The Effectiveness of Home‐Based Phototherapy in Vitiligo: A Systematic Review and Meta‐Analysis

ABSTRACT

Background

Phototherapy with narrowband ultraviolet B (UVB) is an effective treatment for vitiligo. However, it requires frequent in‐office visits. Self‐administered home‐based UVB (HBUVB) therapy with a handheld or panel device is an alternative that has received limited research attention. The purpose of this systematic review and meta‐analysis is to evaluate the effectiveness and safety of HBUVB in the treatment of vitiligo compared to in‐office UVB (IOUVB).

Methods

Searches were conducted on Medline, Cochrane and Embase from inception until 4th of August, 2024. Studies with primary data of patients being treated with a hand‐held UVB device were included in this study. Efficacy as measured in number of patients achieving > 50% or > 75% repigmentation was the primary study outcome. Secondary outcomes analyzed were cost, adverse events and adherence.

Results

Eighteen studies were included with a total of 1341 vitiligo patients. Of these, four studies (148 HBUVB and 143 IOUVB) could be used for the meta‐analysis. HBUVB was not inferior to IOUVB at achieving > 50% (OR 1.04, 95% CI 0.58–1.87) and > 75% repigmentation (OR 1.26, 95% CI 0.65–2.42). Home‐based UVB was not associated with an increase in risk for either erythema or burning sensations when compared to IOUVB. There was a 14% lower rate of discontinuation amongst patients using HBUVB.

Conclusions and Significance/Impact

Home‐based UVB therapy is not inferior to in‐office UVB therapy in efficacy or rate of adverse events and demonstrates favourable treatment adherence. This meta‐analysis provides evidence supporting HBUVB as a safe and effective alternative to IOUVB in patients with vitiligo.

Trial Registration

PROSPERO registration number: (CRD42023478123)

1. Introduction

Vitiligo is a chronic depigmenting skin disorder that causes non‐scaly white patches on the skin, mucosa and hair due to a selective loss of melanocytes. Its pathogenesis is multifactorial and involves oxidative stress and subsequent autoimmune destruction [1, 2]. Vitiligo affects approximately 0.5%–2% of the world's population, with a substantial number of patients not receiving a formal diagnosis [3]. Whilst the disease does not have a significant impact on a patient's physical health, it can create a significant mental and psychosocial burden, especially for those with darker skin [4]. Narrowband UVB phototherapy is a central component of vitiligo treatment. It is usually administered two or three times per week in a healthcare provider's office or clinic with the use of a phototherapy booth [5]. In‐office UVB therefore requires a significant time investment by patients and may not be accessible to those living far away from such clinics.

Home‐based UVB using a portable hand‐held device is a convenient alternative that may improve patient compliance and reduce the workload for clinic staff [6]. A meta‐analysis found that home‐based phototherapy was significantly more effective at maintaining treatment adherence compared to institutional UVB, though there is a paucity of research on the efficacy and safety of home‐based UVB and hand‐held UVB devices.

This systematic review and meta‐analysis aims to review the efficacy, adherence and safety of HBUVB compared with IOUVB.

2. Methods

2.1. Registration and Study Protocol

This study was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines (Data S1).

2.2. Search Strategy

Cochrane Library, OVID Medline and Embase were systematically searched from inception until August 4th, 2024 using the terms ‘vitiligo’, ‘home‐based’ or ‘hand‐held’ and ‘UVB’ in addition to any synonyms. The reference lists of included studies were further screened to identify any potentially relevant studies. All studies with primary data such as case reports, cohort studies and randomised control trials studying vitiligo patients being treated with a home‐based phototherapy machine were included. All reviews and animal studies were excluded.

2.3. Study Selection

The study selection process comprised two stages. Initially, the titles and abstracts of all records were screened independently by the two reviewers (F.X. and H.H.). In cases where the eligibility of a study remained unclear, the full text of the article was retrieved for further assessment. Potentially eligible full‐text articles were obtained and subjected to screening. Any discrepancies between reviewers were resolved through discussion, and if needed, the input of two additional reviewers (C.C. and Z.L.) was sought. For all excluded records in the full‐text review, a documented reason for exclusion was provided (Figure 1).

FIGURE 1.

PRISMA flow diagram.

