrants.

For nail matrix psoriasis, if no nail matrix features are present, a score of 0 is assigned. A score of 1 is assigned if nail matrix features are present in 1 quadrant of the nail, 2 if present in 2 quadrants, 3 if present in 3 quadrants, and 4 if present in 4 quadrants. The nail bed score and nail matrix score are added together to produce a total score for each nail, ranging from 0–8.

Recall period for items.

Current.

Examples of use.

Used in psoriasis randomized controlled trials (RCTs).

Practical Application

How to obtain.

Available from the original article (29).

Method of administration.

Physical examination, recorded on paper or electronically.

Scoring.

Each nail has a possible score of 0–8, with a total possible score of 0–80 for fingernails, or 0–160 if toenails are included.

Score interpretation.

Higher scores represent worse nail disease.

Respondent burden.

Minimal.

Administrative burden.

Takes 5–10 minutes to complete scoring, depending on amount and severity of nail disease.

Translations/adaptations.

An abbreviated “target nail” version of the NAPSI is described in the original article (29). In this adaptation, 1 nail is selected, and the presence or absence of 8 parameters (onycholysis, splinter hemorrhages, hyperkeratosis, oil-drop dyschromia, pitting, leukonychia, crumbling, and red spots in the lunula) is assessed in 4 quadrants, yielding a total possible score of 0–32.

Psychometric Information

Acceptability.

Acceptable for use by dermatologists in psoriasis RCTs but in PsA, either the mNAPSI or the single target nail NAPSI are utilized.

Reliability.

An informal assessment of the NAPSI in its original development study, involving 37 dermatologists, demonstrated good agreement between evaluators (29).

Validity.

Not validated in psoriasis or PsA. A meta-analysis of nail assessment in psoriasis RCTs has reviewed the NAPSI and other measures and identified the need for validation (32).

Ability to detect change.

The NAPSI has shown responsiveness and discrimination in psoriasis RCTs. The target nail NAPSI showed responsiveness in a placebo-controlled study of golimumab in patients with PsA (33).

Critical Appraisal of Overall Value to the Rheumatology Community

The NAPSI is a detailed and highly quantitative instrument used in psoriasis RCTs but not in practice, and often a simpler measure, such as the mNAPSI or nail visual analog scale (VAS), is used in PsA clinical trials, wherein not all examiners are necessarily dermatologists, may be utilized.

MODIFIED NAIL PSORIASIS SEVERITY INDEX (MNAPSI)

Description

Purpose.

To develop a nail scoring method that is simpler and more reliable than the NAPSI.

Content.

For each fingernail, 7 groups of features are evaluated: pitting, onycholysis and oil-drop dyschromia, nail plate crumbling, leukonychia, splinter hemorrhages, hyperkeratosis, and red spots in the lunula. Pitting, onycholysis and oil-drop dyschromia, and crumbling (including fragmentation and horizontal ridging of the nail bed) are graded from 0–3 in severity. Leukonychia, splinter hemorrhages, hyperkeratosis, and red spots in the lunula are graded as either present or absent.

Number of items.

7 groups of features are evaluated for each fingernail.

Response options/scale.

Onycholysis and oil-drop dyschromia are considered together. If no part of the nail has onycholysis or dyschromia, a score of 0 is assigned. If ≤10% of the nail has onycholysis or dyschromia a score of 1 is assigned, if 11–30% is involved a score of 2 is assigned, and if >30% is involved a score of 3 is assigned.

Pitting is scored by the number of pits present in the nail. Only pits distinctly separate from nail plate crumbling are scored. A nail with no pits is assigned a score of 0, a nail with 1–10 pits is assigned a score of 1, 11–49 pits is assigned a score of 2, and ≥50 pits is assigned a score of 3.

Crumbling may be associated with pitting. If no crumbling is present, the nail is assigned a score of 0. If crumbling is present in 1–25% of the nail, a score of 1 is assigned. If 26–50% is involved, a score of 2 is assigned. If >50% is involved, a score of 3 is assigned.

Leukonychia, splinter hemorrhages, hyperkeratosis, and red spots in the lunula are scored only by their presence or absence. A score of 1 indicates the presence of a feature, and a score of 0 indicates absence.

Recall period for items.

Current.

Examples of use.

Used in PsA clinical trials.

Practical Application

How to obtain.

Measure is available from the original article (31).

Method of administration.

Physical examination, recorded on paper or electronically.

Scoring.

0–13 per nail and 0–130 for all fingernails.

Score interpretation.

Higher scores represent worse nail disease.

Respondent burden.

Minimal.

Administrative burden.

Scoring takes <5 minutes to perform.

Psychometric Information

Method of development.

The mNAPSI was developed by rheumatologists with assistance from dermatologists as a measure simpler than the NAPSI for clinical trials.

Acceptability.

Acceptable.

Reliability.

Excellent intraobserver agreement has been demonstrated among PsA patients, with an intraclass correlation coefficient of 0.92 (95% confidence interval 0.87, 0.97) (31).

Validity.

The mNAPSI retains the content validity of the original NAPSI in that it retains all clinically relevant aspects of psoriatic nail disease (30). A significant correlation (P < 0.05) has been found between mNAPSI scores and several other clinical measures of PsA (including physician global PsA disease severity VAS, swollen joint count, tender joint count, and patient global nail severity VAS), providing construct validity (31).

Ability to detect change.

Responsiveness is currently being assessed in PsA clinical trials.

Critical Appraisal of Overall Value to the Rheumatology Community

This system does away with quadrant analysis and is simpler to perform, and is therefore more practical for clinical trials, and demonstrates excellent intra- and interrater reliability (31).

Nail VAS

A simple nail VAS scoring system or overall assessment of mild/moderate/severe has also been employed in PsA trials in order to gain an impression of therapeutic effect.

ENTHESITIS

Enthesitis is characterized by inflammation at sites of tendon, ligament, and joint capsule fiber insertion into bone, and is considered a pathophysiologically important aspect of psoriatic arthritis (PsA), as well as other spondylarthritides (SpA) (34). Recent registry and clinical trial patient sets have found enthesitis in approximately 30–50% of PsA patients (35). Although classically depicted involving the Achilles tendon and plantar fascia insertion sites, enthesitis can involve many parts of the body, including periknee, pelvis, spine (vertebral ligament insertion), rib cage, shoulder, and elbow. Several enthesitis scoring measures have been developed, some originally developed in patients with ankylosing spondylitis (AS). All involve a standard palpation approach, i.e., applying ∼4 kg/cm2 of pressure (enough to blanch the tip of the examiner's fingernail) and ascertaining the presence/absence and, in some indices, severity of tenderness.

MANDER/NEWCASTLE ENTHESITIS INDEX (MEI)

Purpose.

The MEI was originally developed to assess all clinically accessible entheses potentially involved in AS.

Method of development.

Based on clinical experience, the investigators identified a large number of potentially involved enthesial sites. After removing sites that did not produce tenderness on palpation in any of 19 study patients, the instrument specified 66 sites for assessment (36).

Scoring.

A scoring system based on the patient's response to palpation over the entheses is rated from 0–3 (where 0 = no pain, 1 = mild tenderness, 2 = moderate tenderness, and 3 = wince or withdraw). A maximum total score of 90 is possible (36).

Examples of use.

The MEI has not been used in randomized controlled trials (RCTs) because of burden of administration and concern about reliability.

Critical Appraisal of Overall Value to the Rheumatology Community

The instrument has been criticized for the large number of sites examined, rendering it too time consuming for use in clinical trials, as well as overlap of many sites with fibromyalgia tender point sites. Further, the 0–3 scoring system could contribute to greater inter- and intrarater inconsistency. It has never been used in an RCT and therefore has not been evaluated for reliability or responsiveness. However, it is often referred to for the purpose of describing the overall set of potential enthesis sites from which other measures have derived their simpler version. Indeed, the Maastricht scoring system (see below) was derived from the MEI and in the process, a validation exercise for the MEI was performed (16).

