Introduction
On February 6, 2023, southeastern Türkiye was struck by 2 major earthquakes (Mw 7.8 and 7.5) followed by prolonged aftershocks, constituting one of the most devastating natural disasters of the past century.Reference Galasso and Opabola 1 The disaster affected 11 provinces, resulting in extensive loss of life, mass displacement, and severe disruption of health care services. Such large-scale catastrophes, characterized by compromised infrastructure, overcrowding, and impaired hygiene, substantially increase public health risks, particularly for infectious and inflammatory dermatologic conditions.Reference Tunalı, Harman and Özbilgin 2 , Reference Öztaş, Önder and Öztaş 3
Although data on dermatologic presentations after earthquakes remain limited, previous reports from Türkiye, Nepal, Pakistan, Iran, China, and Japan have described changes in parasitic and microbial infections, inflammatory dermatoses, and chronic skin conditions after major disasters.Reference Öztaş, Önder and Öztaş 3 –Reference Stewart and Goodman 19 Most studies, however, have focused on the immediate or short-term period and on a limited spectrum of diagnoses. The present study aimed to compare long-term dermatology outpatient diagnostic frequencies and presentation patterns before and after the 2023 Hatay earthquake at a tertiary referral center. As the data were clinic-based, the findings were interpreted as changes in outpatient diagnostic patterns rather than population-level disease prevalence.
Materials and Methods
This retrospective comparative cohort study was conducted at a tertiary university-affiliated dermatology center (ethics approval: 21/05/2025-27). Dermatology outpatient records were reviewed for the pre-earthquake period (February 6, 2021-February 3, 2023) and the post-earthquake period (May 11, 2023-February 6, 2025). The interval from February 6 through May 10, 2023 was excluded because severe infrastructural disruption prevented reliable assessment of record completeness. The number of potentially missing encounters during this interval could not be reconstructed from the damaged records; therefore, the entire interval was excluded a priori and no imputation was performed.
Data retrieved from the hospital information management system included patient identifier, sex, date of birth, admission date, and ICD-10-coded dermatologic diagnoses. Age at each index visit was recalculated from date of birth and admission date. Descriptive mean age was calculated from elapsed days divided by 365.2425; completed age in years was used for age-group assignment and continuous model adjustment. Age groups were defined as 0-18 years (<19 years), 19-64 years (≥19 and <65 years), and ≥65 years. Diagnoses were established by dermatologists on the basis of clinical evaluation, supported by laboratory or histopathological findings when clinically indicated. Conditions were grouped into 18 etiological or clinicopathological categories: infestations and insect bites; acneiform diseases; hair and nail disorders; sexually transmitted diseases; neutrophilic, eosinophilic and mast-cell diseases; dermatitis; fungal infections; pigmentation disorders; bacterial diseases; autoimmune connective-tissue and rheumatic diseases; vesiculobullous diseases; lichenoid and granulomatous diseases; neoplasms; environmentally induced diseases; erythematosquamous diseases; reactive erythemas; vascular diseases; and viral infections. These categories comprised 69 subdiagnoses. The unit of analysis was the eligible patient-diagnosis record. Repeated presentations for the same diagnosis were represented by the first eligible record, whereas the same patient could contribute separate records for distinct diagnoses or diagnostic categories.
The same hospital information system and ICD-10 framework were used across both study periods. However, ICD-10 coding was not independently revalidated for this study, and inter-rater agreement was not assessed. Accordingly, possible changes in clinical personnel, diagnostic practices, or hospital operations across the disaster period were considered potential sources of classification bias.
Continuous variables are presented as mean ± standard deviation (SD), and categorical variables as number and percentage. For each of the 18 prespecified diagnostic categories, logistic regression estimated the association between study period and the probability that the category was represented among eligible dermatology diagnosis records. Adjusted models included age at the index visit, sex, and calendar month to account for demographic differences and seasonal variation. Because the same patient could contribute more than one distinct diagnosis record, robust standard errors were clustered by patient. Crude and adjusted odds ratios (ORs) with 95% confidence intervals (CIs) are reported. The Benjamini-Hochberg procedure was applied across the 18 primary category-level comparisons to control the false discovery rate at 5%; both P values and FDR-adjusted q values are presented. Subdiagnostic analyses were considered exploratory. A 2-sided P value of <0.05 was considered statistically significant; for the primary category-level comparisons, an FDR-adjusted q value of <0.05 was used to determine statistical significance.
