- Email: [email protected]
Inpatient Financial Burden of Atopic Dermatitis in the United States
Accepted Manuscript Inpatient financial burden of atopic dermatitis in the United States Shanthi Narla, BA, Derek Y. Hsu, BA, Jacob P. Thyssen, MD, PhD, DmSci, Jonathan I. Silverberg, MD, PhD, MPH PII:
S0022-202X(17)31160-0
DOI:
10.1016/j.jid.2017.02.975
Reference:
JID 781
To appear in:
The Journal of Investigative Dermatology
Received Date: 22 December 2016 Revised Date:
6 February 2017
Accepted Date: 7 February 2017
Please cite this article as: Narla S, Hsu DY, Thyssen JP, Silverberg JI, Inpatient financial burden of atopic dermatitis in the United States, The Journal of Investigative Dermatology (2017), doi: 10.1016/ j.jid.2017.02.975. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Page 1 of 23 Title: Inpatient financial burden of atopic dermatitis in the United States Short Title: Hospitalization of AD.
RI PT
Shanthi Narla, BA, Derek Y. Hsu, BA1 , Jacob P. Thyssen, MD, PhD, DmSci2, and Jonathan I. Silverberg, MD, PhD, MPH3
SC
1. Department of Dermatology, Northwestern University Feinberg School of Medicine, Chicago, IL 60611 2. Department of Dermatology and Allergy, Herlev-Gentofte University Hospital, University of Copenhagen, Hellerup, Denmark 3. Departments of Dermatology, Preventive Medicine and Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, 60611
M AN U
Word count: 3,026 Abstract word count: 136 Figures: 2 Tables: 2 Supplemental Tables: 3
TE D
Key Words: atopic dermatitis (AD); burden; cost of care; eczema; hospitalization; inpatient; length of stay; morbidity; mortality; prognosis; Abbreviations used: AD=atopic dermatitis; AD-E=Atopic dermatitis or eczema; ICD-9CM=International Classification of Disease 9th edition Clinical Modification; LOS=length of stay; NIS=Nationwide Inpatient Sample; OR=odds ratio; CI=confidence interval
AC C
EP
Corresponding author: Jonathan I Silverberg, MD, PhD, MPH Department of Dermatology Suite 1600 676 N. St. Clair St. 312-503-4985 Northwestern University Feinberg School of Medicine Chicago, IL 60611 [email protected] JI Silverberg had full access to all the data in the study and takes responsibility for the integrity of the data and accuracy of the data analysis. Study concept and design: JI Silverberg Acquisition of Data: JI Silverberg, D Hsu and S Narla
ACCEPTED MANUSCRIPT Page 2 of 23
AC C
EP
TE D
M AN U
SC
RI PT
Analysis and interpretation of data: S Narla, JI Silverberg, D Hsu, JP Thyssen Drafting of the manuscript: S Narla, JI Silverberg Critical revision of the manuscript for important intellectual content: S Narla, JI Silverberg, D Hsu, JP Thyssen Statistical analysis: JI Silverberg, S Narla and D Hsu Obtained funding: JI Silverberg Administrative technical or material support: None Study supervision: None Financial disclosures: None Funding Support: This publication was made possible with support from the Agency for Healthcare Research and Quality (AHRQ), grant number K12 HS023011, the Dermatology Foundation and American Medical Association Foundation. Design and conduct of the study? No Collection, management, analysis and interpretation of data? No Preparation, review, or approval of the manuscript? No Decision to submit the manuscript for publication? No Conflicts of interest: None
ACCEPTED MANUSCRIPT Page 3 of 23 Abstract: Little is known about the inpatient burden of atopic dermatitis (AD). We sought to determine some risk factors and financial-burden of hospitalizations for AD in the US. Data were analyzed from
RI PT
the 2002-2012 National Inpatient Sample, including of a 20% representative sample of all
hospitalizations in the US. Hospitalization rates for AD or eczema (AD-E) were highest in the
northeast during the winter and south during the summer. Geometric mean cost of care [95% CI] was
SC
lower for a primary diagnosis of AD-E vs. no AD-E in adults ($3,502 [$3,360-$3,651] vs. $6,849 [$6,775-$6,925]; P=0.0004) and children ($2,716 [$2,542-$2,903] vs. $4,488 [$4,302-$4,682];
M AN U
P=0.0004). However, the high prevalence of hospitalization resulted in total inpatient costs of $8,288,083 per-year for adults and $3,333,868 per-year for children. In conclusion, there is a
AC C
EP
TE D
substantial inpatient financial-burden of AD in the US.
ACCEPTED MANUSCRIPT Page 4 of 23 Introduction AD is associated with considerable morbidity and quality of life impairment (Drucker et al., 2017). Patients with AD may require hospitalization for severe intermittent flares or persistent disease
RI PT
refractory to outpatient treatment, sometimes in association with psychiatric comorbidity or inability to properly self-care. However, little is known about the inpatient financial burden of AD in the
SC
United States (US).
The prevalence of AD has increased globally over the past decades (Deckers et al., 2012).
M AN U
Approximately 13% of US children have AD, of which one-third have moderate-severe disease (Silverberg and Simpson, 2013, 2014); 7-10% of US adults have AD (Silverberg et al., 2015, Silverberg and Hanifin, 2013). High prevalences and associated morbidity account for AD being the skin-disease with highest disability-adjusted life-years in the 2010 Global Burden of Disease project
TE D
(Murray et al., 2012, Vos et al., 2012) and considerable cost of care. Older estimates of direct payercosts of AD were $364 million (Lapidus et al., 1993) and $3.8 billion (Ellis et al., 2002) per-year.
EP
Given the rise in AD prevalence in the US over the past 2 decades and large proportion of adults with AD, it is likely that costs of care have increased concomitantly. In addition, the cost-
AC C
burden of hospitalization for AD has not been explored. Overall, only 7% of the US population had an inpatient hospital stay in 2011, which accounted for 29% of all health care expenses, making hospitalization one of the most expensive types of health care treatments (Moore et al., 2006). Interventions aimed at reducing hospitalization for AD may reduce health expenditures and the public-health burden of AD. In the present study, we analyzed the prevalence of hospitalization for AD, length of stay (LOS) and cost of care in the US.
ACCEPTED MANUSCRIPT Page 5 of 23 Results Hospitalization prevalence and trends Overall, there were 87,053,155 discharges in the NIS between 2002-2012. There were 1,541
RI PT
and 5,036 primary admissions for AD and eczema (weighted frequency: 7,344 and 24,036;
prevalence: 15-29 and 63-81 per million hospitalized patients). In particular, there were 582 and 3,748 primary adult admissions (weighted frequency: 2,791 and 17,914; 7.0-14.0 and 47.0-64.0 per
SC
million hospitalized adult patients) and 959 and 1,288 pediatric admissions (weighted frequency: 4,554 and 6,121; 136-271 and 280-314 per million hospitalized pediatric patients) for AD and
M AN U
eczema.
Patients admitted with a primary diagnosis of AD or eczema (AD-E) had significantly higher prevalence of asthma than those without AD-E (24.8% and 11.0% vs. 7.3%; P=0.0004 for both). This
TE D
was true for adults (26.3% and 9.8% vs. 6.8%; P=0.0004 for both) and children (23.9% and 14.6% vs. 13.8%; P=0.0004 and 0.6, respectively). Patients admitted with a primary diagnosis of AD-E had higher prevalences of viral (adults: 2.1% vs. 0.9%; children: 5.2% vs. 4.7%), bacterial (adults: 9.6%
EP
vs. 5.8%; children: 33.3% vs. 6.1%) and any skin infections (adults: 18.9% vs. 4.3%; children: 32.3%
AC C
vs. 3.7%) than those without AD-E (P=0.0004 for all).
Prevalences of hospitalization for AD-E were higher in adults in 2006-2007, 2008-2009, 2010-2011 and 2012 compared with 2002-2003 (P≤0.01 for all) (Figure 1B), but not in children (Figure 1A). Similar results were observed for eczema alone. However, there were no significant differences of the prevalence of hospitalization for AD between 2002-2012.
ACCEPTED MANUSCRIPT Page 6 of 23 Season and region of hospitalization Overall, prevalences of hospitalization for AD-E were highest in spring and summer for adults
RI PT
and children (Table 1).
Hospitalization rates [95% CI] for AD-E were highest in the South (37.3% [34.0, 40.6%]), followed by the Midwest (23.9% [20.8, 27.0%]) and Northeast (23.1% [20.6, 25.6%]) and least in the
SC
West (15.7% [13.5, 18.0%]). Similar regional patterns were observed in adults and children and in models of AD or eczema alone (Supplemental Tables 1-2). Hospitalization rates for AD-E was
in the northeast during the summer.
M AN U
highest in the northeast and lowest in the south during the winter, but highest in the south and lowest
In adults, there were significantly higher odds of hospitalization for AD-E in the northeast
TE D
(bivariate logistic regression; OR [95% CI]: 1.47 [1.26, 1.71]) and midwest (1.28 [1.09, 1.50]). Similar results were found for adults with eczema, but no associations between region and hospitalization for AD-E were observed in children. However, pediatric eczema was associated with
EP
higher odds of hospitalization in the south (1.47 [1.15, 1.87]) and northeast (1.38 [1.07, 1.78]). There
AC C
were no associations between region and hospitalization for AD in children or adults.