LEEDS ENTHESITIS INDEX (LEI)

Description

Purpose.

To assess enthesitis in patients with PsA. Whereas other enthesitis measures described here were developed and/or validated in patients with AS, the LEI was developed specifically for PsA (37).

Content.

Enthesial sites include the bilateral lateral epicondyles, medial femoral condyles, and Achilles tendon insertions.

Number of items.

6 enthesial sites.

Response options/scale.

Presence or absence of tenderness.

Recall period for items.

Current.

Examples of use.

Used in several PsA trials being conducted currently.

Practical Application

How to obtain.

Description of sites can be found in the original article (37) and examination technique is present on the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) web site (http://www.grappanetwork.org).

Method of administration.

Physical examination; results can be recorded on paper document or electronically.

Scoring.

Tenderness on examination is recorded as either present (1) or absent (0) for each of the 6 sites, for an overall score range of 0–6.

Score interpretation.

Higher count represents greater enthesitis burden.

Respondent burden.

Minimal.

Administrative burden.

Takes ∼30 seconds to complete.

Psychometric Information

Method of development.

Similar to the methodology used to develop the Maastricht Ankylosing Spondylitis Enthesis Score (MASES; see below), the 6 sites of the measure were selected based on a stepwise data reduction to identify those sites most commonly involved.

Acceptability.

Acceptable.

Reliability.

In the study involving comparison of measures in the assessment of AS and PsA spondylitis patients, the LEI demonstrated an intraclass correlation coefficient (ICC) of 0.81 (95% confidence interval [95% CI] 0.65, 0.94) (16).

Validity.

In a study comparing the LEI, MEI, MASES, modified Spondyloarthritis Research Consortium of Canada (SPARCC; 8 sites), and Major indices in PsA patients commencing disease-modifying therapy, clinical parameters of disease activity correlated most consistently with the LEI (37).

Ability to detect change.

In the study described above, the greatest effect size at 6 months was demonstrated with the LEI and modified SPARCC, moderate change with the Major, and small for the MASES and MEI (37). In this same exercise, the LEI showed the least floor effect (scoring 0 when MEI was >0) of any of these indices (37).

Critical Appraisal of Overall Value to the Rheumatology Community

In an open-label longitudinal treatment study including several enthesial measures, the LEI showed the closest correlation with other disease activity measures, was responsive, and showed the least floor effect, i.e., indicating its ability to identify the majority of PsA patients with enthesitis (37).

SPONDYLOARTHRITIS RESEARCH CONSORTIUM OF CANADA (SPARCC)

Description

Purpose.

To assess enthesitis in patients with SpA. The SPARCC created a measure for enthesitis in SpA in general (i.e., not limited to PsA or AS).

Content.

Enthesial sites examined include the bilateral Achilles tendons, plantar fascia insertion at the calcaneus, patellar tendon insertion at the base of the patella, quadriceps insertion into the superior border of the patella, supraspinatus insertion into the greater tuberosity of the humerus, and medial and lateral epicondyles.

Number of items.

16 enthesial sites.

Response options/scale.

Presence or absence of tenderness.

Recall period for items.

Current.

Practical Application

How to obtain.

List of sites is in the original article (38). Examination method may be viewed online at www.arthritisdoctor.ca and is also present on the GRAPPA web site (http://www.grappanetwork.org).

Method of administration.

Physical examination; results can be recorded on paper document or electronically.

Scoring.

Tenderness on examination is recorded as either present (1) or absent (0) for each of the 16 sites, for an overall score range of 0–16.

Score interpretation.

Higher count represents greater enthesitis burden.

Respondent burden.

Minimal.

Administrative burden.

Takes ∼2–5 minutes to complete.

Translations/adaptations.

Modified versions with fewer but more commonly involved sites (6 and 8 sites) showed greater responsiveness and were more discriminant between treatment and placebo (6 sites) (38).

Psychometric Information

Method of development.

Selection of enthesitis sites was based on information from published power Doppler ultrasound in SpA patients compared to rheumatoid arthritis patients and healthy controls, and magnetic resonance imaging studies of the shoulder in AS patients. The most frequent enthesitic sites were selected (38).

Acceptability.

Acceptable.

Reliability.

In a study comparing enthesitis indices in patients with AS versus PsA with spondylitis, the 8-site SPARCC index showed an ICC of 0.81 (95% CI 0.64, 0.93) (16).

Validity.

The instrument has not been validated in PsA. In patients with AS, substantial correlations have been observed between the SPARCC enthesitis score and the Bath Ankylosing Spondylitis Disease Activity Index, Bath Ankylosing Spondylitis Function Index, and patient global (38).

Ability to detect change.

In a study of 9 patients randomized to adalimumab, a nonsignificant reduction in the SPARCC index scores was recorded at 12 weeks. The SPARCC scores had decreased further at the 24-week assessment (P = 0.04).

MAASTRICHT ANKYLOSING SPONDYLITIS ENTHESIS SCORE (MASES)

Description

Purpose.

The original purpose was for the assessment of enthesitis in AS, and now is additionally used in PsA and SpA in general.

Content.

Clinical scoring system for enthesitis in SpA, including AS and PsA. Enthesial sites assessed include the bilateral first costochondral joints, seventh costochondral joints, posterior superior iliac spines, anterior superior iliac spines, iliac crests, proximal insertion of Achilles tendons, and the fifth lumbar spinous process.

Number of items.

13 enthesial sites.

Response options/scale.

Presence or absence of tenderness.

Recall period for items.

Current.

Endorsements.

Recommended by the Assessment of SpondyloArthritis international Society for use in randomized controlled trials of AS and SpA.

Examples of use.

Used in AS clinical trials, emerging trials with the recently adopted axial SpA criteria, and PsA trials.

Practical Application

How to obtain.

List of entheses in the original article (39) and examination technique is demonstrated on the GRAPPA web site (http://www.grappanetwork.org).

Method of administration.

Physical examination; results can be recorded on paper document or electronically.

Scoring.

Tenderness on examination is recorded as either present (1) or absent (0) for each of the 13 sites, for an overall score range of 0–13 (39).

Score interpretation.

Higher scores reflect greater enthesitis burden.

Respondent burden.

Minimal.

Administrative burden.

Completion by health professional takes 2–5 minutes.

Psychometric Information

Method of development.

Recognizing that the MEI was too lengthy for use in clinical trials, Heuft-Dorenbosch and colleagues employed the MEI in AS patients over 2 years, and selected the 13 most specific and sensitive sites from that index to constitute the MASES (39).

Acceptability.

Acceptable. The MASES correlates well with the MEI, and the reduction in site number and removal of intensity grading yields a more practical instrument.

Reliability.

In a study in which several enthesitis indices were compared in the evaluation of patients with AS or PsA with spondylitis, moderate intraobserver agreement was demonstrated among PsA patients, with an ICC of 0.56 (95% CI 0.34, 0.82). The ICC for the MASES was greater in patients with AS than in PsA (16).

Validity.

The instrument has not been validated in PsA.

Ability to detect change.

Discrimination and responsiveness have been demonstrated in a trial of golimumab in PsA (33) as well as multiple studies in AS.

Berlin (Major)

This is a 12-site enthesitis index (40) (Table 2) used in studies of AS, and was also evaluated in the International Spondyloarthritis Interobserver Reliability Exercise (INSPIRE) trial, although it showed lower ICC values than the Leeds and SPARCC instruments in PsA patients (16). This instrument has not been used in PsA trials.

Table 2. Enthesial sites assessed in outcome measures for enthesitis (16)*
 MASESMajor (Berlin)SPARCCSan FranciscoPEST (Leeds)4 point
  • *

    MASES = Maastricht Ankylosing Spondylitis Enthesis Score; SPARCC = Spondyloarthritis Research Consortium of Canada; PEST = Psoriasis Epidemiology Screening Tool; X = single site present, not bilateral; R = right; L = left.