Results
Overall Characteristics
A total of 41,012 eligible patient-diagnosis records were included, comprising 19,646 records from the pre-earthquake period and 21,366 from the post-earthquake period. The corresponding observation windows were 728 and 638 days, yielding approximately 27.0 and 33.5 included records per observation day, respectively. Because individual patients could contribute records for distinct diagnoses, these counts represent patient-diagnosis records rather than unique patients. Overall, 17,016 records (41.5%) were from male patients and 23,996 (58.5%) from female patients. Patients aged 0-18, 19-64, and ≥65 years accounted for 9,387 (22.9%), 28,860 (70.4%), and 2,765 (6.7%) records, respectively.
Pre-Earthquake Period Findings
In the pre-earthquake period, 7,641 records (38.9%) were from male patients and 12,005 (61.1%) from female patients. Mean age at the index visit was 32.60 ± 17.64 years. The age distribution was 4,009 records (20.4%) among patients aged 0-18 years, 14,437 (73.5%) among those aged 19-64 years, and 1,200 (6.1%) among those aged ≥65 years. The most frequently represented diagnostic category was acneiform disorders (n = 5,412; 27.55%), followed by fungal infections (n = 3,839; 19.54%), infestations/insect bites (n = 2,164; 11.01%), dermatitis (n = 1,831; 9.32%), and hair and nail disorders (n = 1,439; 7.32%).
Among specific diagnoses before the earthquake (n = 19,646), the most frequent were acne (n = 5,029; 25.60%), dermatophyte infection (n = 3,358; 17.09%), scabies (n = 2,117; 10.78%), seborrheic dermatitis (n = 1,493; 7.60%), and urticaria (n = 1,187; 6.04%). Dermatophyte infection was the leading diagnosis among records from male patients (n = 1,597/7,641; 20.90%), whereas acne was the leading diagnosis among records from female patients (n = 3,791/12,005; 31.58%). By age group, acne was the leading diagnosis among patients aged 0-18 years (n = 1,298/4,009; 32.38%) and 19-64 years (n = 3,706/14,437; 25.67%), whereas dermatophyte infection was the leading diagnosis among those aged ≥65 years (n = 429/1,200; 35.75%).
Post-Earthquake Period Findings
In the post-earthquake period, 9,375 records (43.9%) were from male patients and 11,991 (56.1%) from female patients. Mean age at the index visit was 33.09 ± 18.98 years. Patients aged 0-18 years accounted for 5,378 records (25.2%), those aged 19-64 years for 14,423 (67.5%), and those aged ≥65 years for 1,565 (7.3%). Infestations/insect bites were the most frequently represented category (n = 5,289; 24.75%), followed by fungal infections (n = 4,416; 20.67%), acneiform disorders (n = 3,984; 18.65%), dermatitis (n = 2,328; 10.90%), and sexually transmitted diseases (n = 1,090; 5.10%). The distribution of the major dermatologic diagnostic categories in the pre- and post-earthquake periods is shown in Figure 1.
Distribution of the 18 prespecified dermatologic diagnostic categories before and after the earthquake, ranked by post-earthquake record count. Bars represent the number of eligible diagnosis records during the pre-earthquake (February 6, 2021-February 3, 2023) and post-earthquake (May 11, 2023-February 6, 2025) periods. Counts represent diagnosis records rather than unique patients.
Figure 1 Long description
Horizontal grouped bar chart with pre-earthquake counts shown in blue and post-earthquake counts in orange for 18 categories, ordered by post-earthquake count. Infestations and insect bites increased from 2,164 to 5,289 records and became the largest category. Fungal infections increased from 3,839 to 4,416, dermatitis from 1,831 to 2,328, sexually transmitted diseases from 691 to 1,090, bacterial diseases from 160 to 304, and environmentally induced diseases from 85 to 150. Acneiform disorders decreased from 5,412 to 3,984; hair and nail disorders from 1,439 to 847; neutrophilic/eosinophilic/mast-cell diseases from 1,196 to 797; erythematosquamous diseases from 1,128 to 975; and pigmentation disorders from 775 to 549. Smaller categories also decreased: viral infections 341 to 319, autoimmune connective-tissue/rheumatic diseases 176 to 93, lichenoid and granulomatous diseases 162 to 92, vesiculobullous diseases 157 to 82, vascular diseases 41 to 35, reactive erythemas 21 to 13, and neoplasms 28 to 3. Counts are diagnosis records, not unique patients.