Disposition
The majority of adults with AD-E were classified into minor (70.8%) or moderate (23.5%) likelihood of dying by All Patient Refined Diagnosis-Related Group (AP-DRG), whereas almost all children were classified into minor (97.5%) likelihood of dying. Loss of function severity was most commonly minor (adults: 36.3%; children: 59.8%) and moderate (adults: 48.7%; children: 33.8%).
ACCEPTED MANUSCRIPT Page 7 of 23 The majority of adults (79.7%) and children (96.6%) were discharged routinely (Table 2). Similar results were found in analyses of AD or eczema alone (Supplemental Table 3).
RI PT
LOS
LOS in the hospital was significantly shorter in adults and children with vs. without a primary
diagnosis of AD-E (geometric-mean [95% CI] for adults: 2.7 [ 2.7-2-8] vs. 3.5 [3.5-3.5] days; children: 2.4
M AN U
children with diagnoses of AD or eczema alone.
SC
[2.3-2.5] vs. 2.7 [2.7-2.7] days; P=0.0004) (Figure 1C, 1D). Similar results were found in adults and
Cost of care
The total inflation-adjusted cost of care for hospitalization with a primary diagnosis of AD-E was $127,841,466 (adults: $8,288,083 per-year; children: $3,333,868 per-year). N.B. the actual cost
TE D
is actually higher because 55 observations had missing values for charge and/or cost. The geometricmean [95% CI] cost of hospitalization with AD-E vs. neither was $3,502 [$3,360-$3,651] vs. $6,849 [$6,775-$6,925] for adults and $2,716 [$2,542-$2,903] vs. $4,488 [$4,302-$4,682] for children. Costs
EP
of hospitalization for adults and children with AD-E were higher in 2010-2011 and 2012 compared
AC C
with 2002-2003 (P≤0.01 for all) (Figures 1E, 1F).
ACCEPTED MANUSCRIPT Page 8 of 23 Discussion The present study demonstrates a considerable inpatient financial-burden of AD in the US. Between 2002-2012, the prevalence of hospitalization for AD-E was 2.4- and 5.6-fold higher than for
RI PT
psoriasis (Hsu D. Y. et al., 2016) and pemphigus (Hsu et al., 2015) during the same time period, respectively. The prevalence of hospitalization with a primary diagnosis of AD-E increased in adults between 2002 and 2012, compared with stable or decreasing rates of hospitalization previously
SC
observed in psoriasis (Hsu D. Y. et al., 2016) and pemphigus (Hsu et al., 2015), respectively. Mean inpatients costs per hospitalization for AD were only 60% and 33% of the costs per hospitalization
M AN U
for psoriasis or pemphigus (Hsu et al., 2015, Hsu D. Y. et al., 2016). Lower costs per hospitalization for AD-E were due to 62.5% and 50% shorter LOS than previously observed for pemphigus and psoriasis (Hsu et al., 2015, Hsu D. Y. et al., 2016). In the US, hospitalization primarily for acute care of AD-E is often needed for optimized topical and/or systemic therapy in patients with uncontrolled
TE D
skin disease in the outpatient setting. Inpatient treatments include topical corticosteroid wet-wraps, requiring more aggressive and skilled nursing care, and initiation of systemic immunosuppressants. Few hospitals have dedicated dermatology wards with specialized nurses. Most patients hospitalized
EP
primarily with a dermatologic disorder would be admitted to a general ward. LOS for hospitalization may be shorter for AD-E than psoriasis or pemphigus because of a faster treatment response. The
AC C
acute signs and symptoms of AD-E, i.e. erythema, oozing/weeping, scaling and pruritus, may resolve faster with optimized treatment than vesicobullae and erosions in pemphigus or “lakes of pus” and skin sloughing in generalized pustular psoriasis. Further, biologic therapies used in severe pemphigus and psoriasis may contribute to increased cost per hospitalization compared with AD. Hospitalization for AD was associated with good prognosis, including shorter LOS, few invasive procedures, and only mild-moderate disability. Despite a subset of AD patients being at increased risk for inpatient
ACCEPTED MANUSCRIPT Page 9 of 23 mortality, there were virtually no deaths, suggesting that inpatient management successfully mitigated mortality risk. However, due to the higher prevalence of hospitalization for AD, the total inpatient costs for AD were 40% and 70% higher than for psoriasis and pemphigus during the same time
RI PT
period, respectively (Hsu et al., 2015, Hsu D. Y. et al., 2016). Prevalences and costs of hospitalization for AD significantly increased during the study period without plateauing, indicating that the total-
SC
cost of inpatient care for AD may continue to increase.
There are considerable differences in the delivery and costs of dermatological care in the US
M AN U
versus other countries. Few studies assessed the economic burden of AD internationally, with the majority focusing on outpatient costs of AD. A Taiwanese study of the costs of early childhood atopic disease found 2.9 billion Thai baht ($94 million USD), including 302 million Thai baht ($9.7 million USD) (Ngamphaiboon et al., 2012). AD costs were posited to be lower because of differences in
TE D
climate and AD management in Thailand compared to western countries. A retrospective cohort study of AD costs using data from general practitioners (GP) and one general hospital in the Netherlands found that most patients were seen by a GP in the outpatient setting and few were
EP
referred to the hospital (Verboom et al., 2002). Mean healthcare costs (adjusted to the 1999 US Consumer Price Index) were $71 per-patient-year, mostly from GP visits ($32 USD) and medications,
AC C
mostly corticosteroids ($21 USD). Referral to a hospital-based specialist occurred in only 7.8% of AD patients and was associated with higher costs ($186 USD). Hospitalization for AD was rare, with unspecified costs. Costs of AD care were posited to be low in their cohort because of the low hospitalization rates. A cross-sectional survey of 1523 parents of British children age 1-5 years found mean annual costs per-child with AD of ₤79.59, with only ₤1.96 incurred by caregivers for seeing a private specialist (Emerson et al., 2001). That study did not examine costs of AD hospitalization. A
ACCEPTED MANUSCRIPT Page 10 of 23 survey study of 85 Australian adolescents and adults with AD found a mean A$425 annual out-ofpocket costs used for treatment (range: A$13.50- A$2105.64) (Jenner et al., 2004). Direct payercosts, hospitalization rates and costs were not assessed. Another Australian study using modeling and
RI PT
expert opinion found that inpatient costs accounted for more than half of direct-costs of AD in infants (Su et al., 2012). A study of 40 Indian children with AD in an outpatient hospital setting found a high number of flares (mean=3.0) and outpatient hospital visits (mean=6.5), with high provider-costs,
SC
including consultation, nursing/paramedical staff and infrastructure (mean=948 rupees) (Handa et al., 2015). However, acute inpatient care for AD was not assessed in that study. A UK study of 155
M AN U
outpatients with AD and 10 patients hospitalized with AD in the previous year found mean health services costs of £16.2 and £415 in 2 months, respectively, with hospital consultations accounting for 6% of health services costs (Herd et al., 1996). There are several challenges of comparing costs of hospitalization internationally, including different currencies, periods of study, age groups of patients
TE D
and definitions of AD and hospital care. Even among studies that assessed direct-costs related to hospitalization, most used modeling of representative patients, but few examined actual disease-costs
EP
on the population-level.
AD hospitalization were highest in regions with cold-weather during the fall and winter time,
AC C
but warm-weather during the spring and summer time. These patterns likely reflect the effects of climate and other environmental factors on triggering AD flares. A US population-based study of 91,642 children found the 1-year prevalence of AD to be significantly associated with decreased temperature, humidity, and ultraviolet index, and increased use of indoor heating (Silverberg et al., 2013). A systematic review explored the mechanisms of worsening AD in the wintertime and proposed that low humidity and temperature during the winter impairs skin-barrier function and
ACCEPTED MANUSCRIPT Page 11 of 23 increases susceptibility to mechanical stresses, irritants and allergens (Engebretsen et al., 2016). A German study suggested that AD may flare seasonally in both the winter, secondary to low outdoor temperatures, and summer, secondary to high outdoor temperature and grass pollen counts (Kramer et
RI PT
al., 2005). Sweating is one of the most commonly reported exacerbants of itch in AD (Williams et al., 2004, Yosipovitch et al., 2002). Further, higher rates of Staphylococcus aureus skin infections have been
demonstrated globally in warmer regions (Tong et al., 2011). These may contributed toward AD
SC
flares observed in the Southern US during the summer months. Indeed, patients hospitalized with AD-E had higher rates of cutaneous infections. Further studies are needed to elucidate the interaction
M AN U
between climate, skin flora and infection, and AD flares.
We previously found the code 691.8 to be sufficiently valid for identifying AD in the inpatient setting (Hsu DY et al., 2016). We also found that the ICD-9-CM codes for AD (691.8) and eczema
TE D
(692.9) are used interchangeably, with similar proportions of patients meeting diagnostic criteria for AD (Hanifin and Rajka, 1980) whether they were coded as AD or eczema (Hsu DY et al., 2016). This misclassification is further illustrated by a study of the top diagnoses at the dermatologist from 1993-
EP
2010 National Ambulatory Medical Care Survey that found among patients aged 0–4, 18.4% had AD and 16.1% had eczema (Landis et al., 2014). Therefore, we examined both codes and performed
AC C
sensitivity analyses between codes. Indeed, the code for eczema was used more frequently than the code for AD in all age groups. This is particularly interesting in young children, where other eczematous dermatoses are relatively uncommon compared with AD, such as allergic contact dermatitis and eczematous drug eruptions. It is unlikely that other eczematous etiologies would be confused for AD in young children or that such disorders would commonly result in hospitalization for acute care in any age group. Thus, we believe that most of these cases represent miscoding of AD
ACCEPTED MANUSCRIPT Page 12 of 23 as eczema. Nevertheless, some patients who were coded with eczema may have had other etiologies of eczematous dermatoses, which may contribute to the particularly high proportion of eczema cases observed at ages 40-59 years. Therefore, estimates for AD-E may overestimate the costs of AD,
RI PT
whereas analyses of AD or eczema alone underestimate the costs. There is heterogeneity of
terminology in both the scientific literature (Kantor et al., 2016) and global search engine trends (Xu et al., 2017), including AD, eczema, atopic eczema, etc. Future efforts are needed to harmonize the
SC
nomenclature of AD to reduce misclassification and coding errors.
