Alopecia Areata-Quality of Life Index Questionnaire (Reliability and Validity of the Persian Version) in Comparison to Dermatology Life Quality Index.

BACKGROUND: Alopecia areata (AA) is an autoimmune disease with an incidence of 2% globally and plays a key role in the quality of life (QOL) of patients with AA. It has been recently shown that there are no sufficient disease-specific questionnaires to assess the QOL in patients with AA. AIMS: This study tried to evaluate the validity and reliability of the Persian version of AA-Quality of Life Index (AA-QLI) and compare it with the Dermatology Life Quality Index (DLQI) questionnaire.
MATERIALS AND METHODS: During 1 year, 100 individuals were enrolled in this study and asked to complete the DLQI questionnaire and AA-QLI questionnaire. First of all, we enrolled 25 individuals for evaluating the validity of the Persian version of the questionnaire, and after achieving the proper validity, 75 additional patients were enrolled in this project.
RESULTS: The results showed that the test had an appropriate validity (P < 0.001, R = 0.76), reliability (P < 0.001 , internal stability R = 0.89), and (α = 0.91). In this study, we observed that the scores of both questionnaires are quite close. In this regard, in both questionnaires, females had higher scores in comparison to males (P = 0.03), and also both of them correlated with age, age of onset of disease, and skin involvement percentage.
CONCLUSIONS: The Persian version of the AA-QLI questionnaire is valid and reliable. The QOL of AA patients needs to be considered more seriously. Psychological evaluation of patients is one of the important suggestions in this study. Copyright:

INTRODUCTION

Alopecia areata (AA) is considered as a relapsing, nonscarring, immune-mediated, and chronic condition of hair loss that could happen in all races and genders with a risk of 2% worldwide.[12] AA is classically characterized by a rapid onset, during which individuals experienced hair loss in well-circumscribed patches. This condition mostly affects the scalp or beard area, however, a complete loss of hair on the scalp could be observed in severe types.[3] Entire body involvement and nail changes might be present in some patients.[1]

Previous studies revealed that this condition could increase the prevalence of mood disorders including anxiety and depression.[4] On the other hand, studies have shown that AA is tightly correlated with psychiatric disorders. These studies have demonstrated that psychiatric diseases could increase the risk of relapse or increase the risk of AA. In this regard, stressful life conditions are considered as triggers of episodes of AA. It has been well documented that anxiety and depression induced by this chronic disorder could negatively affect disease' course via stress mediators.[5] Therefore, this disease could play a significant role in the patient's quality of life (QOL). Patients' appearance by AA could alter mental health and social communication.[67] Consequently, QOL tenaciousness has become increasingly paramount in assessing AA rigor and impacts.[8]

There are now several dermatology-concrete questionnaires such as Skindex,[9] Dermatology Life Quality Index (DLQI),[9] Dermatology QOL Scales,[10] and Dermatology Categorical QOL.[11] Some reports evaluated the effectiveness of these indexes, and they showed that AA could significantly induce impairment in QOL in patients who suffered from AA, especially in the area of mental health. However, specific disease questionnaires are better than other questionnaires for the evaluation of QOL. A disease-specific instrument for estimating QOL in AA has been suggested by Fabbrocini et al., in 2013, as named AA-Quality of Life Index (AA-QLI).[6]

Assessment of QOL of Iranian patients with AA with a specific instrument has not been performed. In this regard, in the current manuscript, we tried to evaluate the effectiveness of the AA-QLI questionnaire in Iranian patients with AA.

MATERIALS AND METHODS

After a definite clinical diagnosis of AA by a dermatologist, the design was explained to patients, and patients entered our study. We considered at least 25 patients to appropriately analyze the construct validity, reproducibility, responsiveness, and ceiling/floor effect.

This study was conducted on 100 patients with AA at Razi Dermatology Hospital affiliated to Tehran University of Medical Sciences (TUMS), Tehran, Iran, from August 2017 to August 2018. Patients completed AA-QLI and DLQI questionnaires.

