Prescription Pattern of Systemic Corticosteroids in Dermatology: A Prospective Observational Study from a Tertiary Care Teaching Hospital

1. Introduction

Glucocorticoids are among the most extensively prescribed immunomodulatory drugs in contemporary practice. Acting through both genomic and non-genomic pathways, they exert broad anti-inflammatory and immunosuppressive effects that make them indispensable in moderate-to-severe inflammatory, autoimmune and bullous dermatoses [1]. They anchor initial therapy in pemphigus vulgaris and bullous pemphigoid [2,3], cutaneous lupus [4], severe lichen planus [5], vitiligo [6], alopecia areata [7] and antihistamine-refractory chronic spontaneous urticaria [8]. Their prolonged use, however, carries dose- and duration-dependent hazards — osteoporosis, glucose intolerance, hypothalamic–pituitary–adrenal (HPA)-axis suppression, infection, hypertension and gastrointestinal bleeding [9] — that constrain long-term prescribing.

In India, the ready over-the-counter availability of potent topical corticosteroids has fuelled widespread misuse and the now well-documented topical steroid-damaged face epidemic [10,11,12]. By contrast, comparatively little is known about how systemic corticosteroids are used in day-to-day Indian dermatology practice. Prescribing audits have been reported [13,14,15], but prospective studies that combine patient-reported quality-of-life impact with validated severity indices, formal ADR causality assessment and guideline concordance remain uncommon.

The ICMR Standard Treatment Workflows (STW) for dermatology, released in 2022 [16], offer condition-specific algorithms tailored to Indian practice, yet their real-world uptake has seldom been examined prospectively. The present study pairs an ICMR concordance audit with internationally validated patient-centred instruments — the Dermatology Life Quality Index (DLQI) [17,18], the 5-item Medication Adherence Report Scale (MARS-5) [19,20] and WHO pharmacovigilance methodology [21] — to characterise the prescription pattern of systemic corticosteroids in adult dermatology outpatients along with their indications, drug selection, dosage regimens, duration of therapy, effectiveness and safety. The primary objective was to characterise the prescribing pattern of systemic corticosteroids — including indications, drug selection, dose, route and duration — in adult dermatology outpatients at a tertiary-care teaching hospital. The secondary objectives were: (i) to assess therapeutic effectiveness using validated disease-specific severity indices (clinical improvement defined as ≥50% reduction from baseline) at day 30; (ii) to evaluate the safety profile through ADR documentation and WHO–UMC causality assessment; (iii) to quantify the impact of systemic corticosteroid therapy on health-related quality of life using the DLQI; (iv) to audit physician adherence to the ICMR STW 2022 dermatology workflow; and (v) to measure patient adherence to systemic corticosteroid therapy using the MARS-5.

2. Materials and Methods

2.1. Study Design and Setting

This prospective observational study was conducted in the Department of Dermatology of a tertiary-care teaching hospital in Western India. Eligible patients were recruited consecutively from the dermatology outpatient clinic over a six-month period and followed for 30 days from the day systemic corticosteroid therapy was initiated. The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Ethics Committee of Smt. N.H.L. Municipal Medical College, Ahmedabad, Gujarat, India (protocol code 25/114; date of approval 9 October 2025). Written informed consent was obtained from every participant prior to enrolment. The study is reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement.

Prior to enrolment, the study’s purpose, procedures, potential risks and benefits, the voluntary nature of participation, the right to withdraw at any time without affecting routine dermatological care, and the measures adopted to maintain confidentiality of personal and clinical data were explained to each participant by the investigator in a language the participant understood (English, Hindi or Gujarati). A written informed consent form, approved by the Institutional Review Board, was signed by every participant before any study-related procedure was undertaken. Consent covered both participation in the study and publication of anonymised, aggregated findings; no individually identifiable images of patients are included in this manuscript.

2.2. Study Population

Inclusion criteria: Adults (≥18 years) with a confirmed dermatological diagnosis who were prescribed any systemic corticosteroid by the treating dermatologist and provided written informed consent.