M AN U
Strengths of this study, include a nationally representative sample of >87 million hospitalizations spanning eleven years. Weaknesses of this study include the lack of information about the onset, severity and phenotype of AD, or outpatient treatments and how these impacted hospitalization rates and course. We could not determine if the primary diagnosis of AD was given by
TE D
a dermatologist vs. other physician, or what diagnostic criteria were used. Hospitalizations with a primary diagnosis of AD-E indicate that AD-E was the primary reason for admission. However, we could not determine more specific reasons, such as acute-onset severe flares vs. persistent severe
EP
disease refractory to outpatient treatments. We were also unable to determine how many hospitalizations were due to readmissions. Patients with severe disease or non-adherence to treatment
AC C
may have had multiple admissions. Nevertheless, readmissions of such patients are appropriately reflected in estimates of LOS and cost of hospitalization.
In conclusion, this study demonstrates a considerable inpatient burden of AD in the US, with higher prevalence and total costs than other chronic inflammatory skin disorders. Future research is needed to develop clinical and policy interventions to help reduce hospitalization in AD.
ACCEPTED MANUSCRIPT Page 13 of 23 Methods Data source
RI PT
The 2002-2012 National Inpatient Sample (NIS) was analyzed. The NIS was developed as part of the Healthcare Cost and Utilization Project (HCUP) sponsored by the Agency for Healthcare Research and Quality. Each year of the NIS contains a ~20% stratified representative sample of all
SC
US hospitalizations. All data were de-identified and no attempts were made to identify any of the individuals in the database. All parties with access to the HCUP were compliant to HCUP’s formal
University.
Identification of AD and hospital course
M AN U
data use agreement. The study was approved by the institutional review board at Northwestern
TE D
The databases were searched for a primary discharge diagnosis of AD or eczema using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes 691.8 (AD) and/or 692.9 (eczema). The NIS defined primary diagnosis as the medical condition
EP
principally responsible for the hospitalization of the patient, and was assigned at the time of hospital discharge. Three variables were created for these analyses, including primary diagnosis of: AD,
AC C
eczema, and AD-E. The control group included all hospitalizations without any AD-E and excluded normal pregnancy/delivery, yielding a representative cohort of US hospitalizations.
The Clinical Classifications Software (CCS) codes 128, 197, 3, and 7 were used to identify a diagnosis of asthma, any skin infection, bacterial, and viral infections respectively. ICD-9-CM procedure codes were used to identify use of ventilation (overall, <96 or ≥96 hours) and physical therapy. Mortality risk and functional severity were classified according to the AP-DRG coded in NIS
ACCEPTED MANUSCRIPT Page 14 of 23 based on software developed by the 3M Health Information Services (2008). The AP-DRG incorporates severity of illness subclasses and is used to relate the type of patients treated to the costs
RI PT
incurred by the hospital. Discharge status was assessed.
Statistical analysis
All data analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC). Weighted
SC
prevalences (95% CI) of hospitalization with a primary ICD-9-CM codes of AD and/or eczema
among inpatients were determined. Hospital cost for inpatient care was calculated based on the total-
M AN U
charge of hospitalization and cost-to-charge ratio estimated by HCUP. All costs were adjusted for inflation to the year 2014 according to the Consumer Price Index (Statistics USDoLBoL. CPI Detailed Report June 2015). Summary statistics were generated for LOS, inflation-adjusted cost-ofcare, including sum, geometric mean and 95% CI for hospitalizations with a primary or no diagnosis
TE D
of AD.
All statistical models employed SURVEY procedures, including discharge trend weights,
EP
sample strata accounting for hospital's census region or division, ownership/control, location/teaching and number of beds that were provided by NIS, and clustering by individual hospital. These models
the US.
AC C
allow for representative weighted estimates of frequency and prevalence of hospital discharges across
Complete case-analysis was performed. Missing data were encountered in 78,104 (0.1%) for sex, 0 (0.0%) for age and hospital region. Post-hoc correction for multiple dependent tests was
ACCEPTED MANUSCRIPT Page 15 of 23 performed by minimizing the false discovery rate (Benjamini and Hochberg, 1995). Two-sided
AC C
EP
TE D
M AN U
SC
RI PT
corrected P-values are presented and considered significant if ≤0.05.
ACCEPTED MANUSCRIPT Page 16 of 23 Bibliography Benjamini Y, Hochberg Y. Controlling the false discovery rate - a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol 1995;57:289-300.
RI PT
Deckers IA, McLean S, Linssen S, Mommers M, van Schayck CP, Sheikh A. Investigating
international time trends in the incidence and prevalence of atopic eczema 1990-2010: a systematic review of epidemiological studies. PloS one 2012;7(7):e39803.
SC
Drucker AM, Wang AR, Li WQ, Sevetson E, Block JK, Qureshi AA. The Burden of Atopic
Dermatitis: Summary of a Report for the National Eczema Association. The Journal of
M AN U
investigative dermatology 2017;137(1):26-30.
Ellis CN, Drake LA, Prendergast MM, Abramovits W, Boguniewicz M, Daniel CR, et al. Cost of atopic dermatitis and eczema in the United States. Journal of the American Academy of Dermatology 2002;46(3):361-70.
TE D
Emerson RM, Williams HC, Allen BR. What is the cost of atopic dermatitis in preschool children? The British journal of dermatology 2001;144(3):514-22. Engebretsen KA, Johansen JD, Kezic S, Linneberg A, Thyssen JP. The effect of environmental
EP
humidity and temperature on skin barrier function and dermatitis. Journal of the European Academy of Dermatology and Venereology : JEADV 2016;30(2):223-49.
AC C
Handa S, Jain N, Narang T. Cost of care of atopic dermatitis in India. Indian J Dermatol 2015;60(2):213.
Hanifin J, Rajka G. Diagnostic features of atopic eczema. Acta dermato-venereologica 1980;92:44-7. Herd RM, Tidman MJ, Prescott RJ, Hunter JA. The cost of atopic eczema. The British journal of dermatology 1996;135(1):20-3.
ACCEPTED MANUSCRIPT Page 17 of 23 Hsu D, Brieva J, Silverberg JI. Costs of care for hospitalization for pemphigus in the United States. JAMA dermatology 2015;152(6):645-54. Hsu D, Dalal P, Sable K, Voruganti1 N, Nardone B, West D, et al. Validation of international
RI PT
classification of disease ninth revision codes for atopic dermatitis. Allergy 2016;In Press. Hsu DY, Gordon K, Silverberg JI. The inpatient burden of psoriasis in the United States. Journal of the American Academy of Dermatology 2016;75(1):33-41.
SC
Jenner N, Campbell J, Marks R. Morbidity and cost of atopic eczema in Australia. Australas J Dermatol 2004;45(1):16-22.
M AN U
Kantor R, Thyssen JP, Paller AS, Silverberg JI. Atopic dermatitis, atopic eczema, or eczema? A systematic review, meta-analysis, and recommendation for uniform use of 'atopic dermatitis'. Allergy 2016;71(10):1480-5.
Kramer U, Weidinger S, Darsow U, Mohrenschlager M, Ring J, Behrendt H. Seasonality in symptom
TE D
severity influenced by temperature or grass pollen: results of a panel study in children with eczema. The Journal of investigative dermatology 2005;124(3):514-23. Landis ET, Davis SA, Taheri A, Feldman SR. Top dermatologic diagnoses by age. Dermatol Online J
EP
2014;20(4):22368.
Lapidus CS, Schwarz DF, Honig PJ. Atopic dermatitis in children: who cares? Who pays? Journal of
AC C
the American Academy of Dermatology 1993;28(5 Pt 1):699-703. Moore B, Levit K, Elixhauser A. Costs for Hospital Stays in the United States, 2012: Statistical Brief #181. Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Rockville (MD); 2006.
ACCEPTED MANUSCRIPT Page 18 of 23 Murray CJ, Vos T, Lozano R, Naghavi M, Flaxman AD, Michaud C, et al. Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380(9859):2197-223.
RI PT
Ngamphaiboon J, Kongnakorn T, Detzel P, Sirisomboonwong K, Wasiak R. Direct medical costs associated with atopic diseases among young children in Thailand. Journal of Medical Economics 2012;15(6):1025-35.
SC
Project HCaU. Description of Data Elements: ADRG - All Patient Refined DRG, http://www.hcupus.ahrq.gov/db/vars/siddistnote.jsp?var=adrg; 2008 [accessed August 16, 2016.
M AN U
Silverberg JI, Garg NK, Paller AS, Fishbein AB, Zee PC. Sleep disturbances in adults with eczema are associated with impaired overall health: a US population-based study. The Journal of investigative dermatology 2015;135(1):56-66.