C1/C2   X  
C7/T1   X  
T12/L1   X  
First costochondralR, L     
Seventh costochondralR, L     
Supraspinatus insertion  R, L   
Lateral epicondyle humerus  R, L R, L 
Medial epicondyle humerus  R, L   
Posterior superior iliac spineR, L     
Anterior superior iliac spineR, L  R, L  
Iliac crestR, LR, L    
Fifth lumbar spinous processX  X  
Ischial tuberosity   R, L  
Proximal AchillesR, LR, LR, LR, LR, LR, L
Greater trochanter R, LR, LR, L  
Medial condyle femur R, L  R, L 
Lateral condyle femur R, L    
Insertion plantar fascia R, LR, LR, L R, L
Quadriceps insertion patella  R, L   
Inferior pole patella  R, L   
Tibial tubercle  R, L   

San Francisco

This is a 14-site enthesitis index (41) (Table 2) employed in trials of AS that, like the Major, showed lower ICCs in PsA patients in the INSPIRE study and has not been used in PsA trials (16).

4 Point

The 4 point enthesitis measure includes both Achilles tendon and plantar fascia insertions and may be graded as present or absent or scored on a 0–3 scale of severity. This measure has shown discrimination and responsiveness in 2 infliximab trials in PsA and the etanercept trial known as Psoriasis Randomized Etanercept Study in Subjects with Psoriatic Arthritis (5, 8, 9); however, it did not perform as well as measures with a greater number of sites in a comparative study (16).

DACTYLITIS

Dactylitis, or “sausage digit,” is characterized by swelling of an entire finger due to synovitis, tenosynovitis, enthesitis, and soft tissue edema (12, 13, 42). Its presence helps distinguish psoriatic arthritis (PsA) from other forms of arthritis, and is found in 16–48% of reported PsA cases (43). In PsA clinical trials over the past decade (5, 8, 9, 44), dactylitis has been assessed by having the investigator examine each finger and determine if it is swollen or not (12, 13, 45). On occasion, a severity score of 0–3 (where 0 = no swelling or pain and 3 = severe swelling and pain) has been utilized as well. Both the methods of simple count and scoring have demonstrated responsiveness and discrimination in anti–tumor necrosis factor trials (5, 8, 9). Recently, a more quantitative dactylitis measure, the Leeds Dactylitis Index (LDI), has been developed and is described below.

LEEDS DACTYLITIS INDEX (LDI)

Description

Purpose.

To quantitatively measure dactylitis. The LDI, using a measurement instrument known as a dactylometer, has been recently developed but not yet used in a randomized controlled trial (RCT) (43). In this method, circumference of the affected fingers, circumference of contralateral fingers, and tenderness of affected fingers are all assessed for a total score.

Content.

Evaluation of finger size and pain to assess for the presence of dactylitis.

Number of items.

Evaluation of each of 20 fingers for size and tenderness.

Response options/scale.

See below.

Recall period for items.

Current.

Examples of use.

Currently being used in PsA RCTs.

Practical Application

How to obtain.

The measure is available in the original article (43). The tool developed for measuring digital circumference is available online at www.mie-uk.com.

Method of administration.

The clinician marks which fingers are affected on a diagram displaying fingers and toes. Circumferences of the affected and contralateral fingers are then measured around the proximal phalanx, as close as possible to the web space, using either a measuring tape or a precalibrated loop. The clinician then squeezes the affected fingers with moderate pressure and documents the patient's response: 0 = no tenderness, 1 = tender, 2 = tender and winces, and 3 = tender and withdraws (43).

Scoring.

The ratio of circumference between an affected finger and the contralateral unaffected finger is recorded. If both sides are affected, the circumference of the affected finger is compared to normative data supplied in a table. The tenderness score (0–3) for a finger with dactylitis is recorded, and a total score is generated for each finger. If multiple fingers are affected, each score is added together to produce a total for the patient (38). A difference in digital circumference of ≥10% is used to define a finger with dactylitis.

Score interpretation.

A higher score is associated with worse dactylitis (43).

Respondent burden.

Minimal burden to the patient; no discomfort.

Administrative burden.

Assessment and scoring time by the clinician depends on the number of affected fingers; however, total administration should take <10 minutes (43).

Translations/adaptations.

A later modification of the LDI (referred to as the LDI basic) replaced the original tenderness grading (0–3) with a binary score reflecting either the presence or absence of tenderness (1 or 0, respectively) (45).

Psychometric Information

Method of development.

Developed in a patient study specific for the development of this measure (43).

Acceptability.

Acceptable.

Reliability.

In a study involving 20 rheumatologists expert in PsA and ankylosing spondylitis who examined 20 patients, overall interobserver scores indicated a strong agreement, with an intraclass correlation coefficient of 0.70 (95% confidence interval 0.49, 0.89) (16).

Validity.

Discrimination capability has not yet been assessed and the measure has not been validated.

Ability to detect change.

Responsiveness to change has been demonstrated in a small open study of treated PsA patients (45).

Critical Appraisal of Overall Value to the Rheumatology Community

The strength of this method is its quantitative objectivity. A weakness is that it takes more time to perform than an observational count. When matching fingers are involved, reference values gathered by Helliwell and colleagues in 2005 are used to calculate the dactylitis score. While these values are grouped by sex, they do not control for other variables such as age or body mass index (43).

SPINE ASSESSMENT

The prevalence and impact of spine disease in psoriatic arthritis (PsA) has not been as well characterized as more peripheral manifestations of the disease. Classified as a spondylarthritis based on overlapping clinical and pathologic features with ankylosing spondylitis (AS), spinal inflammation does occur in up to one-half of PsA patients. Manifestations include sacroiliitis, spinal joint and enthesial inflammation, and ankylosis in the form of bridging syndesmophytes. The spinal manifestations of PsA tend to be less severe than those seen in AS. Because spine involvement tends to be mild and inconstant, it has not been systematically assessed in clinical trials of PsA. Rather, it appears that investigators, clinicians, and regulators tend to adopt the research findings of AS and extrapolate to PsA in the spine in the absence of specific clinical trial data from PsA patients.

Commonly used measures in AS trials include the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) (46), Bath Ankylosing Spondylitis Function Index (BASFI) (47), and Bath Ankylosing Spondylitis Metrology Index (48). The BASDAI is a set of 6 visual analog scale (VAS) patient questionnaires regarding fatigue, pain, and stiffness. A study of the BASDAI in PsA showed a high correlation with a VAS of overall arthritis activity, but the correlation was similar in patients with a greater amount of axial disease as compared to patients with a greater amount of peripheral disease (49). Also, it did not correlate with the physician's perception of disease activity or with treatment decisions. Therefore, the BASDAI did not discriminate between axial and peripheral disease activity. Similarly, the BASFI demonstrated a high correlation with other measures of function, such as the Health Assessment Questionnaire (HAQ) disability index (DI) and Short Form 36 (SF-36), but did not discriminate between patients with predominantly axial versus peripheral disease, and therefore offers no advantage over the commonly used HAQ DI and SF-36 (50).

In a study of 10 patients with AS and 10 patients with PsA spondylitis, 10 experts in AS and 10 experts in PsA performed AS spine examinations and found there to be good performance characteristics and interobserver reliability (intraclass correlation coefficient 0.89) using standard AS metrology in both AS and PsA spondylitis patients (51). In a similar exercise, spine measurement techniques, in particular the modified Schober test, lumbar side flexion, and cervical rotation, compared favorably between patients with PsA spondylitis and AS (52).

The recently developed Ankylosing Spondylitis Disease Activity Score (53) was compared to the BASDAI in patients with PsA spondylitis. Both showed good correlation with disease activity measured by patient and physician global and both performed comparably (54). It appears that if measures of axial disease developed for AS are used in trials of PsA patients with axial involvement, the measures will be reasonably reliable, responsive, and discriminative.

