Further refinement of the Patient-Reported Impact of Dermatological Diseases (PRIDD) measure using classical test theory and item response theory

Background Existing dermatology-specific Patient-Reported Outcome Measures (PROMs) do not fully capture the substantial physical, psychological and social impact of dermatological conditions on patients’ lives and are not recommended for use according to the COSMIN criteria. Most were developed with insufficient patient involvement and relied on classical psychometric methods. We are developing the new Patient-Reported Impact of Dermatological Diseases (PRIDD) measure for use in research and clinical practice in partnership with patients. Objectives To examine the factor structure of PRIDD, determine the definitive selection of items for each subscale, and establish structural validity and internal consistency through classical and modern psychometric methods. Methods Two cross-sectional online surveys were conducted. Adults ( ≥ 18 years) worldwide living with a dermatological condition were recruited through the membership network of the International Alliance of Dermatology Patient Organizations (GlobalSkin). They completed the PRIDD questionnaire and a demographics questionnaire via an online survey. We examined missing data and distribution of scores for each item. The factor structure was assessed using confirmatory and exploratory factor analysis (Survey 1). Internal consistency was examined using Cronbach’s α . Rasch measurement theory analyses were conducted, including iterative assessment of rating scale function, fit to the Rasch model, unidimensionality, reliability, local dependence, targeting and differential item functioning (DIF) (Surveys 1 and 2). Results Participants in Surveys 1 and 2 numbered 483 and 504 people, respectively. All items had ≤ 3% missing scores and all five response options were used. A four-factor model showed the best fit. PRIDD and all four subscales were internally consistent but showed some misfit to the Rasch measurement model. Adjustments were made to rectify disordered thresholds, remove misfitting items, local dependency and DIF, and improve targeting. The resulting 16-item version and subscales fit the Rasch model, showed no local dependency or DIF at the test level, and were well targeted.

Assessment of the full impact of dermatological conditions on patients' lives is crucial to effective management.6][7] Many of the issues identified stem from insufficient patient involvement during development and the methodological limitations of classical test theory (CTT). 5e are developing the Patient-Reported Impact of Dermatological Diseases (PRIDD) measure in partnership with patients and using both classic and modern psychometric methods.The PRIDD tool measures the impact of dermatological conditions on the patient's life and is for use in research and clinical practice with adults living with any dermatological condition worldwide.

Development and validation of the Patient-Reported Impact of Dermatological Diseases tool
[10][11] The content validity phase had three key stages: (1) concept elicitation, 12 (2) participatory item reduction 13 and (3) pilot testing. 14,15The resultant 26-item English version of PRIDD, with each item rated on a 5-point scale ranging from 0 (never) to 5 (always), has strong evidence of content validity according to the COSMIN standards. 4The conceptual framework of impact followed a reflective model 12 with five domains of impactphysical, daily life and responsibilities, psychological, social and financial (Figure S2; see Supporting Information) 13 -but is yet to be validated quantitatively. 16he current psychometric testing phase consists of two sequential stages: (4) field testing and (5) testing of the measurement properties.The field test aims to establish structural validity, an important measurement property that describes the 'degree to which the scores of an instrument are an adequate reflection of the dimensionality of the construct to be measured'. 17,18Only once PRIDD is finalized through field testing can its measurement properties be fully established.
Factor analysis, an extension of CTT, and Rasch analysis, part of the item response theory (IRT) family, are the preferred statistical methods to assess structural validity. 19ctor analysis is valuable for identifying the dimensions (or subscales) in a PROM, but cannot establish the psychometric quality of those dimensions.IRT, a modern psychometric method, is a powerful tool to assess PROM psychometrics as it overcomes many of the limitations of CTT.The Rasch model is a unidimensional measurement model that satisfies the fundamental assumptions of IRT, 20,21 meaning it provides a measurement template against which scales can be tested. 22his study (stage 4 of 5) aimed to examine the factor structure of the PRIDD measure, determine the definitive selection of items for each subscale, and establish structural validity and internal consistency through classical and modern psychometric methods.Based on the conceptual framework of impact we hypothesized that the PRIDD tool had five domains.

Study design and setting
We conducted two cross-sectional online surveys.Ethical approval was obtained from Cardiff University School of Healthcare Sciences Ethics Committee (SREC:826).Informed consent was obtained from all participants.

Patients and recruitment
We employed convenience sampling to recruit eligible participants to both surveys through the membership network of the International Alliance of Dermatology Patient Organizations (GlobalSkin, https://globalskin.org/),which is a not-for-profit alliance of over 245 dermatology patient organizations worldwide.The samples were independent of each other.It is best practice to develop and validate a PROM in one language with later cross-cultural translation.PRIDD is being developed initially in English.Participants therefore met the inclusion criteria if they were an adult (≥ 18 years), living with a dermatological condition, and understood English sufficiently to complete the survey independently.We aimed to recruit the recommended sample size of 250-500 for Rasch analysis to each survey. 23Nonparticipation was due to nonresponse.