Among specific diagnoses after the earthquake (N = 21,366), the most frequent were scabies (n = 5,256; 24.60%), dermatophyte infection (n = 3,906; 18.28%), acne (n = 3,620; 16.94%), seborrheic dermatitis (n = 1,727; 8.08%), and viral/anogenital warts (n = 1,083; 5.07%). Scabies was the leading diagnosis among records from male patients (n = 2,616/9,375; 27.90%), whereas acne was the leading diagnosis among records from female patients (n = 2,747/11,991; 22.91%). By age group, scabies was the leading diagnosis among patients aged 0-18 years (n = 1,967/5,378; 36.57%), dermatophyte infection among those aged 19-64 years (n = 2,954/14,423; 20.48%), and scabies among those aged ≥65 years (n = 469/1,565; 29.97%).
Comparison of the Pre- and Post-Earthquake Periods
The demographic composition of eligible diagnosis records differed modestly between the 2 periods. Mean age at the index visit was 0.49 years higher in the post-earthquake period (mean difference 0.49 years; 95% CI 0.13-0.84; Cohen d = 0.03; P = 0.007). The proportion of records from male patients was also higher in the post-earthquake period (crude OR 1.23; 95% CI 1.18-1.28; P < 0.001). Age-group distribution differed between periods (χ2[2] = 176.02; Cramér V = 0.066; P < 0.001), although the effect size was small (Table 1). These comparisons describe the composition of records at the study center and should not be interpreted as changes in population susceptibility, access, or disease prevalence.
Demographic characteristics and outpatient record volume before and after the earthquake
Table 1. Long description
The pre- and post-earthquake observation periods were 728 and 638 days, with 19,646 and 21,366 eligible diagnosis records, equivalent to 27.0 and 33.5 records per day. Mean age was 32.60 ± 17.64 versus 33.09 ± 18.98 years; mean difference 0.49 years (95% CI 0.13-0.84), Cohen d=0.03, p=0.007. Male records were 7,641 (38.9%) versus 9,375 (43.9%), crude OR 1.23 (95% CI 1.18-1.28), p<0.001; female records were 12,005 (61.1%) versus 11,991 (56.1%). Age 0-18 years: 4,009 (20.4%) versus 5,378 (25.2%); age 19-64 years: 14,437 (73.5%) versus 14,423 (67.5%); age ≥65 years: 1,200 (6.1%) versus 1,565 (7.3%). The overall age-group distribution differed, χ²(2)=176.02, Cramér V=0.066, p<0.001. Counts are clinic-based diagnosis records, not population denominators.
Note: Effect estimates compare the post-earthquake with the pre-earthquake period. The age-category P value refers to the overall 2 × 3 chi-square test. Included record counts are not population denominators.
Values are presented as n (%) unless otherwise indicated. Percentages were calculated within each study period.
Abbreviations: CI, confidence interval; OR, odds ratio; SD, standard deviation.
† The χ2 test and corresponding P value represent the overall comparison of the 3-category age distribution.