2.4. Data Collection

Data was extracted by two reviewers (F.X. and H.H.) independently using a standardised spreadsheet. Extracted data included study design, demographic characteristics (age, gender, ethnicity, comorbidities), clinical characteristics (vitiligo type, treatment type) and outcomes data (50% and 75% repigmentation).

2.5. Risk of Bias

Risk of bias of non‐randomised control trials (RCT) was assessed using the ROBINS‐I tool, whilst the risk of bias of RCTs was assessed with the RoB 2 tool (Tables S1 and S2).

2.6. Statistical Synthesis

The proportion of patients achieving > 50% and > 75% repigmentation was obtained and a comparison was made between the home‐based UVB group and the in‐office UVB group using random‐effects meta‐analysis as described by Borenstein et al. [7]. The alpha level was set at 0.05. All analyses were performed on R statistical software. Per‐protocol analysis was also performed alongside Intention‐to‐Treat (Table S4).

Adherence for meta‐analysis was measured by subtracting the number of patients who ceased therapy or were lost to follow up from the total number of patients in each experimental group.

3. Results

HBUVB was assessed in 18 studies including a total of 974 patients. Of these, 5 were randomised controlled trials, 6 were prospective cohort studies, 2 were retrospective studies, and 4 were qualitative surveys. Patients were followed up over a range of 3 to 12 months and were most commonly assessed for repigmentation, adverse effects and compliance (Table 1).

TABLE 1.

Summary of all included studies.

