DLQI

DLQI in Dermatology Trials and NICE Appraisals

In NICE technology appraisals of biologic and small-molecule treatments for psoriasis and atopic dermatitis, the Dermatology Life Quality Index acts as a gatekeeper. DLQI thresholds are written into the rules on who may start treatment and who may continue, so the questionnaire decides which patients enter the treated arm of a cost-effectiveness model and how long they stay there. For a health economist the DLQI therefore matters less as a source of utilities, which a score without preference weights cannot supply, than as the patient-reported half of a response definition. In the dupilumab appraisal (TA534), for example, treatment continues after 16 weeks only for people whose eczema severity score has halved and whose DLQI has fallen by at least 4 points.

How the ten questions are built and scored

Finlay and Khan published the DLQI in Clinical and Experimental Dermatology in 1994. They asked 120 patients with different skin diseases about the impact of their disease and its treatment on their lives, built the questionnaire from those answers, and then gave it to 200 consecutive new patients attending a dermatology clinic.

The developer's page at Cardiff University describes the DLQI as 10 questions about the impact of skin disease on quality of life over the last week, designed for adults aged 16 and over. Each question is scored 0 to 3, the item scores are summed, and the total runs from a minimum of 0 to a maximum of 30. A higher score means quality of life is more impaired. An answer of "not relevant" is scored 0, the same as "not at all".

$$\text{DLQI} = \sum_{i=1}^{10} s_i, \qquad s_i \in \{0, 1, 2, 3\}$$

where $s_i$ is the score on question $i$ and the sum over the ten questions gives a total between 0 and 30.

Finlay and Khan examined test-retest reliability in 53 patients over one week and reported it as high (gamma s = 0.99). Their validation sample also showed that atopic eczema, psoriasis and generalised pruritus had a greater impact on quality of life than acne, basal cell carcinoma and viral warts.

The questionnaire is copyright material. Cardiff University's licensing pages say that no permission or licence is needed when the questionnaires are used only for routine clinical purposes, that licences may be granted without charge to not-for-profit institutions and to academics for non-commercial purposes, and that any other use, including clinical trials, needs a licence with a charge. The licence terms also forbid editing or altering the index, so a study that drops or rewords items is no longer using the licensed index.

Score bands and the minimal clinically important difference

A raw total is hard to read without anchors, so Hongbo and colleagues linked DLQI totals to a patient-rated global question about the effect of the skin condition on their life. They mailed the DLQI and the global question to 3,834 adult general dermatology outpatients and received 1,993 responses, with a mean DLQI of 4.86. They compared several candidate band sets and proposed the one below, which had a kappa coefficient of agreement of 0.489 with the global question.

A DLQI of more than 10, the threshold in the NICE secukinumab eligibility rule (TA350) described below, places a patient in the very large or extremely large effect band. Cardiff University's account of the banding research describes a score above 10 as a major effect on quality of life.

The minimal clinically important difference answers a different question, namely how large a change in one patient's score counts as meaningful. Basra and colleagues studied patients with inflammatory skin diseases in two stages, with 192 completing the first and 107 the second, and used a global rating of change in quality of life as the anchor, an anchor-based MID design. The mean DLQI fell from 9.8 to 7.4, their data gave an estimate of about 3.3 points, and earlier estimates had ranged from 3 to 5. They recommended an MCID of 4 points for inflammatory skin diseases, and Cardiff University's own account of the instrument's development describes the refinement as replacing an MCID of 5 with an MCID of 4.

Eligibility and response rules in NICE technology appraisals

In the NICE appraisals TA350, TA534 and TA681 the DLQI sits beside a clinician-rated severity score. The severity score (the Psoriasis Area and Severity Index, PASI, or the Eczema Area and Severity Index, EASI) records how much skin is affected and how badly, and the DLQI records whether the patient's life has changed. The two together define the population and the response that the appraisal committee relies on.

• Psoriasis, secukinumab (TA350). NICE recommends secukinumab only when the disease is severe, defined by a PASI of 10 or more and a DLQI of more than 10, and when standard systemic treatment and phototherapy have failed or are unsuitable. Treatment stops at 12 weeks unless there has been a 75% reduction in the PASI score (PASI 75), or a 50% reduction (PASI 50) together with a 5-point reduction in DLQI from when treatment started.
• Atopic dermatitis, dupilumab (TA534). Dupilumab is stopped at 16 weeks if the atopic dermatitis has not responded adequately, defined as at least a 50% reduction in EASI (EASI 50) and at least a 4-point reduction in DLQI from when treatment started. The committee concluded that this composite of EASI 50 with a DLQI improvement of at least 4 was appropriate for decision-making, after clinical experts explained that it was more sensitive to treatment change and more clinically relevant than EASI 75.
• Atopic dermatitis, baricitinib (TA681). The committee considered EASI 50 combined with a DLQI improvement of at least 4 the most relevant endpoint for decision-making. It also understood that the EQ-5D often fails to capture quality-of-life improvements for people with skin conditions.

