Patient Demographics
For each study, the majority of participants were female, non-Hispanic, White (except for Study 2 CD interview and online survey, which had a majority African-American participants), and group mean ages ranged from 39–51 years. Participants in Study 1 GCM (N=20) had primarily moderate (75%) psoriasis, as reported by dermatologists. For the Study 2 GCM phase (N=20), all participants (100%) had mild/very mild disease severity. Participants in Study 2 CD interview and online survey phases were generally evenly distributed across disease severities (Table 1).
Table 1
Patient Demographics and Disease Severity
|
Study 1: GCM
(N=20)
|
Study 2: GCM
(N=20)
|
Study 2: CD Interview
(N=20)
|
Study 2: Online Survey (N=198)
|
Sex, n (%)
|
Female
|
11 (55)
|
16 (80)
|
15 (75)
|
111 (56)
|
Male
|
9 (45)
|
4 (20)
|
5 (25)
|
87 (44)
|
Age, years
|
Mean (SD)
|
50.6 (14.4)
|
45.6 (18.4)
|
51.1 (19.0)
|
39.3 (12.7)
|
Min/Max
|
28/85
|
25/84
|
19/80
|
18/78
|
Race, n (%)
|
White/Caucasian
|
17 (85)
|
15 (75)
|
5 (25)
|
77 (39)
|
Black/African-American
|
3 (15)
|
5 (25)
|
11 (55)
|
99 (50)
|
Asian/Asian-American
|
0
|
0
|
1 (5)
|
2 (1)
|
Other
|
0
|
0
|
3 (15)
|
20 (10)
|
Ethnicity, n (%)*
|
Hispanic/Latino
|
0
|
0
|
1 (5)
|
21 (11)
|
Not Hispanic/Latino
|
20 (100)
|
20 (100)
|
18 (90)
|
177 (89)
|
Disease Severity, Clinician-rated, n (%)
|
None/Very mild
|
-
|
-
|
-
|
1 (<1%)
|
Mild
|
0
|
20 (100)
|
5 (25)
|
63 (32)
|
Moderate
|
15 (75)
|
0
|
5 (25)
|
74 (37)
|
Severe
|
5 (25)
|
0
|
4 (20)
|
46 (23)
|
Very Severe
|
0
|
0
|
6 (30)
|
14 (7)
|
*1 missing in Study 2: CD Interview.
GCM Exercise
In Study 1, a total of 142 responses were generated by patients (n=20) and clinicians (n=10) to the prompt about ideal psoriasis treatment. After harmonization, 99 responses were removed primarily due to redundancy and 43 (numbered 1–43) were retained for the sorting and rating tasks (one concept, “be safe for long-term use,” was mentioned by both patients and clinicians and was inadvertently retained as a duplicate) (Table 2). Of the initial 30 participants, 27 participants completed the sorting and rating tasks (patients, n=18; clinicians, n=9; some participants did not sort and some completed <75% of the tasks and were excluded).32 A point map generated from the sorting data through multidimensional scaling had a stress value of 0.1878, which indicated a good fit between the point map and the input similarity matrix. Seven clusters were configured from the point map and specified as domains consisting of related concepts (Figure 1): proximal benefits, relief from symptom manifestations, efficacy/general efficacy, accessibility, safety, convenience, and administration/mode of administration. The domain labels were derived from labels participants attributed to the clusters during sorting. The average importance ratings for domains ranged from 5.92 to 9.16, represented by 1‒5 layers (representing quintiles) as shown in Figure 1.
[Figure 1 here]
In Study 2, patients (n=20) generated 42 concepts from which 37 concepts were retained after harmonization. Sorting and rating were completed by participants (n=19; 1 patient was lost to follow-up). Cluster mapping resulted in 5 domains: efficacy/general efficacy, administration/mode of administration, application characteristics, treatment preferences, and prevention; the average importance ratings ranged from 6.85 to 8.96, represented by 1‒5 layers (representing quintiles). In both studies, patients (but not dermatologists) rated all domains >5 in importance, indicating a higher-than-average importance for all areas of treatment represented by the domains.