Paper	Design/randomised (Country)	Type/activity/chronicity	BSA affected	Locations assessed	n HBUVB/n control	Intervention/HBUVB device	Control	Outcome	Follow‐up length
Total					974/426
Eleftheriadou, 2014 [8]	RCT/Y (United Kingdom)	NSV/spreading or stable/NS	< 25%	Face, neck, trunk, upper limbs, lower limbs, hands, feet	19/10	Hand‐held (Dermfix 1000, Waldmann NB‐UVB 109)	Placebo light	Primary: proportion of patients willing to be randomised Secondary: proportion of patients expressing interest in the trial and eligible to be included, discontinuation rate, treatment satisfaction, adherence to treatment, adverse events, percentage repigmentation, DLQI and cDLQI, 5‐point likert Global improvement in vitiligo, Patient Benefit Index (PBI), colour match of newly pigmented vitiliginous lesions	16 weeks
Tien Guan, 2015 [6]	RCT/Y (Singapore)	Localised/Stable/NS	NS	Face, trunk, upper limbs, lower limbs	22/22	Hand‐held (Daavlin Dermapal system, 5.5 mW/cm2, distance 5 cm)	Institute Excimer	Primary: 1%–75% Repigmentation scale Secondary: compliance, adverse events	12, 24 (3, 6 months)
Liu, 2019 [9]	RCT/Y (China)	Localised/NS/New‐onset (< 3 months)	< 5%	Face, neck, trunk, hands, feet	61/61	Hand‐held (Sigma SH1b)	Hospital NB‐UVB	Primary: percentage repigmentation Secondary: VitiQoL index, adverse events, cost of treatment	20 weeks
Thomas, 2021 [10]	RCT/Y (United Kingdom)	NSV/Active/> 12 months	< 10%	Face, neck, Hands, feet	175/173	Hand‐held +/− topical steroids (Dermfix)	TCS + Placebo light	Primary: Patient‐reported treatment success at 9 months using Vitiligo Noticeability Scale (VNS) Secondary: onset of treatment response assessed by investigators, % repigmentation, quality of life, disease‐specific quality‐of‐life (VitiQoL, Skindex 29), generic quality‐of‐life, adverse reactions, time burden of treatment	12, 24, 36 weeks (3,6,9 months)
Sach, 2021 [11]	Cost‐Analysis (United Kingdom)	N/A	N/A	N/A	N/A	Hand‐held +/− topical steroids (Dermfix)	TCS + Placebo light	Cost analysis for Thomas, 2021 [10]	12, 24, 36 weeks (3,6,9 months)
Lei, 2024 [12]	RCT/Y (China)	NSV/Stable/NS	< 1%	Neck, trunk, back, upper limbs, lower limbs	121 (62 reverse irradiation/59 conventional irradiation)	Reverse Perilesional Irradiation with Sigma SC2B Fluorescent Lamp Tube	Normal irradiation technique	Primary: percentage repigmentation Secondary: change in leukotrichia, Patient‐related VNS score, Treatment related adverse events	12 weeks
Goel, 2012 [13]	Cohort (Prospective)/N (India)	SV, NSV, generalised and focal/NS/NS	NS	Face, trunk, hands, feet, knee, elbow, lip	50/0	Hand‐held (Philips PLS 9 W/01, Dermaindia Spot Phototherapy Unit)	N/A	Primary: percentage repigmentation Secondary: adverse events	6 months
Khandpur, 2020 [14]	Cohort (Prospective)/N (India)	NSV, localised and generalised/stable/NS	< 2%	Face, neck, trunk, upper limbs, lower limbs	10	Hand‐held (V‐Care Meditech UV Comb)	N/A	Primary: percentage repigmentation Secondary: Investigator Global Assessment (IGA), Patient Global Assessment (PGA), Dermatology LQI	24 weeks (6 months)
Shan, 2014 [15]	Cohort (Prospective)/N (China)	NS/NS/NS	NS	Face, neck, trunk, upper limbs, lower limbs, hands, feet	93	Hand‐held (Shanghai Sigma High‐Tech Co SS‐01 UV phototherapy instrument)	N/A	Primary: percentage repigmentation Secondary: adverse events	3, 6, 9,12 months
Singh 2023 [16]	Cohort (Prospective)/N (India)	Localised/stable/NS	< 2%	Head, neck, trunk, upper limb, lower limb	17/14	Hand‐held (V‐Care Meditech UV Comb)	Hospital NB‐UVB	Primary: percentage repigmentation Secondary: quality of life, adverse events, VIS‐22, Patient‐Global‐Assessment (PGA), Investigator Global Assessment (IGA)	16 weeks (4 months)
Zhang, 2019 [17]	Cohort (Prospective)/N (China)	Generalised/Stable/> 6 months	> 2%	Face, trunk hands, feet	48/46	Hand‐held (Shanghai Sigma High‐Tech Co SS‐05AB/PT8/SH1B/SH2B/SH4B)	Hospital NB‐UVB	Primary: percentage repigmentation Secondary: BSA affected, VASI‐reverse rate, VitiQOL, adverse events	24 weeks (6 months)
Ardiana, 2021 [18]	Cohort (Prospective)/N (Indonesia)	NS/NS/NS	NS	N/A	9	Hand‐held NBUVB (Dermapal Daavlin)	N/A	Primary: Minimal Erythema Dose Measurements	N/A
Cline, 2019 [19]	Cohort (Retrospective)/N (USA)	N/A	N/A	N/A	27	Hand‐held (Clarify Medical Home Light Therapy System)	N/A	Primary: adherence (number of treatments administered divided by number of opportunities available), early adherence (calculated by the number who completed at least 7 of 20 treatments)	N/A
Dillon, 2017 [20]	Comparative (Retrospective)/N (USA)	N/A	N/A	N/A	9/9	Multipanel (Panasol 3D NB‐UVB phototherapy unit)	Hospital NB‐UVB	Treatment time, cost, BSA affected	24 weeks (6 months)
Nguyen, 2020 [21]	Retrospective questionnaire/N (USA)	N/A	N/A	N/A	17	Home‐Based NB‐UVB Type Not Specified (NS)	N/A	Patient‐reported use patterns, adverse events, satisfaction (10‐point Likert scale), compliance	N/A
Wind, 2010 [22]	Retrospective questionnaire/N (Netherlands)	NSV/NS/NS	N/A	N/A	57/32	Multipanel (Panasol 3D NB‐UVB phototherapy unit)	Hospital NB‐UVB	Adverse events, time until onset of repigmentation, satisfaction of treatment, compliance	N/A
Haykal, 2006 [23]	Survey/N (Canada)	N/A	N/A	N/A	2	Multipanel and Hand and Foot (SolArc Systems Eight‐Bulb Stand Up Unit/Hand and Feet Unit)	N/A	Reason for picking home‐based phototherapy, satisfaction, patient‐assessed effectiveness, adverse events	N/A
Xu, 2021 [24]	Survey/N (China)	N/A	N/A	N/A	237	Home‐Based NB‐UVB Type Not Specified (NS)	N/A	Primary: progression of vitiligo Secondary: anxiety, depression, stress, education level, course of disease, treatment willingness, gender, age	N/A

