Psychiatric comorbidities on an inpatient dermatology consultation service: A cross‐sectional analysis

Abstract Despite the high prevalence of psychiatric illness in hospitalised dermatology patients, characterisation of psychiatric comorbidities on an inpatient dermatology consultation service in the United States has yet to be performed. To fill this gap in knowledge, we investigated the prevalence of and factors associated with psychiatric illness on the inpatient dermatology consultation service at the University of Southern California. Of the 429 patients seen by the dermatology consultation service between June 2021 to July 2022, 147 (34%) had psychiatric illness (defined as having at least 1 psychiatric diagnosis). Increasing age was associated with a decreased likelihood of psychiatric illness, while housing instability, chronic dermatologic disease, drug reaction, and pruritus without rash were associated with an increased likelihood of psychiatric illness. The high prevalence of psychiatric illness observed in hospitalised dermatology patients emphasises the importance of collaboration between consultant dermatologists and mental health specialists, particularly when specific sociodemographic or disease factors are present.

Charlson Comorbidity Index (CCI).CCI predicts longterm mortality based on existing medical conditions 2 and was used to approximate physical disease burden.We also recorded the dermatologic diagnosis, chronicity, and whether it was the reason for admission.Dermatologic diagnoses were divided into 11 groups (Table 1).Dermatologic disease was defined as acute (symptoms <6 months) or chronic (≥6 months).Psychiatric diagnoses were based on International Classification of Diseases, version 10 codes and primary team documentation.Patients were divided into those with psychiatric illness (≥1 psychiatric diagnosis) and those without psychiatric illness (no psychiatric diagnoses).
Characteristic differences between groups were analysed using t-test and chi-square test.Univariable and multivariable logistic regressions were performed to identify factors associated with psychiatric illness.Since our cohort included patients with drug reactions caused by illicit substances and psychiatric medications, these patients contributed to the association between drug reactions and psychiatric illness.Therefore, a sensitivity analysis was conducted, whereby multivariable analysis was repeated after excluding these patients.P < 0.05 was considered statistically significant.R (version 4.2.2) was used.This study was approved by our institutional review board (HS-18000640).
In other countries, the prevalence of psychiatric illness in hospitalised dermatology patients has been as  b 61 patients were identified as either "other" or "unknown" in their electronic medical record.c 6 patients had unknown disease chronicity.
d "Undifferentiated" referred to patients without a leading diagnosis."Other" included patients who had a likely diagnosis that did not fit into the other categories.

of 4
-LETTER TO THE EDITOR high as 38%. 1 This is comparable to the 34% observed in our patients, although the cited study reported on inpatients at a specialised dermatologic hospital, rather than those seen by a consultation service.The higher likelihood of psychiatric illness with housing instability, dermatologic disease chronicity, and pruritus without rash are consistent with the literature.Although the association between psychiatric illness and all drug reactions has not been previously reported, prior studies have demonstrated connections between psychiatric disease and specific drug reactions, including after Stevens-Johnson syndrome/toxic epidermal necrolysis. 3The lack of association between psoriasis/dermatitis and psychiatric illness could be partially explained by aggregating multiple diseases to form this group.Nationally, the prevalence of comorbid psychiatric illness is 38.7% for psoriasis 4 versus 16.36% for atopic dermatitis. 5he lower likelihood of psychiatric illness with increasing age could reflect that individuals with mental health disorders die a median of 10 years earlier than the general population. 6While unnatural causes of death are higher in individuals with psychiatric illness, the main source of death in this population is from natural causes (eg medical comorbidities), suggesting that lifestyle, social determinants, and barriers to care may also contribute to the survival disparity. 6his study was limited by retrospective nature, small sample size, and single-institution origin.Nevertheless, the high prevalence of psychiatric illness observed in hospitalised dermatology patients emphasises the importance of collaborating with mental health specialists.Future research is warranted to determine how the observed prevalence compares to that seen by consultation services of other specialties.Additionally, while it has been demonstrated that comorbid psychiatric illness increases length of hospitalisation 7 and rate of readmission, 8 it is necessary to determine whether these worse outcomes are seen in dermatology inpatients specifically.

T A B L E 1
Description of patient characteristics according to presence of psychiatric illness.
1 � 16.2 55.8 � 16.5 47.7 � 14.1 <0.001 Note: Data are mean � standard deviation or number (%).Only factors significant on t-test or chi-square tests are shown.a Welch's unequal variance t-test was used for age.Chi-square test was used for all other comparisons.