Table 2
Sorting of the Harmonized Concepts Generated in GCM Exercises
Domain
|
Study 1 GCM (Moderate–Severe Patients):
43 Harmonized Concepts Generated
|
Study 2 GCM (Newly Diagnosed–Mild Patients): 37 Harmonized Concepts Generated
|
Proximal Benefits
|
20. Clear outside symptoms, such as dryness and redness
|
—
|
|
28. Be effective in treating joint disease
|
—
|
|
38. Help control the itching
|
—
|
|
40. Be effective in treating skin disease, including nails and scalp
|
—
|
Relief from Symptom Manifestations
|
8. Free a patient from hiding problem skin
|
—
|
|
19. Consider how much the symptoms affect a patient's daily life
|
—
|
|
22. Free a patient from feeling embarrassed
|
—
|
Efficacy/General Efficacy
|
1. Resist building a tolerance
|
1. Result in the disappearance of rash
|
|
3. Not lose efficacy over time
|
2. Relieve the itching and associated discomfort
|
|
15. Be predictably effective
|
7. Absorb invisibly in my skin
|
|
26. Be strong enough to be effective, but mild enough not to cause other sicknesses or symptoms
|
9. Be effective with little to no side effects
|
|
41. Be effective
|
10. Control the itching
|
|
—
|
11. Work fast and effective for thicker plaques
|
|
—
|
12. Work longer than two hours
|
|
—
|
17. Safe for long term daily use
|
|
—
|
22. Control the itch longer
|
|
—
|
35. Have little to no side effects
|
Accessibility
|
12. Be affordable under a patient's insurance plan
|
—
|
|
34. Be available under a patient's insurance plan
|
—
|
|
35. Not interfere with a patient's occupational responsibilities
|
—
|
Safety
|
4. Be safe for long term use (patient-generated)*
|
—
|
|
5. Not damage a patient's liver
|
—
|
|
10. Not increase the risk of infection, cancer, or other disease
|
—
|
|
14. Be safe to use for those with pre-existing conditions, such as renal failure, liver disease, or internal malignancy
|
—
|
|
17. Be safe for long-term use (clinician-generated)*
|
—
|
|
27. Not lower a patient's immune system
|
—
|
|
30. Be safe to use during pregnancy and lactation
|
—
|
|
36. Have minimal, non-life-threatening side effects
|
—
|
|
37. Consider potential side effects
|
—
|
|
43. Have minimal interactions with other drugs and medications
|
—
|
Convenience
|
2. Be able to adjust dosage to psoriasis severity
|
—
|
|
6. Be simple and quick to administer
|
—
|
|
7. Be able to take with or without a meal
|
—
|
|
13. Not be very time-intensive or consuming
|
—
|
|
16. Have an alternative to injections
|
—
|
|
18. Be easy to include in a patient's normal routine
|
—
|
|
31. Not be messy
|
—
|
|
32. Be convenient and easily managed by a patient
|
—
|
|
33. Be easy to administer
|
—
|
|
42. Be painless to use
|
—
|
Administration/Mode of Administration
|
9. Not require multiple drugs
|
13. Require less frequent applications throughout the day
|
|
11. Be a medication implanted just below the skin
|
14. Is effective using once to twice weekly
|
|
21. Be an oral medication taken once a day
|
15. Be a once a day application that lasts
|
|
23. Be a topical ointment or cream
|
16. Be a once a month injection
|
|
24. Not require lab monitoring
|
25. Be easy to apply to all areas which need treatment
|
|
25. Be a once monthly injection
|
26. Come with gloves for application
|
|
29. Have infrequent dosing
|
31. Be a pill for long term use
|
|
39. Be an oral medication
|
—
|
Application Characteristics
|
—
|
7. Absorb invisibly in my skin
|
|
—
|
24. Not affect the way my hair looks
|
|
—
|
32. Not feel oily on the skin
|
|
—
|
33. Be gentle enough for daily use on my scalp
|
|
—
|
36. Not leave my hair oily
|
|
—
|
37. Not damage clothing
|
Treatment Preferences
|
—
|
18. Be a non-steroidal cream
|
|
—
|
19. Have a hands free applicator
|
|
—
|
20. Double as a safe mechanism for scratching
|
|
—
|
21. Be easy to apply to the scalp
|
|
—
|
23. Not be greasy
|
|
—
|
27. Be an exfoliating solution that can get rid of the scale and flakes
|
|
—
|
28. Be a liquid to give immediate relief
|
|
—
|
29. Be gentle enough to use daily without damaging my hair
|
|
—
|
30. Not involve placing oily drops in my ears
|
Prevention
|
—
|
8. Clear the skin and keep it clear
|
|
—
|
3. Prevent symptoms from occurring
|
|
—
|
4. Prevent plaque buildup on my scalp
|
|
—
|
5. Work from the inside to prevent itching
|
|
—
|
6. Start with managing the triggers, like stress
|
*This concept was inadvertently retained as a duplicate. Concept numbers were assigned by the GlobalMax software after sorting and rating and are not ordered by domain category, pharmacological characteristics or alphabet. The 10 concepts bolded were selected for online survey as described later.
Pattern Matching
In Study 1, following interpretation of the cluster maps, pattern-matching diagrams were generated to compare average domain ratings between patients and clinicians (Figure 2). Overall, there was a high correlation in average domain ratings between patients and clinicians (Spearman r = 0.98). The domain of Accessibility was rated highest by both patients and clinicians. In addition, Safety and General Efficacy domains were also rated highly by both groups. Domains of Convenience and Mode of Administration, although rated lowest by both patients and clinicians, nonetheless had a higher absolute rating from patients than from clinicians.