3.1. Efficacy

Nine of 18 studies reported data on efficacy as percentage repigmentation, with a total of 442 patients (308 patients receiving HBUVB and 134 patients in a control group) (Table 2). Overall, > 50% repigmentation was achieved in 122/308 (39.6%) of patients treated with HBUVB. Six of the 10 studies had a control arm which was either IOUVB (n = 4) or placebo light (n = 2). The four studies with IOUVB patients in the control group were meta‐analysed (Figure 2). Home‐based UVB had similar efficacy and was not inferior to in‐office UVB in achieving either > 50% repigmentation (OR 1.04 95% CI [0.58–1.87]) or > 75% repigmentation (1.26 [0.65–2.42]). Efficacy for HBUVB was typically in similar areas which are more treatment responsive to phototherapy, with face and neck lesions demonstrating the best response [9]. All four studies assessed efficacy in Fitzpatrick Type III–V skin phototypes only, with most assessed patients being Type III. Despite no observed heterogeneity (I ² = 0%), this should be interpreted cautiously given that follow‐up duration varied between 12 and 24 weeks and only 4 studies were included in the meta‐analysis.

TABLE 2.

Studies reporting efficacy as calculated in percentage repigmentation at 3–6 months.

	Home‐based UVB (n = 308)					Control (n = 134)
	< 1%	1%–25%	25%–50%	50%–75%	> 75%	< 1%	1%–25%	25%–50%	50%–75%	> 75%
Eleftheriadou, 2014 [8]	6 (35.3%)	8 (47.1%)	1 (5.9%)	0 (0%)	2 (11.8%)	4 (40%)	6 (60%)	0 (0%)	0 (0%)	0 (0%)
Thomas, 2021 [10]	NS	NS	NS	NS	14 (7.0%)	NS	NS	NS	NS	8 (11.2%)
Liu, 2019 [9]	4 (7.7%)	19 (36.5%)	6 (11.5%)	11 (21.2%)	12 (23.1%)	6 (12.5%)	11 (22.9%)	7 (14.6%)	15 (31.3%)	9 (18.8%)
Tien Guan, 2015 [6]	3 (13.6%)	3 (13.6%)		5 (22.7%)	11 (50.0%)	2 (9.5%)	7 (33.3%)		4 (19.0%)	8 (38.1%)
Singh 2023 [16]	1 (5.0%)	1 (5.0%)	8 (40.0%)	1 (5.0%)	9 (45.0%)	0 (0%)	2 (18.2%)	5 (45.5%)	1 (9.1%)	3 (27.3%)
Khandpur, 2020 [14]	1 (10.0%)	0 (0%)	2 (20.0%)	6 (60.0%)	1 (10.0%)	N/A	N/A	N/A	N/A	N/A
Zhang, 2019 [17]	6 (15.8%)	2 (5.3%)	12 (31.6%)	16 (42.1%)	2 (5.3%)	4 (10%)	1 (2.5%)	17 (42.5%)	15 (37.5%)	3 (7.5%)
Shan, 2014 a [15]	17 (18.3%)	42 (45.2%)	17 (18.3%)	9 (9.7%)	8 (8.6%)	N/A	N/A	N/A	N/A	N/A
Goel, 2012 [13]	5 (10.0%)	7 (14.0%)	15 (30.0%)	23 (46.0%)	0 (0%)	N/A	N/A	N/A	N/A	N/A
Total, (%)	43 (14.0%)	143 (46.4%)		71 (23.1%)	51 (16.6%)	16 (11.9%)	56 (41.8%)		35 (26.1%)	27 (20.1%)

^a Repigmentation assessed at 3 months.

FIGURE 2.

Percentage repigmentation achieved at follow‐up (range 3–6 months) performed as Intention‐to‐Treat (ITT) analysis. (A) > 50% repigmentation and (B) > 75% repigmentation.

3.2. Adherence and Compliance

Four studies assessed adherence rates to HBUVB compared with IOUVB as measured by treatment cessation or loss to follow‐up. Home‐based UVB had an overall adherence rate of 129/144 (89.6%) and demonstrated a favourable adherence profile compared to IOUVB (83.7%) (OR 0.72 95% CI [0.16–3.18]); however, this result was not significant (Figure 3). Moderate heterogeneity was present amongst the studies assessed (I ² = 35%).

FIGURE 3.