Exclusion criteria: Pregnant or lactating women; patients receiving systemic corticosteroids for a non-dermatological indication; those with documented uncontrolled diabetes mellitus, uncontrolled hypertension or active tuberculosis; and individuals unable or unwilling to provide informed consent.

2.3. Sample Size

Consecutive enrolment of all eligible patients during the six-month recruitment window determined the sample size, yielding 82 patients who met the inclusion criteria. No formal a priori power calculation was performed; the cohort was a convenience sample defined by the recruitment window, and the resulting sample size is acknowledged as a limitation.

2.4. Data Collection

A structured case-record form was used to capture demographics, comorbidities, dermatological diagnosis, prescription details (drug, dose, route, frequency, duration and tapering plan) and clinical findings. Diagnoses were consolidated into nine pragmatic categories, defined according to whether a corresponding ICMR STW 2022 workflow was available. Each participant was assessed at baseline (day 0) and again at day 30.

2.5. Outcome Measures

Clinical severity was assessed with disease-specific validated indices: PDAI [22] for pemphigus; BPDAI [23] for bullous pemphigoid; CLASI [24] for cutaneous lupus; EASI [25] and SCORAD [26] for eczematous disorders; SALT [27] for alopecia areata; VASI [28] for vitiligo; and LPSI [29] for lichen planus.

Quality of life was measured with the DLQI [17,18] at baseline and day 30 (a 10-item questionnaire scored 0–30, with higher values indicating greater impairment).

Clinical outcome at day 30 was defined using the disease-specific severity score wherever a validated index applied — improvement as a ≥50% reduction from baseline, stable disease as any change of less than 50%, and worsening as any increase in score, appearance of new lesions or treatment failure requiring escalation. For conditions lacking a validated severity index, physician clinical judgement categorised outcomes as improved, stable or worsened.

Adverse drug reactions were recorded at every visit; each episode was graded for causality using the WHO–UMC scale [30,31] and for severity as mild, moderate or severe. The action taken (therapy continued, dose modified, symptomatic treatment or discontinuation) was captured at the patient level.

Medication adherence was measured with the MARS-5 [19,20] (score range 5–25); a total of 25 was classified as high adherence and any score below 25 as low. Reasons for low adherence were elicited through open-ended questioning, and only spontaneously reported reasons were tabulated.

Post-hoc DLQI domain analysis: For every follow-up completer, the absolute reduction in each of the six DLQI subscales was calculated, and the subscale showing the largest absolute change was designated that patient’s principal domain of improvement. This patient-level analysis was performed post-hoc and is reported descriptively.

Tapering-schedule definitions were operationally defined at two levels. A prescription was recorded as having a “documented taper” whenever the prescriber specified any step-down schedule (dose reduction over defined intervals) at initiation, irrespective of pace. A prescription was classified as an “STW-concordant taper” only when the step-size and total taper duration fell within the range recommended by the ICMR STW 2022 workflow for the relevant indication; tapers that were documented but faster, slower or of different total duration than the STW recommendation were classified as documented-but-non-concordant. This distinction is preserved throughout the Results and Discussion.

ICMR concordance was determined by comparing each prescription against the corresponding ICMR STW 2022 workflow [16]. Conditions not covered by the STW — lichen planus, pityriasis rosea, pityriasis capitis, cutaneous small-vessel vasculitis, leprosy [32], Behçet’s disease, en coup de sabre and certain drug-induced reactions — were excluded from the concordance denominator.