[Figure 2 here]
Cognitive Debriefing
In Study 2, a total of 28 concepts were selected from those generated in the 2 GCM exercises for inclusion in the preliminary TAQ for the CD phase. Participants (N=20) expressed a high level of understanding (80–100%) of all concepts. With regards to relevance, patients regarded all safety- and efficacy-related concepts >80% for relevance; however, some concepts pertaining to administration or aesthetic issues scored lower, eg, “Have a hands-free applicator” was considered relevant by only n=11/20 (55%) participants. Participants chose a 4-point scale as the preferred response option for importance to the concepts, ranging from 0=not at all important to 3=extremely important. All participants who were asked (n=19) stated they would be comfortable showing their clinicians their responses to the concepts and considered their responses would help them communicate with their clinician.
Final Formulation of Treatment Acceptability Questionnaire
Of the 28 concepts cognitively debriefed, 20 were selected for inclusion in the final TAQ. Of the 8 items not included, 6 were considered the least relevant by the participants according to the CD interview (≤ 85% of participants). The other 2 items were considered possibly redundant to other included items (eg. Item 12, “clear the skin and keep it clear” redundant to item 1, “clear outside symptoms”). Combined with the 4-point response option, the TAQ was formulated as a verbal rating scale (Appendix 2). Based on participant suggestions from CD, a few items had wording modified from the original concept. Then, the TAQ was evaluated in an online survey of 198 patients with mild-to-severe psoriasis. For most items, a ceiling effect (≥25% of participants rating them 3=extremely important) was observed; in contrast, only 1 item (Item 20, “Be an injection”) displayed a floor effect (≥25% of participants rating it 0=not at all important).
Psychometric Analysis
The following psychometric analyses were conducted on the TAQ item set to inform the item performance in terms of reliability and validity of the measure.
Inter-Item Correlations
Inter-item correlations for each of the 20 items in the TAQ ranged from –0.2 to 0.85 with the majority scoring <0.7, indicating a substantial level of diversity among the items. Item pairs that correlated at >0.8 are specified in Table 3 and were examined for possible redundancy. Items 2, 3, and 4 are related but nonidentical concepts of safety and effectiveness. Items 6 and 7 are also related concepts of convenience in administration. However, despite the higher correlation, the comparison of items 3 and 10 (safe for long-term use and relieve the itching) are unrelated concepts.
Item-Total Correlations
Item-total correlations were above the accepted threshold of 0.4 in all cases except Item 15: Be effective using once-to-twice weekly (r = 0.346) and Item 20: Be an injection (r = 0.112).
Internal Consistency
The Cronbach’s Alpha for the TAQ total score was 0.895 indicating very high reliability (Table 3). The alphas for the total score with the removal of each of the 20 items of the TAQ ranged from 0.884 to 0.905, suggesting that removal of any one item does not improve the overall alpha for the total score.
Analysis of Known-Groups Validity
Known-groups validity specifically demonstrates the ability of the TAQ to discriminate across independent known groups (eg, clinical severity). Table 3 presents the analysis of known-groups validity for the TAQ using the DLQI and CGI-S from n=197 online survey patients. The TAQ total score has a range from 0 to 60, where higher scores are associated with participant’s viewing multiple treatment features and impacts as important. For the DLQI, a statistically significant monotonic relationship existed with the TAQ, where a higher mean TAQ total score was associated with increasing levels of psoriasis’ effect on a patient’s life as measured by the DLQI (ie, impairment in QoL) (P = .0072). Thus, the higher one scores on the TAQ could be associated with a lower QoL due to psoriasis. However, no statistically significant (P = .8106) difference was observed between the patients’ rating of treatment importance (mean TAQ total score) and the clinician rating of psoriasis severity groups in the CGI-S outcomes.
Table 3
Psychometric Analysis
Selected Comparisons from the TAQ with Strong Correlations (≥ 0.80)
|
Inter-item Correlations
|
Item 2: Be effective in treating psoriasis for skin and other areas
|
Item 3: Be safe for long-term use
|
0.805
|
Item 2: Be effective in treating psoriasis for skin and other areas
|
Item 4: Not lose effectiveness over time
|
0.802
|
Item 3: Be safe for long term use
|
Item 4: Not lose effectiveness over time
|
0.823
|
Item 3: Be safe for long term use
|
Item 10: Relieve the itching
|
0.842
|
Item 6: Be easy to include in your (a patient’s) normal routine
|
Item 7: Be simple and quick to administer
|
0.851
|
Analysis of Known-Groups Validity
|
Known-Group
|
TAQ Total Mean (SD) Score
(n=197)
|
P-value
|
DLQI
|
No or small effect on patient’s life
|
44.3 (12.73)
|
0.0072
|
Moderate effect on patient’s life
|
46.3 (8.38)
|
Very large effect on patient’s life
|
49.2 (7.63)
|
Extremely large effect on patient’s life
|
52.6 (6.38)
|
CGI-S
|
Mild
|
47.2 (10.65)
|
0.8106
|
Moderate
|
46.8 (9.12)
|
Severe
|
47.8 (8.85)
|
CGI-S - Clinician Global Impression – Severity; DLQI - Dermatology Life Quality Index; SD - standard deviation; TAQ - Treatment Acceptability Questionnaire