Comparison of adherence and rate of erythema and burning sensations, Intention To Treat (ITT) analysis. Liu [9] specified that four patients were excluded due to later being found to have the wrong diagnosis and as such was not included for IIT analysis.

In the study by Tien Guan et al. [6], no patients defaulted on HBUVB and only 8% of patients failed to adhere to the intended regime compared to 30% in the IOUVB (excimer lamp) group. The Hi‐light trial calculated adherence by dividing the total number of assigned phototherapy sessions by the number of sessions that actually took place and identified a 74%–81% adherence rate with home‐based UVB [8].

3.3. Adverse Effects

Erythema and burning sensations were the most commonly reported adverse events in both the HBUVB and IOUVB groups. Overall, HBUVB demonstrated similar rates of erythema (OR 1.74, 95% CI [0.28–10.92]) and burning (1.36 [0.16–11.71]) compared with IOUVB, with other adverse events such as oedema and blistering being uncommon overall. Perilesional hyperpigmentation was reported in one study, with 10/52 (19.2%) patients in the HBUVB group experiencing excessive pigmentation and 0/48 (0%) amongst those receiving IOUVB [9]. Substantial hetereogeneity was present between studies, with the most marked being for burning (I ² = 64%). Total number of adverse events can be seen in Table S3.

3.4. Cost

Three studies reported and compared the cost of treatment. Dillon et al. [20] found the cost of HBUVB over a 1‐year period to be $4590, the purchase cost of a Panosol 3D 3‐sided full‐height UVB panel device, and this compared favourably with the cost of IOUVB over the same period, which was $21,270 (based on insurance reimbursements and patient out‐of‐pocket expenses). The difference was highly significant (p < 0.0001, 95% CI [−21,691 to −11,671]). Liu et al. reported that the one‐time payment for the Sigma SH1b hand‐held UVB unit was $202 and the cost of each treatment in hospital (described only as including the office visit charge) was about $11. The authors suggested that the cost of IOUVB would exceed the cost of HBUVB after 7 weeks of treatment for patients in a metropolitan setting in China. Economic evaluation of the Hi‐Light Trial also revealed that combination treatment with corticosteroids may be more cost‐effective in treating vitiligo compared to just corticosteroids alone for hospital services in the UK, with a cost of £1932 (~$2265) per treatment success [11].

4. Discussion

Home‐based UVB therapy with the use of a handheld device or UVB panels appears to be an effective treatment for vitiligo. It is not inferior to traditional UVB therapy in an office‐based setting and is comparable in terms of safety and rate of compliance. Narrowband UVB (NBUVB) remains a first‐line therapy for vitiligo, and a home‐based approach to delivering UVB would seem to be an important strategy for reducing barriers that might prevent timely access to this treatment [29].

The role of phototherapy in the treatment for vitiligo is multifaceted. It halts disease progression through downregulating inflammatory cytokines such as IL‐18 and C‐X‐C motif chemokine ligand 10 (CXCL10) [26], dampening CD8+ T‐cell honing and recruitment as well as inducing T‐cell apoptosis [28]. Concurrent upregulation of IL‐10 also induces expansion of immunosuppressive Treg (T‐regulatory) T‐cell population by inducing differentiation of naive CD4+ T cells via STAT3 and FOXO1 [25]. In addition, repigmentation is achieved through stimulating the migration of melanoblasts in the outer hair root sheath to the epidermis, and their subsequent differentiation into functional melanocytes [1].

The Vitiligo Working Group recommends NBUVB treatment 3 times per week with 10%–20% incremental dose escalation to achieve minimal erythema dose, up to a maximal fluence of 1500 mJ/cm² on the face and 3000 mJ/cm² on body [27]. A novel irradiation technique called ‘reverse perilesional irradiation’ involving irradiating only the perilesional skin with the area of leukoderma covered was found to be more effective in treating leukotrichia affected skin compared to a traditional irradiation technique [12].

To achieve the maximal benefit from NBUVB, prolonged and sustained treatment is needed. HBUVB devices such as the Dermfix 1000MX unit have been independently measured to emit a mean irradiance of 3.8 mW/cm², demonstrating both notable inter‐device heterogeneity and marked reduced intensity with increasing distance if not used in close contact [30]. By contrast, in‐office cabinet NBUVB devices such as the Waldmann UV 7002 cabinet can maintain an average irradiance exceeding 10 mW/cm² [31]. Consequently, to achieve the same fluence (mJ/cm² or J/cm²) as a NBUVB cabinet, treatment with HBUVB, especially handheld devices, likely require longer irradiation duration.