2.6. Statistical Analysis

Data were entered into Microsoft Excel and analysed with IBM SPSS Statistics, version 26. Continuous variables are presented as mean ± SD and categorical variables as frequencies and percentages, with 95% Wilson-score confidence intervals (CIs) reported for the key proportions. The paired t-test compared pre- and post-treatment severity scores and DLQI values. Because each of the eight disease-specific severity indices applies to a distinct condition and the subgroup analyses are exploratory, no correction for multiple comparisons was applied. Between-group continuous comparisons used the independent (Student’s) t-test, with the Shapiro–Wilk test for normality, Levene’s test for equality of variances and Cohen’s d as the effect size; between-group categorical comparisons used Fisher’s exact test. A post-hoc Spearman rank correlation examined the association between total MARS-5 score and DLQI improvement (baseline minus day 30 DLQI), both overall and separately within each disease category, with disease category used only as a stratification variable; these stratified analyses were exploratory and unadjusted for multiple comparisons. A two-sided p < 0.05 was considered statistically significant.

3. Results

The flow of patients through screening, enrolment, follow-up and inclusion in the outcome, adherence and ICMR concordance analyses is presented in Figure 1.

3.1. Patient Demographics and Baseline Characteristics

A total of 82 patients were enrolled; 76 (92.68%) completed day 30 follow-up and 6 (7.32%) were lost to follow-up. Mean age was 42.4 ± 14.9 years (range 18–74), 48 (58.54%) were female and 52 (63.41%) resided in urban areas. Hypertension was documented in 10 patients (12.20%), diabetes mellitus in 8 (9.76%) and thyroid disorder and dyslipidaemia in 3 each (3.66%); overlap was present, and 61 patients (74.39%) had none of the four documented comorbidities. Prior systemic corticosteroid exposure was reported by 22 patients (26.83%). Baseline characteristics are summarised in Table 1.

3.2. Disease Spectrum

Patients were assigned to nine disease categories defined by ICMR STW 2022 workflow availability. Eczematous dermatitis (19 patients, 23.17%), immunobullous disorders (18, 21.95%) and papulosquamous disorders (10, 12.20%) were the three commonest categories; the full distribution is shown in Figure 2.

3.3. Prescription Pattern

Prednisolone was the leading agent prescribed (52/82, 63.41%), followed by Dexamethasone (9/82, 10.98%), Betamethasone (8/82, 9.76%), Methylprednisolone (7/82, 8.54%; 6 as intravenous pulse and 1 as oral pulse) and Hydrocortisone (6/82, 7.32%). The oral route dominated (70/82, 85.37%), with intravenous therapy accounting for 12/82 prescriptions (14.63%).

Doses in the medium Prednisolone-equivalent range (0.5–1 mg/kg/day) were most frequent (35 patients, 42.68%), followed by low-dose regimens (<0.5 mg/kg/day; 18, 21.95%), pulse or mini-pulse regimens (15, 18.29%) and high-dose regimens (>1 mg/kg/day; 14, 17.07%). Treatment duration was most often 4–8 weeks (28/82, 34.15%), and a tapering schedule was documented in 66 of 82 prescriptions (80.49%; 95% CI 70.6–87.6%).

3.4. Clinical Outcome at Day 30

Among the 76 follow-up completers, 56 (73.68%; 95% CI 62.8–82.3%) improved — defined as a ≥50% reduction where a disease-specific severity index was available and by physician assessment otherwise — while 16 (21.05%) remained stable and 4 (5.26%) worsened. Mean DLQI fell from 16.86 ± 4.67 at baseline to 7.24 ± 3.41 at day 30, a reduction of 9.62 points (paired t-test, p < 0.001) that comfortably exceeds the ~4-point minimal clinically important difference. Every disease-specific severity index improved significantly from baseline (paired t-test; full p-values in Table 2).

3.5. DLQI Domain-Wise Improvement

In a post-hoc patient-level review of the 76 follow-up completers, the DLQI subscale showing each patient’s largest absolute reduction was identified as their principal domain of improvement. Symptoms and Feelings led the ranking (15 patients, 19.74%), followed by Work and School (14, 18.42%), Personal Relationships (14, 18.42%), Leisure (12, 15.79%), Daily Activities (11, 14.47%) and Treatment (10, 13.16%). This analysis describes how often each domain surfaced as the leading contributor to a patient’s DLQI improvement.