PATIENT GLOBAL ASSESSMENT

The intent of the patient global assessment is to encompass not only specific disease severity and multidimensional impact, but also take into account the impact of such issues as treatment side effects. Typically, when a patient is asked to judge their global status on a visual analog scale (VAS) or Likert scale, it is in relation to the following question: “In all of the ways that your [insert disease name] affects you [today, or over the past week], how are you?” American College of Rheumatology (ACR) criteria to assess rheumatoid arthritis (RA) (55) express that the patient global assessment of disease activity, by VAS score, should be answered in response to the following statement: “Considering all the ways your arthritis affects you, please mark a vertical line on the scale below to show how you are feeling today.” Zero is considered to be “very good, no symptoms” and 100 is “very poor, severe symptoms.” If responses are offered on a Likert scale, the same question regarding the effect of arthritis today is asked and the patient circles a number ranging from 1, denoting “very good, no symptoms and no limitation of normal activities,” to 5, denoting “very poor, very severe symptoms which are intolerable and inability to carry out normal activities.”

In psoriatic arthritis (PsA), the patient (and the clinician asking the question) may not understand if they should focus their thinking on joint disease, all aspects of musculoskeletal disease, skin and nail disease, or a composite of these.

Illustrating the conundrum about how patient global should be assessed in a PsA trial, 1 recent PsA trial protocol included the above sentence to describe the patient global, but with the instruction that the subject should be asked to consider both joint and skin components in their response to this statement (Mease PJ: unpublished observations), even though asking the question in this way has not been formally validated in the ACR or Disease Activity Score (DAS) criteria. In most PsA protocols, there has been no such clarification and it has been left up to the interpretation of the patient and/or investigator/coordinator.

To address this issue, the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) organized a multicenter study in which patients were asked, in variable order, to rate on a VAS scale their assessment of psoriasis and arthritis (PGA), arthritis alone (PJA), and psoriasis alone (PSA) over the past week (56). For example, the PGA question was, “In all the ways that your psoriasis and arthritis, as a whole, affect you, how would you rate the way you felt over the past week?”

PATIENT GLOBAL FOR PSORIATIC ARTHRITIS

Description

Purpose.

To adequately capture the patient global experience of PsA, taking into consideration the separate domains of musculoskeletal and skin disease.

Content.

Questions asked of the patient about patient global experience.

Number of items.

3 self-reported questions.

Response options/scale.

The patient rates global state related to arthritis, skin, and a combination of these on VAS or numerical rating scales.

Recall period for items.

Past week.

Endorsements.

Endorsed by GRAPPA.

Practical Application

How to obtain.

From the original article (56).

Method of administration.

Self-administered.

Scoring.

In addition to being scored on a VAS, the patient global assessment of arthritis has been scored on a Likert scale.

Score interpretation.

The patient global assessment is used in the ACR, DAS, and Psoriatic Arthritis Response Criteria scoring systems (see below), as well as emerging composite scoring systems.

Respondent burden.

Minimal.

Administrative burden.

Minimal.

Psychometric Information

Method of development.

Evaluation of 319 patients seen in globally distributed clinics of GRAPPA members.

Acceptability.

Acceptable.

Reliability.

All 3 measures, i.e., the PGA, PJA, and PSA, demonstrated good test–retest reliability with intraclass correlation coefficients 0.87 (95% confidence interval [95% CI] 0.83, 0.90), 0.86 (95% CI 0.81, 0.89), and 0.78 (95% CI 0.72, 0.83), respectively. Analysis of the GRAPPA study responses found that the PJA had a B coefficient of 0.63, while the PSA had a B coefficient of 0.30. The regression coefficient B quantifies the effect of each exposure variable by expressing the increase of the outcome variables produced by a unitary increment of the exposure variable (56). This disparity indicates that in general, the study patients considered the arthritis component of their disease to be a greater problem than the psoriasis component.

Validity.

Analysis of the recent GRAPPA study responses shows a statistically significant correlation between higher values on the patient global VAS and the number of joints involved (56).

Ability to detect change.

To be determined based on results of randomized controlled trials (RCTs) currently being conducted. Historically, the PJA, as used in RA, has been responsive and able to discriminate treatment from placebo.

Critical Appraisal of Overall Value to the Rheumatology Community

The recent GRAPPA study found that although the single question addressing both joint and skin disease is a reliable and responsive measure, that since some patients had divergent severity of joint and skin disease, the variable impact of the 2 major aspects of the disease could be better understood by asking all 3 questions, if feasible, in an RCT.

Since the ACR and DAS scoring systems allow for only 1 entry of patient global in the score, based on the study recently published by the GRAPPA group described above (56), it would appear reasonable to use either the dual question (PGA) or the question related to arthritis alone (PJA).

PATIENT PAIN ASSESSMENT

Assessment of pain related to arthritis is part of the core set of measures for the American College of Rheumatology score (55). Patient pain is assessed by means of a 0–100 visual analog scale (VAS), where 0 = “no pain” and 100 = “most severe pain” that day. The minimum clinically important difference of pain VAS is considered to be 10 mm (57).

PHYSICIAN'S GLOBAL ASSESSMENT OF DISEASE ACTIVITY

The physician's global assessment of disease activity is scored in an identical manner to the patient global assessment of disease activity described above, both in visual analog scale format for the American College of Rheumatology scoring system and the Likert scale format for the Psoriatic Arthritis Response Criteria. The same problem regarding whether one focuses on joint or skin activity or both applies, and in recent psoriatic arthritis study protocols, the clinician has been instructed to take into account both the severity of joint and skin disease when making the assessment (Mease PJ: unpublished observations).

PHYSICAL FUNCTION

Physical function has been reliably assessed in psoriatic arthritis (PsA) trials by the Health Assessment Questionnaire (HAQ; see below) (58). This measure contains 20 items divided into 8 domains: dressing and grooming, arising, eating, walking, hygiene, reach, grip, and common daily activities. Subjects rate the degree of difficulty they have had in the past week on a 4-point scale, ranging from 0 (no difficulty) to 3 (unable to do). The highest scores in each category are summed (0–24) and divided by the number of categories scored to yield a score from 0–3. The HAQ instrument has been most extensively utilized and validated in rheumatoid arthritis (RA). In an effort to modify this instrument to be more specific for spondylarthritides in general and PsA specifically, the HAQ for spondylarthritis (59) and the expanded HAQ (60) were developed with questions more specific for spondylarthritis and psoriasis patients. These modifications have not been found to be psychometrically superior to the original HAQ, and so have not been further used in PsA clinical trials. In a large cohort of patients in the Toronto PsA registry, the number of inflamed joints (reflecting disease activity) and deformed joints (reflecting damage) were significantly related to the HAQ score (61). The effect of disease activity on the HAQ declined with duration of disease activity, as has been noted in RA, but there was less evidence that the effect of joint damage on HAQ increased over time. A Rasch analysis showed that the Short Form 36 (SF-36) physical functioning (PF) subscale and the HAQ measure the same physical disability construct, but the SF-36 PF subscale has better distributional properties, scale length, fewer misfitting items, and less differential item functioning (62).

The minimum important difference (MID) of the HAQ, determined using an anchor-based method, calculated using the patient-rated importance of change in a trial of etanercept in PsA (63), was 0.35 (64). Calculations based on distribution-based methods are thought to be a more appropriate approach to analysis of MID than that based on patient-rated satisfaction with change, noted to be 0.3 in a preliminary abstract from this same study (65). In contrast, a study from a single center, using methods based on an overall health status anchor, found the MID to be 0.131 (66). The difference may have been partly accounted for by the difference in disability of patients in the 2 studies, with an average HAQ score of 1.16 in the former study and 0.732 in the latter, as well as trial methodology. Both of the figures derived from these 2 studies contrast with the minimal clinically important difference, a synonymous term for MID, established for RA of 0.22 (67).