Procedure and materials
Survey 1 was open from 1 November to 1 December 2021, and Survey 2 from 29 June to 29 July 2022.Participants were directed to the online platform, which included the participant information sheet, consent form and survey consisting of demographic items and PRIDD (Appendix S1; see Supporting Information), and given at least 4 weeks to respond.

Patient involvement
GlobalSkin conceived of the PRIDD measure, were involved in setting the research priorities and defining research questions, and provided input into study design, conduct and dissemination.Our lead patient co-researchers J.A. and A.F. are named co-authors.

Data analysis strategy
We followed the order of data analysis for field testing set out by the COSMIN group and evaluated the results against their quality criteria for structural validity and internal consistency. 11,18We completed all the steps outlined below on the Survey 1 data.This revealed that further amendments were required.After adjusting PRIDD, we conducted Survey 2, where we repeated many aspects of the analysis.We have noted below which parts of the data analysis used which dataset.We ran multiple iterations of the analysis on the different versions of PRIDD (Table S1; see Supporting Information): analysis of PRIDD V0.1 to V.04 used Survey 1 data and V.05 and V.1 used Survey 2 data.We used Little's test of Missing Completely at Random (MCAR) for missing values. 24

Examination of individual items of the Patient-Reported Impact of Dermatological Diseases tool (Surveys 1 and 2)
The percentage of missing scores was examined for each item using SPSS 27 (IBM; Armonk, NY, USA).Items with ≤ 3% missing scores were deemed 'acceptable' and ≥ 15% 'not acceptable'. 18Distributions of item scores were examined using item means (x) and standard deviations.

Confirmatory factor analysis
Confirmatory factor analysis (CFA) is more appropriate than exploratory factor analysis (EFA) when a conceptual framework is available. 4,18,19,25We performed a CFA with categorical factor indicators applying full information maximum likelihood to missing data using Mplus 8.2 (Muthén & Muthén, Los Angeles, CA, USA). 26,27Mplus determines the number of categories for each factor indicator with a robust weighted least squares estimator (wlsmv). 28Multicollinearity was assessed via bivariate correlations (Spearman's), with r ≤ 0.8 deemed acceptable. 291][32][33] Structural validity was sufficient if the comparative fit index (CFI) or Tucker-Lewis index (TLI) > 0.95, root mean square error of approximation (RMSEA) < 0.06 or standardized root mean square residual (SRMR) < 0.08. 11

Exploratory factor analysis
As the CFA did not support our five-domain conceptual framework, we performed an EFA with listwise deletion on SPSS 27 using the principal factor method with oblique rotation (direct oblimin) to determine the number of dimensions. 18e Kaiser-Meyer-Olkin (KMO) test (KMO > 0.5) and Bartlett's test of sphericity (P < 0.05) were used to confirm the adequacy of the sample and data, respectively.
5][36] As uncertainty remained regarding the number of factors to extract, we followed Costello and Osborne's 37 recommendation to run multiple factor analyses, setting the number of factors to retain manually once at the projected number based on the a priori factor structure, then at the number of factors suggested by the scree test, and finally at numbers above and below those numbers.Item-loading tables were compared and the solution with the most factors and 'cleanest' factor structure (item loadings > 0.3, no or few cross-loadings, and no factors with fewer than three items) was deemed to have the best fit to the data.Residual correlations < 0.1 and factor loadings ≥ 0.5 were deemed acceptable. 30,31,38

Internal consistency (Surveys 1 and 2)
We tested internal consistency for PRIDD and each factor separately with listwise deletion using SPSS 27.Items with inter-item correlations > 0.7 and item-total correlations < 0.3 were candidates for removal.Cronbach's α > 0.7 was deemed acceptable and > 0.9 indicated item redundancy. 11

Rasch analysis (Surveys 1 and 2)
0][41][42][43][44] We tested whether the subscales could be validly combined into an 'overall impact' total score using the subtest approach to obtain R (average latent correlation between the subscales) and A values (the amount of shared variance between the subscales).Joliffe's criteria of eigenvalues > 0.7 36 3 Visual inspection of the scree plot to identify the number of eigenvalues before the slope flattens out 18 4 Parallel analysis 34

Examination of PRIDD items
All items had acceptable levels (≤ 3%) of missing scores (Table S6; see Supporting Information).All item means were close to the centre of the range of possible scores, indicating that the response options detected the full range of the construct, were well worded and had higher variances. 34

PRIDD V0.1
Bivariate correlations for the CFA ranged from 0.2 to 0.76, indicating no multicollinearity. 29Approximate fit to the four-factor model was not achieved (Table 5); therefore, we conducted an EFA.Three factors had eigenvalues > 1, six factors > 0.7 and the scree plot was slightly ambiguous, showing inflexions that would justify retaining two or three factors (Appendix S2; see Supporting Information).The parallel analysis diverged from these results, suggesting a 26-factor model (Table S7; see Supporting Information).We consequently ran six EFAs, setting the number of factors to retain manually at 5, 2, 3, 6, 4 and 1.