3.6. Adverse Drug Reactions

Adverse drug reactions occurred in 11 of 82 patients (13.41%; 95% CI 7.7–22.4%) across 14 episodes. Weight gain (3 patients, 3.66%), hyperglycaemia (2, 2.44%) and dyspepsia or gastritis (2, 2.44%) were the commonest events; Cushingoid features, hypertension, acneiform eruption, insomnia, infection and mood disturbance were each reported once (1.22%), while menstrual irregularity affected 1 of the 48 female patients (2.08%). Per-episode WHO–UMC causality was probable in 9 (64.29%) and possible in 5 (35.71%); severity was mild in 10 (71.43%), moderate in 3 (21.43%) and severe in 1 (7.14%) — the single severe event being an infection that required hospitalisation. At the patient level, treatment was continued in 8 of 11 patients (72.73%), the dose was modified in 2 (18.18%) and therapy was discontinued in 1 (9.09%). The incidence of ADRs did not differ significantly across dose categories (Fisher’s exact test p = 0.213) or disease categories (p = 0.901); however, both comparisons were under-powered given the small number of events.

3.7. Patient-Reported Adherence (MARS-5)

The MARS-5 was completed by the 76 follow-up completers; 58 (76.32%; 95% CI 65.6–84.5%) reported high adherence (score = 25) and 18 (23.68%) reported low adherence (score < 25). Among the 18 low-adherence patients, the spontaneously reported reasons were fear of adverse effects (8/18, 44.44%), symptomatic improvement prompting self-discontinuation (5/18, 27.78%), forgetting doses (3/18, 16.67%) and financial constraints (2/18, 11.11%).

3.8. Effect of Adherence on DLQI Improvement

Substantial day 30 DLQI improvement was observed in both adherence groups. Mean improvement measured 8.97 ± 8.10 points in the high-adherence group (n = 58) and 11.72 ± 7.69 points in the low-adherence group (n = 18); the between-group difference (high − low) was −2.76 points (95% CI −7.06 to 1.55), which did not reach statistical significance (independent t-test, p = 0.206). The numerically larger improvement in the low-adherence group is most plausibly attributable to reverse causation — early symptomatic relief itself prompting some patients to reduce dosing (self-discontinuation on perceived improvement being the second most frequent reason for low adherence; see Section 3.7) — and to the higher baseline DLQI in this group (18.06 ± 4.43 vs. 16.48 ± 4.72), which afforded greater scope for absolute reduction. This finding should not be construed as evidence that incomplete adherence is safe, particularly given that 30 days is too short a window to detect relapse. Detailed comparisons are shown in Table 3.

A post-hoc Spearman analysis revealed a weak inverse correlation between MARS-5 score and DLQI improvement overall (ρ = −0.169, p = 0.145) that was not statistically significant. Interpretation is further constrained by a ceiling effect: 76.32% of patients scored the maximum MARS-5 of 25, substantially reducing rank variability. No disease-category-specific correlation reached statistical significance. The vascular-disorder subgroup could not be analysed because all three patients had identical MARS-5 scores, and the infective subgroup contained only two patients, so a meaningful p-value could not be estimated. These exploratory findings are presented in Table 4 and should be interpreted cautiously given the small subgroup sizes.