In addition to the HAQ and the SF-36 PF subscale, the Disabilities of Arm, Shoulder, and Hand questionnaire has been validated for assessment of upper extremity function and inflammatory disease activity in PsA. The Arthritis Impact Measurement Scales (AIMS) and a revised form of this instrument (AIMS2) have been validated as measures of function in PsA in a PsA registry, but these measures have not been used in PsA clinical trials (68, 69).

HEALTH-RELATED QUALITY OF LIFE

The most commonly used measure of quality of life (QOL) employed in psoriatic arthritis (PsA) trials is the generic QOL instrument, the Medical Outcomes Study Short Form 36 (SF-36) (70). Other commonly used measures have included the Dermatology Life Quality Index (DLQI) (71) and the EuroQol 5-domain (EQ-5D) (72). A PsA-specific instrument, the Psoriatic Arthritis Quality of Life (PsAQOL) measure, has been developed and validated.

The SF-36 is a patient questionnaire assessing 8 domains of health status: physical functioning (PF), pain, vitality, social functioning, psychological functioning, general health perceptions, and role limitations due to physical and emotional problems (70). It also can be subdivided into 2 summary scores, the physical and mental component scores. This instrument has been validated in PsA (73). All domains of the SF-36 demonstrated internal consistency reliability, with Cronbach's α exceeding 0.8. Discriminant validity was demonstrated in that the adjusted SF-36 scale scores for patients with PsA were lower than age- and sex-matched controls. Discriminant and convergent validity were demonstrated by correlation between SF-36 domains and clinical measures of function and pain and measures of disease activity and severity. A more recent study used Rasch analysis to compare the SF-36 and Health Assessment Questionnaire (HAQ) in PsA and rheumatoid arthritis (RA) (62). The Rasch model fit the SF-36 PF subscale scores better than the HAQ with good item separation in both PsA and RA and minimal floor effects in either group. Rasch analysis of the HAQ demonstrated a better span of the HAQ in RA than in PsA, and there were significant floor effects in PsA, with 30% indicating no disability and only 7% of the RA group indicating no disability. The HAQ and SF-36 PF subscale measure the same physical disability construct, and the SF-36 PF subscale has better distributional properties and scale length, fewer misfit items, and less differential item functioning than the HAQ. Another recent study assessed the scaling assumptions, internal reliability, and construct validity of the SF-36 in PsA in an Asian population in Hong Kong, further validating it (62). As previously described, the SF-36 physical component summary score demonstrated superior psychometric properties compared to the HAQ as well as to the Bath Ankylosing Spondylitis Functional Index and Dougados Functional Index (50).

DERMATOLOGY LIFE QUALITY INDEX (DLQI)

Description

Purpose.

The DLQI was developed to measure the disability experienced by patients with different dermatologic conditions. The DLQI is a 10-item questionnaire used in 33 different dermatologic conditions (71). It has been validated in assessment of psoriasis and shows discrimination and responsiveness in PsA trials (5, 8, 9). It is often used in conjunction with the SF-36 to provide more specific assessment of the impact of skin disease.

Content.

Questions assess the effect of a dermatologic condition on a patient's QOL, including impact on work, leisure activities, personal relationships, feelings of embarrassment, etc.

Number of items.

10 items.

Response options/scale.

Answers are based on a 4-point Likert scale. Responses of “not at all,” “a little,” “a lot,” and “very much” are available for each question, and correspond to scores of 0, 1, 2, and 3, respectively. A response of “not relevant” is also offered for select questions (71).

Recall period for items.

7 days.

Examples of use.

Widely used in psoriasis and PsA clinical trials.

Practical Application

How to obtain.

Measure is available from the original article (71).

Method of administration.

Can be self-administered on paper or electronically.

Scoring.

0–3 for each question, yielding a total possible score of 0–30.

Score interpretation.

Higher scores represent a greater effect on QOL.

Respondent burden.

Minimal.

Administrative burden.

Minimal.

Psychometric Information

Method of development.

The DLQI was developed through a study of 120 dermatology patients presenting with a variety of skin conditions. Patients were asked questions about their skin disease and its impact on their lives, and a 10-item questionnaire was formulated based in their answers.

Acceptability.

Acceptable.

Reliability.

Reliability has been demonstrated using a 1-week test–retest method in 53 patients (71). Correlations between individual question scores were very high (γs = 0.95–0.98, P < 0.001), and the correlation between overall DLQI scores was also high (γs = 0.99, P < 0.0001).

Validity.

Construct validity has been demonstrated in a comparison of scores of 200 dermatology outpatients compared to 100 control subjects. In a separate study, the DLQI was cross-validated against other QOL measures such as the SF-36 (74).

Ability to detect change.

The DLQI has demonstrated sensitivity to change in 181 patients following inpatient treatment for their dermatologic conditions (75). It is used widely in psoriasis and PsA clinical trials and shows significant responsiveness.

Critical Appraisal of Overall Value to the Rheumatology Community

The DLQI is often included in PsA randomized controlled trials (RCTs), along with the HAQ, SF-36, and currently, the PsAQOL measure, partly as a measure that dermatologists can relate to when evaluating QOL effects of therapies. It is one of the measures that may be used to assess disease severity when utilizing the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis treatment recommendations grid (76).

EQ-5D

The EQ-5D is comprised of a 5-item set of health status measures and a visual analog scale (VAS), with each of the 5 health states (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression) evaluated from “no problem” to “extreme problem,” scored from 1–3 (72). The VAS is rated from 0 = worst imaginable health status to 100 = best imaginable health status. The EQ-5D has shown discrimination and responsiveness in PsA trials (5, 8, 9).

PSORIATIC ARTHRITIS QUALITY OF LIFE (PSAQOL)

Description

Purpose.

PsA-specific health-related QOL instrument. The PsAQOL, derived from PsA patient interviews and evaluations, has shown reliability and construct validity (77). It is now being employed in RCTs (Mease PJ: unpublished observations) to further assess discrimination and responsiveness.

Number of items.

20 items.

Response options/scale.

Questions are answered as “true” or “not true.”

Recall period for items.

Current.

Examples of use.

Currently being employed in PsA RCTs.

Practical Application

How to obtain.

Contact Stephen McKenna (e-mail: smckenna@galen-research.com).

Method of administration.

Self-administered on paper or electronically.

Scoring.

Each “true” response is 1 point on a 20-point scale, for a possible total score of 0–20.

Score interpretation.

Higher scores indicate worse health-related QOL.

Respondent burden.

1–2 minutes to complete.

Administrative burden.

Minimal.

Psychometric Information

Method of development.

Qualitative interviews were conducted with 48 PsA patients, from which a 51-item questionnaire was generated. Followup surveys of 94 patients reduced the number of relevant questions to 35, and a Rasch analysis of an additional 286 patient surveys identified 20 meaningful items to include in the final PsAQOL questionnaire (77).

Acceptability.

Acceptable.

Reliability.

The PsAQOL has demonstrated excellent test–retest reliability (Spearman's rank correlation coefficient = 0.89) (77).

Validity.

Strong correlations have been found between the PsAQOL and other comparable measures (77). Construct validity of the PsAQOL has also been confirmed in a longitudinal study of 28 patients over 6 months (78).

Ability to detect change.

Responsiveness has been demonstrated at both 3-month and 6-month time points after treatment initiation in PsA patients (78).

Critical Appraisal of Overall Value to the Rheumatology Community

The PsAQOL has been rigorously developed and is specific for patients with PsA. It is now being used for the first time in RCTs, from which more will be learned about its performance characteristics.

In a series of PsA patient focus groups, transcribed texts of the discussions were divided into meaning units from which concepts were extracted and mapped using the International Classification of Functioning, Disability and Health as a frame of reference (79). Multiple commonly used measures of function and QOL, including those described herein, demonstrated significant gaps in coverage of key concepts important to patients in any one instrument, suggesting the need to use several instruments to more adequately address these concepts, as well as the need for more research on development of more comprehensive instruments to measure function and QOL.