Silverberg JI, Hanifin J, Simpson EL. Climatic factors are associated with childhood eczema
9.
TE D
prevalence in the United States. The Journal of investigative dermatology 2013;133(7):1752-
Silverberg JI, Hanifin JM. Adult eczema prevalence and associations with asthma and other health
EP
and demographic factors: a US population-based study. The Journal of allergy and clinical immunology 2013;132(5):1132-8.
AC C
Silverberg JI, Simpson EL. Association between severe eczema in children and multiple comorbid conditions and increased healthcare utilization. Pediatric allergy and immunology : official publication of the European Society of Pediatric Allergy and Immunology 2013;24(5):476-86. Silverberg JI, Simpson EL. Associations of childhood eczema severity: a US population-based study. Dermatitis : contact, atopic, occupational, drug : official journal of the American Contact Dermatitis Society, North American Contact Dermatitis Group 2014;25(3):107-14.
ACCEPTED MANUSCRIPT Page 19 of 23 Su J, Prescott S, Sinn J, Tang M, Smith P, Heine RG, et al. Cost-effectiveness of partially-hydrolyzed formula for prevention of atopic dermatitis in Australia. J Med Econ 2012;15(6):1064-77. Tong SY, Steer AC, Jenney AW, Carapetis JR. Community-associated methicillin-resistant
RI PT
Staphylococcus aureus skin infections in the tropics. Dermatol Clin 2011;29(1):21-32.
Verboom P, Hakkaart-Van L, Sturkenboom M, De Zeeuw R, Menke H, Rutten F. The cost of atopic dermatitis in the Netherlands: an international comparison. The British journal of dermatology
SC
2002;147(4):716-24.
Vos T, Flaxman AD, Naghavi M, Lozano R, Michaud C, Ezzati M, et al. Years lived with disability
M AN U
(YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380(9859):2163-96. Williams JR, Burr ML, Williams HC. Factors influencing atopic dermatitis-a questionnaire survey of schoolchildren's perceptions. The British journal of dermatology 2004;150(6):1154-61.
TE D
Xu S, Thyssen J, Paller A, Silverberg J. Eczema, atopic dermatitis or atopic eczema? An analysis of global search engine trends. Dermatitis : contact, atopic, occupational, drug : official journal of the American Contact Dermatitis Society, North American Contact Dermatitis Group
EP
2017;In Press.
Yosipovitch G, Goon AT, Wee J, Chan YH, Zucker I, Goh CL. Itch characteristics in Chinese
AC C
patients with atopic dermatitis using a new questionnaire for the assessment of pruritus. International journal of dermatology 2002;41(4):212-6.
ACCEPTED MANUSCRIPT Page 20 of 23 Figure Legends. Figure 1. Prevalence of hospitalization for adults and children, length of stay of hospitalization for adults and children and cost of hospitalization for adults and children with atopic
RI PT
dermatitis or eczema. (A) Prevalence of hospitalization per million hospitalized adults for a primary diagnosis of atopic dermatitis (dotted), eczema (dashed), and atopic dermatitis and eczema combined (solid) during the years of 2002-2003, 2004-2005, 2006-2007, 2008-2009, 2010-2011, and 2012. (B)
SC
Prevalence of hospitalization per million hospitalized children for a primary diagnosis of atopic
dermatitis (dotted), eczema (dashed), and atopic dermatitis and eczema combined (solid) during the
M AN U
years of 2002-2003, 2004-2005, 2006-2007, 2008-2009, 2010-2011, and 2012.
Geometric mean LOS in (C) adults or (D) children with (circles) or without (squares) a primary diagnosis of atopic dermatitis (dashed line), eczema (dotted line), and atopic dermatitis and eczema combined (solid line) during the years of 2002-2003, 2004-2005, 2006-2007, 2008-2009, 2010-2011,
TE D
and 2012.
Geometric mean cost in (E) adults or (F) children with (circles) or without (squares) a primary diagnosis of atopic dermatitis (dashed line), eczema (dotted line), and atopic dermatitis and eczema
EP
combined (solid line) during the years of 2002-2003, 2004-2005, 2006-2007, 2008-2009, 2010-2011,
AC C
and 2012.
Figure 2. Weighted frequency of hospitalization for adults and children with atopic dermatitis or eczema.
ACCEPTED MANUSCRIPT Page 21 of 23
Summer Fall
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Seasonal and regional variation of hospitalization with a primary diagnosis of atopic dermatitis or eczema. Primary inpatient diagnosis of AD or eczema No Yes Variable Wtd % Prev Wtd % Prev Crude OR P Freq [95% CI] Freq [95% CI] [95% CI] Adults Season 25.5 23.2 0.88 Winter 70,338,854 4,400 .01 [25.5, 25.6] [21.9, 24.5] [0.80, 0.96] 25.0 25.9 4,908 1.00 Spring 68,902,498 [25.0, 25.0] [24.5, 27.3] 24.8 26.5 1.03 5,019 .6 Summer 68,336,209 [24.8, 24.8] [25.1, 27.9] [0.94, 1.13] 24.6 24.4 0.96 Fall 67,906,561 4,626 .5 [24.6, 24.7] [23.0, 25.8] [0.87, 1.05] Hospital region 20.3 25.1 1.47 Northeast 61,096,853 5,195 .0004 [19.1, 21.6] [22.7, 27.5] [1.26, 1.71] 23.5 25.2 1.28 Midwest 70,543,693 5,226 .01 [22.2, 24.7] [22.7, 27.7] [1.09, 1.50] 38.5 34.8 1.08 South 115,671,661 7,204 .5 [36.8, 40.1] [32.1, 37.5] [0.93, 1.24] 17.7 14.9 West 53,177,599 3,081 1.00 [16.6, 18.8] [12.9, 16.8] Children Season 28.9 23.8 0.71 Winter 60,621,151 2,2287 .0004 [28.7, 29.2] [22.0, 25.7] [0.63, 0.79] 27.6 23.6 Spring 4,943,979 2,644 [25.6, 29.5] 1.00 [23.5, 23.7] 4,707,635
5,234,233
22.5 [22.3, 22.6] 25.0 [24.9, 25.1]
2,449 2,216
25.5 [23.7, 27.3] 23.1 [21.1, 25.1]
0.97 [0.87, 1.09] 0.79 [0.69, 0.90]
19.2 [14.8, 23.5] 21.4 [14.1, 28.7] 42.1 [35.0, 49.2]
1.28 [0.95, 1.71] 1.26 [0.83, 1.90] 1.33 [0.99, 1.78]
.7 .002
Hospital region Northeast
4,221,843
Midwest
4,785,291
South
8,891,062
18.6 [15.9, 21.2] 21.0 [18.0, 24.1] 39.1 [35.4, 42.7]
2,048 2,285 4,495
.2 .4 .1
ACCEPTED MANUSCRIPT Page 22 of 23 21.3 [18.4, 24.3]
1,846
17.3 [12.3, 22.3]
1.00
-
EP
TE D
M AN U
SC
RI PT
4,852,928
AC C
West
ACCEPTED MANUSCRIPT Page 23 of 23
AC C
EP
TE D
M AN U
SC
RI PT
Table 2. Procedures, mortality risk, and loss of function risk for adults and children admitted with a primary diagnosis of atopic dermatitis or eczema. Adults Children Variable % Prev Wtd % Prev Wtd Freq [95% CI] Freq [95% CI] Ventilation Yes 29 0.1 [0.0, 0.2] 6 0.1 [0.0, 0.2] No 20,677 99.9 [99.8, 100.0] 10,669 99.9 [99.8, 100.0] Ventilation <96h Yes 19 0.1 [0.0, 0.2] 0 0 [0.0, 0.0] No 20,687 99.9 [99.8, 100.0] 10,675 100.0 [100.0, 100.0] Ventilation >96h Yes 10 0.05 [0.00, 0.1] 6 0.1 [0.0, 0.2] No 20,696 99.95 [99.9, 100.0] 10,669 99.9 [99.8, 100.0] Physical Therapy Yes 75 0.4 [0.2, 0.5] 91 0.9 [0.2, 1.5] No 20,630 99.6 [99.5, 99.8] 10,584 99.1 [98.5, 99.8] Mortality risk Minor 14,510 70.8 [69.3, 72.4] 10,336 97.5 [96.7, 98.2] Moderate 4,816 23.5 [22.1, 24.9] 230 2.2 [1.4, 2.9] Major 1,039 5.1 [4.4, 5.7] 23 0.2 [0.0, 0.4] Extreme 119 0.6 [0.4, 0.8] 6 0.1 [0.0, 0.2] Loss of function severity Minor 7,405 36.3 [34.5, 37.8] 6,341 59.8 [56.2, 63.4] Moderate 9,972 48.7 [47.1, 50.2] 3,584 33.8 [30.6, 37.0] Major 2,895 14.1 [13.0, 15.3] 640 6.0 [4.8, 7.3] Extreme 212 1.0 [0.7, 1.3] 29 0.3 [0.1, 0.5] Patient disposition at discharge Routine 16,488 79.7 [78.4, 81.0] 10,316 96.6 [95.7, 97.6] Transfer, short-term hospital 210 1.0 [0.7, 1.3] 97 0.9 [0.5, 1.3] Transfer, SNF, ICF or other facility 1,977 9.6 [8.6, 10.5] 28 0.3 [0.1, 0.5] Home health care 1,656 8.0 [7.2, 8.8] 216 2.0 [1.3, 2.8] Left against medical advice 317 1.5 [1.2, 1.9] 18 0.2 [0.0, 0.3] Died 42 0.2 [0.1, 0.3] 0 0.0 [0.0, 0.0]
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0022-202X(17)31160-0
DOI:
10.1016/j.jid.2017.02.975
Reference:
JID 781
To appear in:
The Journal of Investigative Dermatology
Received Date: 22 December 2016 Revised Date:
6 February 2017
Accepted Date: 7 February 2017
Please cite this article as: Narla S, Hsu DY, Thyssen JP, Silverberg JI, Inpatient financial burden of atopic dermatitis in the United States, The Journal of Investigative Dermatology (2017), doi: 10.1016/ j.jid.2017.02.975. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Page 1 of 23 Title: Inpatient financial burden of atopic dermatitis in the United States Short Title: Hospitalization of AD.