3.9. ICMR Standard Treatment Workflow Concordance

Of the 82 prescriptions, 59 (71.95%) were for conditions covered by the ICMR STW 2022 dermatology workflow; the remaining 23 — lichen planus (n = 8), drug-induced reactions (n = 6: BFDE 1, DIMR 1, DRESS 4), cutaneous small-vessel vasculitis (n = 3), Type-2 lepra reaction (n = 2), Behçet’s disease (n = 1), en coup de sabre (n = 1), pityriasis rosea (n = 1) and pityriasis capitis (n = 1) — fell outside the workflow and were excluded from the concordance analysis. Among the 59 evaluable prescriptions, 40 (67.80%; 95% CI 55.1–78.3%) were fully concordant with all core ICMR STW prescribing parameters, while 19 (32.20%) deviated on at least one parameter, with substantial overlap across categories: drug selection (n = 8), dose (n = 8), non-concordant taper regimen (n = 9) and missing baseline random blood glucose documentation (n = 12). Drug-selection non-concordance (8/59, 13.56%) spanned three condition categories: discoid lupus erythematosus (n = 5), moderate eczema (n = 2) and urticaria (n = 1); the clinical rationale and STW-recommended alternatives are addressed in the Discussion. Concordance by component is presented in Table 5.

4. Discussion

4.1. Prescribing Patterns in Context

The predominance of oral Prednisolone (63.41%) in our cohort mirrors earlier Indian corticosteroid-prescribing audits [13,14,15] and is in keeping with the chronic course of most dermatological indications and Prednisolone’s suitability for graded taper. It is encouraging that a documented tapering schedule accompanied 80.49% of prescriptions, pointing to prescriber awareness of HPA-axis suppression risk. Intravenous pulse Methylprednisolone, reserved for severe immunobullous disease, was consistent with international consensus recommendations [2,3].

4.2. Clinical Outcomes and Quality of Life

The day 30 clinical improvement rate of 73.68% is in line with previously reported Indian and international short-term outcomes for systemic corticosteroids in inflammatory dermatoses [2,4,8], and the 9.62-point DLQI reduction reflects a clinically meaningful gain [17,18]. Every disease-specific severity index improved significantly from baseline, with mean percentage reductions ranging from 20.42% (VASI, vitiligo) to 59.08% (SCORAD, atopic dermatitis); immunobullous, papulosquamous and eczematous conditions all achieved clinically meaningful improvements of approximately 48–57% (PDAI 54.99%, BPDAI 48.53%, CLASI 54.38%, EASI 56.88%, LPSI 52.11%, SALT 48.00%). The comparatively modest VASI reduction (28.30 → 22.52; 20.42%, p = 0.022) is expected given the protracted timecourse of vitiligo repigmentation, with additional improvement anticipated over subsequent months [6].

4.3. Domain-Wise Quality-of-Life Improvement

Symptoms and Feelings was most often the principal domain of improvement (15/76; 19.74%), followed by Work and School and Personal Relationships (14/76 each; 18.42%) — a pattern with a coherent clinical logic. Symptoms and Feelings captures itch, soreness, pain and embarrassment, precisely the sensory and affective burden that the direct anti-inflammatory action of systemic corticosteroids relieves most rapidly. Improvement in work, schooling and interpersonal functioning typically follows as a downstream benefit of symptomatic relief. The Treatment domain — reflecting the inconvenience and disruption of therapy itself — led in the fewest patients (10/76; 13.16%), consistent with the real-world burden of tapering, laboratory monitoring and adverse-effect concerns. Because each patient was assigned to a single principal domain, the analysis constitutes a descriptive ranking rather than a quantification of each domain’s contribution to the 9.62-point total DLQI improvement; even so, the ranking accords with the established role of systemic corticosteroids in delivering rapid relief of the most distressing features of inflammatory dermatoses.

4.4. Safety Profile

The patient-level ADR rate of 13.41% falls within published estimates for short-term systemic corticosteroid exposure (typically 10–20%) [9] and is substantially lower than rates seen with prolonged use. The leading ADRs — weight gain, hyperglycaemia and dyspepsia — reproduce the pattern reported internationally [9]. Ten of the 14 episodes (71.43%) were mild, 3 (21.43%) moderate and 1 (7.14%) severe; the single severe event (infection requiring hospitalisation) was managed successfully with antimicrobials and with no fatalities. The WHO–UMC causality distribution (probable in 9/14, possible in 5/14) lends confidence to the ADR attributions [21,30,31]. A longer follow-up interval would be needed to detect long-term toxicities such as osteoporosis and cataract, which cannot be captured within 30 days.