PRIDD V0.2
The six-factor model was the 'cleanest' (Table S8; see Supporting Information) and item clustering suggested the following underlying concepts: All six factors showed at least some misfit to the Rasch model (Table S9; see Supporting Information).Five factors showed local dependency (Appendix S4; see Supporting Information).Correction involved removing three items and grouping six other items into three testlets.This produced three factors with fewer than three items, the minimum recommended number of items in a scale. 18,37Given the conceptual overlap of the remaining items, Appearance-related Concerns was combined with Negative Affect to create Psychological Impact, and Interpersonal Relationships and Identity were combined to make Social Impact.This resulted in a 22-item, four-factor model (PRIDD V0.3).Threshold ordering Rating scales function optimally when thresholds are ordered.Thresholds correspond with the threshold points between two different scores on the rating scale, in this case 'never' to 'always'.At the threshold point, it is equally likely to obtain either score (i.e. the probability of scoring 2 or 3 on the item is 50/50).This is demonstrated by category probability curves where each curve shows a distinct peak which illustrates the position along the continuum where the categories are most likely to be selected. 40,41This indicates that respondents are able to discriminate between response options. 41,42isordered thresholds indicate that an item is not working properly as the response categories are not progressing in a logical order.In this case, even when the probability of selecting a particular response option is at its highest, it is still more likely that another option will be selected.We examined threshold ordering visually using the threshold map.The category probability curves of disordered thresholds were visually inspected to determine whether the item response options were functioning optimally and whether rescoring was indicated.

2
Tests of fit Model fit was acceptable if the item-trait interaction, reported as χ 2 , was nonsignificant (P > 0.05) and the item and person residuals had x̄ ≈ 0 and SD ≈ 1.Individual items and persons were regarded as misfitting if their residuals fell outside the range of ± 2.5.Individual items were also tested by χ 2 and F-tests.

3
Unidimensionality Strict unidimensionality was confirmed with a series of t-tests reporting significantly different person estimates in < 5% of cases (or the lower bound of the 95% confidence interval < 5%).

4
Local independence Local dependency among the items was assessed via the residual correlations using a cut-point of the average plus 0.

PRIDD V0.3
This four-factor model achieved approximate fit and met the COSMIN criteria for structural validity (Table 5).All factors were internally consistent (Table S10; see Supporting Information) but showed some misfit to the Rasch model (Table S11; see Supporting Information).This improved on removal of two items: 'I have been hiding, covering or concealing my condition' (Psychological Impact) and 'I have  been excluded, stigmatized or discriminated against by others' (Social Impact).

PRIDD V0.4
In this 20-item version, all factors were strictly unidimensional and had no evidence of differential item functioning (DIF), except for Life Responsibilities Impact, although this cancelled out at test level (Appendix S5; see Supporting Information).That is, the effects from the item exhibiting DIF for those with an inflammatory condition was cancelled out by the item exhibiting DIF for those with a noninflammatory condition.The Person-Item Threshold Distribution graphs indicated that the addition of an item capturing more 'severe' impact to each factor would improve targeting (Figure S3; see Supporting Information).We added four items based on patient prioritization of items during the previous Delphi study (Table S12; see Supporting Information). 13tigma ('I have been excluded, stigmatized or discriminated against by others') emerged as an important impact during the content validity phase 15 but was not captured by any of the included Social Impact items.While this item was not required for this subscale to fit the Rasch model, we decided to retain this item for further testing alongside the additional item above.This resulted in a 24-item, four-factor version (PRIDD V0.5).

PRIDD V0.5
The first three additional items plus the stigma item showed disordered thresholds.Combining adjacent response categories improved fit to the model (Figure 1).All factors demonstrated at least some misfit to the model (Table 6), with some also showing breaches of unidimensionality, local dependency (Appendix S6; see Supporting Information) and DIF (Appendix S7; see Supporting Information).

PRIDD V1
Removal of nine items across the four factors (Table S13; see Supporting Information) resulted in a 16-item PRIDD (PRIDD V1) with each dimension (Table 6) and item (Table 7) showing fit to the Rasch model, strict unidimensionality, no local dependence (Appendix S8; see Supporting Information), no DIF at the test level (Appendix S9; see Supporting Information), and was well targeted (Figure 2).
The R (0.84) and A values (0.95) demonstrated high average pairwise correlation and very high levels of common variance between the four subscales, indicating that summing the four subscales to obtain an 'overall impact' total score was valid.

PRIDD scoring
PRIDD total (0-63) and subscale scores are obtained in a two-step process by summing item scores and transforming these raw, ordinal level scores to interval level data using a conversion table (full instructions in Appendix S10; see Supporting Information).