The DLQI therefore acts on the cost side of these models as much as the outcome side. A biological therapy is costly to continue, and a stopping rule tied to DLQI response limits continued spending to the patients judged to benefit. The psoriasis rule and the atopic dermatitis rule use different DLQI thresholds (5 points and 4 points), so a responder in one appraisal is not automatically a responder in the other.

Why DLQI scores cannot enter a QALY calculation directly

The DLQI total is a count of item scores with no preference weights, so equal steps are not known to be equally valued and a score of 0 is not anchored to full health. It cannot serve as a health state utility value in a cost-utility analysis. In an appraisal such as TA534 the DLQI fixes who counts as a responder, while the QALY gains in the model need utilities from a preference-based measure.

When a dermatology study has collected the DLQI but no preference-based measure, a mapping function can predict EQ-5D utilities from DLQI responses. Davison and colleagues built such an algorithm from the British Association of Dermatologists Biologic Interventions Register (BADBIR), a pharmacovigilance register of patients with moderate to severe psoriasis on systemic therapies, using 22,085 observations in which the DLQI and the EQ-5D-3L were recorded on the same day. They tested six regression methods that predict the EQ-5D-3L index directly and two that predict responses to each EQ-5D-3L domain. The two measures were moderately correlated (r = -0.47), the best-performing model was ordinary least squares with the DLQI items, age and sex as explanatory variables, and the published tool includes algorithms that need fewer variables, such as the DLQI total alone. The general design choices behind such work are described under mapping study.

Illustrative worked example of a DLQI stopping rule in a model

All numbers in this example are illustrative and do not come from any appraisal. A cohort of 200 adults with moderate to severe atopic dermatitis starts a biologic. At the 16-week review, 120 reach EASI 50. Of these, 90 also have a DLQI reduction of at least 4 points and 30 do not. Continuing treatment for the rest of the year costs GBP 12,000 per patient, and the model sets treatment costs after stopping to zero. Compared with stopping, continuing gives an EQ-5D utility gain worth 0.20 QALYs per patient over the year in those who met both criteria, and 0.05 QALYs per patient in those who reached EASI 50 without the DLQI improvement.

1. Classifying one patient. A patient whose DLQI was 14 at baseline and 9 at week 16 has a reduction of 14 − 9 = 5 points. That passes the 4-point DLQI element of the TA534 rule and also the 5-point DLQI element of the TA350 rule, and the score now sits in the moderate effect band where the baseline sat in the very large effect band. A patient whose score fell from 14 to 11 has a reduction of 3 points and fails both.

2. Costs under each continuation rule. Under the composite rule 90 patients continue, costing 90 × 12,000 = 1,080,000. If continuation required EASI 50 alone, 120 patients would continue, costing 120 × 12,000 = 1,440,000. The extra cost of the looser rule is 1,440,000 − 1,080,000 = 360,000.

3. QALYs under each rule. The composite rule yields 90 × 0.20 = 18 QALYs. The looser rule adds the 30 partial responders, 30 × 0.05 = 1.5 QALYs, for a total of 18 + 1.5 = 19.5 QALYs.

4. Cost per QALY of continuing the partial responders.

$$\frac{\Delta C}{\Delta Q} = \frac{360000}{1.5} = 240000$$

where $\Delta C$ is the extra cost in GBP of continuing treatment in the 30 partial responders and $\Delta Q$ is the QALY gain from doing so.

Continuing treatment in people who reach EASI 50 without a meaningful DLQI improvement would cost GBP 240,000 per QALY gained in this illustration. Continuing the 90 full responders costs 1,080,000 / 18 = 60,000, or GBP 60,000 per QALY against stopping, so the partial responders are four times as costly per QALY. The DLQI component of the rule removes those patients, which is how a non-preference-based score can change an incremental cost-effectiveness ratio without ever being used as a utility. The QALY gains in steps 3 and 4 still come from a preference-based measure.

How the DLQI differs from nearby measures and concepts

The DLQI is a patient-reported outcome measure, and its content is specific to skin disease. The DLQI records health-related quality of life as the patient sees it, which separates the questionnaire from PASI and EASI, both rated by a clinician. The DLQI also differs from a generic preference-based measure such as the EQ-5D, which covers mobility, self-care, usual activities, pain or discomfort and anxiety or depression with no skin-specific content but yields utilities. Dermatology economics, the wider application of economic evaluation to skin conditions, uses the DLQI as one outcome among several.

The DLQI also has relatives from the same developers that are separate instruments. The Children's Dermatology Life Quality Index (CDLQI) is for children aged 4 to 16 and has its own scoring bands, and Cardiff University advises that its results are not combined with adult DLQI data. The Family Dermatology Life Quality Index is designed for adult family members or partners of people with skin disease, so it records a family spillover and is not a measure of the patient's own quality of life.