It is recommended that HBUVB therapy is administered under the guidance of a dermatologist or trained clinic staff, who provide education on the safe use of the device and arrange regular follow‐up and support [32]. Nonetheless, it is the patient themselves who must administer the treatment, and therefore a high level of adherence is required for optimal results to be achieved, and this necessitates a treatment that is affordable, convenient and safe. When guided appropriately, incremental dose escalation and cumulative fluence follow the same principles for both HBUVB and in‐office NBUVB phototherapy. However, because handheld devices deliver lower irradiance, the total irradiation time per exposure required to achieve an equivalent fluence is proportionally greater.

A highly attractive feature of HBUVB is its low cost compared to IOUVB. The cost of administering UVB in the office setting includes staff salary and facility costs and the cost of purchasing the UVB machine. In addition to treatment costs, there are also travel and loss‐of‐productivity costs borne by the patient. In many countries, treatment costs for phototherapy are fully reimbursed by private or public insurers. Unlike HBUVB, the cost of IOUVB accrues with the number of treatments. The cost of HBUVB therapy is mainly the cost to the patient of purchasing the device. There are no treatment costs to institutions or insurers. Full sized panels such as the Panosol 3D 3‐sided unit are not available in many countries, however smaller hand‐held devices are widely available and generally cost less than $400. There may be maintenance costs associated with owning a UVB device, however it has been observed that bulb replacement is usually not required within a year [20]. Regarding productivity loss, it is expected that home‐based treatments can be performed outside of working hours. With these factored in, HBUVB devices may prove financially advantageous to patients themselves in the long‐term [33, 34]. Given that a single session of institutional NBUVB can generate a cost of £8.50 (~$10) for the institution including both equipment running costs and staff‐related expenses, a course of treatment consisting of three sessions per week for 12 weeks would cost an estimated £306 (~$360) for the institution with over 80% of the cost being staff‐related and just ~4% relating to maintenance and energy costs [35]. In contrast, a hand‐held device such as the Dermfix 1000MX incurs a one‐off cost of approximately £220 ($258) for the patient and can be used for continuous treatment. There is a paucity of health economic research assessing the benefit of HBUVB, however considering the actual cost of treatment (rather than the out‐of‐pocket cost to the patient) HBUVB appears to offer a significant cost benefit compared to IOUVB over an extended course of treatment.

Unlike IOUVB therapy, which requires frequent visits to a dedicated clinic over a period of months or years, HBUVB is readily accessible, and therefore it overcomes the barriers presented by the need for travel. During the COVID‐19 pandemic, a survey of expert dermatologists found that 80% would be comfortable recommending HBUVB therapy for vitiligo [32]. Adherence amongst vitiligo patients to prescribed HBUVB therapy was found in one study to be 92%, compared to 70% for excimer lamp treatment in a clinic [6]. High patient satisfaction has been reported with the use of HBUVB in the treatment of many other cutaneous conditions including psoriasis, atopic dermatitis and cutaneous T‐cell lymphoma [19, 28, 36, 37, 38]. Whilst the compact form of hand‐held devices makes them highly accessible, this may impact their feasibility for patients with greater body surface area (BSA) involvement or generalised and active vitiligo due to difficulties with application. Patients included in the meta‐analysis barring Zhang et al. [17] had localised involvement only and did not assess efficacy or adherence in patients with generalised vitiligo. We therefore recommend that future studies assess the efficacy and safety of HBUVB in patients with greater BSA involvement and to explore the use of larger home‐based devices to facilitate HBUVB in these patients.