Discussion
This field test study represents the fourth of five steps in PRIDD's development and validation.It examined PRIDD's factor structure and established the definitive selection of items for each subscale.The findings and resultant adjustments produced the final 16-item PRIDD within four domains.
This study further supported PRIDD's content validity, feasibility and refined the conceptual framework of the impact of dermatological conditions. 12,14While we found support for each of the original five domains of impact; the data indicated a four-factor model (Figure 3) consisting of Physical Impact, Life Responsibilities Impact (combining the previous daily life and responsibilities and financial impacts domains), Psychological Impact and Social Impact subscales.This validated conceptual framework provides clinicians and researchers with a valuable, theoretically coherent framework for understanding and measuring the impact of dermatological conditions.The multidimensionality demonstrates that holistic, multidisciplinary approaches are fundamental to high-quality, personalized dermatological care.PRIDD total and subscales scores can indicate targets for interventions and guide shared decision-making and referral to appropriate specialists such as psychologists and occupational therapists.
Structural validity is an important psychometric property and unidimensionality, the assumption that a scale  (or subscale) measures a single construct, is fundamental to this.There is generally poor or mixed evidence for unidimensionality across many of the commonly used dermatology-specific PROMs.This is often because unidimensionality has been assumed but not tested or because of a reliance on CTT methods over IRT methods.For example, the three subscales of the Skindex-29 -emotions, symptoms and functioning -were established using Cronbach's α and correlations. 45However, a Rasch analysis found evidence of unidimensionality for only two subscales: symptoms and combined emotions and functioning. 46It is also often erroneously assumed that PROM subscales can automatically be summed to obtain a total score.This study provides empirical evidence that PRIDD and all four subscales are unidimensional and can be validly combined into a total score.This means that PRIDD can not only provide a single score of overall impact, but the subscales can be used individually to distinguish among the domains of impact, making it a powerful and versatile tool.Existing dermatology-specific PROMs were developed with a small number of patients in one country and subsequently used globally, with limited revalidation.As PRIDD is intended for worldwide use with adults with a dermatological condition, we recruited geographically diverse samples throughout all stages of PRIDD's development to enhance transferability.The final version of PRIDD, tested with this heterogenous study sample, showed no DIF, indicating that it is well worded and appropriate for use with the global dermatology patient population.These results attest to the value of developing and validating PROMs with an inclusive, patient-centred approach and following best practices, including a rigorous pilot test, use of both CTT and modern psychometric methods, and combining participatory and statistical methods of item reduction.
Researchers, clinicians and regulatory agencies should choose measurement instruments based on their quality.The next and final step in PRIDD's development will be a study to test the remaining measurement properties, interpretability information (e.g.minimally important change) 47 and comparison with other well-used measures.Subsequently, PRIDD will undergo cultural translation and linguistic validation, and be used to collect global data on the life impact of dermatological conditions.It will also be beneficial to revalidate PRIDD's measurement properties in a sample of patients not involved in the original development and validation.
This study met the highest standards for tests of structural validity (Table S14; see Supporting Information) and internal consistency outlined by COSMIN (Table S15; see Supporting Information). 9We recruited a diverse sample enabling us to test PRIDD's performance across a range of subgroups.However, as participants were primarily recruited through patient organizations and the sample was predominantly White and well educated, these results may not be representative of the dermatology patient population.As we used the same recruitment methods for both surveys, it is possible that there was overlap between samples, but we were unable to verify this.We were also not able to check that all participants were sufficiently proficient in English.
Using a combination of classical and modern psychometric methods enabled us to select the best statistical methods for assessing structural validity and internal consistency.Data triangulation from the multiple methods used -CFA, EFA, parallel and Rasch analysis -provide strong support for the conceptual framework.The sample was relatively large for factor and Rasch analyses, increasing the generalizability of the conclusions. 34Existing dermatology-specific PROMs use ordinal, Likert-type responses, meaning that intervals between successive points on the scales are not intrinsically equal. 48This leads to challenges in comparing intervention efficacy across patients with PROM scores on different portions of the scale. 49Using Rasch analysis allowed us to transform PRIDD's ordinal responses into interval level scores.This optimizes the level of quantitative information that can be obtained, including the calculation of mean and change scores without the restrictions of nonparametric, representational measurement, 46 and enables valid comparison of scores across the scale. 39We therefore recommend using the transformed rather than the raw scores, although the latter may be more feasible in routine practice.
While we previously employed participatory methods to prioritize items for inclusion in PRIDD, this study used purely statistical techniques of item reduction, which may reduce item content coverage and consequently content validity.We recommend that PROM developers consider involving patients during final item selection to ensure a good balance between face validity and psychometric performance and ensure the final PROM is acceptable to the target population.(GRIDD) project.GlobalSkin is a not-for-profit alliance of dermatology patient organizations worldwide.Participants were recruited through GlobalSkin's network of patient organizations.
Cosentyx is recommended by NICE as an option for the treatment of moderate to severe HS in adults who have not responded to conventional systemic treatment (subject to eligibility criteria) 6