Limitations and common misreadings

Most problems with the DLQI in economic work come from treating its total as more precise or more general than it is. Several recur.

• "Not relevant" answers. Standard scoring treats "not relevant" as no impact, which can understate impairment in patients for whom work, sport or sexual activity is not relevant. Rencz and colleagues reported in 2020 that the DLQI-R scoring, which adjusts the total for the number of "not relevant" responses, improved the discriminatory power of the questionnaire in psoriasis, pemphigus and morphea.
• Treating the total as one dimension. A single total assumes the ten items measure one underlying construct, an assumption that validation evidence for the condition being modelled should support. Basra and colleagues reviewed 272 articles on the DLQI published between 1994 and the end of 2007, which gives a starting point for that evidence.
• Reading bands as an interval scale. The Hongbo bands rest on agreement with one global question (kappa 0.489), so a move of one band does not represent a fixed amount of quality of life, and two patients with the same change in total may have crossed different bands.
• Assuming a mapped utility is a measured one. With a correlation of -0.47 between the DLQI and the EQ-5D-3L in the BADBIR data, mapped utilities carry prediction error, and an analysis that uses them should show how far the results depend on the mapping.
• Assuming one DLQI threshold for every appraisal. Basra and colleagues recommended 4 points for inflammatory skin diseases, the 2015 psoriasis rule in TA350 uses a 5-point DLQI reduction and the 2018 atopic dermatitis rule in TA534 uses 4 points. The threshold in a model should follow the appraisal and population being modelled.

Sources

• Basra MKA, Fenech R, Gatt RM, Salek MS, Finlay AY. The Dermatology Life Quality Index 1994-2007: a comprehensive review of validation data and clinical results. British Journal of Dermatology. 2008;159(5):997-1035. doi:10.1111/j.1365-2133.2008.08832.x
• Basra MKA, Salek MS, Camilleri L, Sturkey R, Finlay AY. Determining the minimal clinically important difference and responsiveness of the Dermatology Life Quality Index (DLQI): further data. Dermatology. 2015;230(1):27-33. doi:10.1159/000365390
• Cardiff University School of Medicine. Dermatology Life Quality Index. https://www.cardiff.ac.uk/medicine/resources/quality-of-life-questionnaires/dermatology-life-quality-index (accessed 17 October 2026)
• Cardiff University School of Medicine. Children's Dermatology Life Quality Index. https://www.cardiff.ac.uk/medicine/resources/quality-of-life-questionnaires/childrens-dermatology-life-quality-index (accessed 17 October 2026)
• Cardiff University School of Medicine. Global adoption of the Dermatology Life Quality Index into clinical practice (research impact case). https://www.cardiff.ac.uk/medicine/research/impact/global-adoption-of-the-dermatology-life-quality-index-into-clinical-practice (accessed 17 October 2026)
• Cardiff University. Dermatology Quality of Life licensing, request a licence and not-for-profit licence terms. https://sites.cardiff.ac.uk/dqlf/ (accessed 17 October 2026)
• Davison NJ, Thompson AJ, Turner AJ, et al. Generating EQ-5D-3L utility scores from the Dermatology Life Quality Index: a mapping study in patients with psoriasis. Value in Health. 2018;21(8):1010-1018. doi:10.1016/j.jval.2017.10.024
• Finlay AY, Khan GK. Dermatology Life Quality Index (DLQI), a simple practical measure for routine clinical use. Clinical and Experimental Dermatology. 1994;19(3):210-216. doi:10.1111/j.1365-2230.1994.tb01167.x
• Hongbo Y, Thomas CL, Harrison MA, Salek MS, Finlay AY. Translating the science of quality of life into practice: what do Dermatology Life Quality Index scores mean? Journal of Investigative Dermatology. 2005;125(4):659-664. https://pubmed.ncbi.nlm.nih.gov/16185263/
• National Institute for Health and Care Excellence. Secukinumab for treating moderate to severe plaque psoriasis (TA350). Published July 2015. https://www.nice.org.uk/guidance/ta350
• National Institute for Health and Care Excellence. Dupilumab for treating moderate to severe atopic dermatitis (TA534). Published 1 August 2018. https://www.nice.org.uk/guidance/ta534
• National Institute for Health and Care Excellence. Baricitinib for treating moderate to severe atopic dermatitis (TA681). Published 3 March 2021. https://www.nice.org.uk/guidance/ta681
• Rencz F, Gergely LH, Wikonkál N, et al. DLQI-R scoring improves the discriminatory power of the Dermatology Life Quality Index in patients with psoriasis, pemphigus and morphea. British Journal of Dermatology. 2020;182(5):1167-1175. doi:10.1111/bjd.18435