Comment on “UV Protection Habits and Preferences in Patients With Distinct Cutaneous Immune‐Mediated Inflammatory Diseases”

To the Editor,

Álvarez Bobillo and colleagues address a real and under‐studied gap: how patients with cutaneous immune‐mediated inflammatory diseases (IMIDs) actually practice UV protection, and whether their habits align with the expert recommendations issued by Passeron et al. [1, 2]. Mapping behavior against a published guideline, disease by disease, is a useful contribution, and the finding that vitiligo patients show the highest adherence while atopic dermatitis patients struggle most is clinically plausible. We wish to flag a tension between what the design can support and what the paper concludes.

First, and most importantly, the study is explicitly a pilot with 20 participants per group, yet it culminates in a normative recommendation framework (table 4) carrying formal levels of evidence and grades of recommendation [2]. This is a category error. A single‐center pilot characterizing habits cannot generate graded clinical recommendations; the evidence grades in table 4 derive from the cited external literature, not from the present data, and presenting them within this paper's results risks implying the cohort validated them. The recommendations may well be sound, but they are inherited, not demonstrated here, and the framing should make that explicit.

Second, the between‐group inferential statistics are fragile. With 20 patients per arm and six arms, the contingency cells underlying several “significant” comparisons are very small. For example, the daily‐photoprotection comparison (p = 0.242) is non‐significant, yet the narrative still ranks groups as though the differences were real; conversely, several p < 0.05 results (hat use, p = 0.003; format preference, p = 0.004) rest on cells with single‐digit counts where chi‐square approximations are unreliable and Fisher's exact test is mandatory. The paper does not consistently indicate which test was applied to which comparison. With this many simultaneous group contrasts across numerous questionnaire items, the absence of any correction for multiple comparisons means several nominally significant findings are likely to be false positives.

Third, the headline “compliance” figures conflate distinct constructs. Table 3 mixes awareness items (“80% of psoriasis patients were aware that sunburn could worsen their disease”) with behavioral items (“50% aware that regular sun exposure is recommended”), and the denominator shifts (the tetracycline recommendation is scored over only the six HS patients receiving tetracyclines, reported as “100%”). Aggregating knowledge and behavior under a single “compliance” label, and reporting 100% from a denominator of six, overstates the coherence of the data. These should be disaggregated, and small‐denominator percentages reported as raw fractions.

Fourth, the control group warrants more caution than the limitations section affords. Companions of dermatology patients are not a community sample; they are plausibly more sun‐aware than the general population through proximity to dermatological care, which would bias the IMID‐versus‐control contrasts toward the null and undercut the claim that IMID patients are distinctively more protective. The authors note this as selection bias but do not carry it into the interpretation.

None of this undermines the study's value as hypothesis‐generating groundwork. The behavioral signal—low daily‐sunscreen use, near‐absent reapplication, poor oral‐photo protection awareness—is informative and consistent with adjacent literature [3]. But the conclusions, particularly the graded recommendation table, are written with a confidence the pilot design cannot sustain. We would encourage the authors to reframe Table 4 as a proposal synthesized from prior evidence rather than an output of this cohort, to report small‐cell comparisons as exact fractions with appropriate tests, and to temper the disease‐specific compliance rankings to reflect the wide uncertainty around 20‐patient subgroups. Confirmation in an adequately powered, multi‐center sample with a community‐based control group is needed before these disease‐specific patterns guide practice.

Funding

The author has nothing to report.

Conflicts of Interest

The author declares no conflicts of interest.

Linked Articles

This article is linked to Saber et al. papers. To view this article, visit https://doi.org/10.1111/phpp.70110.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.