Cosentyx® (secukinumab)
is available for eligible patients with moderate to severe hidradenitis suppurativa (HS)* 1,2 The primary endpoint was met for Cosentyx 300 mg Q2W in both SUNRISE and SUNSHINE (p=0.015 and p=0.007, respectively) and was met for Cosentyx 300 mg Q4W in SUNRISE (p=0.002), but not in SUNSHINE. 4e most frequently reported adverse reactions are upper respiratory tract infections (17.1%) (most frequently nasopharyngitis, rhinitis). 1,2 new safety signals observed in HS trials 3 The most frequently reported adverse events in SUNSHINE and SUNRISE were headache, nasopharyngitis and worsening of hidradenitis up to Week 16. 3 Please consult the SmPC before prescribing.
Cosentyx can help to provide fast relief and lasting control for your eligible patients with HS 3 *Cosentyx is indicated in adult patients with moderate to severe HS (acne inversa) with an inadequate response to conventional HS therapy. 1,2Please see above for the licensed dermatology indications.† HiSCR50: ≥50% decrease in abscesses and inflammatory nodules count with no increase in the number of abscesses and/or in the number of draining fistulae relative to baseline at Week 16.In HS study 1 HiSCR50 was 41.8% and 45.0% in the Q4W arm (n=180) and Q2W arm (n=181), respectively.he percentage of patients who started with moderate or severe pain and had mild or no pain was 65.3% in the Cosentyx group and 80.9% in the placebo group for the Q2W dosing regimen.The percentage of patients who started with moderate or severe pain and had mild or no pain at Week 52 was 70.1% in the Cosentyx group and 64.8% in the placebo group for the Q4W dosing regimen. 3§ Flare, a prespecified exploratory endpoint, is defined as at least a 25% increase in AN count with a minimum increase of 2 in absolute AN count relative to baseline.In the Q4W arm, 360 patients were evaluable at Week 16 and 278 patients were evaluable at Week 52, 27.3% of patients experienced flares at Week 52.In the Q2W arm, 361 and 289 were evaluable at Week 16 and Week 52, respectively with 20.4% of patients experiencing flares at Week 52.Indications: Treatment of: moderate to severe plaque psoriasis in adults, children and adolescents from the age of 6 years who are candidates for systemic therapy; active psoriatic arthritis in adults (alone or in combination with methotrexate) who have responded inadequately to disease-modifying anti-rheumatic drug therapy; active ankylosing spondylitis in adults who have responded inadequately to conventional therapy; active nonradiographic axial spondyloarthritis (nr-axSpA) with objective signs of inflammation as indicated by elevated C-reactive protein (CRP) and/or magnetic resonance imaging (MRI) evidence in adults who have responded inadequately to non-steroidal anti-inflammatory drugs; active enthesitisrelated arthritis and juvenile psoriatic arthritis in patients 6 years and older (alone or in combination with methotrexate) whose disease has responded inadequately to, or who cannot tolerate, conventional therapy; active moderate to severe hidradenitis suppurativa (acne inversa) in adults with an inadequate response to conventional systemic HS therapy.Presentations: Cosentyx 150 mg solution for injection in pre-filled pen; Cosentyx 300 mg solution for injection in pre-filled pen.Dosage & Administration: Administered by subcutaneous injection at weeks 0, 1, 2, 3 and 4, followed by monthly maintenance dosing.Consider discontinuation if no response after 16 weeks of treatment.Each 150 mg dose is given as one injection of 150 mg.Each 300 mg dose is given as two injections of 150 mg or one injection of 300 mg.If possible avoid areas of the skin showing psoriasis.
Plaque Psoriasis: Adult recommended dose is 300 mg monthly.Based on clinical response, a maintenance dose of 300 mg every 2 weeks may provide additional benefit for patients with a body weight of 90 kg or higher.Adolescents and children from the age of 6 years: if weight ≥ 50 kg, recommended dose is 150 mg (may be increased to 300 mg as some patients may derive additional benefit from the higher dose).If weight < 50 kg, recommended dose is 75 mg.However, 150mg solution for injection in pre-filled pen is not indicated for administration of this dose and no suitable alternative formulation is available.Psoriatic Arthritis: For patients with concomitant moderate to severe plaque psoriasis see adult plaque psoriasis recommendation.For patients who are anti-TNFα inadequate responders, the recommended dose is 300 mg, 150 mg in other patients.Can be increased to 300 mg based on clinical response.Adverse events should be reported.Reporting forms and information can be found at www.mhra.gov.uk/yellowcard.
Adverse events should also be reported to Novartis via uk.patientsafety@novartis.com or online through the pharmacovigilance intake (PVI) tool at www.novartis.com/report.
If you have a question about the product, please contact Medical Information on 01276 698370 or by email at medinfo.uk@novartis.com