RI PT
Shanthi Narla, BA, Derek Y. Hsu, BA1 , Jacob P. Thyssen, MD, PhD, DmSci2, and Jonathan I. Silverberg, MD, PhD, MPH3
SC
1. Department of Dermatology, Northwestern University Feinberg School of Medicine, Chicago, IL 60611 2. Department of Dermatology and Allergy, Herlev-Gentofte University Hospital, University of Copenhagen, Hellerup, Denmark 3. Departments of Dermatology, Preventive Medicine and Medical Social Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, 60611
M AN U
Word count: 3,026 Abstract word count: 136 Figures: 2 Tables: 2 Supplemental Tables: 3
TE D
Key Words: atopic dermatitis (AD); burden; cost of care; eczema; hospitalization; inpatient; length of stay; morbidity; mortality; prognosis; Abbreviations used: AD=atopic dermatitis; AD-E=Atopic dermatitis or eczema; ICD-9CM=International Classification of Disease 9th edition Clinical Modification; LOS=length of stay; NIS=Nationwide Inpatient Sample; OR=odds ratio; CI=confidence interval
AC C
EP
Corresponding author: Jonathan I Silverberg, MD, PhD, MPH Department of Dermatology Suite 1600 676 N. St. Clair St. 312-503-4985 Northwestern University Feinberg School of Medicine Chicago, IL 60611 [email protected] JI Silverberg had full access to all the data in the study and takes responsibility for the integrity of the data and accuracy of the data analysis. Study concept and design: JI Silverberg Acquisition of Data: JI Silverberg, D Hsu and S Narla
ACCEPTED MANUSCRIPT Page 2 of 23
AC C
EP
TE D
M AN U
SC
RI PT
Analysis and interpretation of data: S Narla, JI Silverberg, D Hsu, JP Thyssen Drafting of the manuscript: S Narla, JI Silverberg Critical revision of the manuscript for important intellectual content: S Narla, JI Silverberg, D Hsu, JP Thyssen Statistical analysis: JI Silverberg, S Narla and D Hsu Obtained funding: JI Silverberg Administrative technical or material support: None Study supervision: None Financial disclosures: None Funding Support: This publication was made possible with support from the Agency for Healthcare Research and Quality (AHRQ), grant number K12 HS023011, the Dermatology Foundation and American Medical Association Foundation. Design and conduct of the study? No Collection, management, analysis and interpretation of data? No Preparation, review, or approval of the manuscript? No Decision to submit the manuscript for publication? No Conflicts of interest: None
ACCEPTED MANUSCRIPT Page 3 of 23 Abstract: Little is known about the inpatient burden of atopic dermatitis (AD). We sought to determine some risk factors and financial-burden of hospitalizations for AD in the US. Data were analyzed from
RI PT
the 2002-2012 National Inpatient Sample, including of a 20% representative sample of all
hospitalizations in the US. Hospitalization rates for AD or eczema (AD-E) were highest in the
northeast during the winter and south during the summer. Geometric mean cost of care [95% CI] was
SC
lower for a primary diagnosis of AD-E vs. no AD-E in adults ($3,502 [$3,360-$3,651] vs. $6,849 [$6,775-$6,925]; P=0.0004) and children ($2,716 [$2,542-$2,903] vs. $4,488 [$4,302-$4,682];
M AN U
P=0.0004). However, the high prevalence of hospitalization resulted in total inpatient costs of $8,288,083 per-year for adults and $3,333,868 per-year for children. In conclusion, there is a
AC C
EP
TE D
substantial inpatient financial-burden of AD in the US.
ACCEPTED MANUSCRIPT Page 4 of 23 Introduction AD is associated with considerable morbidity and quality of life impairment (Drucker et al., 2017). Patients with AD may require hospitalization for severe intermittent flares or persistent disease
RI PT
refractory to outpatient treatment, sometimes in association with psychiatric comorbidity or inability to properly self-care. However, little is known about the inpatient financial burden of AD in the
SC
United States (US).
The prevalence of AD has increased globally over the past decades (Deckers et al., 2012).
M AN U
Approximately 13% of US children have AD, of which one-third have moderate-severe disease (Silverberg and Simpson, 2013, 2014); 7-10% of US adults have AD (Silverberg et al., 2015, Silverberg and Hanifin, 2013). High prevalences and associated morbidity account for AD being the skin-disease with highest disability-adjusted life-years in the 2010 Global Burden of Disease project
TE D
(Murray et al., 2012, Vos et al., 2012) and considerable cost of care. Older estimates of direct payercosts of AD were $364 million (Lapidus et al., 1993) and $3.8 billion (Ellis et al., 2002) per-year.
EP
Given the rise in AD prevalence in the US over the past 2 decades and large proportion of adults with AD, it is likely that costs of care have increased concomitantly. In addition, the cost-
AC C
burden of hospitalization for AD has not been explored. Overall, only 7% of the US population had an inpatient hospital stay in 2011, which accounted for 29% of all health care expenses, making hospitalization one of the most expensive types of health care treatments (Moore et al., 2006). Interventions aimed at reducing hospitalization for AD may reduce health expenditures and the public-health burden of AD. In the present study, we analyzed the prevalence of hospitalization for AD, length of stay (LOS) and cost of care in the US.
ACCEPTED MANUSCRIPT Page 5 of 23 Results Hospitalization prevalence and trends Overall, there were 87,053,155 discharges in the NIS between 2002-2012. There were 1,541
RI PT
and 5,036 primary admissions for AD and eczema (weighted frequency: 7,344 and 24,036;
prevalence: 15-29 and 63-81 per million hospitalized patients). In particular, there were 582 and 3,748 primary adult admissions (weighted frequency: 2,791 and 17,914; 7.0-14.0 and 47.0-64.0 per
SC
million hospitalized adult patients) and 959 and 1,288 pediatric admissions (weighted frequency: 4,554 and 6,121; 136-271 and 280-314 per million hospitalized pediatric patients) for AD and
M AN U
eczema.
Patients admitted with a primary diagnosis of AD or eczema (AD-E) had significantly higher prevalence of asthma than those without AD-E (24.8% and 11.0% vs. 7.3%; P=0.0004 for both). This
TE D
was true for adults (26.3% and 9.8% vs. 6.8%; P=0.0004 for both) and children (23.9% and 14.6% vs. 13.8%; P=0.0004 and 0.6, respectively). Patients admitted with a primary diagnosis of AD-E had higher prevalences of viral (adults: 2.1% vs. 0.9%; children: 5.2% vs. 4.7%), bacterial (adults: 9.6%
EP
vs. 5.8%; children: 33.3% vs. 6.1%) and any skin infections (adults: 18.9% vs. 4.3%; children: 32.3%
AC C
vs. 3.7%) than those without AD-E (P=0.0004 for all).
Prevalences of hospitalization for AD-E were higher in adults in 2006-2007, 2008-2009, 2010-2011 and 2012 compared with 2002-2003 (P≤0.01 for all) (Figure 1B), but not in children (Figure 1A). Similar results were observed for eczema alone. However, there were no significant differences of the prevalence of hospitalization for AD between 2002-2012.
ACCEPTED MANUSCRIPT Page 6 of 23 Season and region of hospitalization Overall, prevalences of hospitalization for AD-E were highest in spring and summer for adults
RI PT
and children (Table 1).
Hospitalization rates [95% CI] for AD-E were highest in the South (37.3% [34.0, 40.6%]), followed by the Midwest (23.9% [20.8, 27.0%]) and Northeast (23.1% [20.6, 25.6%]) and least in the
SC
West (15.7% [13.5, 18.0%]). Similar regional patterns were observed in adults and children and in models of AD or eczema alone (Supplemental Tables 1-2). Hospitalization rates for AD-E was
in the northeast during the summer.
M AN U
highest in the northeast and lowest in the south during the winter, but highest in the south and lowest
In adults, there were significantly higher odds of hospitalization for AD-E in the northeast
TE D
(bivariate logistic regression; OR [95% CI]: 1.47 [1.26, 1.71]) and midwest (1.28 [1.09, 1.50]). Similar results were found for adults with eczema, but no associations between region and hospitalization for AD-E were observed in children. However, pediatric eczema was associated with
EP
higher odds of hospitalization in the south (1.47 [1.15, 1.87]) and northeast (1.38 [1.07, 1.78]). There
AC C
were no associations between region and hospitalization for AD in children or adults.
Disposition
The majority of adults with AD-E were classified into minor (70.8%) or moderate (23.5%) likelihood of dying by All Patient Refined Diagnosis-Related Group (AP-DRG), whereas almost all children were classified into minor (97.5%) likelihood of dying. Loss of function severity was most commonly minor (adults: 36.3%; children: 59.8%) and moderate (adults: 48.7%; children: 33.8%).