The inclusion criteria were all patients who were above 16 years old, and the exclusion criteria were the patients who had any other stigmatizing disorders or disabilities such as visible cutaneous deformities and mental retardation or psychiatric disorders which prevent the right judgment of the person.

For the translation and cross-cultural adaptation of the AA-QLI questionnaire, we used WHO recommendations.[12] This study was conducted in two steps:

The AA-QLI questionnaire for the first time in Iran was translated by forward–backward translation method by two fluent English speakers who were blind, and after validation and reliability, then, Persian version was translated to English by another bilingual person who had not read the original form of a questionnaire (back-to-back translation). Then, the final form of the English-translated version questionnaire was matched with the original English version. The differences were found and reviewed by translators, and some words in the Persian version were replaced by new words to take the main meaning of the words of the original AA-QLI questionnaire. Finally, a psychiatrist, a dermatologist, and a community and preventive medicine specialist confirmed the final version of the questionnaire.

The questionnaire has two sections. In the first section, questions about sex, age, educational level, age of onset of the disease, the time of the last relapse, the extent of the scalp involvement, duration of the disease, anatomical location, and type of disease, nail involvement, the time between disease onset and starting the treatment with diphencyprone, smoking, and history of previous illness were questioned. The second part (AA-QLI) has 21 questions that measure alopecia effects on a person. Patients answered questions based on the effect that illness had on the person last month. These questions are divided into three parts: (1) subjective symptom, (2) relationship, and (3) objective sign. The patient would answer each question as 1 (no matter) to 4 (fully involved), which results in a total score of “21–84.” The higher the score marks the higher impact of AA in QOL.

The DLQI questionnaire has been used to assess the QOL of the patients with skin disease for several years, and patients respond to the questions given the impact of the illness in a recent week. DLQI, which was presented by Finlay and Khan,[9] is a self-explanatory survey which comprises ten inquiries. The DLQI is determined by summing the score of each inquiry bringing about a conceivable score of 0–30. The higher score means the lower QOL. Furthermore, a valid Persian of this questionnaire was used to evaluate the QOL in patients with AA.[13] The questions can be classified under six headings items: symptoms and feelings (questions 1–2), daily activities (questions 3–4), leisure (questions 5–6), and personal relationships (questions 8–9), each item with maximum score 6, and work and school (question 7) and treatment (question 10), each item with maximum score 3. In order to help the clinical interpretation of the DLQI scores, a banding system has been validated. According to this system, DLQI scores 0–1 = no affect at all, 2–5 = small effect, 6–10 = moderate effect, 11–20 = very large effect, and DLQI score of 21–30 = extremely large effect on patient's life.

For the pilot study, 25 patients who were referred to the AA clinic of Razi Hospital and agreed to participate in this study were selected. These patients filled the AA-QLI questionnaire in the first session and 2–3 weeks later. In these two times, the main investigator of the study answered all of the comments and questions of the patients.

After the first pilot study, essential revisions were made based on the opinion of the research team on findings (e.g., change the arrangement of complicated sentences and use the simple words that were more comprehensible).

For statistical analysis, SPSS version 24 (Chicago, IL, USA) was used. Continuous variables were described using mean and standard deviation. Furthermore, demographical data were reported as frequencies and percentages. For comparison of continuous variables, we used independent samples t-test for normally distributed data and nonparametric Mann–Whitney test for variables showing skewed distribution. For qualitative variables, the Chi-square test was used for each contingency table. The correlation between continuous variables was assessed using Pearson's and Spearman's correlation coefficients. Furthermore, data with a P value of lower than 0.2 were included in the multivariable linear regression model to adjust the effect of potentially confounding factors. It should be noted that in all analyses, P < 0.05 was considered as a statistically significant difference.

This study has been performed as the thesis project for the medical degree, and ethical approval was provided by the TUMS ethics committee.