In models adjusted for age, sex, and calendar month, infestations/insect bites were substantially more frequently represented in the post-earthquake diagnostic mix (aOR 2.55, 95% CI 2.41-2.70; q < 0.001). Dermatitis (aOR 1.17, 95% CI 1.10-1.25; q < 0.001), bacterial diseases (aOR 1.58, 95% CI 1.30-1.92; q < 0.001), environmentally induced diseases (aOR 1.51, 95% CI 1.16-1.97; q = 0.003), and sexually transmitted diseases (aOR 1.50, 95% CI 1.36-1.66; q < 0.001) were also more frequently represented. Conversely, acneiform disorders (aOR 0.61, 95% CI 0.59-0.64), hair and nail disorders (aOR 0.52, 95% CI 0.48-0.57), neutrophilic/eosinophilic/mast-cell diseases (aOR 0.59, 95% CI 0.54-0.65), pigmentation disorders (aOR 0.68, 95% CI 0.61-0.75), autoimmune connective-tissue and rheumatic diseases (aOR 0.49, 95% CI 0.38-0.62), vesiculobullous diseases (aOR 0.47, 95% CI 0.36-0.60), lichenoid and granulomatous diseases (aOR 0.50, 95% CI 0.39-0.65), erythematosquamous diseases (aOR 0.80, 95% CI 0.73-0.86), and neoplasms (aOR 0.11, 95% CI 0.03-0.37) were less frequently represented (all q < 0.001). Viral infections were also less frequently represented, but the association was smaller (aOR 0.84, 95% CI 0.72-0.99; q = 0.040). Although fungal infections were more frequent in the crude comparison, this association was attenuated and was no longer statistically significant after adjustment (aOR 1.04, 95% CI 0.99-1.09; q = 0.149). Reactive erythemas and vascular diseases showed no statistically significant adjusted associations with study period. Adjusted category-level associations are summarized in Figure 2 and presented numerically in Table 2.
Adjusted associations between the post-earthquake period and dermatologic diagnostic categories. Adjusted odds ratios compare the post-earthquake period with the pre-earthquake period for the 18 prespecified diagnostic categories. Points represent adjusted odds ratios, and horizontal lines indicate 95% confidence intervals. Estimates were obtained from logistic regression models adjusted for age at the index visit, sex, and calendar month, using patient-clustered robust standard errors. The dashed vertical line denotes the null value of 1.0. Abbreviations: aOR, adjusted odds ratio; CI, confidence interval.
Figure 2 Long description
Forest plot on a logarithmic scale. The dashed vertical line at 1.0 represents no association; points are adjusted odds ratios (aORs) and horizontal lines are 95% confidence intervals. Post- versus pre-earthquake aORs were: infestations/insect bites 2.55 (2.41-2.70); acneiform disorders 0.61 (0.59-0.64); hair/nail disorders 0.52 (0.48-0.57); sexually transmitted diseases 1.50 (1.36-1.66); neutrophilic/eosinophilic/mast-cell diseases 0.59 (0.54-0.65); dermatitis 1.17 (1.10-1.25); fungal infections 1.04 (0.99-1.09); pigmentation disorders 0.68 (0.61-0.75); bacterial diseases 1.58 (1.30-1.92); autoimmune connective-tissue/rheumatic diseases 0.49 (0.38-0.62); vesiculobullous diseases 0.47 (0.36-0.60); lichenoid/granulomatous diseases 0.50 (0.39-0.65); neoplasms 0.11 (0.03-0.37); environmentally induced diseases 1.51 (1.16-1.97); erythematosquamous diseases 0.80 (0.73-0.86); reactive erythemas 0.58 (0.30-1.11); vascular diseases 0.72 (0.47-1.10); and viral infections 0.84 (0.72-0.99). Confidence intervals crossed 1.0 for fungal infections, reactive erythemas, and vascular diseases. Models adjusted for age, sex, and calendar month with patient-clustered robust standard errors.
Main dermatologic diagnostic categories before and after the earthquake
Table 2. Long description
The table reports pre- versus post-earthquake counts and percentages, crude odds ratios, adjusted odds ratios (aORs), p values, and FDR-adjusted q values for 18 categories. Models adjusted for age, sex, and calendar month with patient-clustered robust standard errors. Higher post-earthquake representation occurred for infestations/insect bites, 2,164 (11.01%) versus 5,289 (24.75%), aOR 2.55 (2.41-2.70); sexually transmitted diseases, 691 (3.52%) versus 1,090 (5.10%), aOR 1.50 (1.36-1.66); dermatitis, 1,831 (9.32%) versus 2,328 (10.90%), aOR 1.17 (1.10-1.25); bacterial diseases, 160 (0.81%) versus 304 (1.42%), aOR 1.58 (1.30-1.92); and environmentally induced diseases, 85 (0.43%) versus 150 (0.70%), aOR 1.51 (1.16-1.97). These associations remained significant after FDR correction (q≤0.003). Lower post-earthquake representation was found for acneiform disorders (aOR 0.61), hair/nail disorders (0.52), neutrophilic/eosinophilic/mast-cell diseases (0.59), pigmentation disorders (0.68), autoimmune connective-tissue/rheumatic diseases (0.49), vesiculobullous diseases (0.47), lichenoid/granulomatous diseases (0.50), neoplasms (0.11), erythematosquamous diseases (0.80), and viral infections (0.84); all q≤0.040. Fungal infections showed no adjusted difference: 3,839 (19.54%) versus 4,416 (20.67%), aOR 1.04 (0.99-1.09), q=0.149. Reactive erythemas and vascular diseases were also nonsignificant: aOR 0.58 (0.30-1.11), q=0.112, and 0.72 (0.47-1.10), q=0.139. All 41,012 records were included; no imputation was performed.