A previous meta‐analysis suggested a higher risk of adverse events when using HBUVB compared to IOUVB based on two included studies at the time [39]. This current meta‐analysis re‐analysed this data with two additional studies [16, 17] which both demonstrated that HBUVB patients are not at a higher risk of adverse events compared to IOUVB and collectively adverse effects are similar between both groups. Whilst the safety profile of HBUVB appears to be comparable to IOUVB, intentional or accidental overuse of HBUVB devices has been observed and this may result in sporadic adverse effects. The ‘overenthusiastic’ use of a handheld devices was noted by Tien Guan et al. [6] as the cause of one case of phototherapy burn in the HBUVB group. In another study, incorrect dosing frequency led to grade 3 erythema in a patient using HBUVB [1]. Overall, phototherapy burn was observed in 16% of patients receiving HBUVB compared with 10.6% of patients treated with IOUVB (OR 1.36), however due to the heterogeneity of the studies analysed, further research is needed to confirm a difference in the risk of adverse effects. Patient factors, such as the ability to follow instructions and to safely operate a HBUVB device must be taken into consideration and patients at risk of inappropriate or excessive use of HBUVB may be better suited to the controlled delivery of UVB in a hospital or clinic environment. NBUVB use has not been associated with increased long‐term adverse events such as cutaneous malignancies, though long‐term studies on HBUVB have not been performed to determine the risk of non‐supervised use [40, 41]. Therefore, HBUVB should be carefully considered for eye and genital involvement and patients should be followed up as per normal NBUVB use (e.g., every 12 weeks) to monitor for adverse events.

It is important to note that these results are limited by the heterogeneity of the studies included due to differences in device brands and modality (e.g., panel vs. hand‐held vs. excimer) and individual treatment intensity. There is a need for more research, especially into the long‐term efficacy and safety of HBUVB in the treatment of vitiligo, and its possible benefits or shortfalls compared with IOUVB. Only 16 studies in the literature met our inclusion criteria for analysis and of these, only four were randomised control‐trials (RCT). In contrast, a Cochrane review found 32 RCTs investigating the effect of phototherapy on atopic eczema [37]. Furthermore, most studies included came from a limited geographic area and assessed only patients with Fitzpatrick types III–V skin. Thomas et al. [10] was the only study to compare the difference in efficacy of HBUVB between lighter (Type I–III) and darker (IV–VI) skin types, with no difference in efficacy demonstrated. No studies have assessed differences in adverse events between lighter and darker skin types. As such, the generalisabilty of this meta‐analysis is limited primarily to patients with Fitzpatrick skin types III–V skin and there is a need for further research on hand‐held NBUVB devices in more ethnically and geographically diverse populations.

Efficacy outcomes should be interpreted with caution, as repigmentation extent is known to be greater at 24 weeks compared to 12 weeks in quartile‐based evaluations [42]. In this review, included studies had follow up durations ranging from 12 to 24 weeks but were analysed together, which may limit comparability. Standardising follow‐up periods in future trials would improve the generalisability of findings. Moreover, given that HBUVB is typically reserved for localised, stable vitiligo and IOUVB for more widespread or active disease, further research should delineate the extent of body surface area involvement and body sites in which HBUVB may demonstrate non‐inferiority to IOUVB. Finally, as adjuvant therapies such as topical corticosteroids and calcineurin inhibitors are known to enhance the efficacy of phototherapy in vitiligo [10], future studies should evaluate these combination strategies to better reflect real‐world clinical practice.

5. Conclusion

The findings of our meta‐analysis support the notion that HBUVB has similar efficacy to IOUVB and that HBUVB is relatively safe and may have favourable adherence rates. Being both affordable and convenient, HBUVB offers a crucial treatment option for individuals with vitiligo who might otherwise lack access to in‐office treatment.

Author Contributions

Firdavis Xireaili: conceptualisation (supporting), methodology (equal), literature search (lead), abstract screening (equal), full‐text screening (equal), data extraction (equal), data curation (equal), formal analysis (lead), writing of final draft (equal), revisions (lead). Hieu Ha: abstract screening (equal), full‐text screening (equal), data extraction (equal), data curation (supporting), writing of final draft (equal). Zhao Feng Liu: literature search (supporting), methodology (supporting). Lawrence Lin: literature search (supporting), methodology (supporting), abstract screening (supporting). Michelle Min Hsiao: literature search (supporting), abstract screening (supporting), full‐text screen (supporting). Adrian Mar: review of manuscript (equal), supervision (equal). Christopher Chew: conceptualisation (lead), methodology (equal), literature search (supporting), data curation (equal), formal analysis (supporting), writing of final draft (supporting), revisions (supporting), review of manuscript (equal), supervision (equal).

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Tables S1–S4: phpp70079‐sup‐0001‐TablesS1‐S4.docx.

Data S2: phpp70079‐sup‐0002‐DataS1.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.