RESULTS

The analysis of the internal consistency of every domain of pilot data, Cronbach's alpha method, was used to estimate the reliability of the questionnaire. The final reliability coefficient for the questionnaire was estimated 0.7 which showed high internal consistency. In evaluating the construct validity and identifying the factors in the questions, exploratory factor analysis and then a confirmatory factor analysis were used first. According to the Kaiser-Meyer-Olkin Measure Scale (80%) and the significance of Bartlett's test of sphericity at a 99% level, data adequacy was confirmed for factor analysis.

These indicators indicate the appropriateness of the test psychometric properties. Two factors were named from the correlation matrix between the extraction questionnaire and the varimax rotation. To re-evaluate the construct validity of this scale, factor analysis was performed the test had an appropriate validity (P < 0.001, R = 0.76), reliability (P < 0.001, internal stability R = 0.89), and (α = 0.91).

One hundred patients with the mean age of 29.24 ± 8.31 participated in this cross-sectional study. In the first part of this study, we analyzed the demographical data [Table 1]. We observed that 65 individuals (65%) were male and 35 individuals (35%) were female. The Severity of Alopecia Tool (SALT) score was 47.11 ± 26.49. The mean age of disease onset was estimated at 22.09 ± 9.10, and the mean period of treatment was 9.07 ± 7.99 months [Table 1].

Table 1

Baseline characteristics of alopecia areata patients

	AA patients, n (%)
Gender, n (%)
Female	35 (35)
Male	65 (65)
Age (years)*	29.24±8.31
Duration of disease (years)*	22.09±9.10
Duration of treatment (months)*	9.07±7.99
The distance between the lesions and treatment (months)*	44.87
SALT score*	47.11±26.49
Nail involvement	24 (24)
Underlying diseases	20 (20)
Marital status
Single	42 (42)
Married	58 (58)
Divorced	0
Educational level
Elementary	4 (4)
intermediate education	11 (11)
Diploma	43 (43)
Associate degree	19 (19)
Licentiate’s degree	18 (18)
Master of science and higher	5 (5)
Smoking status
Yes	22 (22)
No	78 (78)
Positive family history	27 (27)

*age of disease onset. n – Number; SALT score –Severity of Alopecia Tool score

The mean total DLQI score was 10.69 ± 5.93. Furthermore, the total mean score of AA-QLI was 48.04 ± 9.75 including the mean score of 21.54 ± 5.37 in subjective symptoms, 19.58 ± 4.8 in the relationship, and 6.95 ± 2.32 in subjective part.

In this study, male individuals had a score of 10.71, and females had a score of 11.66 in total DLQI (P = 0.03). In a relationship and objective parts of AA-QLI and QLI total, t-test analysis showed that the mean score of females is significantly higher than males, respectively (P = 0.04, 0.04, and 0.03) [Table 2].

Table 2

Association between Alopecia Areata-Quality of Life Index scores and Dermatology Life Quality Index total scores and basic characteristics in alopecia areata patients

	AA-QLI				DLQI total
	Subjective	Relationship	Objective	Total
Sex
Female	23.43±4.83	20.57±4.27	8.71±2.07	49.7±18.32	11.66±5.66
Male	21.6±5.68	19.53±5.15	7.03±2.33	47.22±10.51	10.7±6.11
P	0.05	0.04	0.04	0.03	0.03
Marital status
Single	21.29±5.66	20.88±5.43	6.67±2.29	48.82±11.57	11.26±6.04
Married	21.58±5.13	18.6±4.19	7.12±2.2	47.3±8.24	10.25±5.91
P	0.7	0.02	0.3	0.4	o. 4
Educational level
Elementary	29.5±7	20±7.12	8.25±0.96	57.75±14.71	12±7.62
intermediate education	20.73±1.79	20.55±4.39	4.73±2.61	46±5.51	7.45±6.25
Diploma	22.26±5.74	19.56±5.78	6.95±2.16	48.77±11.92	12.19±5.66
Associate degree	20.21±4.5	20.26±2.47	7.47±1.39	47.95±5.23	9±4.77
Licentiate’s degree	18.94±4.22	17.33±3.65	7.28±2.61	43.56±6.2	8.83±5.77
Master of science and higher	25.20±5.59	22.80±4.27	7±1.41	55±8.8	17±3.87
Total	21.54±5.37	19.58±4.84	6.92±2.23	48.04±9.76	10.69±5.63
P	0.002	0.2	0.01	0.04	0.007
Type
Patchy alopecia	19.41±4.39	18.59±4.88	7.22±2.47	44.86±7.8	12.19±5.66
Ophiasis	20.6±7.67	19±5.61	6.6±2.19	45.22±9.38	9±4.77
Alopecia totalis	21.09±5.33	17.59±3.39	6.18±2.06	46.2±14.38	7.45±6.25
Alopecia universalis	23.56±5.26	21.49±4.86	7.12±2.11	52.17±9.22	12±7.62
Total	21.54±5.37	19.58±4.84	6.92±2.23	48.4±9.76	8.83±5.57
P	0.009	0.08	0.3	0.05	0.9
Nail involvement
Yes	22.58±5.7	21.29±5.45	7.58±1.50	51.46±10.40	10.46±5.37
No	21.21±5.26	19.04±4.54	6.71±2.39	46.96±9.36	10.76±6.13
P	0.2	0.04	0.05	0.04	0.8
Family history
Yes	20.22±5.92	20.26±4.6	6.81±2.86	47.3±9.61	10.11±5.96
No	22.03±5.12	19.33±4.93	6.96±2.06	48.32±9.86	10.90±5.94
P	0.1	0.3	0.7	0.6	0.5
Underlying disease
Yes	23.95±6.04	20.48±6.35	7±2.24	51.43±12.3	11.48±5.68
No	20.90±5.03	19.34±4.37	6.90±2.25	47.14±8.84	10.48±6.01
P	0.02	0.3	0.8	0.07	0.4

*P<0.05 is considered significant. DLQI – Dermatology Life Quality Index; AA-QLI – Alopecia Areata-Quality of Life Index

In a relationship and QLI total, t-test analysis showed a significant difference between patients with and without nail involvement (P = 0.04).

In the subjective part of the AA-QLI score, the analysis showed a significant difference among various types of alopecia (P = 0.009) but not in the relationship, objective, and total parts. Furthermore, the analysis failed to reveal any significant difference among various types of alopecia based on DLQI [Table 2].

Finally, we observed that DLQI total scores were correlated with age, age onset of disease, SALT score, and the time long between the onset of disease and the beginning of treatment [Table 3]. Evaluation of the correlation between QOL scores based on the DLQI questionnaire and AA-QLI questionnaire was shown [Figure 1].

Table 3

The correlation between the overall score of the Dermatology Life Quality Index and Alopecia Areata-Quality of Life Index questionnaires and the independent variables

	AA-QLI				DLQI total
	Subjective	Relationship	Objective	Total
Age
Pearson correlation	0.7	0.6	0.55	0.6	0.6
P	0.02	0.02	0.01	0.01	0.01
Age of onset of the disease
Pearson correlation	0.6	0.5	0.8	0.8	0.63
P	0.03	0.02	0.04	0.04	0.02
SALT score
Pearson correlation	0.5	0.8	0.42	0.45	0.35
P	0.04	0.03	0.02	0.02	0.03
Treatment duration
Pearson correlation	0.8	0.34	0.41	0.5	0.45
P	0.2	0.2	0.3	0.3	0.2
The distance between the lesions and treatment
Pearson correlation	0.34	0.6	0.4	0.5	0.48
P	0.03	0.04	0.01	0.02	0.04

SALT score –Severity of Alopecia Tool score; DLQI – Dermatology Life Quality Index; AA-QLI – Alopecia Areata-Quality of Life Index

Figure 1

Evaluation of the correlation between the quality of life score based on the Dermatology Life Quality Index questionnaire and Alopecia Areata-Quality of Life Index questionnaire

DISCUSSION

AA is an autoimmune disease with an incidence of 2% globally, which significantly impairs the QOL of affected individuals. It has been well documented the correlation between AA and psychological disorders, as they can be a trigger for initiation or exacerbation of AA.[414151617] This study tried to show the impact of AA on the patient's QOL and evaluate the validity of the AA-QLI questionnaire.

In a meta-analysis that was run by Rencz et al.,[18] studies were analyzed, representing a total of 2530 adult AA patients. AA-QLI, AA-QOL, and AA Symptom Impact Scale were evaluated in this study. The mean pooled DLQI score of AA patients was 6.3. Compared to age- and gender-matched controls, the meta-analysis revealed significantly reduced QOL across role-emotional, mental health, and vitality domains (P < 0.001). However, this study suggested that controversial results were found regarding the association between QOL questionnaires. The newly developed AA-specific measures seem very promising; however, a more extensive assessment of validity and reliability is needed.

Our study found that the mean total DLQI score was 10.69 ± 5.93 which showed moderate to the very large effect of AA on the patient's QOL.

Our findings showed that females with AA suffered from psychological disorders more common than males in both DLQI and AA-QLI questioners. This could be explained by the fact that appearance is more important for females than males and change of it would be more distressing for females. However, based on the review of Alzolibani et al., men seem to be more severely affected by AA than women and also had more prolonged disease duration.[19] Female patients had higher total DLQI scores and more impairment in symptoms and feelings, leisure, and treatment parameters in our study. Furthermore, in multivariable analysis, there was not any significant association between sex and disease duration with total DLQI scores. Unlike our findings, Al-Mutairi and Eldin revealed that sex and duration of illness did not impact total DLQI scores.[20]

Our result revealed that patients with lower educational levels had higher scores both in total DLQI and AA-QLI, unlike Abedini et al. findings.[21]

Furthermore, we showed that the mean scores could significantly alter in AA-QLI based on the type of alopecia. As it turned out from this project, the progression of the disease from patchy to universalis could lead to a decrease in QOL scores in the AA-QLI questionnaire.[212223]

A 7-year study by Al-Mutairi and Eldin showed that there was a significant correlation between the mean DLQI score and the severity of the disease, and this study suggested that severe forms of AA had a major impact on the psychosocial well-being of the patients.[20] Ghajarzadeh et al. (2011) showed that the mean DLQI score of AA patients was 6.4, and also, they found higher DLQI scores in psoriasis compared to AA and vitiligo.[24] Furthermore, these findings suggest that life is impaired in AA. Apart from clinical severity, other variables associated with higher DLQI scores were sex and disease duration in bivariate analysis. In this study, we showed that both DLQI and AA-QLI were correlated with genders, age, age onset of disease, and time long between the onset of disease and the beginning of treatment. These results are in line with past studies which are revealed that these items could play a key role in both DLQI and AA-QLI. Furthermore, in this study, we revealed that the Persian version of AA-QLI had an appropriate validity (P < 0.001, R = 0.76), reliability (P < 0.001, internal stability R = 0.89), and (α = 0.91).

Analysis in AA-QLI relationship showed a significant difference between single and married patients and also with and without nail involvement, but DLQI total score analysis failed to show any significant difference between these groups. Fabbrocini et al. research revealed that relationship has a major impact than subjective symptoms and objective signs on the QLI, which is inconsistent with our findings.[6] Our results revealed that subjective symptoms and objective signs had a higher correlation with the DLQI total score.

AA-QLI questionnaire was used for the first time in Iranian patients to assess the QOL and evaluating the validity and reliability. Therefore, large sample size and multicentric studies should be carried out at different times of the disease recurrence to provide a more accurate assessment of patients in terms of different aspects of the questionnaire.

CONCLUSIONS

Based on the present study, the Persian version of AA-QLI for estimating the QOL of patients with AA was developed. There was a significant correlation between the mean AA-QLI and DLQI scores. Female patients and lower educational levels had higher total scores. Psychological evaluation of patients is one of the important suggestions in this study.

Financial support and sponsorship

This study was supported financially by the Tehran University of Medical Sciences, Tehran, Iran.

Conflicts of interest

There are no conflicts of interest.