FATIGUE

Fatigue has increasingly been recognized as an important clinical dimension by patients with rheumatic diseases (80). Fatigue has been determined to be an important clinical domain in psoriatic arthritis (PsA), independent of and not fully explained by other domains such as pain, tender and swollen joint count, patient global, and function (81). Several scales have been used to assess fatigue in rheumatic diseases, including the Multidimensional Assessment of Fatigue scale (82), the Multidimensional Fatigue Index (83), the Fatigue Severity Scale (FSS) (84), the Functional Assessment of Chronic Illness Therapy–Fatigue (FACIT-F) scale (85), the Brief Fatigue Inventory (86), the vitality scale of the Short Form 36 (70), and the visual analog scale for fatigue (87). A modified version of the FSS has been validated in PsA (88). Changes in fatigue were correlated with changes in disease activity using this instrument (89). The FACIT-F was responsive to change in a trial of adalimumab in PsA (90). The FACIT-F was validated in a Toronto PsA cohort and correlated well with the modified FSS, showing high internal consistency, test–retest reliability, and criterion and construct validity (91).

FUNCTIONAL ASSESSMENT OF CHRONIC ILLNESS THERAPY–FATIGUE (FACIT-F)

Description

Purpose.

The FACIT measurement system was originally developed to assess health-related quality of life in patients with chronic illnesses (92, 93). The additional questions of the FACIT-F survey were compiled to assess anemia-related fatigue (94).

Number of items.

13 items.

Response options/scale.

Answers are based on a 5-point Likert scale. Responses of “not at all,” “a little,” “somewhat,” “quite a bit,” and “very much” are available for each question, and correspond to scores of 0, 1, 2, 3, and 4, respectively.

Recall period for items.

7 days.

Practical Application

How to obtain.

The questionnaire can be found online at http://www.facit.org.

Method of administration.

Self-administered on paper or electronically.

Scoring.

Each question scores between 0 and 4, with a total score range from 0–52.

Respondent burden.

Minimal.

Administrative burden.

Minimal.

Psychometric Information

Method of development.

Each FACIT measure is developed through interview-based item generation and survey-based item reduction (93).

Acceptability.

Acceptable.

Reliability.

High test–retest reliability of the FACIT-F was demonstrated in a Toronto study of 135 patients with PsA. FACIT-F surveys given to patients 1 week apart yielded an intraclass correlation coefficient of 0.95 (91).

Validity.

The measure was validated in the same Toronto PsA cohort. It correlated well with the modified FSS (−0.79; 95% confidence interval −0.85, −0.72) and showed high internal consistency (Cronbach's α = 0.96), as well as criterion and construct validity (91).

Ability to detect change.

The FACIT-F was responsive to change in a trial of adalimumab in PsA (90).

Critical Appraisal of Overall Value to the Rheumatology Community

The domain of fatigue, as mentioned, is increasingly recognized as an important domain to assess in patients with inflammatory arthritis conditions. No single measure has emerged as a favored instrument, although several, as described, have been shown to be reliable, responsive, and discriminative. The area of fatigue assessment is still in evolution in PsA.

PATIENT-REPORTED OUTCOMES MEASUREMENT INFORMATION SYSTEM (PROMIS)

In 2004, the National Institutes of Health Roadmap initiative PROMIS was launched to develop a new generation of patient-reported outcome measures for chronic illness facilitated by the methods of item-response theory and computer adaptive testing. PROMIS is intended to be a publicly available updateable repository of well-calibrated items facilitating the assessment of numerous clinically relevant domains across disease states with precision and minimal patient burden (95). The intent is for patients to be able to self-report about disease activity, function, and quality of life using online questionnaires, centrally and securely stored electronically. This is a public–private partnership intended to facilitate disease state registries and clinical trials. It is anticipated that this methodology will be used in the assessment of psoriatic arthritis as well as other rheumatic diseases, hence the description in this article. The PROMIS web site is www.nihpromis.org.

IMAGING

Although it is beyond the scope of this article to describe in detail imaging assessments in psoriatic arthritis (PsA), providing a few words about the approach to imaging in PsA trials and steering the reader to reviews on this subject are appropriate. Radiographic imaging is used for diagnostic purposes and to assess joint damage. Ultrasound (US) and magnetic resonance imaging (MRI) can be used for these same purposes, and to assess for the presence of inflammation through detection of soft tissue changes such as synovitis, enthesitis, excess fluid, and increased blood flow (US power Doppler). Additionally, MRI can detect inflammatory changes in bone by detecting bone edema.

In terms of radiographic assessment, the most commonly employed methodology used in PsA clinical trials is the Sharp/van der Heijde method (hands, wrists, and feet) modified for PsA to include the distal interphalangeal joints. A detailed review of this and other radiographic methods used in PsA is provided in a review article by Mease and van der Heijde (96) and other reviews (97). This method has shown excellent response and discrimination characteristics in PsA clinical trials (5, 8, 9, 44).

Although not conducted in PsA clinical trials as of yet, spine radiographic methods used in ankylosing spondylitis, the Bath Ankylosing Spondylitis Radiology Index and the modified Stoke Ankylosing Spondylitis Spine Score, have been validated in PsA (98), and a modification of these scales has been proposed for PsA (99).

Portable US technology is increasingly used in clinical practice for both diagnostic purposes (detecting synovitis, joint damage, enthesitis, and dactylitis) as well as for guiding therapeutic injection (100, 101). It is being used in an exploratory fashion in clinical trials. US scoring methods for assessing joints, e.g., synovitis and erosions, have been developed for rheumatoid arthritis (RA) and are being explored in PsA (102). In addition, scoring methods for enthesitis are being assessed in PsA (103, 104). These efforts are being coordinated through the Outcome Measures in Rheumatology (OMERACT) Ultrasound Working Group.

MRI is increasingly being used as a sensitive imaging technology to detect structural damage and to detect inflammatory change of soft tissue and bone. It is useful both for peripheral pathology (arthritis, enthesitis, and dactylitis) and is particularly useful to assess inflammation and damage in the spine and sacroiliac joints where US is not accurate. An OMERACT MRI working group has developed a scoring system for RA, the Rheumatoid Arthritis Magnetic Resonance Imaging Scoring (RAMRIS) system (105), which is now being adapted for PsA (106). The RAMRIS system was employed and showed responsiveness and discrimination in a trial of abatacept in PsA (107).

COMPOSITE MEASURES OF MULTIPLE DOMAINS IN PSORIATIC ARTHRITIS

Composite response measures, which focus on joint disease as well as patient global/pain, clinician global, and acute-phase reactant (American College of Rheumatology [ACR] response criteria); patient global ± acute-phase reactant (Disease Activity Score [DAS], DAS in 28 joints [DAS28]); and patient/clinician global (Psoriatic Arthritis Response Criteria [PsARC]), have shown reliable discriminant and response characteristics in randomized controlled trials (RCTs), but have not been formally validated in psoriatic arthritis (PsA) (5, 12–14). In the past decade, the ACR 20% response criteria (ACR20) have typically been employed as the primary outcome measure of PsA RCTs, and the ACR50 and ACR70, DAS or DAS28, and PsARC have been secondary measures. The ACR and DAS criteria are described in detail in the rheumatoid arthritis section of this article.

Recognizing that PsA is a complex disease that not only involves the domains noted above, but also enthesitis, dactylitis, spine, and skin and nail disease, several groups, including the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis and Outcome Measures in Rheumatology, are working on development of composite measures of disease severity and response to therapy that take into account most, if not all, of these domains. The work on development of these measures is discussed below.

PSORIATIC ARTHRITIS RESPONSE CRITERIA (PSARC)

Description

Purpose.

The PsARC was developed as a PsA-specific composite responder index specifically for a study of sulfasalazine in PsA (108). It was first named the PsARC in a subsequent trial of etanercept in PsA (63).

Content.

Tender and swollen joint count and patient and physician global assessment.

Number of items.

4 items.

Response options/scale.

To achieve response, a patient has to achieve 2 of the following, one of which has to be a tender (68) and swollen (66) joint count, and no worsening of any measure: tender or swollen joint count improvement of ≥30% and/or patient global or physician global improvement of at least 1 point on a 5-point Likert scale.

Recall period for items.

At the time of evaluation.

Endorsements.

Recommended for use by the European regulatory agency, the European Medicines Agency. However, since there is a tendency for a higher placebo response with this measure and it has lower performance characteristics than the ACR criteria (14), it is not used as the primary outcome measure in PsA RCTs but rather is used as a secondary measure.

Examples of use.

It has been used in most PsA clinical trials, as mentioned, as a secondary measure: Mease PJ. Psoriatic arthritis: update on pathophysiology, assessment and management. Ann Rheum Dis 2011;70 Suppl:i77–84 (5).

Mease PJ, Antoni CE. Psoriatic arthritis treatment: biological response modifiers. Ann Rheum Dis 2005;64 Suppl:ii78–82 (8).

Mease PJ. Psoriatic arthritis: pharmacotherapy update. Curr Rheumatol Rep 2010;12:272–80 (9).

Practical Application

How to obtain.

Simple scale as described above; no permission necessary.

Method of administration.

Physical examination of joints scored on paper, as a Likert scale or patient and physician global.

Scoring.

Tender joint count is assessed in 68 joints and swollen joint count in 66. Earlier studies have also employed higher joint counts in PsA, but the current standard is use of the 68/66 joint count. Patient global Likert scale (5 points) is completed in response to the question: “Considering all the ways your arthritis affects you, how are you feeling today?” On the Likert scale, 1 = “very good, no symptoms, and no limitation of normal activity” and 5 = “very poor, very severe symptoms which are intolerable, and inability to carry out normal activities.” Physician global assessment involves the same question asked about the patient and assessment on a similarly scored Likert scale.

Score interpretation.

Response is achieved if at least 2 of the 4 items are achieved: tender and/or swollen joint count has improved by at least 30% (at least one of these required) and/or patient or physician global has improved by at least 1 point, and no item has worsened.

Respondent burden.

Minimal.

Administrative burden.

Minimal; takes ∼2 minutes if someone to record is present, or slightly longer if not.

Translations/adaptations.

A modification of the PsARC using patient and physician global improvement measured by a visual analog scale has been employed and considered acceptable by regulatory agencies (90).

Psychometric Information

Method of development.

Derived from expert opinion.

Acceptability.

Acceptable.

Reliability.

Not assessed.

Validity.

The discrimination capability of the PsARC has been evaluated. In a trial of infliximab in PsA, the response rate in the active treatment arm was 82% and in placebo was 30%, yielding a χ2 value of 27.9 (14). In an etanercept trial, the similar figures were 90% and 33%, respectively, with a χ2 value of 19.3. The PsARC has not been otherwise formally validated.

Ability to detect change.

The ability of individual elements of the PsARC to detect change has been determined. The responsiveness of the quantitative tender and swollen joint count is summarized above. The patient global assessment Likert scale in an etanercept PsA trial demonstrated a standardized response mean (SRM) of −1.51 in the treatment group and 0.48 in placebo, with a t value of 5.8 (14). The physician global assessment Likert scale showed SRMs of −1.98 and −0.34, respectively, for a t value of 7.0 (14).

Critical Appraisal of Overall Value to the Rheumatology Community

Although specifically developed for PsA, the PsARC is limited in that it does not include domains such as enthesitis, dactylitis, or skin disease assessment; tends to display a high placebo response rate; and is primarily used as a secondary measure in clinical trials.

EMERGING COMPOSITE MEASURES: PSORIATIC ARTHRITIS JOINT ACTIVITY INDEX (PSAJAI)

Method of development.

The PsAJAI was developed in a project in which data from 3 trials of anti–tumor necrosis factor (anti-TNF) agents in psoriatic arthritis (PsA) were analyzed to create models, based primarily on statistical considerations and some clinical input, which best distinguished active drug from placebo (109, 110). Note that in this analysis, addition of the Psoriasis Area and Severity Index (PASI) was problematic in that not all patients in these trials could be assessed for the PASI given low skin scores. Anti-TNF therapy had a large impact on the PASI score; therefore, it was recommended that skin be scored separately. From the same data, response criteria currently used for PsA were examined and logistic regression models based on the individual components of these response criteria were analyzed. The PsAJAI, modeled as the American College of Rheumatology 30% response criteria (ACR30), performed better than the ACR20 and Psoriatic Arthritis Response Criteria (PsARC), and was comparable to previously developed models.

Content.

The PsAJAI is the weighted sum of 30% improvement in 6 measures with weights of 2 given to tender joint count, C-reactive protein (CRP) level, and physician global assessment of disease activity. Weights of 1 are given to the remaining 30% improvement measures, including pain, patient global assessment of disease activity, and Health Assessment Questionnaire (110).

Number of items.

6 items.

Recall period for items.

Current.

Examples of use.

The PsAJAI has not yet been use in a PsA randomized controlled trial (RCT).

Practical Application

How to obtain.

Obtainable from the original articles (106, 107).

Method of administration.

Patient questionnaires and physical examination, recorded on paper or electronically.

Acceptability.

Not assessed as of yet.

Reliability.

Not determined.

Validity.

Not assessed.

Ability to detect change.

When applied to a different RCT data set than the 3 from which it was derived, a response rate of 75.6% was noted, similar to a PsARC response of 70% and an ACR20 response of 78% (111).

DISEASE ACTIVITY IN PSORIATIC ARTHRITIS (DAPSA)

Method of development.

A Viennese group collected cross-sectional clinical and laboratory data on 105 patients with PsA and performed principal component analysis on those clinical and laboratory variables recommended by the Outcome Measures in Rheumatology (OMERACT) module (112) (Figure 1). Four principal components were derived: patient global and pain visual analog scale (VAS) scores, tender and swollen joint counts, acute-phase reactant (CRP level), and skin, although the latter did not reach statistical significance. The group then studied the existing composite measures and determined that these domains were best served by using the disease activity index for the assessment of reactive arthritis (113). Further, it was determined that the 68/66 joint count outperformed the 28 joint count. This measure was renamed the DAPSA score.

Content.

Swollen joint count, tender joint count, patient global, pain, and CRP level.

Number of items.

5 items.

Recall period for items.

Current.

Practical Application

How to obtain.

Derived from the original article (108).

Method of administration.

Patient VAS, physical examination, and laboratory, recorded on paper or electronically.

Scoring.

Sum of patient global and pain VAS in centimeters, numerical swollen and tender joint count of 66 and 68 joints, respectively, and CRP level in mg/dl.

Score interpretation.

Higher scores reflect more severe disease activity.

Respondent burden.

Minimal.

Administrative burden.

Takes <5 minutes to obtain.

Psychometric Information

Acceptability.

Acceptable.

Reliability.

Not yet assessed, although components are commonly used in RCTs and have performed reliably in similar measures.

Validity.

The DAPSA was tested in PsA patients starting a new disease-modifying antirheumatic drug treatment (n = 99) and in the data set of a phase III trial of infliximab (Infliximab Multinational Psoriatic Arthritis Controlled Trial 2) (114, 115). The instrument correlated highly with other measures, including the Disease Activity Score in 28 joints (DAS28), Simplified Disease Activity Index, and Clinical Disease Activity Index.

Ability to detect change.

Effect sizes were high (>0.8) for the active treatment arm and low for placebo in the retrospective analysis of the infliximab data set (P = 2.56 × 10−10), suggesting high discriminant capability (114).

COMPOSITE PSORIATIC DISEASE ACTIVITY INDEX (CPDAI)

Method of development.

The CPDAI is based on a grid proposed by the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) to guide treatment decisions in PsA (76, 116).

Content.

Disease involvement is assessed in up to 5 domains: peripheral joints, skin, enthesial, dactylitis, and spinal manifestations. For each domain, individual instruments are used to assess the extent of disease activity as well as the impact on patient function and health-related quality of life.

How to obtain.

Available from the original article (117).

Method of administration.

Measures used are patient self-administered, physical examination, and laboratory, recorded on paper or electronically.

Scoring.

Domains are scored from 0–3, with empirical cutoffs for disease severity/activity proposed in each largely based on the literature (Table 3). Individual domain scores are summed to give an overall composite score (range 0–15).

Table 3. Modification of the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis grid proposed for the Composite Psoriatic Disease Activity Index (116)*
 Not involved (0)Mild (1)Moderate (2)Severe (3)
  • *

    HAQ = Health Assessment Questionnaire (58); PASI = Psoriasis Area and Severity Index (19); DLQI = Dermatology Quality of Life Index (71); BASDAI = Bath Ankylosing Spondylitis Disease Activity Index (46); ASQOL = Ankylosing Spondylitis Quality of Life Questionnaire (125).

  • HAQ only counted if clinical involvement of domain (joint/enthesis/dactylitis) is present.

Peripheral arthritis ≤4 joints (swollen or tender); normal function (HAQ <0.5)≤4 joints but function impaired; or >4 joints, normal function>4 joints and function impaired
Skin disease PASI ≤10 and DLQI ≤10PASI ≤10 but DLQI >10; or PASI >10 but DLQI ≤10PASI >10 and DLQI >10
Enthesitis ≤3 sites; normal function (HAQ <0.5)≤3 sites but function impaired; or >3 sites but normal function>3 sites and function impaired
Dactylitis ≤3 fingers; normal function (HAQ <0.5)≤3 fingers but function impaired; or >3 fingers but normal function>3 fingers and has function impaired
Spinal disease BASDAI <4; normal function (ASQOL <6)BASDAI >4 but normal function; BASDAI <4 but function impairedBASDAI >4 and function impaired

Score interpretation.

Higher scores correspond to more severe disease activity.

Acceptability.

Acceptable.

Reliability.

The reliability of the instrument has not been assessed per se. The measures that constitute the scoring system have been routinely used in clinical trials and have been proven to be reliable.

Validity.

The CPDAI demonstrates significant correlation with patient (r = 0.777) and physician global (r = 0.809) assessments and discriminates well between effectively and ineffectively treated patients (116).

Ability to detect change.

In a cohort of 25 patients in whom treatment was changed, the median CPDAI score had decreased from 8.5 at baseline to 5.5 at 3 months of followup (P = 0.02), with a standardized response mean of 0.60.

Critical Appraisal of Overall Value to the Rheumatology Community

The CPDAI has recently been compared to the DAPSA using the Psoriasis Randomized Etanercept Study in Subjects with Psoriatic Arthritis (PRESTA) study data set (118). In PRESTA, 752 patients were randomized to a double-blind, 2-period study that evaluated the safety and efficacy of 2 doses of etanercept on skin and musculoskeletal disease. Both the CPDAI and DAPSA were effective in determining treatment response in patients treated with etanercept for active psoriasis and PsA. Joint responses were equally determined by both composite scores; however, the CPDAI, which encompasses other domains such as skin, enthesitis, and dactylitis, was the only composite score that could distinguish global treatment response between the 2 etanercept doses. This suggests that the CPDAI is a more sensitive instrument to detect change in domains beyond joints and patient global, particularly in domains such as enthesitis, dactylitis, and the skin, which are important multidimensional components of PsA.

GRACE Project

In addition to critical evaluation of the emerging composite measures described above, the GRAPPA has engaged in a long-term project known as the GRACE Project, in which more than 450 patients with PsA are being evaluated in a multicenter study by members of GRAPPA (Helliwell P: unpublished observations). These patients are being evaluated with multiple measures of all of the clinical domains of PsA longitudinally, capturing information about disease activity and decisions to change therapy in order to determine the individual performance characteristics of measures as well as their ability to provide composite information about PsA as a whole. An attempt is being made to develop a scoring system in which all domains are represented to comprehensively reflect the disease. However, it is acknowledged that this can be a challenge in that musculoskeletal disease activity and response to therapy may be divergent from skin and nail activity and response, and it is critical that the measure be sensitive to disease severity in each domain and not allow “dilution” of a domain in attempting to be comprehensive. It is anticipated that the comparative evaluation of emerging composite measures, including any that are crafted from the GRACE Project, will be presented and voted upon at the OMERACT 12 meeting in May 2012 (Mease PJ: unpublished observations).

Minimal Disease Activity (MDA)

Several studies have utilized remission criteria used in rheumatoid arthritis (RA) to evaluate the ability to achieve this state in PsA. Saber et al comparatively evaluated PsA and RA patients in a single-center study regarding the ability to achieve a DAS28-defined remission (<2.6). They found that a greater percentage of PsA patients than RA patients were able to achieve this degree of response, using a variety of treatment approaches (119). Cantini et al, studying Italian cohorts of PsA and RA patients using the stricter modified ACR criteria for remission from 1981 (120), also found that a greater percentage of PsA patients than RA patients could achieve the state of remission and could remain in that state for a longer period of time (121). These studies suggest that it may be less difficult to aim for sustained remission in PsA than RA. However, these groups have used “joint-centered” definitions of remission, which may be a less comprehensive approach to evaluation of PsA, therefore leading to a GRAPPA project, led by Coates et al, to construct a PsA-specific definition of MDA. Hypothetical cases were evaluated by GRAPPA members and the subsequent analysis by Coates et al resulted in the definition of MDA criteria for PsA shown in Table 4 (122). These criteria were validated by assessing patients in a patient cohort in Toronto (123) and in interventional trial data sets (124). The development of this instrument is a step toward “treatment to target” in PsA, i.e., the goal of achieving remission or low disease activity state.

Table 4. Minimal disease activity (MDA) criteria in psoriatic arthritis (122)*
  • *

    PASI = Psoriasis Area and Severity Index; BSA = body surface area; VAS = visual analog scale; HAQ = Health

A patient is classified as in MDA when they meet 5 of 7 of the following criteria:
 Tender joint count ≤1
 Swollen joint count ≤1
 PASI ≤1 or BSA ≤3
 Patient pain VAS ≤15
 Patient global activity VAS ≤20
 HAQ ≤0.5
 Tender enthesial points ≤1

DISCUSSION

Although relatively young compared to assessment of rheumatoid arthritis and psoriasis, the field of assessment of psoriatic arthritis (PsA) has rapidly evolved over the past decade due to the need for valid and reliable assessments in clinical trials of multiple emerging therapeutic agents in PsA as well as growing interest in the disease state. A key factor has been the collaborative endeavors of rheumatologists and dermatologists in the international research consortia, Group for Research and Assessment of Psoriasis and Psoriatic Arthritis, in league with Outcome Measures in Rheumatology. Core domains to be assessed include joint inflammation and damage, enthesitis, dactylitis, skin and nail disease, spondylitis, function, and quality of life. Measures for these individual domains and where available, their performance characteristics, have been described in this article. Additionally, composite measures of disease state and response that have been used in clinical trials and efforts to improve upon these measures that are underway have been described. It is anticipated that as these single-domain and composite measures become codified, simpler and practical measures will evolve for use in clinical practice, allowing for more precise evaluation of disease activity and response to therapy with the goal of achieving remission or minimal disease activity.

AUTHOR CONTRIBUTIONS

Dr. Mease drafted the article, revised it critically for important intellectual content, and approved the final version to be published.

Acknowledgements

The author would like to thank Kelly Dundon for assistance in the preparation of the manuscript.

Ancillary