Profile of dermatology inpatients and admissions over a four year period in a tertiary level government teaching hospital in North India

Plain Language Summary

Although dermatology is primarily an out-patient specialty, a subset of patients needs hospital admission for management. We retrospectively reviewed the characteristics of 1664 patients who were admitted a total of 2032 times to the dermatology ward of the All India Institute of Medical Sciences, New Delhi, over a period of four years (2014-2017). The most common diagnoses were immunobullous disorders and connective tissue diseases accounting for roughly half of the admissions, followed by infections, psoriasis and reactive arthritis. The average duration of admission was about two weeks and complications developed in about 10% of the admissions during hospital stay. Re-admission was needed in 15% of the patients, especially those with immunobullous disorders and psoriasis. Adult patients, those who were admitted for more than four weeks, those who had comorbidities, and those who developed a complication during the hospital stay also had a greater chance of being re-admitted.

Introduction

Dermatology is primarily an out-patient clinical specialty. Although skin diseases are generally non-life threatening, they are associated with significant morbidity and a negative impact on the quality of life. Many hospitals do not have dedicated in-patient services for patients with severe skin diseases, and such patients are often admitted under the care of internists with dermatologists providing consultation services.¹ There is a paucity of literature regarding the profile of patients with skin diseases requiring hospital admission. This study reviews the characteristics of patients admitted over a four-year period to the dermatology ward of our institute.

Methods

We did a retrospective chart analysis of the discharge sheets of patients admitted in the dermatology inpatient ward of the All India Institute of Medical Sciences, New Delhi, India, a government tertiary care teaching hospital that caters largely to the population of North India. The dermatology ward has 29 beds, four of which are earmarked for providing isolation facilities for patients with extensive skin barrier loss. The ward is also equipped with two mechanical ventilators.

Discharge summaries of patients admitted in the dermatology ward between January 1, 2014 and December 31, 2017 were retrieved from the department’s electronic database. Patients admitted exclusively for day care were not included in the study. Variables analyzed for all admissions included patient age, gender, duration of admission stay, primary dermatological diagnosis, comorbidities, complications, and inter-specialty consults during the hospital stay. We also looked at the number of re-admissions, primary dermatological diagnosis for re-admissions, reason for re-admission and the interval from the date of discharge to the date of re-admission.

Results

During the four-year study period, a total of 3,20,741 patients (approximately 80,000 patients per year) visited our out-patient department. There were 2831 admissions during this period with 48 deaths (1.7%). The discharge records of 2032 (71.8%) admissions for 1664 patients could be retrieved from our database for analysis, of whom 842 (51%) were males accounting for 1036 (51%) admissions. There were 280 (17%) children (age <18 years) and 1384 adults (83%) accounting for 327 (16%) and 1705 (84%) admissions respectively. The mean age of the patients at admissions was 34.8 ± 17.2 years (range 0.1–91 years). The year-wise breakup of admission numbers and patient characteristicsis shown in Table 1.

Profile of hospital re-admissions

Two hundred and fifty six patients (15.4%) were re-admitted one or more times (range 1-10) for a total of 368 (18.1%) re-admissions. These 256 patients accounted for 624 (30.7%) of the 2032 admissions, and included 181 (70.7%) who were re-admitted once, 56 (21.9%) who were re-admitted twice, 13 (5.1%) who were re-admitted thrice and 6 (2.3%) patients who were re-admitted four times or more (range 4–10). The mean interval between the date of discharge and re-admissions (n = 368) was 154.16 ± 193.97 days (median 70, IQR 24–211.5, range 2–1015 days). The 30-day and 1-year re-admission rates were 5.3% (108/2032) and 15.8% (322/2032), respectively. The time interval between discharge and re-admission is shown in Figure 1.

Close

Figure 1:

Break-up of time intervals between discharge and re-admission

Export to PPT

Among the re-admissions, immunobullous disorders (n=134/368, 36.4%) were the most common diagnosis, followed by connective tissue diseases (n=68, 18.5%), reactive arthritis (n=40, 10.9%), psoriasis (n=34, 9.2%) and erythrodermas (n=16, 4.3%). A detailed list of the reason for re-admissions is given in Table 4.