Note: Adjusted ORs were estimated using logistic regression models including completed age at the index visit, sex, and calendar month, with robust standard errors clustered by patient. The Benjamini-Hochberg procedure was applied across the 18 prespecified category-level comparisons. Individual patients could contribute separate records for distinct diagnoses or diagnostic categories. Adjusted models included all 41,012 eligible records; no imputation was performed.
At the subdiagnostic level, acne, dermatophyte infection, scabies, seborrheic dermatitis, and urticaria were the leading recorded diagnoses before the earthquake. After the earthquake, scabies became the leading diagnosis, followed by dermatophyte infection, acne, seborrheic dermatitis, and viral/anogenital warts. Dermatophyte infection was the leading diagnosis among records from male patients before the earthquake, whereas scabies was the leading diagnosis afterward; acne remained the leading diagnosis among records from female patients in both periods. Age-stratified distributions of the leading diagnoses are presented descriptively in Table 3.
Leading dermatologic diagnoses by age group before and after the earthquake
Table 3. Long description
For patients aged 0-18 years, the leading diagnosis changed from acne before the earthquake, 1,298 of 4,009 records (32.38%), to scabies after the earthquake, 1,967 of 5,378 (36.57%). For ages 19-64 years, it changed from acne, 3,706 of 14,437 (25.67%), to dermatophyte infection, 2,954 of 14,423 (20.48%). For ages ≥65 years, it changed from dermatophyte infection, 429 of 1,200 (35.75%), to scabies, 469 of 1,565 (29.97%). Percentages are clinic-based within each age group and period. This table is descriptive; no inferential p values were calculated.
Note: Percentages are clinic-based within age group and period; denominators are the corresponding age-group record counts. Age groups use completed age at the index visit: 0-18 years (<19), 19-64 years (≥19 and <65), and ≥65 years. This table is descriptive; inferential P values were not calculated because the leading diagnosis differs between periods and the table does not represent a single prespecified binary comparison.
Overall, the adjusted analyses demonstrated a redistribution of clinic-based diagnostic records toward infestations/insect bites and selected infectious and environmentally associated categories in the post-earthquake period. The fungal-infection association was attenuated after adjustment, underscoring the contribution of demographic and seasonal composition. These findings characterize changes in the diagnostic composition of dermatology outpatient records and do not represent population-level changes in disease prevalence or incidence.
Limitations
This study has several limitations. First, the retrospective before-and-after design cannot establish causality, and temporal changes may be explained by population displacement, migration, seasonal variation, altered health care access, referral pathways, clinic capacity, or health care-seeking behavior. Second, the analysis used clinic-based patient-diagnosis records rather than a population denominator; the reported proportions therefore describe diagnostic composition and utilization, not incidence or prevalence. Care delivered through temporary clinics, field hospitals, primary-care facilities, and institutions outside the affected area was not captured, potentially altering the case mix and introducing selection bias. Individual patients could contribute more than one record when distinct diagnoses or diagnostic categories were represented. Patient-clustered robust standard errors were used to account for within-patient dependence, but residual correlation and unmeasured confounding remain possible. Third, records from February 6 to May 10, 2023 were excluded because documentation was incomplete and the number of potentially missing encounters could not be reconstructed; therefore, the study does not represent the acute post-earthquake phase. No imputation was performed. After date-format harmonization, all 41,012 eligible records had valid age information and were included in adjusted models. Fourth, diagnoses were derived from routine ICD-10 coding and were not independently revalidated. Changes in personnel, coding practices, or hospital operations could have introduced classification bias. Fifth, age, sex, and calendar month were available for adjustment, but migration, housing, socioeconomic conditions, environmental exposures, psychological stress, and individual health care access were unavailable. Finally, sparse counts in some categories, particularly neoplasms and reactive erythemas, produced imprecise estimates and require cautious interpretation. The Benjamini-Hochberg procedure reduced, but did not eliminate, the risk of chance findings across multiple comparisons.