4.5. Adherence Behaviour

The high level of self-reported adherence (76.32% scoring MARS-5 = 25) suggests that most patients appreciated the importance of completing their systemic corticosteroid course. The leading reason for non-adherence — fear of adverse effects (8/18, 44.44%) — echoes international research on corticosteroid-related illness perception [10,11] and points to a clear counselling target. Premature self-discontinuation on perceived improvement (5/18, 27.78%) and financial constraints (2/18, 11.11%) are recurring themes in Indian outpatient practice. Together, these findings argue for structured patient education at the point of corticosteroid initiation, addressing (i) the expected timeline of benefit, (ii) the importance of completing the taper schedule and (iii) the dose-dependent and largely reversible nature of most ADRs.

4.6. Effect of Adherence on Quality-of-Life Improvement

Low MARS-5 adherence was not linked to poorer 30-day DLQI improvement; the mean improvement was, if anything, numerically greater in the low-adherence group (11.72 vs. 8.97 points), although the difference (−2.76 points; 95% CI −7.06 to 1.55; p = 0.206) was not statistically significant and the effect size was small (Cohen’s d = −0.34). Consistent with this, the overall Spearman correlation between MARS-5 score and DLQI improvement was weak and non-significant (ρ = −0.169, p = 0.145), and none of the disease-category-specific correlations reached significance; interpretation is further constrained by the MARS-5 ceiling effect (76.32% of patients scored 25/25), which markedly attenuates rank variability and the discriminating power of the correlation. Four factors plausibly account for this trend, none of which supports the view that low adherence is benign. First, as noted in Section 4.5, symptomatic improvement prompting self-reduction was the leading reason for low adherence; in these patients DLQI improvement preceded the drop in adherence, a form of reverse causality. Second, the low-adherence group had a higher baseline DLQI (18.06 vs. 16.48), leaving greater room for absolute reduction through regression to the mean. Third, with only 18 patients in the low-adherence group the analysis was clearly under-powered — the 95% CI extends from a 7-point disadvantage to a 1.6-point advantage for high adherence, so a clinically meaningful effect cannot be excluded. Fourth, 30 days of follow-up is too brief to capture relapse, the outcome most likely to follow premature dose reduction in immunobullous and connective-tissue disorders. The clinical implication is not to abandon counselling but to sharpen it: patients should be told explicitly that early symptomatic improvement is not disease remission, and that completing the taper is essential to prevent relapse even when quality of life has substantially improved.

4.7. ICMR Concordance and Drug-Selection Deviations

Prescriptions deviating from the ICMR STW 2022 parameters concentrated in four recurring domains that together define a clear educational agenda. Drug-selection non-concordance (8/59; 13.56%) clustered around three clinically distinct patterns. The most consequential was the use of systemic corticosteroids in five DLE patients despite the ICMR STW 2022 recommendation of Hydroxychloroquine 5 mg/kg/day plus topical corticosteroids as first-line therapy [16] — a position aligned with broader international guidance that positions antimalarials as the cornerstone of first-line systemic therapy for cutaneous lupus, with systemic corticosteroids reserved for antimalarial failure [4]. At our centre this pattern may reflect the rapid symptomatic relief provided by systemic corticosteroids, the 4–8 week lead-time required for Hydroxychloroquine to take effect or unfamiliarity with the STW recommendation. Two further deviations involved moderate eczema managed outside the topical/step-up first-line pathway (severe eczema cases genuinely warranting systemic corticosteroids were not counted as deviations), and one involved a urticaria patient treated outside the STW antihistamine-led algorithm. Beyond drug selection, three additional patterns of non-concordance merit attention. First, baseline random blood glucose was not documented in 12 of the 59 evaluable prescriptions (20.34%), the single largest gap in the audit. Given that hyperglycaemia is among the earliest and most predictable metabolic effects of systemic corticosteroids — and was the second most frequent ADR in this cohort — routine pre-treatment RBG documentation is both easy to implement and clinically meaningful, and its absence in one in five prescriptions represents a low-cost, high-yield target for correction. Second, tapering regimens were not concordant with the STW recommendation in 9 of 59 prescriptions (15.25%), even though a taper of some form was documented in 80.49% of the full cohort. This dissociation between “any taper” (which was generally present) and “STW-concordant taper” (which was less consistently achieved) suggests that residual variability lies in the specifics of taper design — pace, step-size and total duration — rather than in the principle of tapering itself. Third, dose non-concordance occurred in 8 of 59 prescriptions (13.56%), with dose choices falling outside the Prednisolone-equivalent range specified by the STW for the relevant indication. Taken together, these deviations — drug selection, missing baseline RBG, non-concordant taper regimen and dose non-concordance — delineate a focused set of departmental educational priorities: Hydroxychloroquine-first for DLE, topical-led step-up therapy for moderate eczema and high-dose antihistamine-first pathways for urticaria (reserving systemic corticosteroids for refractory or severe acute presentations); routine documentation of baseline random blood glucose to complement the already comprehensively recorded baseline blood pressure and weight; and closer alignment of taper pace and dose selection with the condition-specific ranges recommended in the STW.