Figure 2
Figure 2 Person-Item Threshold Distribution graphs for PRIDD V1 'Overall Impact' and the Physical Impact, Life Responsibilities Impact, Psychological Impact and Social Impact subscales.
Field testing of PRIDD, R. Pattinson et al.
Improved outcomes lasted through Week 52 (observed data with no statistical testing)[3][4][5] Cosentyx licensed indications in dermatology: Cosentyx is indicated for the treatment of moderate to severe plaque psoriasis in adults, children and adolescents from the age of 6 years who are candidates for systemic therapy; active moderate to severe HS (acne inversa) in adults with an inadequate response to conventional systemic HS therapy.For full indications, please see the SmPC.1,2SUNSHINE AND SUNRISE: Two randomised, double-blind, multicentre, Phase III trials: SUNSHINE and SUNRISE (Cosentyx 300 mg Q4W, n=360 or Cosentyx 300 mg Q2W, n=361).The primary endpoint for both SUNSHINE and SUNRISE studies in adult patients with moderate to severe HS was the clinical response (as measured by HiSCR), defined as a decrease in abscess and inflammatory nodule count by 50% or more with no increase in the number of abscesses or draining fistulae compared with baseline, of Cosentyx versus placebo at Week 16, assessed in the overall population.Clinical response was sustained to Week 52 in both trials.4

Table 1
Goodness-of-fit criteria for confirmatory factor analysis models

Table 2
Criteria to determine number of factors extracted in the exploratory factor analysis

Table 3
Steps in the iterative Rasch analysis DIF occurs when members from different groups who have the same level of the latent trait (i.e.impact) have a different probability of giving a certain response to an item.DIF was tested by: • gender (male or female) • age group (years; four equal groups: Survey 1: 18-36, 37-55, 56-74, 75-90; Survey 2: 18-37, 38-57, 58-77, 78+) • inflammatory type (inflammatory and noninflammatory).Inflammatory type was chosen over discrete diseases as DIF analysis can handle no more than four categories.For categorization of diseases see Table S2 • highest qualification • English as a first language (yes or no) A statistically significant Bonferroni-adjusted P-value indicated DIF.6 Targeting We visually inspected the Person-Item Threshold Distribution graphs and reported the x̄ person location value.Mean person locations within + 0.5 logits of the mean item location (i.e.0 logits) suggested acceptable targeting. 44Downloaded from https://academic.oup.com/bjd/article/190/5/718/7462691 by Acquisitions user on 17 June 2024 Field testing of PRIDD, R. Pattinson et al.

Table 4
Participant characteristics

Table 5
Fit indices of the confirmatory factor analysis models for PRIDD V0.1, V0.3 and V1 a Higher-order models tested the factors with 'overall impact' as the second-factor order.CFI, comparative fit index; CI, confidence interval; df, degrees of freedom; PRIDD, Patient-Reported Impact of Dermatological Diseases; RMSEA, root mean square error of approximation; SRMR, standardized root mean square residual; TLI, Tucker-Lewis index; V, version.

Analysis No. of items Valid n (no. of extremes) a Item fit residual Person fit residual Overall χ 2 interaction PSI α Unidimensionality t-tests (CI) x̄SD x̄SD Value df P Proportion significance
a Extremes: people scoring either maximally or minimally across the complete item set.α, Cronbach's α; CI, confidence interval; df, degrees of freedom; PRIDD, Patient-Reported Impact of Dermatological Diseases; PSI, Person Separation Index

Table 7
Individual item fit of PRIDD V1 a Bonferroni-adjusted probability value.PRIDD, Patient-Reported Impact of Dermatological Diseases.