Discussion
This single-center before-and-after study identified a sustained redistribution of dermatology outpatient diagnostic records following the 2023 Hatay earthquake. The principal finding was the higher representation of infestations and insect bites after adjustment for age, sex, and calendar month. Dermatitis, bacterial diseases, environmentally induced diseases, and sexually transmitted diseases were also represented more frequently, whereas several chronic, autoimmune, neoplastic, and follow-up-dependent categories were less frequent. Importantly, the modest crude difference in fungal infections did not remain statistically significant after demographic and seasonal adjustment.
Communicable dermatoses, particularly scabies, showed the clearest post-earthquake increase in the outpatient diagnostic mix. Similar post-disaster patterns have been reported after the Düzce and Van earthquakes in Türkiye, the 2015 Nepal earthquake, the Bam earthquake in Iran, and the Wenchuan earthquake in China.Reference Öztaş, Önder and Öztaş 3 , Reference Bayramgürler, Bilen and Namlı 6 –Reference Renchan, Tangyou and Chunxue 13 , Reference Sharifi, Poursmaelian and Aflatoonian 15 In the 2023 Türkiye earthquake context, Yücesoy et al.Reference Yucesoy, Temiz and Ak 8 and Doğramacı et al.Reference Doğramacı, Akbay and Aydın 20 also documented a prominent burden of scabies, while Beyoğlu and GoklerReference Beyoğlu and Gokler 21 described scabies in tent-city populations. These observations are biologically and operationally compatible with overcrowding, disrupted hygiene, and temporary housing; however, the present study did not measure housing density, sanitation, water access, or individual exposure. Accordingly, these mechanisms should be regarded as plausible explanations rather than demonstrated causes.
Sexually transmitted diseases were more frequently represented after the earthquake after adjustment (aOR 1.50, 95% CI 1.36-1.66), with viral/anogenital warts accounting for most records in this category. This clinic-based association may reflect changes in diagnostic composition, coding, care-seeking, referral pathways, or service availability and should not be interpreted as a population-level increase in sexually transmitted infections.
Conversely, several chronic, follow-up-dependent, autoimmune, and neoplastic diagnoses were recorded less frequently after the earthquake. The marked reduction in cutaneous neoplasm presentations should not be interpreted as a decrease in underlying cancer incidence. It may instead reflect changes in access, referral pathways, patient priorities, or migration. Comparable concerns arose during the COVID-19 health care disruption: Mostafavi Zadeh et al.Reference Mostafavi Zadeh, Rezaei and Barahimi 22 developed a systematic review protocol specifically to evaluate how restrictions in health care access and diagnostic pathways affected skin-cancer diagnosis. Although a pandemic differs from an earthquake, both contexts illustrate how large-scale crises can alter detection and presentation patterns independently of underlying disease biology.
The demographic composition of clinic diagnosis records also changed, with higher proportions of records from male patients, patients aged 0-18 years, and patients aged ≥65 years, alongside a lower proportion from those aged 19-64 years. Mean age at the index visit was only 0.49 years higher, with a negligible standardized difference. Adjustment for age, sex, and calendar month was therefore important. The attenuation of the fungal-infection association after adjustment illustrates that crude pre/post comparisons can partly reflect demographic and seasonal differences rather than an independent period association.