5. Strengths and Limitations

Strengths. This is the first prospective Indian dermatology study to combine an ICMR STW 2022 concordance audit with patient-reported outcomes (DLQI, MARS-5), validated disease-specific severity indices, and formal WHO–UMC causality assessment, which is the novelty of the study in an Indian setting.

The study has made good use of several validated tools like DLQI, MARS-5, and several disease severity indices, which increases the scientific credibility of the study. This paper has a practical impact for clinicians, who will actually learn where prescribing deviates from guidelines. Follow-up retention high at 92.68%, and the study captured both encouraging signals (short-term efficacy, tapering practice, low ADR rate, high adherence) and actionable gaps (drug-selection deviations, missing baseline metabolic screening and taper-regimen non-concordance).

Limitations. Sample size is the largest weakness of the study, with 82 patients, out of which 76 completed follow-ups. This study is also a single tertiary care center pilot study; further studies with a larger sample size and a longer follow-up are the future plan of the authors. Another limitation which the authors want to acknowledge is a shorter 30-day follow-up with respect to the use of systemic corticosteroids, where complications can arise at 3 months, 6 months, and 1 year. Further long-term safety studies are also planned in the future. The authors acknowledge that several subgroup analyses on such tiny samples could reduce the meaningfulness of the study.

6. Conclusions

Systemic corticosteroids were used across a wide spectrum of dermatological conditions at our tertiary-care centre, with oral Prednisolone predominating (63.41%) and a tapering schedule documented in 80.49% of prescriptions — both encouraging indicators of rational prescribing. Short-term outcomes were favourable: 73.68% of patients improved at day 30, and mean DLQI improved by 9.62 points. ADRs occurred in 11 of 82 patients (13.41%) across 14 episodes, of which 10 (71.43%) were mild, and the single severe event resolved without sequelae. Self-reported adherence was high (76.32%), with fear of ADRs identified as the leading reason for non-adherence. Concordance with the ICMR STW 2022 stood at 67.80% among covered prescriptions, and the residual gaps clustered around a small and coherent set of areas: drug-selection deviations (8/59, 13.56%) in DLE, moderate eczema and urticaria; missing baseline random blood glucose documentation (12/59, 20.34%); non-concordant taper regimens (9/59, 15.25%); and dose non-concordance (8/59, 13.56%). Together, these findings define a focused and actionable educational agenda — spanning first-line drug selection in the three flagged conditions, baseline metabolic screening prior to corticosteroid initiation, and closer alignment of dose- and taper-regimen choices with STW recommendations. Larger multicentre studies with longer follow-up are needed to confirm long-term safety, generalisability and guideline-drift patterns in Indian dermatology.