ACCEPTED MANUSCRIPT Page 7 of 23 The majority of adults (79.7%) and children (96.6%) were discharged routinely (Table 2). Similar results were found in analyses of AD or eczema alone (Supplemental Table 3).
RI PT
LOS
LOS in the hospital was significantly shorter in adults and children with vs. without a primary
diagnosis of AD-E (geometric-mean [95% CI] for adults: 2.7 [ 2.7-2-8] vs. 3.5 [3.5-3.5] days; children: 2.4
M AN U
children with diagnoses of AD or eczema alone.
SC
[2.3-2.5] vs. 2.7 [2.7-2.7] days; P=0.0004) (Figure 1C, 1D). Similar results were found in adults and
Cost of care
The total inflation-adjusted cost of care for hospitalization with a primary diagnosis of AD-E was $127,841,466 (adults: $8,288,083 per-year; children: $3,333,868 per-year). N.B. the actual cost
TE D
is actually higher because 55 observations had missing values for charge and/or cost. The geometricmean [95% CI] cost of hospitalization with AD-E vs. neither was $3,502 [$3,360-$3,651] vs. $6,849 [$6,775-$6,925] for adults and $2,716 [$2,542-$2,903] vs. $4,488 [$4,302-$4,682] for children. Costs
EP
of hospitalization for adults and children with AD-E were higher in 2010-2011 and 2012 compared
AC C
with 2002-2003 (P≤0.01 for all) (Figures 1E, 1F).
ACCEPTED MANUSCRIPT Page 8 of 23 Discussion The present study demonstrates a considerable inpatient financial-burden of AD in the US. Between 2002-2012, the prevalence of hospitalization for AD-E was 2.4- and 5.6-fold higher than for
RI PT
psoriasis (Hsu D. Y. et al., 2016) and pemphigus (Hsu et al., 2015) during the same time period, respectively. The prevalence of hospitalization with a primary diagnosis of AD-E increased in adults between 2002 and 2012, compared with stable or decreasing rates of hospitalization previously
SC
observed in psoriasis (Hsu D. Y. et al., 2016) and pemphigus (Hsu et al., 2015), respectively. Mean inpatients costs per hospitalization for AD were only 60% and 33% of the costs per hospitalization
M AN U
for psoriasis or pemphigus (Hsu et al., 2015, Hsu D. Y. et al., 2016). Lower costs per hospitalization for AD-E were due to 62.5% and 50% shorter LOS than previously observed for pemphigus and psoriasis (Hsu et al., 2015, Hsu D. Y. et al., 2016). In the US, hospitalization primarily for acute care of AD-E is often needed for optimized topical and/or systemic therapy in patients with uncontrolled
TE D
skin disease in the outpatient setting. Inpatient treatments include topical corticosteroid wet-wraps, requiring more aggressive and skilled nursing care, and initiation of systemic immunosuppressants. Few hospitals have dedicated dermatology wards with specialized nurses. Most patients hospitalized
EP
primarily with a dermatologic disorder would be admitted to a general ward. LOS for hospitalization may be shorter for AD-E than psoriasis or pemphigus because of a faster treatment response. The
AC C
acute signs and symptoms of AD-E, i.e. erythema, oozing/weeping, scaling and pruritus, may resolve faster with optimized treatment than vesicobullae and erosions in pemphigus or “lakes of pus” and skin sloughing in generalized pustular psoriasis. Further, biologic therapies used in severe pemphigus and psoriasis may contribute to increased cost per hospitalization compared with AD. Hospitalization for AD was associated with good prognosis, including shorter LOS, few invasive procedures, and only mild-moderate disability. Despite a subset of AD patients being at increased risk for inpatient
ACCEPTED MANUSCRIPT Page 9 of 23 mortality, there were virtually no deaths, suggesting that inpatient management successfully mitigated mortality risk. However, due to the higher prevalence of hospitalization for AD, the total inpatient costs for AD were 40% and 70% higher than for psoriasis and pemphigus during the same time
RI PT
period, respectively (Hsu et al., 2015, Hsu D. Y. et al., 2016). Prevalences and costs of hospitalization for AD significantly increased during the study period without plateauing, indicating that the total-
SC
cost of inpatient care for AD may continue to increase.
There are considerable differences in the delivery and costs of dermatological care in the US
M AN U
versus other countries. Few studies assessed the economic burden of AD internationally, with the majority focusing on outpatient costs of AD. A Taiwanese study of the costs of early childhood atopic disease found 2.9 billion Thai baht ($94 million USD), including 302 million Thai baht ($9.7 million USD) (Ngamphaiboon et al., 2012). AD costs were posited to be lower because of differences in
TE D
climate and AD management in Thailand compared to western countries. A retrospective cohort study of AD costs using data from general practitioners (GP) and one general hospital in the Netherlands found that most patients were seen by a GP in the outpatient setting and few were
EP
referred to the hospital (Verboom et al., 2002). Mean healthcare costs (adjusted to the 1999 US Consumer Price Index) were $71 per-patient-year, mostly from GP visits ($32 USD) and medications,
AC C
mostly corticosteroids ($21 USD). Referral to a hospital-based specialist occurred in only 7.8% of AD patients and was associated with higher costs ($186 USD). Hospitalization for AD was rare, with unspecified costs. Costs of AD care were posited to be low in their cohort because of the low hospitalization rates. A cross-sectional survey of 1523 parents of British children age 1-5 years found mean annual costs per-child with AD of ₤79.59, with only ₤1.96 incurred by caregivers for seeing a private specialist (Emerson et al., 2001). That study did not examine costs of AD hospitalization. A
ACCEPTED MANUSCRIPT Page 10 of 23 survey study of 85 Australian adolescents and adults with AD found a mean A$425 annual out-ofpocket costs used for treatment (range: A$13.50- A$2105.64) (Jenner et al., 2004). Direct payercosts, hospitalization rates and costs were not assessed. Another Australian study using modeling and
RI PT
expert opinion found that inpatient costs accounted for more than half of direct-costs of AD in infants (Su et al., 2012). A study of 40 Indian children with AD in an outpatient hospital setting found a high number of flares (mean=3.0) and outpatient hospital visits (mean=6.5), with high provider-costs,
SC
including consultation, nursing/paramedical staff and infrastructure (mean=948 rupees) (Handa et al., 2015). However, acute inpatient care for AD was not assessed in that study. A UK study of 155
M AN U
outpatients with AD and 10 patients hospitalized with AD in the previous year found mean health services costs of £16.2 and £415 in 2 months, respectively, with hospital consultations accounting for 6% of health services costs (Herd et al., 1996). There are several challenges of comparing costs of hospitalization internationally, including different currencies, periods of study, age groups of patients
TE D
and definitions of AD and hospital care. Even among studies that assessed direct-costs related to hospitalization, most used modeling of representative patients, but few examined actual disease-costs
EP
on the population-level.
AD hospitalization were highest in regions with cold-weather during the fall and winter time,
AC C
but warm-weather during the spring and summer time. These patterns likely reflect the effects of climate and other environmental factors on triggering AD flares. A US population-based study of 91,642 children found the 1-year prevalence of AD to be significantly associated with decreased temperature, humidity, and ultraviolet index, and increased use of indoor heating (Silverberg et al., 2013). A systematic review explored the mechanisms of worsening AD in the wintertime and proposed that low humidity and temperature during the winter impairs skin-barrier function and
ACCEPTED MANUSCRIPT Page 11 of 23 increases susceptibility to mechanical stresses, irritants and allergens (Engebretsen et al., 2016). A German study suggested that AD may flare seasonally in both the winter, secondary to low outdoor temperatures, and summer, secondary to high outdoor temperature and grass pollen counts (Kramer et
RI PT
al., 2005). Sweating is one of the most commonly reported exacerbants of itch in AD (Williams et al., 2004, Yosipovitch et al., 2002). Further, higher rates of Staphylococcus aureus skin infections have been
demonstrated globally in warmer regions (Tong et al., 2011). These may contributed toward AD
SC
flares observed in the Southern US during the summer months. Indeed, patients hospitalized with AD-E had higher rates of cutaneous infections. Further studies are needed to elucidate the interaction
M AN U
between climate, skin flora and infection, and AD flares.
We previously found the code 691.8 to be sufficiently valid for identifying AD in the inpatient setting (Hsu DY et al., 2016). We also found that the ICD-9-CM codes for AD (691.8) and eczema
TE D
(692.9) are used interchangeably, with similar proportions of patients meeting diagnostic criteria for AD (Hanifin and Rajka, 1980) whether they were coded as AD or eczema (Hsu DY et al., 2016). This misclassification is further illustrated by a study of the top diagnoses at the dermatologist from 1993-
EP
2010 National Ambulatory Medical Care Survey that found among patients aged 0–4, 18.4% had AD and 16.1% had eczema (Landis et al., 2014). Therefore, we examined both codes and performed
AC C
sensitivity analyses between codes. Indeed, the code for eczema was used more frequently than the code for AD in all age groups. This is particularly interesting in young children, where other eczematous dermatoses are relatively uncommon compared with AD, such as allergic contact dermatitis and eczematous drug eruptions. It is unlikely that other eczematous etiologies would be confused for AD in young children or that such disorders would commonly result in hospitalization for acute care in any age group. Thus, we believe that most of these cases represent miscoding of AD
ACCEPTED MANUSCRIPT Page 12 of 23 as eczema. Nevertheless, some patients who were coded with eczema may have had other etiologies of eczematous dermatoses, which may contribute to the particularly high proportion of eczema cases observed at ages 40-59 years. Therefore, estimates for AD-E may overestimate the costs of AD,
RI PT
whereas analyses of AD or eczema alone underestimate the costs. There is heterogeneity of
terminology in both the scientific literature (Kantor et al., 2016) and global search engine trends (Xu et al., 2017), including AD, eczema, atopic eczema, etc. Future efforts are needed to harmonize the
SC
nomenclature of AD to reduce misclassification and coding errors.