¶5 Cosentyx is approved for use in eligible patients with HS 1,2
This promotional material has been created and funded by Novartis Pharmaceuticals UK Ltd. for UK healthcare professionals only.
5    ¶Observed data from full analysis set.Number of patients with no increase from baseline from Week 16 to Week 52 in patients with at least one draining fistulae at baseline.82.6% in Q4W arm (n=218), 80.7% in Q2W arm (n=239).5Abbreviations:AN,abscessandinflammatory nodule; HiSCR, hidradenitis suppurativa clinical response; HS, hidradenitis suppurativa; Q2W, every 2 weeks; Q4W, every 4 weeks; SmPC, summary of product characteristics.References: 1. Cosentyx® (secukinumab) GB Summary of Product Characteristics; 2. Cosentyx® (secukinumab) NI Summary of Product Characteristics; 3. Kimball AB, et al.Lancet 2023;401(10378):747-761 and supplementary appendix; 4. Novartis Data on File.SUNNY clinical programme post-hoc analysis of skin pain severity.March 2023; 5. Novartis Data on File.Draining fistulas; 6.National Institute for Health and Care Excellence.Secukinumab for treating moderate to severe hidradenitis suppurativa.Available at: https://www.nice.org.uk/guidance/ta935[AccessedApril 2024].Prescribing information and adverse event reporting can be found on the next page.UK | April 2024 | 421564 of patients achieved HiSCR50 at Week 16 in both trials †1,2 of patients were flare free at Week 52 §3 of patients who started with moderate or severe pain had only mild or no pain at Week 52 ‡4of patients had no increase in draining tunnels at Week 52 From the age of 6 years, if weight ≥ 50 kg, recommended dose is 150 mg.If weight < 50 kg, recommended dose is 75 mg.However, 150mg solution for injection in pre-filled pen is not indicated for administration of this dose and no suitable alternative formulation is available.Hidradenitis suppurativa: Recommended dose is 300 mg monthly.Based on clinical response, the maintenance dose can be increased to 300 mg every 2 weeks.Contraindications: Hypersensitivity to the active substance or excipients.Clinically important, active infection.Warnings & Precautions: Infections: Potential to increase risk of infections; serious infections have been observed.Caution in patients with chronic infection or history of recurrent infection.Advise patients to seek medical advice if signs/symptoms of infection occur.Monitor patients with serious infection closely and do not administer Cosentyx until the infection resolves.Non-serious mucocutaneous candida infections were more frequently reported for secukinumab than placebo in the psoriasis clinical studies.Should not be given to patients with active tuberculosis (TB).Consider anti-tuberculosis therapy before starting Cosentyx in patients with latent TB.Inflammatory bowel disease (including Crohn's disease and ulcerative colitis): New cases or exacerbations of inflammatory bowel disease have been reported with secukinumab.Secukinumab, is not recommended in patients with inflammatory bowel disease.If a patient develops signs and symptoms of inflammatory bowel disease or experiences an exacerbation of pre-existing inflammatory bowel disease, secukinumab should be discontinued and appropriate medical management should be initiated.Hypersensitivity reactions: Rare cases of anaphylactic reactions have been observed.If an anaphylactic or serious allergic reactions occur, discontinue immediately and initiate appropriate therapy.Vaccinations: Do not give live vaccines concurrently with Cosentyx; inactivated or non-live vaccinations may be given.Paediatric patients should receive all age appropriate immunisations before treatment with Cosentyx.Latex-Sensitive Individuals: The removable needle cap of the 150mg pre-filled pen contains a derivative of natural rubber latex.Concomitant immunosuppressive therapy: Combination with immunosuppressants, including biologics, or phototherapy has not been evaluated in psoriasis studies.Cosentyx was given concomitantly with methotrexate, sulfasalazine and/or corticosteroids in arthritis studies.Caution when considering concomitant use of other immunosuppressants.Interactions: Live vaccines should not be given concurrently with secukinumab.No interaction between Cosentyx and midazolam (CYP3A4 substrate) seen in adult psoriasis study.No interaction between Cosentyx and methotrexate and/or corticosteroids seen in arthritis studies.Fertility, pregnancy and lactation: Women of childbearing potential: Use an effective method of contraception during and for at least 20 weeks after treatment.Pregnancy: Preferably avoid use of Cosentyx in pregnancy.Breast feeding: It is not known if secukinumab is excreted in human breast milk.A clinical decision should be made on continuation of breast feeding during Cosentyx treatment (and up to 20 weeks after discontinuation) based on benefit of breast feeding to the child and benefit of Cosentyx therapy to the woman.Fertility: Effect on human fertility not evaluated.Adverse Reactions: Very Common (≥1/10): Upper respiratory tract infection.Adverse events should be reported.Reporting forms and information can be found at www.mhra.gov.uk/yellowcard.Adverse events should also be reported to Novartis via uk.patientsafety@novartis.com or online through the pharmacovigilance intake (PVI) tool at www.novartis.com/reportPotential to increase risk of infections; serious infections have been observed.Caution in patients with chronic infection or history of recurrent infection.Advise patients to seek medical advice if signs/symptoms of infection occur.Monitor patients with serious infection closely and do not administer Cosentyx until the infection resolves.Non-serious mucocutaneous candida infections were more frequently reported for secukinumab in the psoriasis clinical studies.Should not be given to patients with active