Table 4: Reasons for re-admission (n=338)*

Reason	Number (%)
Administering subsequent doses of pulsed steroids or biologics	137 (37.5%)
Exacerbation of the primary dermatological disease	121 (32.8%
Re-evaluation of the disease	17 (4.6%)
Treatment-related complications	16 (4.3%)
Oral drug challenge	7 (1.9%)
Unrelated to the primary diagnosis	40 (10.9%)
Infections	22
Comorbidities	9
Miscellaneous causes	9
*Reason for re-admission not available	30 (8.1%)

Table 5 shows patient and admission variables associated with re-admissions. Adults (OR 1.54), patients with long (> 4 weeks) hospital stays (OR 1.73), patients with immunobullous disorders (OR 1.72) or psoriasis (OR 1.62), and those with comorbidities (OR 1.46-1.59) or complications (OR 1.93) were more likely to be re-admitted. However, patients with drug reactions were significantly less likely to be re-admitted (OR 0.34).

Table 5: Comparison of baseline admission profiles between patients with and without re-admissions

Baseline parameters	Patients with re-admission (n=256)	Patients without re-admissions (n=1408)	P-value	OR (95% CI)	P-value
Mean age (range) in years	35.24±15.5 (0.6–76)	34.88±18.03 (0.1–91)	0.802	-
Age group (years)			0.005*
Children (<18)	33 (12.9%)	247 (17.5%)		1.00
Adults (18–59)	207 (80.9%)	1004 (71.3%)		1.54 (1.04–2.29)	0.030*
Elderly (≥60)	16 (6.3%)	157 (11.2%)		0.76 (0.41–1.43)	0.399
Gender			0.730
Male	127 (49.6%)	715 (50.8%)		1.00
Female	129 (50.4%)	693 (49.2%)		1.05 (0.80–1.37)	0.730
Mean duration of admission (median, range) in days	14.89±11.26 (12, 1–75)	13.89±11.31 (11, 1–111)	0.193	-
Duration of admission (weeks)			0.047*
<2	162 (63.2%)	955 (67.8%)		1.00
2–4	64 (25%)	351(24.9%)		1.07 (0.79–1.47)	0.652
>4	30 (11.7%)	102 (7.2%)		1.73 (1.12–2.69)	0.014*
Number of patients with comorbidities	110/228 (48.2%)	484/1262 (38.4%)	0.005*
0	118 (51.7%)	778 (61.6%)		1.00
1	79 (34.6%)	356 (28.2%)	0.017*	1.46 (1.07–1.99)	0.017*
>1	31 (113.6%)	128 (10.1%)		1.59 (1.03–2.47)	0.036*
Number of patients with complications	32/210 (15.2%)	103/1206 (8.5%)	<0.002*
0	178 (84.8%)	1103 (91.5%)	<0.002*	1.00
≥1	32 (15.2%)	103 (8.6%)		1.93 (1.26–2.95)	0.003*
Number of patients with interspecialty consults	140/210 (66.6%)	750/1214 (61.8%)	0.176
0	70 (33.3%)	464 (38.2%)		1.00
1	89 (42.4%)	468 (38.6%)	0.387	1.26 (0.89–1.77)	0.179
>1	51 (24.3%)	282 (23.2%)		1.19 (0.81–1.77)	0.362
Diagnosis
Immunobullous disorders	92 (35.9%)	346 (24.6%)	<0.001*	1.72 (1.29–2.28)	<0.001*
Connective tissue diseases	53 (20.7%)	288 (20.5%)	0.933	1.02 (0.73–1.41)	0.928
Psoriasis	26 (10.2%)	92 (6.5%)	0.046*	1.62 (1.02–2.55)	0.039*
Reactive arthritis	13 (5.1%)	39 (2.8%)	0.075	1.88 (0.98–3.56)	0.054
Cutaneous infections	06 (2.3%)	74 (5.3%)	0.055	0.43 (0.18–1.01)	0.051
Cutaneous adverse drug reactions	05 (1.9%)	77 (5.5%)	0.017*	0.34 (0.14–0.86)	0.022*
Erythroderma	13 (5.1%)	45 (3.2%)	0.138	1.62 (0.86–3.05)	0.134
Genodermatoses	05 (1.9%)	58 (4.1%)	0.109	0.46 (0.18–1.16)	0.103
Leprosy	07 (2.7%)	49 (3.5%)	0.706	0.78 (0.35–1.74)	0.544
Dermatitis	06 (2.3%)	48 (3.4%)	0.448	0.68 (0.28–1.61)	0.379
Drug provocation	02 (0.8%)	35 (2.5%)	0.106	0.31 (0.07–1.29)	0.108
Infantile hemangioma and vascular malformations	02 (0.8%)	37 (2.6%)	0.074	0.29 (0.07–1.22)	0.091
Vasculitis	03 (1.2%)	21 (1.5%)	1.000	0.78 (0.23–2.64)	0.694
Chronic urticaria	02 (0.8%)	25 (1.8%)	0.416	0.44 (0.10–1.86)	0.260
Sexually transmitted infections	02 (0.8%)	19 (1.3%)	0.759	0.58 (0.13–2.48)	0.459
Cutaneous lymphomas	01 (0.4%)	16 (1.1%)	0.496	0.34 (0.04–2.58)	0.298
Cutaneous malignancies	00 (0%)	10 (0.7%)	0.376	0.26 (0.02–4.44)	0.352