M AN U
Strengths of this study, include a nationally representative sample of >87 million hospitalizations spanning eleven years. Weaknesses of this study include the lack of information about the onset, severity and phenotype of AD, or outpatient treatments and how these impacted hospitalization rates and course. We could not determine if the primary diagnosis of AD was given by
TE D
a dermatologist vs. other physician, or what diagnostic criteria were used. Hospitalizations with a primary diagnosis of AD-E indicate that AD-E was the primary reason for admission. However, we could not determine more specific reasons, such as acute-onset severe flares vs. persistent severe
EP
disease refractory to outpatient treatments. We were also unable to determine how many hospitalizations were due to readmissions. Patients with severe disease or non-adherence to treatment
AC C
may have had multiple admissions. Nevertheless, readmissions of such patients are appropriately reflected in estimates of LOS and cost of hospitalization.
In conclusion, this study demonstrates a considerable inpatient burden of AD in the US, with higher prevalence and total costs than other chronic inflammatory skin disorders. Future research is needed to develop clinical and policy interventions to help reduce hospitalization in AD.
ACCEPTED MANUSCRIPT Page 13 of 23 Methods Data source
RI PT
The 2002-2012 National Inpatient Sample (NIS) was analyzed. The NIS was developed as part of the Healthcare Cost and Utilization Project (HCUP) sponsored by the Agency for Healthcare Research and Quality. Each year of the NIS contains a ~20% stratified representative sample of all
SC
US hospitalizations. All data were de-identified and no attempts were made to identify any of the individuals in the database. All parties with access to the HCUP were compliant to HCUP’s formal
University.
Identification of AD and hospital course
M AN U
data use agreement. The study was approved by the institutional review board at Northwestern
TE D
The databases were searched for a primary discharge diagnosis of AD or eczema using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes 691.8 (AD) and/or 692.9 (eczema). The NIS defined primary diagnosis as the medical condition
EP
principally responsible for the hospitalization of the patient, and was assigned at the time of hospital discharge. Three variables were created for these analyses, including primary diagnosis of: AD,
AC C
eczema, and AD-E. The control group included all hospitalizations without any AD-E and excluded normal pregnancy/delivery, yielding a representative cohort of US hospitalizations.
The Clinical Classifications Software (CCS) codes 128, 197, 3, and 7 were used to identify a diagnosis of asthma, any skin infection, bacterial, and viral infections respectively. ICD-9-CM procedure codes were used to identify use of ventilation (overall, <96 or ≥96 hours) and physical therapy. Mortality risk and functional severity were classified according to the AP-DRG coded in NIS
ACCEPTED MANUSCRIPT Page 14 of 23 based on software developed by the 3M Health Information Services (2008). The AP-DRG incorporates severity of illness subclasses and is used to relate the type of patients treated to the costs
RI PT
incurred by the hospital. Discharge status was assessed.
Statistical analysis
All data analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC). Weighted
SC
prevalences (95% CI) of hospitalization with a primary ICD-9-CM codes of AD and/or eczema
among inpatients were determined. Hospital cost for inpatient care was calculated based on the total-
M AN U
charge of hospitalization and cost-to-charge ratio estimated by HCUP. All costs were adjusted for inflation to the year 2014 according to the Consumer Price Index (Statistics USDoLBoL. CPI Detailed Report June 2015). Summary statistics were generated for LOS, inflation-adjusted cost-ofcare, including sum, geometric mean and 95% CI for hospitalizations with a primary or no diagnosis
TE D
of AD.
All statistical models employed SURVEY procedures, including discharge trend weights,
EP
sample strata accounting for hospital's census region or division, ownership/control, location/teaching and number of beds that were provided by NIS, and clustering by individual hospital. These models
the US.
AC C
allow for representative weighted estimates of frequency and prevalence of hospital discharges across
Complete case-analysis was performed. Missing data were encountered in 78,104 (0.1%) for sex, 0 (0.0%) for age and hospital region. Post-hoc correction for multiple dependent tests was
ACCEPTED MANUSCRIPT Page 15 of 23 performed by minimizing the false discovery rate (Benjamini and Hochberg, 1995). Two-sided
AC C
EP
TE D
M AN U
SC
RI PT
corrected P-values are presented and considered significant if ≤0.05.
ACCEPTED MANUSCRIPT Page 16 of 23 Bibliography Benjamini Y, Hochberg Y. Controlling the false discovery rate - a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol 1995;57:289-300.
RI PT
Deckers IA, McLean S, Linssen S, Mommers M, van Schayck CP, Sheikh A. Investigating
international time trends in the incidence and prevalence of atopic eczema 1990-2010: a systematic review of epidemiological studies. PloS one 2012;7(7):e39803.
SC
Drucker AM, Wang AR, Li WQ, Sevetson E, Block JK, Qureshi AA. The Burden of Atopic
Dermatitis: Summary of a Report for the National Eczema Association. The Journal of
M AN U
investigative dermatology 2017;137(1):26-30.
Ellis CN, Drake LA, Prendergast MM, Abramovits W, Boguniewicz M, Daniel CR, et al. Cost of atopic dermatitis and eczema in the United States. Journal of the American Academy of Dermatology 2002;46(3):361-70.
TE D
Emerson RM, Williams HC, Allen BR. What is the cost of atopic dermatitis in preschool children? The British journal of dermatology 2001;144(3):514-22. Engebretsen KA, Johansen JD, Kezic S, Linneberg A, Thyssen JP. The effect of environmental
EP
humidity and temperature on skin barrier function and dermatitis. Journal of the European Academy of Dermatology and Venereology : JEADV 2016;30(2):223-49.
AC C
Handa S, Jain N, Narang T. Cost of care of atopic dermatitis in India. Indian J Dermatol 2015;60(2):213.
Hanifin J, Rajka G. Diagnostic features of atopic eczema. Acta dermato-venereologica 1980;92:44-7. Herd RM, Tidman MJ, Prescott RJ, Hunter JA. The cost of atopic eczema. The British journal of dermatology 1996;135(1):20-3.
ACCEPTED MANUSCRIPT Page 17 of 23 Hsu D, Brieva J, Silverberg JI. Costs of care for hospitalization for pemphigus in the United States. JAMA dermatology 2015;152(6):645-54. Hsu D, Dalal P, Sable K, Voruganti1 N, Nardone B, West D, et al. Validation of international
RI PT
classification of disease ninth revision codes for atopic dermatitis. Allergy 2016;In Press. Hsu DY, Gordon K, Silverberg JI. The inpatient burden of psoriasis in the United States. Journal of the American Academy of Dermatology 2016;75(1):33-41.
SC
Jenner N, Campbell J, Marks R. Morbidity and cost of atopic eczema in Australia. Australas J Dermatol 2004;45(1):16-22.
M AN U
Kantor R, Thyssen JP, Paller AS, Silverberg JI. Atopic dermatitis, atopic eczema, or eczema? A systematic review, meta-analysis, and recommendation for uniform use of 'atopic dermatitis'. Allergy 2016;71(10):1480-5.
Kramer U, Weidinger S, Darsow U, Mohrenschlager M, Ring J, Behrendt H. Seasonality in symptom
TE D
severity influenced by temperature or grass pollen: results of a panel study in children with eczema. The Journal of investigative dermatology 2005;124(3):514-23. Landis ET, Davis SA, Taheri A, Feldman SR. Top dermatologic diagnoses by age. Dermatol Online J
EP
2014;20(4):22368.
Lapidus CS, Schwarz DF, Honig PJ. Atopic dermatitis in children: who cares? Who pays? Journal of
AC C
the American Academy of Dermatology 1993;28(5 Pt 1):699-703. Moore B, Levit K, Elixhauser A. Costs for Hospital Stays in the United States, 2012: Statistical Brief #181. Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Rockville (MD); 2006.
ACCEPTED MANUSCRIPT Page 18 of 23 Murray CJ, Vos T, Lozano R, Naghavi M, Flaxman AD, Michaud C, et al. Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380(9859):2197-223.
RI PT
Ngamphaiboon J, Kongnakorn T, Detzel P, Sirisomboonwong K, Wasiak R. Direct medical costs associated with atopic diseases among young children in Thailand. Journal of Medical Economics 2012;15(6):1025-35.
SC
Project HCaU. Description of Data Elements: ADRG - All Patient Refined DRG, http://www.hcupus.ahrq.gov/db/vars/siddistnote.jsp?var=adrg; 2008 [accessed August 16, 2016.
M AN U
Silverberg JI, Garg NK, Paller AS, Fishbein AB, Zee PC. Sleep disturbances in adults with eczema are associated with impaired overall health: a US population-based study. The Journal of investigative dermatology 2015;135(1):56-66.
Silverberg JI, Hanifin J, Simpson EL. Climatic factors are associated with childhood eczema
9.