tuberculosis (TB).Consider anti-tuberculosis therapy before starting Cosentyx in patients with latent TB.Inflammatory bowel disease (including Crohn's disease and ulcerative colitis): New cases or exacerbations of inflammatory bowel disease have been reported with secukinumab.Secukinumab, is not recommended in patients with inflammatory bowel disease.If a patient develops signs and symptoms of inflammatory bowel disease or experiences an exacerbation of pre-existing inflammatory bowel disease, secukinumab should be discontinued and appropriate medical management should be initiated.Hypersensitivity reactions: Rare cases of anaphylactic reactions have been observed.If an anaphylactic or serious allergic reactions occur, discontinue immediately and initiate appropriate therapy.Vaccinations: Do not give live vaccines concurrently with Cosentyx; inactivated or non-live vaccinations may be given.Paediatric patients should receive all age appropriate immunisations before treatment with Cosentyx.Latex-Sensitive Individuals: The removable needle cap of the 75mg and 150 mg pre-filled syringe and 150mg pre-filled pen contains a derivative of natural rubber latex.Concomitant immunosuppressive therapy: Combination with immunosuppressants, including biologics, or phototherapy has not been evaluated in psoriasis studies.Cosentyx was given concomitantly with methotrexate, sulfasalazine and/or corticosteroids in arthritis studies.Caution when considering concomitant use of other immunosuppressants.Interactions: Live vaccines should not be given concurrently with secukinumab.No interaction between Cosentyx and midazolam (CYP3A4 substrate) seen in adult psoriasis study.Oral herpes, headache, rhinorrhoea, diarrhoea, nausea, fatigue.Uncommon (≥1/1,000 to <1/100): Oral candidiasis, lower respiratory tract infections, neutropenia, inflammatory bowel disease.Rare (≥1/10,000 to <1/1,000): anaphylactic reactions, exfoliative dermatitis (psoriasis patients), hypersensitivity vasculitis.Not known: Mucosal and cutaneous candidiasis (including oesophageal candidiasis).Infections: Most infections were nonserious and mild to moderate upper respiratory tract infections, e.g.nasopharyngitis, and did not necessitate treatment discontinuation.There was an increase in mucosal and cutaneous (including oesophageal) candidiasis, but cases were mild or moderate in severity, non-serious, responsive to standard treatment and did not necessitate treatment discontinuation.Serious infections occurred in a small proportion of patients (0.015 serious infections reported per patient year of follow up).Neutropenia: Neutropenia was more frequent with secukinumab than placebo, but most cases were mild, transient and reversible.Rare cases of neutropenia CTCAE Grade 4 were reported.Hypersensitivity reactions: Urticaria and rare cases of anaphylactic reactions were seen.Immunogenicity: Less than 1% of patients treated with Cosentyx developed antibodies to secukinumab up to 52 weeks of treatment.Other Adverse Effects: The list of adverse events is not exhaustive, please consult the SmPC for a detailed listing of all adverse events before prescribing.Legal Category: POM.MA Number & List Price: PLGB 00101/1205 -75 mg pre-filled syringe x 1 -£304.70;PLGB 00101/1029 -150 mg pre-filled pen x2 £1,218.78;PLGB 00101/1030 -150 mg pre-filled syringe x2 £1,218.78;PLGB 00101/1198 -300 mg pre-filled pen x 1 £1218.78.PI Last Revised: June 2023.Full prescribing information, (SmPC) is available from: Novartis Pharmaceuticals UK Limited, 2nd Floor, The WestWorks Building, White City Place, 195 Wood Lane, London, W12 7FQ.Telephone: (01276) 692255.
Ankylosing Spondylitis: Recommended dose 150 mg.Can be increased to 300 mg based on clinical response.nr-axSpA: Recommended dose 150 mg.Enthesitis-related arthritis and juvenile psoriatic arthritis: provide additional benefit for patients with a body weight of 90 kg or higher.Adolescents and children from the age of 6 years: if weight ≥ 50 kg, recommended dose is 150 mg (may be increased to 300 mg as some patients may derive additional benefit from the higher dose).If weight < 50 kg, recommended dose is 75 mg.Psoriatic Arthritis: For patients with concomitant moderate to severe plaque psoriasis see adult plaque psoriasis recommendation.For patients who are anti-TNFα inadequate responders, the recommended dose is 300 mg, 150 mg in other patients.Can be increased to 300 mg based on clinical response.Ankylosing Spondylitis: Recommended dose 150 mg.Can be increased to 300 mg based on clinical response.nr-axSpA: Recommended dose 150 mg.Enthesitis-related arthritis and juvenile psoriatic arthritis: From the age of 6 years, if weight ≥ 50 kg, recommended dose is 150 mg.If weight < 50 kg, recommended dose is 75 mg.Hidradenitis suppurativa: Recommended dose is 300 mg monthly.Based on clinical response, the maintenance dose can be increased to 300 mg every 2 weeks.Contraindications: Hypersensitivity to the active substance or excipients.Clinically important, active infection.Warnings & Precautions: Infections: No interaction between Cosentyx and methotrexate and/or corticosteroids seen in arthritis studies.Fertility, pregnancy and lactation: Women of childbearing potential: Use an effective method of contraception during and for at least 20 weeks after treatment.Pregnancy: Preferably avoid use of Cosentyx in pregnancy.Breast feeding: It is not known if secukinumab is excreted in human breast milk.A clinical decision should be made on continuation of breast feeding during Cosentyx treatment (and up to 20 weeks after discontinuation) based on benefit of breast feeding to the child and benefit of Cosentyx therapy to the woman.Fertility: Effect on human fertility not evaluated.Adverse Reactions: Very Common (≥1/10): Upper respiratory tract infection.Common (≥1/100 to <1/10):