CI: Confidence interval, OR: Odds ratio

Discussion

This study provides detailed information regarding the profile of admissions in the dermatology ward of a tertiary center in North India. Immunobullous disorders and connective tissue diseases were the most common diagnoses, accounting for almost half (48.5%) of the admissions. Other common diagnoses included psoriasis, infections, reactive arthritis and drug reactions. Of the immunobullous disorders, pemphigus vulgaris was the most common (455/576, 78.9%), while in the connective tissue disease group, systemic sclerosis (including overlap syndromes) was most frequently seen (272/409, 66.5%).

The pattern of diagnoses in our inpatients was quite different from that seen in the West where dermatitis, psoriasis, chronic wounds and skin neoplasms figure prominently.^2-6 Data regarding the profile of admitted dermatology patients from India and Asia are sparse, but as in the present study, immunobullous disorders were the most common in reports from East India⁷ and Iran⁸. The higher proportion of connective tissue diseases and reactive arthritis in our study may, at least in part, be due to the lack of in-patient rheumatology services at our hospital.

Infections, including leprosy and sexually transmitted infections, constituted 8.8% of admissions in our study whereas in the other Indian study by Sen et al⁷, they comprised 19.5% of all admissions. Dermatitis, reported to be the most common admission diagnoses (16–44%) in studies from the West,^2,4-6 made up only 2.9% of our admissions, and about 5% in the study by Sen et al.⁷ Similarly, cutaneous lymphomas and malignancies comprised only 1.3% of our admissions in comparison to 6–36% of admissions in the West.^2,3,6 Notably, about 2% of our admissions had an uncertain diagnosis even at discharge, reflecting the complex patient profile and diagnostic dilemmas that present to us.

The mean hospital stay of 14 days in our study was similar to the average 13 days hospital stay reported from Brazil.⁴ Shorter hospital stays of 11, 10, 7 and 4 days respectively were noted in studies from UK⁶, Australia⁹, Spain³, and the USA² but longer hospital stays of over 20 days were reported in the study by Sen et al⁷ and from South Africa.⁵ Our patients with cutaneous lymphomas and cutaneous malignancies had long hospital stays of over 21 days owing to multiple diagnostic and staging investigations. Patients with reactive arthritis also recorded long hospital stays (> 20 days) for administration of biologics and multiple inter-speciality consultations with rheumatology and pain management specialties. However, our patients with immunobullous disorders had relatively short hospital stays averaging 11.8 days, in contrast to the long stays of 22 days in other reports.^4,5,7

About 40% of our patients had one or more comorbidities. Diabetes mellitus and hypertension were the most frequent comorbidities.⁴ The complication rate of 10% was much lower than that (25%) reported in the study by Sen et al.⁷ Hospital-acquired infection were seen in 5% (89/1742) of our admissions which is similar to that (6%) reported in the Brazilian study, ⁴ and were most frequent among patients with immunobullous disorders. Sepsis was documented in 0.75% our admissions. This is much lower than that (4.6%) reported in an earlier study from our ward in 2004–2006.¹⁰ Although this difference could be due to better hospital infection control practices currently followed, it is more likely that since the focus of the previous study was sepsis in the dermatology ward, it resulted in a more sensitive recording of sepsis-related data. Blood stream infections accounted for a third of all hospital acquired infections (translating into about 2% of all admissions) in the Brazilian study.⁴

About 15% (256/1664) of our patients were re-admitted, making up 30% (624/2032) of our total admissions. The re-admissions alone accounted for about 18% (368/2032) of our total admissions. The majority of these re-admissions (88%) were related to the primary dermatology diagnosis, while the remaining (12%) were re-admitted for unrelated reasons. In our study, 30-day re-admission rates were 5%, while they were 1.8% in a study from Spain,³ and 9.8% and 12.3% in two US studies.^11,12 The one-year re-admission rate was 12.5% in the Spanish study in comparison to 16% in our study. Many patients with chronic skin illnesses discontinue treatment after initial recovery without further consultation, leading to disease flare-ups requiring re-admissions. One way to reduce the re-admissions is to ensure proper counseling at the time of discharge. Telephonic contacts with discharged patients a few weeks after discharge may help prevent such outcomes.

Patients with immunobullous disorders and psoriasis were more likely to be re-admitted than other conditions. The course of these diseases is marked with exacerbations or relapses, necessitating frequent hospitalizations. Psoriasis was reported as a frequent cause of re-admission in the study from Spain,³ while high re-admission rates were noted in patients with cutaneous lymphomas, connective tissue diseases and graft-versus-host disease in reports from the USA.^11,12 A longer stay during the first hospitalization and occurrence of complications were also associated with a higher re-admission rate in our study. The presence of comorbidities was another risk factor for re-admission.^11,12 Children were less likely to require re-admissions, and this may be related to the less frequent occurrence of immunobullous disorders and psoriasis (associated with higher rates of re-admission) in this age group. There were no gender differences for re-admission in our study but Arnold et al.¹¹ noted that females were more likely to be re-admitted. Arnold et al.¹¹ also noted that other factors such as having public medical insurance, belonging to a low-income community or admission in a large hospital were associated with higher rates of re-admission,¹¹ but we did not examine these variables.

In-patient dermatology services have been on the decline in the West.^1,13 The need for dermatology wards has recently become the subject of debate, particularly with the development of newer, more efficacious treatments including biologics for severe skin diseases. It has been suggested that patients with skin diseases requiring in-patient care may be admitted under other specialist units with dermatologists providing consultative services.^13,14 Although patients with skin disease as part of multisystem disease (e.g., an autoimmune connective tissue disease) or having significant comorbidities (e.g., uncontrolled diabetes or end-stage renal disease) may be better managed in a general medical ward with a visiting consultant dermatologist, there is no disputing the fact that patients with severe primary cutaneous disease such as psoriasis, dermatitis or pemphigus would be better managed in a dermatology ward with the dermatologist in direct control. Patients requiring oral drug provocation or those with a complex skin disease that is a diagnostic challenge would be best managed in specialist dermatology wards. The quality-of-life of patients with severe dermatological illnesses has been shown to significantly improve after hospital admission.^15,16

Being a tertiary care teaching institute, patients with severe and complex diseases are often referred to us who get admitted, while some patients are admitted for academic reasons also. Apart from the disease profile and severity, there may be administrative and other factors influencing dermatology admissions. Our institute being a government hospital with minimal healthcare charges may also influence admissions; hospitalizations are often needed for arranging or indenting medications and biologicals for poor patients. Our institute caters to the population from the entire North India and patients coming from far off places requiring multiple dermatosurgery or ablative sessions are also often admitted. Other reasons for admission may include facilitating cross-consultations with other departments and admissions owing to lack of inpatient facilites in allied specialties (such as rheumatology).

Conclusion

Our study provides a glimpse of profile of patients admitted in a dermatology ward of a tertiary care center in North India. The pattern of admission diagnoses differed from that reported in western studies. A significant proportion of patients get re-admitted. Certain diagnoses, patient demographic characteristics and background medical profile and those requiring a longer hospital stay or developing complications are more likely to be re-admitted.