Previous studies have suggested that psychological stress after earthquakes can exacerbate atopic dermatitis, urticaria, psoriasis, vitiligo, and other inflammatory dermatoses.Reference Kodama, Horikawa and Suzuki 4 , Reference Lugović-Mihić, Meštrović-Štefekov and Cvitanović 5 , Reference Bhattarai, Rauniar and Rijal 7 , Reference Karn 10 , Reference Bari, Ishfaq and Butt 12 , Reference Stewart and Goodman 19 In the present cohort, age-specific increases were observed for acne, seborrheic dermatitis, and atopic dermatitis in some groups, while psoriasis and vitiligo presentations decreased. Because psychological stress was not measured, these changes cannot be attributed to stress. They may instead reflect a combination of disease activity, altered follow-up, migration, treatment continuity at other centers, and changes in health care priorities. The contrast with reports from Nepal, Pakistan, Japan, and other disaster settings underscores that post-disaster dermatologic patterns are context-dependent and shaped by both exposures and health care-system function.Reference Bhattarai, Rauniar and Rijal 7 , Reference Karn 10 –Reference Bari, Ishfaq and Butt 12 , Reference Miura, Ohtsuka and Yamamoto 16
We also observed numerically fewer presentations for hair disorders, autoimmune vesiculobullous and connective-tissue diseases, granulomatous and lichenoid disorders, neoplasms, and reactive erythemas. Such reductions may be compatible with disruption of longitudinal care and prioritization of acute needs, but the study cannot determine whether patients deferred care, received care elsewhere, or left the region. After the 1999 Düzce earthquake, Kavak et al.Reference Kavak, Yeşildağ and Parlak 14 reported no increase in alopecia areata visits, whereas Yücesoy et al.Reference Yucesoy, Temiz and Ak 8 described seasonal variation in hair disorders after the 2023 earthquake. These differing findings further support cautious interpretation of single-center outpatient data.
Post-disaster health care utilization is distributed across multiple settings. Studies of evacuated survivors and emergency medical team facilities after the 2023 Türkiye earthquakes documented use of emergency departments, outpatient clinics, and disaster-response facilities.Reference İlhan, Eroğlu and Çanak 23 , Reference Maddah, Vadi and Aloğlu 24 Accordingly, patients may also have sought care at temporary clinics, field hospitals, primary-care centers, or institutions outside the affected area, changing the case mix at our tertiary center and introducing selection bias.
Among patients aged ≥65 years, actinic keratosis accounted for a larger share of presentations in the post-earthquake period. The original hypothesis that this reflected increased sun exposure following housing destruction is plausible but was not directly tested. The finding should therefore be regarded as an observation requiring confirmation with exposure data rather than evidence of a causal pathway.
Overall, the findings indicate a substantial change in the case mix of dermatology outpatient records during the post-earthquake period, characterized primarily by the increased representation of scabies and other infestation-related diagnoses and the lower representation of several chronic and follow-up-dependent conditions. Because population denominators, migration patterns, individual exposures, and health care use outside our center were unavailable, these findings should be interpreted as clinic-based temporal associations rather than changes in population-level disease incidence or causal effects of the earthquake. Disaster-response planning should therefore address both the control of communicable dermatoses and the continuity of care for chronic diseases and suspected cutaneous neoplasms.
Conclusions
In this retrospective single-center comparison, the post-earthquake period was associated with a redistribution of dermatology outpatient diagnosis records, most prominently toward infestations and insect bites, alongside a lower representation of several chronic and follow-up-dependent conditions. Several associations persisted after adjustment for age, sex, and calendar month, whereas the crude increase in fungal infections was attenuated and was no longer statistically significant. These findings support the consideration of dermatologic surveillance, infection-control measures, and continuity pathways for chronic and high-risk dermatologic conditions in disaster preparedness and recovery planning. However, they should be interpreted as clinic-based temporal associations rather than population-level changes in incidence or prevalence or causal effects of the earthquake.
Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Author contribution
Selen Taner Akbay: Conceptualization, data curation, writing—original draft preparation, methodology, investigation, validation, formal analysis. Selen Aşkar: Data curation, writing—original draft preparation, methodology, investigation. Asena Çiğdem Doğramacı: Supervision, data curation.
Funding statement
None.
Competing interests
None.
Ethical approval
This study was approved by the Clinical Research Ethics Committee of Hatay Mustafa Kemal University (21/05/2025-27) and conducted in accordance with the principles of the Declaration of Helsinki.
Ethics statement
The requirement for informed consent was waived by the Clinical Research Ethics Committee of Hatay Mustafa Kemal University because of the retrospective design and use of anonymized data.
Declaration of Generative AI use
During the preparation of this manuscript, the authors used ChatGPT (OpenAI, GPT-5) solely to improve academic language, grammar, clarity, and readability. All AI-assisted revisions were critically reviewed and approved by the authors, who take full responsibility for the final content.
Open access