TE D
prevalence in the United States. The Journal of investigative dermatology 2013;133(7):1752-
Silverberg JI, Hanifin JM. Adult eczema prevalence and associations with asthma and other health
EP
and demographic factors: a US population-based study. The Journal of allergy and clinical immunology 2013;132(5):1132-8.
AC C
Silverberg JI, Simpson EL. Association between severe eczema in children and multiple comorbid conditions and increased healthcare utilization. Pediatric allergy and immunology : official publication of the European Society of Pediatric Allergy and Immunology 2013;24(5):476-86. Silverberg JI, Simpson EL. Associations of childhood eczema severity: a US population-based study. Dermatitis : contact, atopic, occupational, drug : official journal of the American Contact Dermatitis Society, North American Contact Dermatitis Group 2014;25(3):107-14.
ACCEPTED MANUSCRIPT Page 19 of 23 Su J, Prescott S, Sinn J, Tang M, Smith P, Heine RG, et al. Cost-effectiveness of partially-hydrolyzed formula for prevention of atopic dermatitis in Australia. J Med Econ 2012;15(6):1064-77. Tong SY, Steer AC, Jenney AW, Carapetis JR. Community-associated methicillin-resistant
RI PT
Staphylococcus aureus skin infections in the tropics. Dermatol Clin 2011;29(1):21-32.
Verboom P, Hakkaart-Van L, Sturkenboom M, De Zeeuw R, Menke H, Rutten F. The cost of atopic dermatitis in the Netherlands: an international comparison. The British journal of dermatology
SC
2002;147(4):716-24.
Vos T, Flaxman AD, Naghavi M, Lozano R, Michaud C, Ezzati M, et al. Years lived with disability
M AN U
(YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380(9859):2163-96. Williams JR, Burr ML, Williams HC. Factors influencing atopic dermatitis-a questionnaire survey of schoolchildren's perceptions. The British journal of dermatology 2004;150(6):1154-61.
TE D
Xu S, Thyssen J, Paller A, Silverberg J. Eczema, atopic dermatitis or atopic eczema? An analysis of global search engine trends. Dermatitis : contact, atopic, occupational, drug : official journal of the American Contact Dermatitis Society, North American Contact Dermatitis Group
EP
2017;In Press.
Yosipovitch G, Goon AT, Wee J, Chan YH, Zucker I, Goh CL. Itch characteristics in Chinese
AC C
patients with atopic dermatitis using a new questionnaire for the assessment of pruritus. International journal of dermatology 2002;41(4):212-6.
ACCEPTED MANUSCRIPT Page 20 of 23 Figure Legends. Figure 1. Prevalence of hospitalization for adults and children, length of stay of hospitalization for adults and children and cost of hospitalization for adults and children with atopic
RI PT
dermatitis or eczema. (A) Prevalence of hospitalization per million hospitalized adults for a primary diagnosis of atopic dermatitis (dotted), eczema (dashed), and atopic dermatitis and eczema combined (solid) during the years of 2002-2003, 2004-2005, 2006-2007, 2008-2009, 2010-2011, and 2012. (B)
SC
Prevalence of hospitalization per million hospitalized children for a primary diagnosis of atopic
dermatitis (dotted), eczema (dashed), and atopic dermatitis and eczema combined (solid) during the
M AN U
years of 2002-2003, 2004-2005, 2006-2007, 2008-2009, 2010-2011, and 2012.
Geometric mean LOS in (C) adults or (D) children with (circles) or without (squares) a primary diagnosis of atopic dermatitis (dashed line), eczema (dotted line), and atopic dermatitis and eczema combined (solid line) during the years of 2002-2003, 2004-2005, 2006-2007, 2008-2009, 2010-2011,
TE D
and 2012.
Geometric mean cost in (E) adults or (F) children with (circles) or without (squares) a primary diagnosis of atopic dermatitis (dashed line), eczema (dotted line), and atopic dermatitis and eczema
EP
combined (solid line) during the years of 2002-2003, 2004-2005, 2006-2007, 2008-2009, 2010-2011,
AC C
and 2012.
Figure 2. Weighted frequency of hospitalization for adults and children with atopic dermatitis or eczema.
ACCEPTED MANUSCRIPT Page 21 of 23
Summer Fall
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Seasonal and regional variation of hospitalization with a primary diagnosis of atopic dermatitis or eczema. Primary inpatient diagnosis of AD or eczema No Yes Variable Wtd % Prev Wtd % Prev Crude OR P Freq [95% CI] Freq [95% CI] [95% CI] Adults Season 25.5 23.2 0.88 Winter 70,338,854 4,400 .01 [25.5, 25.6] [21.9, 24.5] [0.80, 0.96] 25.0 25.9 4,908 1.00 Spring 68,902,498 [25.0, 25.0] [24.5, 27.3] 24.8 26.5 1.03 5,019 .6 Summer 68,336,209 [24.8, 24.8] [25.1, 27.9] [0.94, 1.13] 24.6 24.4 0.96 Fall 67,906,561 4,626 .5 [24.6, 24.7] [23.0, 25.8] [0.87, 1.05] Hospital region 20.3 25.1 1.47 Northeast 61,096,853 5,195 .0004 [19.1, 21.6] [22.7, 27.5] [1.26, 1.71] 23.5 25.2 1.28 Midwest 70,543,693 5,226 .01 [22.2, 24.7] [22.7, 27.7] [1.09, 1.50] 38.5 34.8 1.08 South 115,671,661 7,204 .5 [36.8, 40.1] [32.1, 37.5] [0.93, 1.24] 17.7 14.9 West 53,177,599 3,081 1.00 [16.6, 18.8] [12.9, 16.8] Children Season 28.9 23.8 0.71 Winter 60,621,151 2,2287 .0004 [28.7, 29.2] [22.0, 25.7] [0.63, 0.79] 27.6 23.6 Spring 4,943,979 2,644 [25.6, 29.5] 1.00 [23.5, 23.7] 4,707,635
5,234,233
22.5 [22.3, 22.6] 25.0 [24.9, 25.1]
2,449 2,216
25.5 [23.7, 27.3] 23.1 [21.1, 25.1]
0.97 [0.87, 1.09] 0.79 [0.69, 0.90]
19.2 [14.8, 23.5] 21.4 [14.1, 28.7] 42.1 [35.0, 49.2]
1.28 [0.95, 1.71] 1.26 [0.83, 1.90] 1.33 [0.99, 1.78]
.7 .002
Hospital region Northeast
4,221,843
Midwest
4,785,291
South
8,891,062
18.6 [15.9, 21.2] 21.0 [18.0, 24.1] 39.1 [35.4, 42.7]
2,048 2,285 4,495
.2 .4 .1
ACCEPTED MANUSCRIPT Page 22 of 23 21.3 [18.4, 24.3]
1,846
17.3 [12.3, 22.3]
1.00
-
EP
TE D
M AN U
SC
RI PT
4,852,928
AC C
West
ACCEPTED MANUSCRIPT Page 23 of 23
AC C
EP
TE D
M AN U
SC
RI PT
Table 2. Procedures, mortality risk, and loss of function risk for adults and children admitted with a primary diagnosis of atopic dermatitis or eczema. Adults Children Variable % Prev Wtd % Prev Wtd Freq [95% CI] Freq [95% CI] Ventilation Yes 29 0.1 [0.0, 0.2] 6 0.1 [0.0, 0.2] No 20,677 99.9 [99.8, 100.0] 10,669 99.9 [99.8, 100.0] Ventilation <96h Yes 19 0.1 [0.0, 0.2] 0 0 [0.0, 0.0] No 20,687 99.9 [99.8, 100.0] 10,675 100.0 [100.0, 100.0] Ventilation >96h Yes 10 0.05 [0.00, 0.1] 6 0.1 [0.0, 0.2] No 20,696 99.95 [99.9, 100.0] 10,669 99.9 [99.8, 100.0] Physical Therapy Yes 75 0.4 [0.2, 0.5] 91 0.9 [0.2, 1.5] No 20,630 99.6 [99.5, 99.8] 10,584 99.1 [98.5, 99.8] Mortality risk Minor 14,510 70.8 [69.3, 72.4] 10,336 97.5 [96.7, 98.2] Moderate 4,816 23.5 [22.1, 24.9] 230 2.2 [1.4, 2.9] Major 1,039 5.1 [4.4, 5.7] 23 0.2 [0.0, 0.4] Extreme 119 0.6 [0.4, 0.8] 6 0.1 [0.0, 0.2] Loss of function severity Minor 7,405 36.3 [34.5, 37.8] 6,341 59.8 [56.2, 63.4] Moderate 9,972 48.7 [47.1, 50.2] 3,584 33.8 [30.6, 37.0] Major 2,895 14.1 [13.0, 15.3] 640 6.0 [4.8, 7.3] Extreme 212 1.0 [0.7, 1.3] 29 0.3 [0.1, 0.5] Patient disposition at discharge Routine 16,488 79.7 [78.4, 81.0] 10,316 96.6 [95.7, 97.6] Transfer, short-term hospital 210 1.0 [0.7, 1.3] 97 0.9 [0.5, 1.3] Transfer, SNF, ICF or other facility 1,977 9.6 [8.6, 10.5] 28 0.3 [0.1, 0.5] Home health care 1,656 8.0 [7.2, 8.8] 216 2.0 [1.3, 2.8] Left against medical advice 317 1.5 [1.2, 1.9] 18 0.2 [0.0, 0.3] Died 42 0.2 [0.1, 0.3] 0 0.0 [0.0, 0.0]
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT