Mortality, length of stay and cost of hospitalization among patients with systemic sclerosis: results from the National Inpatient Sample

Mortality, length of stay and cost of hospitalization among patients with systemic sclerosis:... Abstract Objectives To evaluate the hospitalizations and define the factors associated with in-hospital mortality, longer length of stay (LOS) and higher hospital costs among SSc hospitalizations. Methods We used the National Inpatient Sample (2012–13) to identify adult hospitalizations with SSc, excluding patients with concomitant diagnosis of RA and systemic lupus. We calculated rates of hospitalization, in-hospital mortality, LOS and hospital costs. Factors associated with these outcomes were evaluated by univariate and backward stepwise multivariate logistic regression. Results There were 9731 hospitalizations in the sample representing an estimated 48 655 hospitalizations nationwide with SSc (0.09%), and the inpatient mortality rate was 5%. Patients were predominantly older (mean age 63.2 years), female (82.2%) and Caucasian (71.5%). Infections were the most common primary diagnoses among SSc hospitalizations (17.4%) and among those who died (32.7%). Acute renal failure [adjusted odds ratio (aOR) = 4.3, 95% CI: 3.3, 5.6] and aspiration (aOR= 3.5, 95% CI: 2.5, 4.9) were strongly associated with in-hospital mortality. The median (interquartile range) LOS was 4 days (−2, 7), and the median (interquartile range) cost was $8885 (−5169, 15921). While hospital from the West region, acute renal failure, acute bowel obstruction and aspiration (aOR > 2.0 with P < 0.0001 for all) seem to predict higher cost of hospitalization, pulmonary fibrosis, myositis and any type of infection in addition to the same factors, except the West region (aOR > 2.0 with P < 0.0001 for all), were associated with longer LOS. Conclusion Infections are currently the most common diagnoses among SSc hospitalizations and in-hospital deaths. This emphasizes the importance of being vigilant in prevention and early treatment of infections in SSc patients. systemic sclerosis, scleroderma, inpatient, in-hospital, mortality, hospitalization, national database, National Inpatient Sample, length of stay, cost Rheumatology key messages Infection is an important comorbid condition among hospitalized patients with SSc. Infection, acute renal failure and aspiration events are associated with significantly higher mortality in SSc. Focusing on infection prevention might make significant improvement in outcome among hospitalized SSc patients. Introduction SSc (scleroderma) is a chronic rheumatic CTD characterized by autoimmunity with predominantly vascular and fibrotic changes of the skin and visceral organs [1]. It is associated with significant morbidity and higher rates of mortality with standardized mortality rates ranging from 1.05 to 5.40 across various studies [2]. These rates vary by region as do clinical manifestations such as extent of skin and visceral organ involvement among geographically different populations. Although not much improvement has been made in discovering an effective disease-modifying agent for scleroderma, we have seen a decreasing trend in overall mortality and a change in the primary cause of SSc-related mortality from scleroderma renal crisis, in the pre-angiotensin-converting enzyme inhibitor era, to pulmonary-related complications currently [3–5]. Most of the studies looking at mortality or survival among SSc patients have done so among outpatient observational cohorts, which may not fully capture causes of mortality. Few have looked at the status and predictors of mortality among hospitalized patients [6–10]. The most recent comprehensive study of SSc hospitalizations by Chung et al. [7] a decade ago established CTD (SSc) as the principal cause of hospitalization and in-hospital mortality. The objective of our study was to identify current causes and factors associated with in-hospital mortality, length of stay (LOS) and hospitalization costs among patients with SSc by using a large database of hospitalized patients and comparing results to previous reports. Methods Database used We used the National Inpatient Sample (NIS), a part of the Healthcare Cost Utilization Project (HCUP), developed through a Federal State–Industry partnership sponsored by the Agency for Healthcare Research and Quality [11]. We utilized data for the years 2012 and 2013 to perform a cross-sectional analysis of discharge-level administrative data. NIS is the largest publicly available all-payer inpatient health care database in the USA. Each observation in this database represents a unique hospitalization. NIS provides weights for each observation in the sample that can be used to make national (weighted) estimates of hospital inpatient stays. It represents a 20% stratified sample of all discharges in the USA and contains data from >7 million unweighted and an estimated >35 million weighted hospital stays each year. Data for NIS 2012 and 2013 includes discharges from 44 participating states and represents >95% of the US population. NIS provides clinical, demographic and socio-economic information about each observation including up to 25 diagnosis codes (1 primary and 24 secondary), race, in-hospital death, LOS, hospital charges, hospital location, size, teaching status, primary insurance of the patient, transfer-in or -out status, etc. The NIS (HCUP) data are de-identified; thus, this study did not require institutional review board approval or patient consent. More details about this can be accessed from the HCUP website [11]. Selection of patients We identified adults (⩾18 years) with a discharge diagnosis of SSc (primary or secondary) based on the International Classification of Diseases version 9, Clinical Modification (ICD-9-CM) code 710.1 in any diagnosis position. An inpatient diagnosis code has been shown in previous work to have a positive predictive value of 76% for the diagnosis of SSc in administrative databases [12]. It identifies patients with limited SSc as well as dcSSc, but excludes morphea and other forms of localized scleroderma. To improve the specificity of the diagnosis code, we also excluded patients with a concomitant diagnosis code for RA (714.0, 714.1 and 714.2) or SLE (710.0). Hospitalization percentages and indications for hospitalization We used weighted estimates to calculate the proportion of all hospitalizations nationwide due to SSc during the years 2012–13. Indication for hospitalization was evaluated based on the primary diagnosis code and was examined among all patients with SSc and the subgroup of patients with SSc who died. The diagnoses were selected based on relevant ICD-9-CM codes and clinical classification software codes as supplied by HCUP (supplementary Table S1, available at Rheumatology online). Clinical classification software is a diagnosis and procedure categorization scheme where ICD-9-CM’s multitude of codes are collapsed into a smaller number of clinically meaningful categories that are more useful for presenting descriptive statistics than are individual ICD-9-CM codes [13]. While reporting the most common primary diagnoses associated with SSc hospitalization and deaths, we grouped the diagnoses into clinically relevant categories (beyond clinical classification software codes) based on pathologic as well as systemic involvement. Outcomes Among hospitalized patients with SSc, we evaluated the following outcomes: in-hospital mortality, LOS and cost. LOS and cost were treated as binomial variables and measurements greater than the 90th percentile for the overall SSc hospitalizations were considered as prolonged LOS or higher cost because of the highly skewed nature of these measures and clinical relevance of identifying the longest and most costly admissions [14]. Patients who were transferred out to a different acute care setting were excluded from these analyses as NIS does not enable identification of associated hospitalizations. Therefore, mortality, LOS and cost of the entire hospitalization episode cannot be determined by these high-risk patients. HCUP-NIS supplies hospital charges for each observation as well as the cost-to-charge ratio for each hospital in the database in a separate file. The product of the charge and corresponding cost-to-charge ratio gives the estimated cost (dollars spent) for each hospitalization. Factors associated with in-hospital mortality, longer LOS and higher cost of hospitalization Factors of interest included age (categorized as 18–44, 45–64, ≥65 years), sex, race (categorized as Caucasian, African American and Other), various relevant clinical factors, transfer status, primary payer, patient’s income quartile based on zip code, and hospital characteristic by region, locale, teaching status and size. When we identified infections as a leading cause of mortality in the population we were studying, we went back and also looked at the frequency of chemotherapy reception among these patients (considering ICD-9 codes for chemotherapy, immunotherapy infusion or central catheter placement as a surrogate of chemotherapy likely representing CYC or other infusions; supplementary Table S2, available at Rheumatology online). Comorbidities were defined based on Elixhauser Comorbidity Software as supplied by HCUP, which assigns variables that identify comorbidities in hospital discharge records using an algorithm based on ICD-9-CM codes [15]. These include 29 comorbidity measures encompassing all organ systems, and data on the presence of these conditions are supplied in a separate data file (severity file) by HCUP. Statistical reporting and modelling Continuous variables were expressed as mean (s.d.), or median with interquartile range for skewed data (LOS and cost). Categorical variables were expressed as percentages. Backward stepwise multivariate logistic regression excluding variables with P > 0.2 and forcing age and sex into the final model was used to identify independent factors associated with in-hospital mortality, prolonged LOS and high cost. Logistic regression estimates were reported as odds ratio in univariate analyses and as adjusted odds ratio (aOR) in multivariate analysis. Since in-hospital death necessarily impacts and could shorten LOS, a sensitivity analysis was done repeating multivariate regression after excluding patients who died in the hospital, although patients who died on average had longer LOS. All statistical analyses were conducted using statistical software STATA version 13.0 (College Station, TX, USA) and accounted for the stratified sampling design of NIS. A two-sided P < 0.05 was deemed to be statistically significant for all analyses. Results Overall hospitalizations with SSc We identified a total of 9731 hospitalizations with SSc diagnosis from the HCUP-NIS sample data for the years 2012–13, representing 48 655 hospitalizations (0.09%) nationwide. We estimated $719 million expenditure for those hospitalizations, which represented 0.12% of total cost for all hospitalizations within those years. SSc hospitalizations had a mean age of 63.3 years and almost 50% were ≥65 years of age. Female sex (82.2%) and Caucasian race (71.5%) were predominant. Most hospitalizations had Medicare (62.3%) as their primary insurance payer and the majority of the hospitalizations were emergent (83.3%). Distribution of other patient-, disease- and hospital-related characteristics are summarized in Table 1. Table 1 Baseline characteristics of hospitalized patients with diagnosis of SSc Characteristics Estimation or distribution P-valuea Overall SSc (sample n = 9731) Survived hospitalization (sample n = 9246) Died during hospitalization (sample n = 485) Age, mean( s.d.), years 63.3 (13.8) 63.1 (13.8) 66.0 (13.7) <0.001 Age category, %     18–44 years 9.4 9.5 7.8 0.19     45–64 years 41.4 41.7 35.1 0.003     ≥65 years 49.2 48.8 57.1 <0.001 Sex, male, % 17.8 17.7 19.8 0.24 Race, %     Caucasian/White 71.5 71.8 65.8 0.01     African-American 14.3 14.1 17.3 0.06     Othersb 14.2 14.1 16.9 0.13 Insurance type/status, %     Medicare 62.3 62.1 66.2 0.06     Medicaid 8.2 8.3 6.6 0.16     Private insurance 25.5 25.7 22.9 0.16     Self-pay 1.8 1.9 1.4 0.46     Uninsured 0.2 0.2 0.4 0.44     Otherc 2.0 2.0 2.5 0.49 Type of admission, %     Emergent admission 83.3 82.8 92.8 <0.001     Transfer statusd, %     Transferred in 8.7 8.2 17.4 <0.001 Hospital region, %     South 34.3 34.3 34.2 0.15     Northeast 22.1 22.0 24.7 0.001     Midwest 23.8 24.2 17.7 0.97     West 19.7 19.5 23.3 0.07 Hospital teaching status, %     Teaching 57.9 57.8 60.6 0.21 Hospital location, %     Urban 90.7 90.6 92.2 0.21     Small 12.5 12.5 12.4 0.93 Hospital bed size, %     Medium 24.6 24.5 26.2 0.44     Large 62.9 62.9 61.4 0.52 Household income quartile, %     Quartile 1( $1–38 999) 25.8 25.8 23.8 0.32     Quartile 2( $39 000–47 999) 24.5 24.7 19.8 0.009     Quartile 3( $48 000–62 999) 24.5 24.4 25.5 0.58     Quartile 4( $63 000+) 25.3 25.0 30.8 0.006 Number of comorbid conditions, median( IQR) 3 (2–4) 3 (2–4) 4 (2–5) Hypertension 55.4 55.7 Fluid and electrolytes disorders (55.3) – Fluid and electrolytes disorders 33.7 32.6 Hypertension (50.1) – Top 5 comorbid conditionse, % Deficiency anaemias 28.7 28.6 Pulmonary circulation disorders (36.9) – Hypothyroidism 24.1 24.3 Deficiency anemias (30.7) – Chronic pulmonary disease 23.3 23.4 Congestive heart failure (29.5) – Number of diagnoses codes mentioned, median (IQR) 14 (10–18) 14 (9–18) 18 (15–24) <0.001 Number of procedure codes mentioned, median (IQR) 1 (0–3) 1 (0–3) 3 (1–6) <0.001 Relevant clinical conditions, %     Oesophageal dysfunction 47.7 48.0 41.7 0.005     Hypertension 57.8 58.2 51.1 0.002     Overall infections 32.7 31.0 63.9 <0.001         Opportunistic infections 0.9 0.9 0.6 0.48         Other infections 31.8 30.2 63.3 <0.001     Pulmonary arterial hypertension 26.8 25.9 43.3 <0.001     Renal failure (acute and unspecified) 26.5 25.1 53.2 <0.001     Digital ulcer/gangrene/Raynaud’s 25.6 25.7 23.5 0.29     Congestive heart failure 23.7 23 38.4 <0.001     Coronary artery disease 22.2 21.9 27.8 0.004     Chronic kidney disease 18.3 17.9 26.8 <0.001     Diabetes mellitus 17.9 18.1 13.8 0.008     Respiratory failure 17.3 17.3 72.4 <0.001     Pulmonary fibrosis/interstitial lung disease 15.9 15.3 27.4 <0.001     Cachexia/weight loss/FTT 8.7 8.1 20 <0.001     PEMC 8.4 8 16.1 <0.001     Arrhythmia 4.7 4.5 8.7 0.001     Acute bowel obstruction 4.2 4.2 4.5 0.71     Aspiration 4.2 3.6 15.1 <0.001     Stroke/transient ischaemic attack 4 3.9 6 0.08     Liver disease 2.7 2.7 4.1 0.12     Myositis 0.8 0.8 0.4 0.21 Characteristics Estimation or distribution P-valuea Overall SSc (sample n = 9731) Survived hospitalization (sample n = 9246) Died during hospitalization (sample n = 485) Age, mean( s.d.), years 63.3 (13.8) 63.1 (13.8) 66.0 (13.7) <0.001 Age category, %     18–44 years 9.4 9.5 7.8 0.19     45–64 years 41.4 41.7 35.1 0.003     ≥65 years 49.2 48.8 57.1 <0.001 Sex, male, % 17.8 17.7 19.8 0.24 Race, %     Caucasian/White 71.5 71.8 65.8 0.01     African-American 14.3 14.1 17.3 0.06     Othersb 14.2 14.1 16.9 0.13 Insurance type/status, %     Medicare 62.3 62.1 66.2 0.06     Medicaid 8.2 8.3 6.6 0.16     Private insurance 25.5 25.7 22.9 0.16     Self-pay 1.8 1.9 1.4 0.46     Uninsured 0.2 0.2 0.4 0.44     Otherc 2.0 2.0 2.5 0.49 Type of admission, %     Emergent admission 83.3 82.8 92.8 <0.001     Transfer statusd, %     Transferred in 8.7 8.2 17.4 <0.001 Hospital region, %     South 34.3 34.3 34.2 0.15     Northeast 22.1 22.0 24.7 0.001     Midwest 23.8 24.2 17.7 0.97     West 19.7 19.5 23.3 0.07 Hospital teaching status, %     Teaching 57.9 57.8 60.6 0.21 Hospital location, %     Urban 90.7 90.6 92.2 0.21     Small 12.5 12.5 12.4 0.93 Hospital bed size, %     Medium 24.6 24.5 26.2 0.44     Large 62.9 62.9 61.4 0.52 Household income quartile, %     Quartile 1( $1–38 999) 25.8 25.8 23.8 0.32     Quartile 2( $39 000–47 999) 24.5 24.7 19.8 0.009     Quartile 3( $48 000–62 999) 24.5 24.4 25.5 0.58     Quartile 4( $63 000+) 25.3 25.0 30.8 0.006 Number of comorbid conditions, median( IQR) 3 (2–4) 3 (2–4) 4 (2–5) Hypertension 55.4 55.7 Fluid and electrolytes disorders (55.3) – Fluid and electrolytes disorders 33.7 32.6 Hypertension (50.1) – Top 5 comorbid conditionse, % Deficiency anaemias 28.7 28.6 Pulmonary circulation disorders (36.9) – Hypothyroidism 24.1 24.3 Deficiency anemias (30.7) – Chronic pulmonary disease 23.3 23.4 Congestive heart failure (29.5) – Number of diagnoses codes mentioned, median (IQR) 14 (10–18) 14 (9–18) 18 (15–24) <0.001 Number of procedure codes mentioned, median (IQR) 1 (0–3) 1 (0–3) 3 (1–6) <0.001 Relevant clinical conditions, %     Oesophageal dysfunction 47.7 48.0 41.7 0.005     Hypertension 57.8 58.2 51.1 0.002     Overall infections 32.7 31.0 63.9 <0.001         Opportunistic infections 0.9 0.9 0.6 0.48         Other infections 31.8 30.2 63.3 <0.001     Pulmonary arterial hypertension 26.8 25.9 43.3 <0.001     Renal failure (acute and unspecified) 26.5 25.1 53.2 <0.001     Digital ulcer/gangrene/Raynaud’s 25.6 25.7 23.5 0.29     Congestive heart failure 23.7 23 38.4 <0.001     Coronary artery disease 22.2 21.9 27.8 0.004     Chronic kidney disease 18.3 17.9 26.8 <0.001     Diabetes mellitus 17.9 18.1 13.8 0.008     Respiratory failure 17.3 17.3 72.4 <0.001     Pulmonary fibrosis/interstitial lung disease 15.9 15.3 27.4 <0.001     Cachexia/weight loss/FTT 8.7 8.1 20 <0.001     PEMC 8.4 8 16.1 <0.001     Arrhythmia 4.7 4.5 8.7 0.001     Acute bowel obstruction 4.2 4.2 4.5 0.71     Aspiration 4.2 3.6 15.1 <0.001     Stroke/transient ischaemic attack 4 3.9 6 0.08     Liver disease 2.7 2.7 4.1 0.12     Myositis 0.8 0.8 0.4 0.21 a P-value represents the difference between those who survived and died; statistically significant values are in bold. b Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. c Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programs. d Transfer from other emergency, hospital, office or nursing home. e Comorbid conditions are counted as defined by Elixhauser Comorbidity Software algorithm provided by Healthcare Cost and Utilization Project. IQR: interquartile range; FTT: failure to thrive; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy. Table 1 Baseline characteristics of hospitalized patients with diagnosis of SSc Characteristics Estimation or distribution P-valuea Overall SSc (sample n = 9731) Survived hospitalization (sample n = 9246) Died during hospitalization (sample n = 485) Age, mean( s.d.), years 63.3 (13.8) 63.1 (13.8) 66.0 (13.7) <0.001 Age category, %     18–44 years 9.4 9.5 7.8 0.19     45–64 years 41.4 41.7 35.1 0.003     ≥65 years 49.2 48.8 57.1 <0.001 Sex, male, % 17.8 17.7 19.8 0.24 Race, %     Caucasian/White 71.5 71.8 65.8 0.01     African-American 14.3 14.1 17.3 0.06     Othersb 14.2 14.1 16.9 0.13 Insurance type/status, %     Medicare 62.3 62.1 66.2 0.06     Medicaid 8.2 8.3 6.6 0.16     Private insurance 25.5 25.7 22.9 0.16     Self-pay 1.8 1.9 1.4 0.46     Uninsured 0.2 0.2 0.4 0.44     Otherc 2.0 2.0 2.5 0.49 Type of admission, %     Emergent admission 83.3 82.8 92.8 <0.001     Transfer statusd, %     Transferred in 8.7 8.2 17.4 <0.001 Hospital region, %     South 34.3 34.3 34.2 0.15     Northeast 22.1 22.0 24.7 0.001     Midwest 23.8 24.2 17.7 0.97     West 19.7 19.5 23.3 0.07 Hospital teaching status, %     Teaching 57.9 57.8 60.6 0.21 Hospital location, %     Urban 90.7 90.6 92.2 0.21     Small 12.5 12.5 12.4 0.93 Hospital bed size, %     Medium 24.6 24.5 26.2 0.44     Large 62.9 62.9 61.4 0.52 Household income quartile, %     Quartile 1( $1–38 999) 25.8 25.8 23.8 0.32     Quartile 2( $39 000–47 999) 24.5 24.7 19.8 0.009     Quartile 3( $48 000–62 999) 24.5 24.4 25.5 0.58     Quartile 4( $63 000+) 25.3 25.0 30.8 0.006 Number of comorbid conditions, median( IQR) 3 (2–4) 3 (2–4) 4 (2–5) Hypertension 55.4 55.7 Fluid and electrolytes disorders (55.3) – Fluid and electrolytes disorders 33.7 32.6 Hypertension (50.1) – Top 5 comorbid conditionse, % Deficiency anaemias 28.7 28.6 Pulmonary circulation disorders (36.9) – Hypothyroidism 24.1 24.3 Deficiency anemias (30.7) – Chronic pulmonary disease 23.3 23.4 Congestive heart failure (29.5) – Number of diagnoses codes mentioned, median (IQR) 14 (10–18) 14 (9–18) 18 (15–24) <0.001 Number of procedure codes mentioned, median (IQR) 1 (0–3) 1 (0–3) 3 (1–6) <0.001 Relevant clinical conditions, %     Oesophageal dysfunction 47.7 48.0 41.7 0.005     Hypertension 57.8 58.2 51.1 0.002     Overall infections 32.7 31.0 63.9 <0.001         Opportunistic infections 0.9 0.9 0.6 0.48         Other infections 31.8 30.2 63.3 <0.001     Pulmonary arterial hypertension 26.8 25.9 43.3 <0.001     Renal failure (acute and unspecified) 26.5 25.1 53.2 <0.001     Digital ulcer/gangrene/Raynaud’s 25.6 25.7 23.5 0.29     Congestive heart failure 23.7 23 38.4 <0.001     Coronary artery disease 22.2 21.9 27.8 0.004     Chronic kidney disease 18.3 17.9 26.8 <0.001     Diabetes mellitus 17.9 18.1 13.8 0.008     Respiratory failure 17.3 17.3 72.4 <0.001     Pulmonary fibrosis/interstitial lung disease 15.9 15.3 27.4 <0.001     Cachexia/weight loss/FTT 8.7 8.1 20 <0.001     PEMC 8.4 8 16.1 <0.001     Arrhythmia 4.7 4.5 8.7 0.001     Acute bowel obstruction 4.2 4.2 4.5 0.71     Aspiration 4.2 3.6 15.1 <0.001     Stroke/transient ischaemic attack 4 3.9 6 0.08     Liver disease 2.7 2.7 4.1 0.12     Myositis 0.8 0.8 0.4 0.21 Characteristics Estimation or distribution P-valuea Overall SSc (sample n = 9731) Survived hospitalization (sample n = 9246) Died during hospitalization (sample n = 485) Age, mean( s.d.), years 63.3 (13.8) 63.1 (13.8) 66.0 (13.7) <0.001 Age category, %     18–44 years 9.4 9.5 7.8 0.19     45–64 years 41.4 41.7 35.1 0.003     ≥65 years 49.2 48.8 57.1 <0.001 Sex, male, % 17.8 17.7 19.8 0.24 Race, %     Caucasian/White 71.5 71.8 65.8 0.01     African-American 14.3 14.1 17.3 0.06     Othersb 14.2 14.1 16.9 0.13 Insurance type/status, %     Medicare 62.3 62.1 66.2 0.06     Medicaid 8.2 8.3 6.6 0.16     Private insurance 25.5 25.7 22.9 0.16     Self-pay 1.8 1.9 1.4 0.46     Uninsured 0.2 0.2 0.4 0.44     Otherc 2.0 2.0 2.5 0.49 Type of admission, %     Emergent admission 83.3 82.8 92.8 <0.001     Transfer statusd, %     Transferred in 8.7 8.2 17.4 <0.001 Hospital region, %     South 34.3 34.3 34.2 0.15     Northeast 22.1 22.0 24.7 0.001     Midwest 23.8 24.2 17.7 0.97     West 19.7 19.5 23.3 0.07 Hospital teaching status, %     Teaching 57.9 57.8 60.6 0.21 Hospital location, %     Urban 90.7 90.6 92.2 0.21     Small 12.5 12.5 12.4 0.93 Hospital bed size, %     Medium 24.6 24.5 26.2 0.44     Large 62.9 62.9 61.4 0.52 Household income quartile, %     Quartile 1( $1–38 999) 25.8 25.8 23.8 0.32     Quartile 2( $39 000–47 999) 24.5 24.7 19.8 0.009     Quartile 3( $48 000–62 999) 24.5 24.4 25.5 0.58     Quartile 4( $63 000+) 25.3 25.0 30.8 0.006 Number of comorbid conditions, median( IQR) 3 (2–4) 3 (2–4) 4 (2–5) Hypertension 55.4 55.7 Fluid and electrolytes disorders (55.3) – Fluid and electrolytes disorders 33.7 32.6 Hypertension (50.1) – Top 5 comorbid conditionse, % Deficiency anaemias 28.7 28.6 Pulmonary circulation disorders (36.9) – Hypothyroidism 24.1 24.3 Deficiency anemias (30.7) – Chronic pulmonary disease 23.3 23.4 Congestive heart failure (29.5) – Number of diagnoses codes mentioned, median (IQR) 14 (10–18) 14 (9–18) 18 (15–24) <0.001 Number of procedure codes mentioned, median (IQR) 1 (0–3) 1 (0–3) 3 (1–6) <0.001 Relevant clinical conditions, %     Oesophageal dysfunction 47.7 48.0 41.7 0.005     Hypertension 57.8 58.2 51.1 0.002     Overall infections 32.7 31.0 63.9 <0.001         Opportunistic infections 0.9 0.9 0.6 0.48         Other infections 31.8 30.2 63.3 <0.001     Pulmonary arterial hypertension 26.8 25.9 43.3 <0.001     Renal failure (acute and unspecified) 26.5 25.1 53.2 <0.001     Digital ulcer/gangrene/Raynaud’s 25.6 25.7 23.5 0.29     Congestive heart failure 23.7 23 38.4 <0.001     Coronary artery disease 22.2 21.9 27.8 0.004     Chronic kidney disease 18.3 17.9 26.8 <0.001     Diabetes mellitus 17.9 18.1 13.8 0.008     Respiratory failure 17.3 17.3 72.4 <0.001     Pulmonary fibrosis/interstitial lung disease 15.9 15.3 27.4 <0.001     Cachexia/weight loss/FTT 8.7 8.1 20 <0.001     PEMC 8.4 8 16.1 <0.001     Arrhythmia 4.7 4.5 8.7 0.001     Acute bowel obstruction 4.2 4.2 4.5 0.71     Aspiration 4.2 3.6 15.1 <0.001     Stroke/transient ischaemic attack 4 3.9 6 0.08     Liver disease 2.7 2.7 4.1 0.12     Myositis 0.8 0.8 0.4 0.21 a P-value represents the difference between those who survived and died; statistically significant values are in bold. b Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. c Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programs. d Transfer from other emergency, hospital, office or nursing home. e Comorbid conditions are counted as defined by Elixhauser Comorbidity Software algorithm provided by Healthcare Cost and Utilization Project. IQR: interquartile range; FTT: failure to thrive; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy. Principal diagnoses associated with SSc hospitalizations in patients with and without inpatient mortality The most common reasons for hospitalization based on primary diagnosis codes (collapsed into categories) were infection/septicaemia (17.4%) followed by involvement of cardiovascular (15.9%), gastrointestinal (13.3%), musculoskeletal (includes CTDs) (12.2%), pulmonary (10.9%) and nervous (3.8%) systems among the SSc patients. Among SSc patients who died in the hospital, infection/septicaemia remained the most common primary diagnosis (32.7%) followed by pulmonary (20.0%), cardiovascular (15.7%), musculoskeletal (12.0%) and gastrointestinal (4.8%) system involvement (Fig. 1). When the primary diagnoses are not collapsed into categories, the most common reasons for hospitalization were CTD (includes CTDs other than RA and SLE) (6.35%) followed by respiratory infections (5.96%) and septicaemia (5.44%), while the most common primary diagnoses among SSc patients who died in hospital were septicaemia (23.35%) followed by respiratory failure/insufficiency/arrest (11.98%) followed by CTDs (10.33%) (supplementary Tables S3 and S4, available at Rheumatology online). Fig. 1 View largeDownload slide Primary diagnosis category of SSc patients: overall and among those who died Fig. 1 View largeDownload slide Primary diagnosis category of SSc patients: overall and among those who died Outcomes: in-hospital mortality, hospital LOS and cost of hospitalization Table 2 summarizes the above outcomes among overall hospitalizations by age, race and sex subgroups. Overall mortality was 5%. Mortality was 5.8% among older patients ≥65 years, 5.6% among males and 6.1% among African-Americans. A lower mortality was found among Caucasians (4.6%, P < 0.05). Median LOS was similar across all age, race and sex categories [4 days, interquartile range (2–7)] other than slightly longer LOS among African-Americans [5 days (3–8)]. Cost of hospitalization was also similar in the different subgroups [overall median (interquartile range) cost $8993, (5259–16 420)]. Overall cost of SSc hospitalizations for the years 2012 and 2013 were $357 million and $361 million, respectively. Table 2 Hospitalization proportion, mortality rate, length of stay and hospitalization cost by age category, race and sex Hospitalization composition Total SSc hospitalizations Age group (years) Sexa Race 18–44 45–64 ≥65 Male Female Caucasians AA Othersb Hospitalization counts, n (N = weighted counts) 9731 (48 655) 916 (4580) 4030 (20 150) 4785 (23 925) 1729 (8645) 8001 (40 004) 6586 (32 930) 1313 (65 650) 1312 (6560) Hospitalization percentage 0.093 0.048 0.11 0.095 0.035 0.14 0.095 0.093 0.094 Race, % of hospitalization (% race)     Caucasian/White 0.095 (71.5) 0.042 (50.1) 0.113 (65.1) 0.1 (81.1) 0.035 (69.2) 0.151 (72.0) – – –     AA 0.093 (14.3) 0.068 (28.6) 0.126 (19.3) 0.07 (7.1) 0.04 (15.8) 0.14 (14.0) – – –     Others 0.094 (14.2) 0.051 (21.3) 0.123 (15.5) 0.10 (11.8) 0.04 (14.9) 0.151 (14.0) – – –     Mortality rate, % 5.0 4.1 4.2 5.8 5.6 4.9 4.62 6.10 5.95     LOS, median (IQR), days 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 5 (3–8) 4 (2–7)     Cost per hospitalization, median (IQR), $ 8885 (5169– 15 921) 8567 (4883– 15 629) 9341 (5363– 17 097) 8791 (5236– 15 969) 9435 (5552– 17 619) 8916 (5201– 16 501) 8842 (5194– 5933) 8932 (5165– 16 514) 9839 (5634– 18 281) Hospitalization composition Total SSc hospitalizations Age group (years) Sexa Race 18–44 45–64 ≥65 Male Female Caucasians AA Othersb Hospitalization counts, n (N = weighted counts) 9731 (48 655) 916 (4580) 4030 (20 150) 4785 (23 925) 1729 (8645) 8001 (40 004) 6586 (32 930) 1313 (65 650) 1312 (6560) Hospitalization percentage 0.093 0.048 0.11 0.095 0.035 0.14 0.095 0.093 0.094 Race, % of hospitalization (% race)     Caucasian/White 0.095 (71.5) 0.042 (50.1) 0.113 (65.1) 0.1 (81.1) 0.035 (69.2) 0.151 (72.0) – – –     AA 0.093 (14.3) 0.068 (28.6) 0.126 (19.3) 0.07 (7.1) 0.04 (15.8) 0.14 (14.0) – – –     Others 0.094 (14.2) 0.051 (21.3) 0.123 (15.5) 0.10 (11.8) 0.04 (14.9) 0.151 (14.0) – – –     Mortality rate, % 5.0 4.1 4.2 5.8 5.6 4.9 4.62 6.10 5.95     LOS, median (IQR), days 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 5 (3–8) 4 (2–7)     Cost per hospitalization, median (IQR), $ 8885 (5169– 15 921) 8567 (4883– 15 629) 9341 (5363– 17 097) 8791 (5236– 15 969) 9435 (5552– 17 619) 8916 (5201– 16 501) 8842 (5194– 5933) 8932 (5165– 16 514) 9839 (5634– 18 281) a One observation in the sample had missing data for sex. b Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. AA: African American; LOS: length of stay; IQR: interquartile range. Table 2 Hospitalization proportion, mortality rate, length of stay and hospitalization cost by age category, race and sex Hospitalization composition Total SSc hospitalizations Age group (years) Sexa Race 18–44 45–64 ≥65 Male Female Caucasians AA Othersb Hospitalization counts, n (N = weighted counts) 9731 (48 655) 916 (4580) 4030 (20 150) 4785 (23 925) 1729 (8645) 8001 (40 004) 6586 (32 930) 1313 (65 650) 1312 (6560) Hospitalization percentage 0.093 0.048 0.11 0.095 0.035 0.14 0.095 0.093 0.094 Race, % of hospitalization (% race)     Caucasian/White 0.095 (71.5) 0.042 (50.1) 0.113 (65.1) 0.1 (81.1) 0.035 (69.2) 0.151 (72.0) – – –     AA 0.093 (14.3) 0.068 (28.6) 0.126 (19.3) 0.07 (7.1) 0.04 (15.8) 0.14 (14.0) – – –     Others 0.094 (14.2) 0.051 (21.3) 0.123 (15.5) 0.10 (11.8) 0.04 (14.9) 0.151 (14.0) – – –     Mortality rate, % 5.0 4.1 4.2 5.8 5.6 4.9 4.62 6.10 5.95     LOS, median (IQR), days 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 5 (3–8) 4 (2–7)     Cost per hospitalization, median (IQR), $ 8885 (5169– 15 921) 8567 (4883– 15 629) 9341 (5363– 17 097) 8791 (5236– 15 969) 9435 (5552– 17 619) 8916 (5201– 16 501) 8842 (5194– 5933) 8932 (5165– 16 514) 9839 (5634– 18 281) Hospitalization composition Total SSc hospitalizations Age group (years) Sexa Race 18–44 45–64 ≥65 Male Female Caucasians AA Othersb Hospitalization counts, n (N = weighted counts) 9731 (48 655) 916 (4580) 4030 (20 150) 4785 (23 925) 1729 (8645) 8001 (40 004) 6586 (32 930) 1313 (65 650) 1312 (6560) Hospitalization percentage 0.093 0.048 0.11 0.095 0.035 0.14 0.095 0.093 0.094 Race, % of hospitalization (% race)     Caucasian/White 0.095 (71.5) 0.042 (50.1) 0.113 (65.1) 0.1 (81.1) 0.035 (69.2) 0.151 (72.0) – – –     AA 0.093 (14.3) 0.068 (28.6) 0.126 (19.3) 0.07 (7.1) 0.04 (15.8) 0.14 (14.0) – – –     Others 0.094 (14.2) 0.051 (21.3) 0.123 (15.5) 0.10 (11.8) 0.04 (14.9) 0.151 (14.0) – – –     Mortality rate, % 5.0 4.1 4.2 5.8 5.6 4.9 4.62 6.10 5.95     LOS, median (IQR), days 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 5 (3–8) 4 (2–7)     Cost per hospitalization, median (IQR), $ 8885 (5169– 15 921) 8567 (4883– 15 629) 9341 (5363– 17 097) 8791 (5236– 15 969) 9435 (5552– 17 619) 8916 (5201– 16 501) 8842 (5194– 5933) 8932 (5165– 16 514) 9839 (5634– 18 281) a One observation in the sample had missing data for sex. b Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. AA: African American; LOS: length of stay; IQR: interquartile range. Factors associated with in-hospital mortality, longer hospital LOS and higher hospitalization cost Tables 3–5 summarize the findings of univariate and backward stepwise multivariate logistic regressions to identify factors associated with higher likelihood of inpatient mortality, longer hospital LOS and higher hospitalization cost, respectively. Age and sex categories were not statistically associated with any of the outcomes studies. While African American race was associated with higher odds of mortality (aOR = 1.46, 95% CI: 1.08, 1.98, P = 0.01), races other than Caucasian and African American had higher cost of hospitalization (aOR = 1.53, 95% CI: 1.24, 1.89, P < 0.0001). Many of the relevant clinical factors such as pulmonary fibrosis, pulmonary arterial hypertension (PAH), acute renal failure, pericarditis/endocarditis/myocarditis/cardiomyopathy and aspiration were positively associated with all of the outcomes. Of these, acute renal failure (although it might not always be a result of SSc process) had the strongest association with inpatient mortality (aOR = 4.31, 95% CI: 3.32, 5.60) and aspiration had strongest association with both longer LOS (aOR = 2.88, 95% CI: 2.20, 3.77, P < 0.05) and higher cost (aOR = 2.29, 95% CI: 1.72, 3.04, P < 0.0001). Acute bowel obstruction, on the other hand, was associated with longer LOS only (aOR = 2.66, 95% CI: 1.98, 3.57). Interestingly, diabetes mellitus (aOR = 0.62, 95% CI: 0.46, 0.84), hypertension (aOR = 0.71, 95% CI: 0.57, 0.89) and oesophageal dysfunction (aOR = 0.72, 95% CI: 0.59, 0.89) were significantly (P < 0.05) associated with lower odds of inpatient mortality in both univariate as well as multivariate analyses. Diagnosis of chronic kidney disease, however, had higher odds of mortality during univariate analysis (aOR = 1.7, 95% CI: 1.39, 2.08, P < 0.001) but lower in multivariate analysis (aOR = 0.52, 95% CI: 0.39, 0.69, P < 0.0001). Patients who died had significantly higher proportions of PAH (43.3%), respiratory failure (72.4%) and congestive heart failure (CHF) (38.4%) (Table 1). Table 3 Factors associated with in-hospital mortality among patients with systemic sclerosis: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.01 (0.7, 1.46) 0.95 1.10 (0.73, 1.66) 0.64     ≥65 years 1.42 (0.99, 2.02) 0.054 1.35 (0.89, 2.06) 0.16 Male sex 1.16 (0.92, 1.45) 0.21 1.07 (0.81, 1.4) 0.64 Race (ref. Caucasian)     African-American 1.33 (1.04, 1.7) 0.02 1.46 (1.08, 1.98) 0.01     Othersa 1.30 (0.99, 1.69) 0.06 1.20 (0.89, 1.61) 0.24     Transfer status (ref. not transferred in) 2.41 <0.0001 2.03 (1.52, 2.71) <0.0001 Primary payer status (ref. Medicare)     Medicaid 0.75 (0.52, 1.1) 0.14 – –     Private insurance 0.84 (0.67, 1.05) 0.12 – –     Self-pay 0.73 (0.34, 1.58) 0.43 – –     No charge 2.04 (0.46, 9.0) 0.34 – –     Otherb 1.19 (0.66, 2.17) 0.57 – – Income quartile (ref. quartile 1)     Quartile 2 0.87 (0.66, 1.15) 0.33 0.86 (0.63, 1.18) 0.36     Quartile 3 1.14 (0.87, 1.48) 0.34 1.10 (0.82, 1.49) 0.52     Quartile 4 1.35 (1.05, 1.73) 0.02 1.41 (1.07, 1.87) 0.02 Hospital region (ref. south)     Northeast 1.14 (0.9, 1.44) 0.29 – –     Midwest 0.74 (0.56, 0.98) 0.03 – –     West 1.20 (0.92, 1.56) 0.17 – – Urban hospital (ref. rural) 1.18 (0.84, 1.66) 0.33 – – Teaching hospital (ref. non-teaching) 1.10 (0.91, 1.33) 0.31 – – Hospital size (ref. small)     Medium 1.06 (0.77, 1.47) 0.71 – –     Large 0.96 (0.72, 1.28) 0.79 – – Relevant clinical conditions     Aspiration 4.92 (3.73, 6.51) <0.0001 3.52 (2.51, 4.94) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.33 (0.42, 4.21) 0.62 0.91 (0.22, 3.76) 0.90         Other infections 4.07 (3.35, 4.94) <0.0001 3.36 (2.73, 4.14) <0.0001     Acute renal failure 3.47 (2.88, 4.17) <0.0001 4.31 (3.32, 5.6) <0.0001     Pulmonary fibrosis 2.95 (1.61, 5.42) <0.0001 2.23 (1.15, 4.3) 0.02     Weight loss/cachexia 2.88 (2.27, 3.66) <0.0001 2.31 (1.74, 3.05) <0.0001     PEMC 2.26 (1.75, 2.91) <0.0001 1.6 (1.19, 2.16) <0.0001     Pulmonary arterial hypertension 2.22 (1.85, 2.66) <0.0001 1.82 (1.47, 2.25) <0.0001     Congestive heart failure 2.12 (1.76, 2.55) <0.0001 – –     Arrhythmia 2.05 (1.48, 2.84) <0.0001 1.53 (1.04, 2.25) 0.03     Chronic kidney disease 1.70 (1.39, 2.08) <0.0001 0.52 (0.39, 0.69) <0.0001     Chronic liver disease 1.60 (1, 2.58) 0.05 – –     Stroke/TIA 1.56 (1.04, 2.35) 0.03 1.86 (1.17, 2.95) 0.01     Coronary artery disease 1.38 (1.12, 1.7) <0.0001 1.3 (1.02, 1.66) 0.03     Acute bowel obstruction 1.12 (0.72, 1.73) 0.62 – –     Raynaud’s/ulcer/gangrene 0.89 (0.71, 1.11) 0.29 – –     Oesophageal dysfunction 0.77 (0.64, 0.92) <0.0001 0.72 (0.59, 0.89) <0.0001     Hypertension 0.75 (0.63, 0.9) <0.0001 0.71 (0.57, 0.89) <0.0001     Diabetes mellitus 0.72 (0.56, 0.94) 0.02 0.62 (0.46, 0.84) <0.0001     Myositis 0.53 (0.13, 2.2) 0.39 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.01 (0.7, 1.46) 0.95 1.10 (0.73, 1.66) 0.64     ≥65 years 1.42 (0.99, 2.02) 0.054 1.35 (0.89, 2.06) 0.16 Male sex 1.16 (0.92, 1.45) 0.21 1.07 (0.81, 1.4) 0.64 Race (ref. Caucasian)     African-American 1.33 (1.04, 1.7) 0.02 1.46 (1.08, 1.98) 0.01     Othersa 1.30 (0.99, 1.69) 0.06 1.20 (0.89, 1.61) 0.24     Transfer status (ref. not transferred in) 2.41 <0.0001 2.03 (1.52, 2.71) <0.0001 Primary payer status (ref. Medicare)     Medicaid 0.75 (0.52, 1.1) 0.14 – –     Private insurance 0.84 (0.67, 1.05) 0.12 – –     Self-pay 0.73 (0.34, 1.58) 0.43 – –     No charge 2.04 (0.46, 9.0) 0.34 – –     Otherb 1.19 (0.66, 2.17) 0.57 – – Income quartile (ref. quartile 1)     Quartile 2 0.87 (0.66, 1.15) 0.33 0.86 (0.63, 1.18) 0.36     Quartile 3 1.14 (0.87, 1.48) 0.34 1.10 (0.82, 1.49) 0.52     Quartile 4 1.35 (1.05, 1.73) 0.02 1.41 (1.07, 1.87) 0.02 Hospital region (ref. south)     Northeast 1.14 (0.9, 1.44) 0.29 – –     Midwest 0.74 (0.56, 0.98) 0.03 – –     West 1.20 (0.92, 1.56) 0.17 – – Urban hospital (ref. rural) 1.18 (0.84, 1.66) 0.33 – – Teaching hospital (ref. non-teaching) 1.10 (0.91, 1.33) 0.31 – – Hospital size (ref. small)     Medium 1.06 (0.77, 1.47) 0.71 – –     Large 0.96 (0.72, 1.28) 0.79 – – Relevant clinical conditions     Aspiration 4.92 (3.73, 6.51) <0.0001 3.52 (2.51, 4.94) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.33 (0.42, 4.21) 0.62 0.91 (0.22, 3.76) 0.90         Other infections 4.07 (3.35, 4.94) <0.0001 3.36 (2.73, 4.14) <0.0001     Acute renal failure 3.47 (2.88, 4.17) <0.0001 4.31 (3.32, 5.6) <0.0001     Pulmonary fibrosis 2.95 (1.61, 5.42) <0.0001 2.23 (1.15, 4.3) 0.02     Weight loss/cachexia 2.88 (2.27, 3.66) <0.0001 2.31 (1.74, 3.05) <0.0001     PEMC 2.26 (1.75, 2.91) <0.0001 1.6 (1.19, 2.16) <0.0001     Pulmonary arterial hypertension 2.22 (1.85, 2.66) <0.0001 1.82 (1.47, 2.25) <0.0001     Congestive heart failure 2.12 (1.76, 2.55) <0.0001 – –     Arrhythmia 2.05 (1.48, 2.84) <0.0001 1.53 (1.04, 2.25) 0.03     Chronic kidney disease 1.70 (1.39, 2.08) <0.0001 0.52 (0.39, 0.69) <0.0001     Chronic liver disease 1.60 (1, 2.58) 0.05 – –     Stroke/TIA 1.56 (1.04, 2.35) 0.03 1.86 (1.17, 2.95) 0.01     Coronary artery disease 1.38 (1.12, 1.7) <0.0001 1.3 (1.02, 1.66) 0.03     Acute bowel obstruction 1.12 (0.72, 1.73) 0.62 – –     Raynaud’s/ulcer/gangrene 0.89 (0.71, 1.11) 0.29 – –     Oesophageal dysfunction 0.77 (0.64, 0.92) <0.0001 0.72 (0.59, 0.89) <0.0001     Hypertension 0.75 (0.63, 0.9) <0.0001 0.71 (0.57, 0.89) <0.0001     Diabetes mellitus 0.72 (0.56, 0.94) 0.02 0.62 (0.46, 0.84) <0.0001     Myositis 0.53 (0.13, 2.2) 0.39 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker's Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programmes. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; TIA: transient ischaemic attack. Table 3 Factors associated with in-hospital mortality among patients with systemic sclerosis: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.01 (0.7, 1.46) 0.95 1.10 (0.73, 1.66) 0.64     ≥65 years 1.42 (0.99, 2.02) 0.054 1.35 (0.89, 2.06) 0.16 Male sex 1.16 (0.92, 1.45) 0.21 1.07 (0.81, 1.4) 0.64 Race (ref. Caucasian)     African-American 1.33 (1.04, 1.7) 0.02 1.46 (1.08, 1.98) 0.01     Othersa 1.30 (0.99, 1.69) 0.06 1.20 (0.89, 1.61) 0.24     Transfer status (ref. not transferred in) 2.41 <0.0001 2.03 (1.52, 2.71) <0.0001 Primary payer status (ref. Medicare)     Medicaid 0.75 (0.52, 1.1) 0.14 – –     Private insurance 0.84 (0.67, 1.05) 0.12 – –     Self-pay 0.73 (0.34, 1.58) 0.43 – –     No charge 2.04 (0.46, 9.0) 0.34 – –     Otherb 1.19 (0.66, 2.17) 0.57 – – Income quartile (ref. quartile 1)     Quartile 2 0.87 (0.66, 1.15) 0.33 0.86 (0.63, 1.18) 0.36     Quartile 3 1.14 (0.87, 1.48) 0.34 1.10 (0.82, 1.49) 0.52     Quartile 4 1.35 (1.05, 1.73) 0.02 1.41 (1.07, 1.87) 0.02 Hospital region (ref. south)     Northeast 1.14 (0.9, 1.44) 0.29 – –     Midwest 0.74 (0.56, 0.98) 0.03 – –     West 1.20 (0.92, 1.56) 0.17 – – Urban hospital (ref. rural) 1.18 (0.84, 1.66) 0.33 – – Teaching hospital (ref. non-teaching) 1.10 (0.91, 1.33) 0.31 – – Hospital size (ref. small)     Medium 1.06 (0.77, 1.47) 0.71 – –     Large 0.96 (0.72, 1.28) 0.79 – – Relevant clinical conditions     Aspiration 4.92 (3.73, 6.51) <0.0001 3.52 (2.51, 4.94) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.33 (0.42, 4.21) 0.62 0.91 (0.22, 3.76) 0.90         Other infections 4.07 (3.35, 4.94) <0.0001 3.36 (2.73, 4.14) <0.0001     Acute renal failure 3.47 (2.88, 4.17) <0.0001 4.31 (3.32, 5.6) <0.0001     Pulmonary fibrosis 2.95 (1.61, 5.42) <0.0001 2.23 (1.15, 4.3) 0.02     Weight loss/cachexia 2.88 (2.27, 3.66) <0.0001 2.31 (1.74, 3.05) <0.0001     PEMC 2.26 (1.75, 2.91) <0.0001 1.6 (1.19, 2.16) <0.0001     Pulmonary arterial hypertension 2.22 (1.85, 2.66) <0.0001 1.82 (1.47, 2.25) <0.0001     Congestive heart failure 2.12 (1.76, 2.55) <0.0001 – –     Arrhythmia 2.05 (1.48, 2.84) <0.0001 1.53 (1.04, 2.25) 0.03     Chronic kidney disease 1.70 (1.39, 2.08) <0.0001 0.52 (0.39, 0.69) <0.0001     Chronic liver disease 1.60 (1, 2.58) 0.05 – –     Stroke/TIA 1.56 (1.04, 2.35) 0.03 1.86 (1.17, 2.95) 0.01     Coronary artery disease 1.38 (1.12, 1.7) <0.0001 1.3 (1.02, 1.66) 0.03     Acute bowel obstruction 1.12 (0.72, 1.73) 0.62 – –     Raynaud’s/ulcer/gangrene 0.89 (0.71, 1.11) 0.29 – –     Oesophageal dysfunction 0.77 (0.64, 0.92) <0.0001 0.72 (0.59, 0.89) <0.0001     Hypertension 0.75 (0.63, 0.9) <0.0001 0.71 (0.57, 0.89) <0.0001     Diabetes mellitus 0.72 (0.56, 0.94) 0.02 0.62 (0.46, 0.84) <0.0001     Myositis 0.53 (0.13, 2.2) 0.39 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.01 (0.7, 1.46) 0.95 1.10 (0.73, 1.66) 0.64     ≥65 years 1.42 (0.99, 2.02) 0.054 1.35 (0.89, 2.06) 0.16 Male sex 1.16 (0.92, 1.45) 0.21 1.07 (0.81, 1.4) 0.64 Race (ref. Caucasian)     African-American 1.33 (1.04, 1.7) 0.02 1.46 (1.08, 1.98) 0.01     Othersa 1.30 (0.99, 1.69) 0.06 1.20 (0.89, 1.61) 0.24     Transfer status (ref. not transferred in) 2.41 <0.0001 2.03 (1.52, 2.71) <0.0001 Primary payer status (ref. Medicare)     Medicaid 0.75 (0.52, 1.1) 0.14 – –     Private insurance 0.84 (0.67, 1.05) 0.12 – –     Self-pay 0.73 (0.34, 1.58) 0.43 – –     No charge 2.04 (0.46, 9.0) 0.34 – –     Otherb 1.19 (0.66, 2.17) 0.57 – – Income quartile (ref. quartile 1)     Quartile 2 0.87 (0.66, 1.15) 0.33 0.86 (0.63, 1.18) 0.36     Quartile 3 1.14 (0.87, 1.48) 0.34 1.10 (0.82, 1.49) 0.52     Quartile 4 1.35 (1.05, 1.73) 0.02 1.41 (1.07, 1.87) 0.02 Hospital region (ref. south)     Northeast 1.14 (0.9, 1.44) 0.29 – –     Midwest 0.74 (0.56, 0.98) 0.03 – –     West 1.20 (0.92, 1.56) 0.17 – – Urban hospital (ref. rural) 1.18 (0.84, 1.66) 0.33 – – Teaching hospital (ref. non-teaching) 1.10 (0.91, 1.33) 0.31 – – Hospital size (ref. small)     Medium 1.06 (0.77, 1.47) 0.71 – –     Large 0.96 (0.72, 1.28) 0.79 – – Relevant clinical conditions     Aspiration 4.92 (3.73, 6.51) <0.0001 3.52 (2.51, 4.94) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.33 (0.42, 4.21) 0.62 0.91 (0.22, 3.76) 0.90         Other infections 4.07 (3.35, 4.94) <0.0001 3.36 (2.73, 4.14) <0.0001     Acute renal failure 3.47 (2.88, 4.17) <0.0001 4.31 (3.32, 5.6) <0.0001     Pulmonary fibrosis 2.95 (1.61, 5.42) <0.0001 2.23 (1.15, 4.3) 0.02     Weight loss/cachexia 2.88 (2.27, 3.66) <0.0001 2.31 (1.74, 3.05) <0.0001     PEMC 2.26 (1.75, 2.91) <0.0001 1.6 (1.19, 2.16) <0.0001     Pulmonary arterial hypertension 2.22 (1.85, 2.66) <0.0001 1.82 (1.47, 2.25) <0.0001     Congestive heart failure 2.12 (1.76, 2.55) <0.0001 – –     Arrhythmia 2.05 (1.48, 2.84) <0.0001 1.53 (1.04, 2.25) 0.03     Chronic kidney disease 1.70 (1.39, 2.08) <0.0001 0.52 (0.39, 0.69) <0.0001     Chronic liver disease 1.60 (1, 2.58) 0.05 – –     Stroke/TIA 1.56 (1.04, 2.35) 0.03 1.86 (1.17, 2.95) 0.01     Coronary artery disease 1.38 (1.12, 1.7) <0.0001 1.3 (1.02, 1.66) 0.03     Acute bowel obstruction 1.12 (0.72, 1.73) 0.62 – –     Raynaud’s/ulcer/gangrene 0.89 (0.71, 1.11) 0.29 – –     Oesophageal dysfunction 0.77 (0.64, 0.92) <0.0001 0.72 (0.59, 0.89) <0.0001     Hypertension 0.75 (0.63, 0.9) <0.0001 0.71 (0.57, 0.89) <0.0001     Diabetes mellitus 0.72 (0.56, 0.94) 0.02 0.62 (0.46, 0.84) <0.0001     Myositis 0.53 (0.13, 2.2) 0.39 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker's Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programmes. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; TIA: transient ischaemic attack. Table 4 Factors associated with longer length of stay (>90th percentile) among patients with SSc: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.18 (0.91, 1.52) 0.21 1.23 (0.93, 1.61) 0.14     ≥65 years 0.92 (0.71, 1.19) 0.53 0.85 (0.64, 1.12) 0.25 Male sex 1.27 (1.07, 1.51) 0.01 1.11 (0.92, 1.35) 0.27 Race (ref. Caucasian)     African-American 1.33 (1.09, 1.62) 0.01 – –     Othersa 1.31 (1.07, 1.6) 0.01 – – Transfer status (ref. not transferred in) 2.55 (2.11, 3.09) <0.0001 2.09 (1.68, 2.59) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.31 (1.02, 1.69) 0.04 – –     Private insurance 1.02 (0.85, 1.21) 0.85 – –     Self-pay 1.17 (0.72, 1.9) 0.54 – –     No charge 2.68 (0.93, 7.72) 0.07 – –     Otherb 1.27 (0.8, 2.01) 0.30 – – Income quartile (ref. quartile 1)     Quartile 2 0.88 (0.72, 1.08) 0.22 – –     Quartile 3 1.07 (0.87, 1.31) 0.53 – –     Quartile 4 1.04 (0.85, 1.26) 0.70 – – Hospital region (ref. south)     Northeast 1.05 (0.86, 1.28) 0.62 – –     Midwest 0.82 (0.67, 1.02) 0.08 – –     West 0.92 (0.74, 1.13) 0.41 – – Urban hospital (ref. rural) 2.67 (1.88, 3.79) <0.0001 1.97 (1.36, 2.86) <0.0001 Teaching hospital (ref. non-teaching) 1.69 (1.45, 1.99) <0.0001 1.34 (1.12, 1.59) <0.0001 Hospital size (ref. small)     Medium 0.97 (0.73, 1.3) 0.85 – –     Large 1.32 (1.02, 1.7) 0.03 – – Relevant clinical conditions     Aspiration 3.79 (2.98, 4.81) <0.0001 2.88 (2.2, 3.77) <0.0001     Infection (ref. no infection)         Opportunistic infections 2.40 (1.27, 4.53) 0.01 2.21 (1.13, 4.33) 0.02         Other infections 3.11 (2.7, 3.59) <0.0001 2.8 (2.41, 3.25) <0.0001     Weight loss/cachexia 2.99 (2.48, 3.61) <0.0001 2.41 (1.95, 2.97) <0.0001     Myositis 2.61 (1.49, 4.58) <0.0001 2.17 (1.21, 3.9) 0.01     Pulmonary fibrosis 2.58 (1.55, 4.27) <0.0001 2.01 (1.14, 3.55) 0.02     Acute bowel obstruction 2.43 (1.87, 3.15) <0.0001 2.66 (1.98, 3.57) <0.0001     Acute renal failure 2.33 (2.01, 2.71) <0.0001 2.24 (1.8, 2.78) <0.0001     Arrhythmia 2.15 (1.67, 2.77) <0.0001 1.62 (1.22, 2.16) <0.0001     PEMC 1.98 (1.61, 2.44) <0.0001 1.42 (1.13, 1.78) <0.0001     Congestive heart failure 1.81 (1.56, 2.1) <0.0001 1.44 (1.22, 1.71) <0.0001     Chronic kidney disease 1.58 (1.34, 1.88) <0.0001 0.73 (0.57, 0.95) 0.02     Pulmonary arterial hypertension 1.56 (1.34, 1.81) <0.0001 1.23 (1.04, 1.45) 0.02     Chronic liver disease 1.45 (1.02, 2.08) 0.04 – –     Diabetes mellitus 1.12 (0.94, 1.34) 0.21 – –     Raynaud’s/ulcer/gangrene 1.10 (0.93, 1.28) 0.26 – –     Hypertension 1.10 (0.96, 1.28) 0.17 – –     Coronary artery disease 1.00 (0.85, 1.19) 0.99 – –     Stroke/TIA 0.98 (0.69, 1.4) 0.9 – –     Oesophageal dysfunction 0.95 (0.83, 1.09) 0.44 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.18 (0.91, 1.52) 0.21 1.23 (0.93, 1.61) 0.14     ≥65 years 0.92 (0.71, 1.19) 0.53 0.85 (0.64, 1.12) 0.25 Male sex 1.27 (1.07, 1.51) 0.01 1.11 (0.92, 1.35) 0.27 Race (ref. Caucasian)     African-American 1.33 (1.09, 1.62) 0.01 – –     Othersa 1.31 (1.07, 1.6) 0.01 – – Transfer status (ref. not transferred in) 2.55 (2.11, 3.09) <0.0001 2.09 (1.68, 2.59) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.31 (1.02, 1.69) 0.04 – –     Private insurance 1.02 (0.85, 1.21) 0.85 – –     Self-pay 1.17 (0.72, 1.9) 0.54 – –     No charge 2.68 (0.93, 7.72) 0.07 – –     Otherb 1.27 (0.8, 2.01) 0.30 – – Income quartile (ref. quartile 1)     Quartile 2 0.88 (0.72, 1.08) 0.22 – –     Quartile 3 1.07 (0.87, 1.31) 0.53 – –     Quartile 4 1.04 (0.85, 1.26) 0.70 – – Hospital region (ref. south)     Northeast 1.05 (0.86, 1.28) 0.62 – –     Midwest 0.82 (0.67, 1.02) 0.08 – –     West 0.92 (0.74, 1.13) 0.41 – – Urban hospital (ref. rural) 2.67 (1.88, 3.79) <0.0001 1.97 (1.36, 2.86) <0.0001 Teaching hospital (ref. non-teaching) 1.69 (1.45, 1.99) <0.0001 1.34 (1.12, 1.59) <0.0001 Hospital size (ref. small)     Medium 0.97 (0.73, 1.3) 0.85 – –     Large 1.32 (1.02, 1.7) 0.03 – – Relevant clinical conditions     Aspiration 3.79 (2.98, 4.81) <0.0001 2.88 (2.2, 3.77) <0.0001     Infection (ref. no infection)         Opportunistic infections 2.40 (1.27, 4.53) 0.01 2.21 (1.13, 4.33) 0.02         Other infections 3.11 (2.7, 3.59) <0.0001 2.8 (2.41, 3.25) <0.0001     Weight loss/cachexia 2.99 (2.48, 3.61) <0.0001 2.41 (1.95, 2.97) <0.0001     Myositis 2.61 (1.49, 4.58) <0.0001 2.17 (1.21, 3.9) 0.01     Pulmonary fibrosis 2.58 (1.55, 4.27) <0.0001 2.01 (1.14, 3.55) 0.02     Acute bowel obstruction 2.43 (1.87, 3.15) <0.0001 2.66 (1.98, 3.57) <0.0001     Acute renal failure 2.33 (2.01, 2.71) <0.0001 2.24 (1.8, 2.78) <0.0001     Arrhythmia 2.15 (1.67, 2.77) <0.0001 1.62 (1.22, 2.16) <0.0001     PEMC 1.98 (1.61, 2.44) <0.0001 1.42 (1.13, 1.78) <0.0001     Congestive heart failure 1.81 (1.56, 2.1) <0.0001 1.44 (1.22, 1.71) <0.0001     Chronic kidney disease 1.58 (1.34, 1.88) <0.0001 0.73 (0.57, 0.95) 0.02     Pulmonary arterial hypertension 1.56 (1.34, 1.81) <0.0001 1.23 (1.04, 1.45) 0.02     Chronic liver disease 1.45 (1.02, 2.08) 0.04 – –     Diabetes mellitus 1.12 (0.94, 1.34) 0.21 – –     Raynaud’s/ulcer/gangrene 1.10 (0.93, 1.28) 0.26 – –     Hypertension 1.10 (0.96, 1.28) 0.17 – –     Coronary artery disease 1.00 (0.85, 1.19) 0.99 – –     Stroke/TIA 0.98 (0.69, 1.4) 0.9 – –     Oesophageal dysfunction 0.95 (0.83, 1.09) 0.44 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programmes. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; ref.: Reference group; TIA: transient ischaemic attack. Table 4 Factors associated with longer length of stay (>90th percentile) among patients with SSc: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.18 (0.91, 1.52) 0.21 1.23 (0.93, 1.61) 0.14     ≥65 years 0.92 (0.71, 1.19) 0.53 0.85 (0.64, 1.12) 0.25 Male sex 1.27 (1.07, 1.51) 0.01 1.11 (0.92, 1.35) 0.27 Race (ref. Caucasian)     African-American 1.33 (1.09, 1.62) 0.01 – –     Othersa 1.31 (1.07, 1.6) 0.01 – – Transfer status (ref. not transferred in) 2.55 (2.11, 3.09) <0.0001 2.09 (1.68, 2.59) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.31 (1.02, 1.69) 0.04 – –     Private insurance 1.02 (0.85, 1.21) 0.85 – –     Self-pay 1.17 (0.72, 1.9) 0.54 – –     No charge 2.68 (0.93, 7.72) 0.07 – –     Otherb 1.27 (0.8, 2.01) 0.30 – – Income quartile (ref. quartile 1)     Quartile 2 0.88 (0.72, 1.08) 0.22 – –     Quartile 3 1.07 (0.87, 1.31) 0.53 – –     Quartile 4 1.04 (0.85, 1.26) 0.70 – – Hospital region (ref. south)     Northeast 1.05 (0.86, 1.28) 0.62 – –     Midwest 0.82 (0.67, 1.02) 0.08 – –     West 0.92 (0.74, 1.13) 0.41 – – Urban hospital (ref. rural) 2.67 (1.88, 3.79) <0.0001 1.97 (1.36, 2.86) <0.0001 Teaching hospital (ref. non-teaching) 1.69 (1.45, 1.99) <0.0001 1.34 (1.12, 1.59) <0.0001 Hospital size (ref. small)     Medium 0.97 (0.73, 1.3) 0.85 – –     Large 1.32 (1.02, 1.7) 0.03 – – Relevant clinical conditions     Aspiration 3.79 (2.98, 4.81) <0.0001 2.88 (2.2, 3.77) <0.0001     Infection (ref. no infection)         Opportunistic infections 2.40 (1.27, 4.53) 0.01 2.21 (1.13, 4.33) 0.02         Other infections 3.11 (2.7, 3.59) <0.0001 2.8 (2.41, 3.25) <0.0001     Weight loss/cachexia 2.99 (2.48, 3.61) <0.0001 2.41 (1.95, 2.97) <0.0001     Myositis 2.61 (1.49, 4.58) <0.0001 2.17 (1.21, 3.9) 0.01     Pulmonary fibrosis 2.58 (1.55, 4.27) <0.0001 2.01 (1.14, 3.55) 0.02     Acute bowel obstruction 2.43 (1.87, 3.15) <0.0001 2.66 (1.98, 3.57) <0.0001     Acute renal failure 2.33 (2.01, 2.71) <0.0001 2.24 (1.8, 2.78) <0.0001     Arrhythmia 2.15 (1.67, 2.77) <0.0001 1.62 (1.22, 2.16) <0.0001     PEMC 1.98 (1.61, 2.44) <0.0001 1.42 (1.13, 1.78) <0.0001     Congestive heart failure 1.81 (1.56, 2.1) <0.0001 1.44 (1.22, 1.71) <0.0001     Chronic kidney disease 1.58 (1.34, 1.88) <0.0001 0.73 (0.57, 0.95) 0.02     Pulmonary arterial hypertension 1.56 (1.34, 1.81) <0.0001 1.23 (1.04, 1.45) 0.02     Chronic liver disease 1.45 (1.02, 2.08) 0.04 – –     Diabetes mellitus 1.12 (0.94, 1.34) 0.21 – –     Raynaud’s/ulcer/gangrene 1.10 (0.93, 1.28) 0.26 – –     Hypertension 1.10 (0.96, 1.28) 0.17 – –     Coronary artery disease 1.00 (0.85, 1.19) 0.99 – –     Stroke/TIA 0.98 (0.69, 1.4) 0.9 – –     Oesophageal dysfunction 0.95 (0.83, 1.09) 0.44 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.18 (0.91, 1.52) 0.21 1.23 (0.93, 1.61) 0.14     ≥65 years 0.92 (0.71, 1.19) 0.53 0.85 (0.64, 1.12) 0.25 Male sex 1.27 (1.07, 1.51) 0.01 1.11 (0.92, 1.35) 0.27 Race (ref. Caucasian)     African-American 1.33 (1.09, 1.62) 0.01 – –     Othersa 1.31 (1.07, 1.6) 0.01 – – Transfer status (ref. not transferred in) 2.55 (2.11, 3.09) <0.0001 2.09 (1.68, 2.59) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.31 (1.02, 1.69) 0.04 – –     Private insurance 1.02 (0.85, 1.21) 0.85 – –     Self-pay 1.17 (0.72, 1.9) 0.54 – –     No charge 2.68 (0.93, 7.72) 0.07 – –     Otherb 1.27 (0.8, 2.01) 0.30 – – Income quartile (ref. quartile 1)     Quartile 2 0.88 (0.72, 1.08) 0.22 – –     Quartile 3 1.07 (0.87, 1.31) 0.53 – –     Quartile 4 1.04 (0.85, 1.26) 0.70 – – Hospital region (ref. south)     Northeast 1.05 (0.86, 1.28) 0.62 – –     Midwest 0.82 (0.67, 1.02) 0.08 – –     West 0.92 (0.74, 1.13) 0.41 – – Urban hospital (ref. rural) 2.67 (1.88, 3.79) <0.0001 1.97 (1.36, 2.86) <0.0001 Teaching hospital (ref. non-teaching) 1.69 (1.45, 1.99) <0.0001 1.34 (1.12, 1.59) <0.0001 Hospital size (ref. small)     Medium 0.97 (0.73, 1.3) 0.85 – –     Large 1.32 (1.02, 1.7) 0.03 – – Relevant clinical conditions     Aspiration 3.79 (2.98, 4.81) <0.0001 2.88 (2.2, 3.77) <0.0001     Infection (ref. no infection)         Opportunistic infections 2.40 (1.27, 4.53) 0.01 2.21 (1.13, 4.33) 0.02         Other infections 3.11 (2.7, 3.59) <0.0001 2.8 (2.41, 3.25) <0.0001     Weight loss/cachexia 2.99 (2.48, 3.61) <0.0001 2.41 (1.95, 2.97) <0.0001     Myositis 2.61 (1.49, 4.58) <0.0001 2.17 (1.21, 3.9) 0.01     Pulmonary fibrosis 2.58 (1.55, 4.27) <0.0001 2.01 (1.14, 3.55) 0.02     Acute bowel obstruction 2.43 (1.87, 3.15) <0.0001 2.66 (1.98, 3.57) <0.0001     Acute renal failure 2.33 (2.01, 2.71) <0.0001 2.24 (1.8, 2.78) <0.0001     Arrhythmia 2.15 (1.67, 2.77) <0.0001 1.62 (1.22, 2.16) <0.0001     PEMC 1.98 (1.61, 2.44) <0.0001 1.42 (1.13, 1.78) <0.0001     Congestive heart failure 1.81 (1.56, 2.1) <0.0001 1.44 (1.22, 1.71) <0.0001     Chronic kidney disease 1.58 (1.34, 1.88) <0.0001 0.73 (0.57, 0.95) 0.02     Pulmonary arterial hypertension 1.56 (1.34, 1.81) <0.0001 1.23 (1.04, 1.45) 0.02     Chronic liver disease 1.45 (1.02, 2.08) 0.04 – –     Diabetes mellitus 1.12 (0.94, 1.34) 0.21 – –     Raynaud’s/ulcer/gangrene 1.10 (0.93, 1.28) 0.26 – –     Hypertension 1.10 (0.96, 1.28) 0.17 – –     Coronary artery disease 1.00 (0.85, 1.19) 0.99 – –     Stroke/TIA 0.98 (0.69, 1.4) 0.9 – –     Oesophageal dysfunction 0.95 (0.83, 1.09) 0.44 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programmes. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; ref.: Reference group; TIA: transient ischaemic attack. Table 5 Factors associated with higher cost of hospitalization (>90th percentile) among patients with SSc: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.23 (0.96, 1.57) 0.10 1.21 (0.92, 1.6) 0.17     ≥65 years 0.96 (0.75, 1.24) 0.78 0.89 (0.67¸ 1.19) 0.43 Male sex 1.21 (1.03¸ 1.44) 0.02 1.11 (0.92¸ 1.34) 0.27 Race (ref. Caucasian)     African-American 1.24 (1.01, 1.52) 0.04 1.22 (0.97, 1.54) 0.09     Othersa 1.96 (1.62, 2.37) <0.0001 1.53 (1.24, 1.89) <0.0001 Transfer status (ref. not transferred in) 2.05 (1.68¸ 2.5) <0.0001 1.80 (1.44, 2.26) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.17 (0.92, 1.5) 0.20 – –     Private insurance 1.21 (1.04, 1.41) 0.02 – –     Self-pay 0.77 (0.46, 1.31) 0.34 – –     No charge 2.17 (0.76¸ 6.25) 0.15 – –     Otherb 0.87 (0.52, 1.44) 0.58 – – Income quartile (ref. quartile 1)     Quartile 2 1.10 (0.9, 1.34) 0.38 1.16 (0.93, 1.45) 0.18     Quartile 3 1.48 (1.21, 1.81) <0.0001 1.42 (1.13, 1.79) <0.0001     Quartile 4 1.78 (1.44, 2.19) <0.0001 1.80 (1.43, 2.28) <0.0001 Hospital region (ref. south)     Northeast 1.22 (0.97, 1.55) 0.09 1.03 (0.81, 1.31) 0.80     Midwest 1.22 (0.96, 1.55) 0.10 1.19 (0.92, 1.53) 0.20     West 3.00 (2.39, 3.77) <0.0001 2.98 (2.34, 3.78) <0.0001 Urban hospital (ref. rural) 3.19 (2.16, 4.71) <0.0001 – – Teaching hospital (ref. non-teaching) 1.91 (1.6, 2.27) <0.0001 1.90 (1.57, 2.31) <0.0001 Hospital size (ref. small)     Medium 1.06 (0.79, 1.43) 0.69 – –     Large 1.20 (0.92, 1.57) 0.18 – – Relevant clinical conditions     Myositis 2.83 (1.69, 4.77) <0.0001 – –     Aspiration 2.81 (2.19, 3.6) <0.0001 2.29 (1.72, 3.04) <0.0001     Pulmonary fibrosis 2.59 (1.61, 4.17) <0.0001 1.83 (1.06, 3.15) 0.03     Arrhythmia 2.42 (1.93, 3.05) <0.0001 1.95 (1.51, 2.52) <0.0001     Chronic liver disease 2.22 (1.6, 3.09) <0.0001 1.67 (1.18, 2.36) <0.0001     Acute renal failure 2.17 (1.89, 2.48) <0.0001 2.31 (1.85, 2.89) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.89 (1.01, 3.55) 0.047 1.39 (0.71, 2.71) 0.33         Other infections 2.08 (1.81, 2.39) <0.0001 1.87 (1.61, 2.18) <0.0001     PEMC 1.98 (1.63, 2.4) <0.0001 1.61 (1.29, 2) <0.0001     Acute bowel obstruction 1.93 (1.48, 2.51) <0.0001 2.32 (1.71, 3.15) <0.0001     Weight loss/cachexia 1.80 (1.47, 2.19) <0.0001 1.5 (1.2, 1.89) <0.0001     Chronic kidney disease 1.53 (1.31, 1.79) <0.0001 0.69 (0.54, 0.89) <0.0001     Pulmonary arterial hypertension 1.50 (1.29, 1.74) <0.0001 1.3 (1.1, 1.53) <0.0001     Congestive heart failure 1.49 (1.28, 1.72) <0.0001 – –     Stroke/TIA 1.40 (1.05, 1.87) 0.02 1.66 (1.21, 2.29) <0.0001     Diabetes mellitus 1.32 (1.12, 1.54) <0.0001 1.23 (1.03, 1.47) 0.02     Raynaud’s/ulcer/gangrene 1.21 (1.05, 1.4) 0.01 1.27 (1.08, 1.5) <0.0001     Coronary artery disease 1.19 (1.03, 1.38) 0.02 1.27 (1.07, 1.51) 0.01     Hypertension 1.17 (1.03, 1.33) 0.02 1.19 (1.03, 1.39) 0.02     Oesophageal dysfunction 0.96 (0.85, 1.09) 0.53 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.23 (0.96, 1.57) 0.10 1.21 (0.92, 1.6) 0.17     ≥65 years 0.96 (0.75, 1.24) 0.78 0.89 (0.67¸ 1.19) 0.43 Male sex 1.21 (1.03¸ 1.44) 0.02 1.11 (0.92¸ 1.34) 0.27 Race (ref. Caucasian)     African-American 1.24 (1.01, 1.52) 0.04 1.22 (0.97, 1.54) 0.09     Othersa 1.96 (1.62, 2.37) <0.0001 1.53 (1.24, 1.89) <0.0001 Transfer status (ref. not transferred in) 2.05 (1.68¸ 2.5) <0.0001 1.80 (1.44, 2.26) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.17 (0.92, 1.5) 0.20 – –     Private insurance 1.21 (1.04, 1.41) 0.02 – –     Self-pay 0.77 (0.46, 1.31) 0.34 – –     No charge 2.17 (0.76¸ 6.25) 0.15 – –     Otherb 0.87 (0.52, 1.44) 0.58 – – Income quartile (ref. quartile 1)     Quartile 2 1.10 (0.9, 1.34) 0.38 1.16 (0.93, 1.45) 0.18     Quartile 3 1.48 (1.21, 1.81) <0.0001 1.42 (1.13, 1.79) <0.0001     Quartile 4 1.78 (1.44, 2.19) <0.0001 1.80 (1.43, 2.28) <0.0001 Hospital region (ref. south)     Northeast 1.22 (0.97, 1.55) 0.09 1.03 (0.81, 1.31) 0.80     Midwest 1.22 (0.96, 1.55) 0.10 1.19 (0.92, 1.53) 0.20     West 3.00 (2.39, 3.77) <0.0001 2.98 (2.34, 3.78) <0.0001 Urban hospital (ref. rural) 3.19 (2.16, 4.71) <0.0001 – – Teaching hospital (ref. non-teaching) 1.91 (1.6, 2.27) <0.0001 1.90 (1.57, 2.31) <0.0001 Hospital size (ref. small)     Medium 1.06 (0.79, 1.43) 0.69 – –     Large 1.20 (0.92, 1.57) 0.18 – – Relevant clinical conditions     Myositis 2.83 (1.69, 4.77) <0.0001 – –     Aspiration 2.81 (2.19, 3.6) <0.0001 2.29 (1.72, 3.04) <0.0001     Pulmonary fibrosis 2.59 (1.61, 4.17) <0.0001 1.83 (1.06, 3.15) 0.03     Arrhythmia 2.42 (1.93, 3.05) <0.0001 1.95 (1.51, 2.52) <0.0001     Chronic liver disease 2.22 (1.6, 3.09) <0.0001 1.67 (1.18, 2.36) <0.0001     Acute renal failure 2.17 (1.89, 2.48) <0.0001 2.31 (1.85, 2.89) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.89 (1.01, 3.55) 0.047 1.39 (0.71, 2.71) 0.33         Other infections 2.08 (1.81, 2.39) <0.0001 1.87 (1.61, 2.18) <0.0001     PEMC 1.98 (1.63, 2.4) <0.0001 1.61 (1.29, 2) <0.0001     Acute bowel obstruction 1.93 (1.48, 2.51) <0.0001 2.32 (1.71, 3.15) <0.0001     Weight loss/cachexia 1.80 (1.47, 2.19) <0.0001 1.5 (1.2, 1.89) <0.0001     Chronic kidney disease 1.53 (1.31, 1.79) <0.0001 0.69 (0.54, 0.89) <0.0001     Pulmonary arterial hypertension 1.50 (1.29, 1.74) <0.0001 1.3 (1.1, 1.53) <0.0001     Congestive heart failure 1.49 (1.28, 1.72) <0.0001 – –     Stroke/TIA 1.40 (1.05, 1.87) 0.02 1.66 (1.21, 2.29) <0.0001     Diabetes mellitus 1.32 (1.12, 1.54) <0.0001 1.23 (1.03, 1.47) 0.02     Raynaud’s/ulcer/gangrene 1.21 (1.05, 1.4) 0.01 1.27 (1.08, 1.5) <0.0001     Coronary artery disease 1.19 (1.03, 1.38) 0.02 1.27 (1.07, 1.51) 0.01     Hypertension 1.17 (1.03, 1.33) 0.02 1.19 (1.03, 1.39) 0.02     Oesophageal dysfunction 0.96 (0.85, 1.09) 0.53 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programs. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; TIA: transient ischaemic attack. Table 5 Factors associated with higher cost of hospitalization (>90th percentile) among patients with SSc: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.23 (0.96, 1.57) 0.10 1.21 (0.92, 1.6) 0.17     ≥65 years 0.96 (0.75, 1.24) 0.78 0.89 (0.67¸ 1.19) 0.43 Male sex 1.21 (1.03¸ 1.44) 0.02 1.11 (0.92¸ 1.34) 0.27 Race (ref. Caucasian)     African-American 1.24 (1.01, 1.52) 0.04 1.22 (0.97, 1.54) 0.09     Othersa 1.96 (1.62, 2.37) <0.0001 1.53 (1.24, 1.89) <0.0001 Transfer status (ref. not transferred in) 2.05 (1.68¸ 2.5) <0.0001 1.80 (1.44, 2.26) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.17 (0.92, 1.5) 0.20 – –     Private insurance 1.21 (1.04, 1.41) 0.02 – –     Self-pay 0.77 (0.46, 1.31) 0.34 – –     No charge 2.17 (0.76¸ 6.25) 0.15 – –     Otherb 0.87 (0.52, 1.44) 0.58 – – Income quartile (ref. quartile 1)     Quartile 2 1.10 (0.9, 1.34) 0.38 1.16 (0.93, 1.45) 0.18     Quartile 3 1.48 (1.21, 1.81) <0.0001 1.42 (1.13, 1.79) <0.0001     Quartile 4 1.78 (1.44, 2.19) <0.0001 1.80 (1.43, 2.28) <0.0001 Hospital region (ref. south)     Northeast 1.22 (0.97, 1.55) 0.09 1.03 (0.81, 1.31) 0.80     Midwest 1.22 (0.96, 1.55) 0.10 1.19 (0.92, 1.53) 0.20     West 3.00 (2.39, 3.77) <0.0001 2.98 (2.34, 3.78) <0.0001 Urban hospital (ref. rural) 3.19 (2.16, 4.71) <0.0001 – – Teaching hospital (ref. non-teaching) 1.91 (1.6, 2.27) <0.0001 1.90 (1.57, 2.31) <0.0001 Hospital size (ref. small)     Medium 1.06 (0.79, 1.43) 0.69 – –     Large 1.20 (0.92, 1.57) 0.18 – – Relevant clinical conditions     Myositis 2.83 (1.69, 4.77) <0.0001 – –     Aspiration 2.81 (2.19, 3.6) <0.0001 2.29 (1.72, 3.04) <0.0001     Pulmonary fibrosis 2.59 (1.61, 4.17) <0.0001 1.83 (1.06, 3.15) 0.03     Arrhythmia 2.42 (1.93, 3.05) <0.0001 1.95 (1.51, 2.52) <0.0001     Chronic liver disease 2.22 (1.6, 3.09) <0.0001 1.67 (1.18, 2.36) <0.0001     Acute renal failure 2.17 (1.89, 2.48) <0.0001 2.31 (1.85, 2.89) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.89 (1.01, 3.55) 0.047 1.39 (0.71, 2.71) 0.33         Other infections 2.08 (1.81, 2.39) <0.0001 1.87 (1.61, 2.18) <0.0001     PEMC 1.98 (1.63, 2.4) <0.0001 1.61 (1.29, 2) <0.0001     Acute bowel obstruction 1.93 (1.48, 2.51) <0.0001 2.32 (1.71, 3.15) <0.0001     Weight loss/cachexia 1.80 (1.47, 2.19) <0.0001 1.5 (1.2, 1.89) <0.0001     Chronic kidney disease 1.53 (1.31, 1.79) <0.0001 0.69 (0.54, 0.89) <0.0001     Pulmonary arterial hypertension 1.50 (1.29, 1.74) <0.0001 1.3 (1.1, 1.53) <0.0001     Congestive heart failure 1.49 (1.28, 1.72) <0.0001 – –     Stroke/TIA 1.40 (1.05, 1.87) 0.02 1.66 (1.21, 2.29) <0.0001     Diabetes mellitus 1.32 (1.12, 1.54) <0.0001 1.23 (1.03, 1.47) 0.02     Raynaud’s/ulcer/gangrene 1.21 (1.05, 1.4) 0.01 1.27 (1.08, 1.5) <0.0001     Coronary artery disease 1.19 (1.03, 1.38) 0.02 1.27 (1.07, 1.51) 0.01     Hypertension 1.17 (1.03, 1.33) 0.02 1.19 (1.03, 1.39) 0.02     Oesophageal dysfunction 0.96 (0.85, 1.09) 0.53 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.23 (0.96, 1.57) 0.10 1.21 (0.92, 1.6) 0.17     ≥65 years 0.96 (0.75, 1.24) 0.78 0.89 (0.67¸ 1.19) 0.43 Male sex 1.21 (1.03¸ 1.44) 0.02 1.11 (0.92¸ 1.34) 0.27 Race (ref. Caucasian)     African-American 1.24 (1.01, 1.52) 0.04 1.22 (0.97, 1.54) 0.09     Othersa 1.96 (1.62, 2.37) <0.0001 1.53 (1.24, 1.89) <0.0001 Transfer status (ref. not transferred in) 2.05 (1.68¸ 2.5) <0.0001 1.80 (1.44, 2.26) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.17 (0.92, 1.5) 0.20 – –     Private insurance 1.21 (1.04, 1.41) 0.02 – –     Self-pay 0.77 (0.46, 1.31) 0.34 – –     No charge 2.17 (0.76¸ 6.25) 0.15 – –     Otherb 0.87 (0.52, 1.44) 0.58 – – Income quartile (ref. quartile 1)     Quartile 2 1.10 (0.9, 1.34) 0.38 1.16 (0.93, 1.45) 0.18     Quartile 3 1.48 (1.21, 1.81) <0.0001 1.42 (1.13, 1.79) <0.0001     Quartile 4 1.78 (1.44, 2.19) <0.0001 1.80 (1.43, 2.28) <0.0001 Hospital region (ref. south)     Northeast 1.22 (0.97, 1.55) 0.09 1.03 (0.81, 1.31) 0.80     Midwest 1.22 (0.96, 1.55) 0.10 1.19 (0.92, 1.53) 0.20     West 3.00 (2.39, 3.77) <0.0001 2.98 (2.34, 3.78) <0.0001 Urban hospital (ref. rural) 3.19 (2.16, 4.71) <0.0001 – – Teaching hospital (ref. non-teaching) 1.91 (1.6, 2.27) <0.0001 1.90 (1.57, 2.31) <0.0001 Hospital size (ref. small)     Medium 1.06 (0.79, 1.43) 0.69 – –     Large 1.20 (0.92, 1.57) 0.18 – – Relevant clinical conditions     Myositis 2.83 (1.69, 4.77) <0.0001 – –     Aspiration 2.81 (2.19, 3.6) <0.0001 2.29 (1.72, 3.04) <0.0001     Pulmonary fibrosis 2.59 (1.61, 4.17) <0.0001 1.83 (1.06, 3.15) 0.03     Arrhythmia 2.42 (1.93, 3.05) <0.0001 1.95 (1.51, 2.52) <0.0001     Chronic liver disease 2.22 (1.6, 3.09) <0.0001 1.67 (1.18, 2.36) <0.0001     Acute renal failure 2.17 (1.89, 2.48) <0.0001 2.31 (1.85, 2.89) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.89 (1.01, 3.55) 0.047 1.39 (0.71, 2.71) 0.33         Other infections 2.08 (1.81, 2.39) <0.0001 1.87 (1.61, 2.18) <0.0001     PEMC 1.98 (1.63, 2.4) <0.0001 1.61 (1.29, 2) <0.0001     Acute bowel obstruction 1.93 (1.48, 2.51) <0.0001 2.32 (1.71, 3.15) <0.0001     Weight loss/cachexia 1.80 (1.47, 2.19) <0.0001 1.5 (1.2, 1.89) <0.0001     Chronic kidney disease 1.53 (1.31, 1.79) <0.0001 0.69 (0.54, 0.89) <0.0001     Pulmonary arterial hypertension 1.50 (1.29, 1.74) <0.0001 1.3 (1.1, 1.53) <0.0001     Congestive heart failure 1.49 (1.28, 1.72) <0.0001 – –     Stroke/TIA 1.40 (1.05, 1.87) 0.02 1.66 (1.21, 2.29) <0.0001     Diabetes mellitus 1.32 (1.12, 1.54) <0.0001 1.23 (1.03, 1.47) 0.02     Raynaud’s/ulcer/gangrene 1.21 (1.05, 1.4) 0.01 1.27 (1.08, 1.5) <0.0001     Coronary artery disease 1.19 (1.03, 1.38) 0.02 1.27 (1.07, 1.51) 0.01     Hypertension 1.17 (1.03, 1.33) 0.02 1.19 (1.03, 1.39) 0.02     Oesophageal dysfunction 0.96 (0.85, 1.09) 0.53 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programs. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; TIA: transient ischaemic attack. Among other studied clinical factors, CHF was significantly associated with longer LOS (aOR = 1.44, 95% CI: 1.22, 1.71) but not with mortality or cost. Arrhythmia, weight loss/cachexia and transfer-in status had higher odds of mortality, longer LOS and higher cost. Infection other than opportunistic had a very strong association with higher odds of mortality (aOR = 3.36, 95% CI: 2.73, 4.41, P < 0.0001) and comparable to that of acute renal failure and aspiration. Interestingly, higher income quartiles were associated with higher odds of mortality (quartile 4, aOR = 1.41, 95% CI: 1.07, 1.87, P = 0.02) and higher cost (quartile 3 aOR = 1.48, 95% CI: 1.13, 1.79; quartile 4 aOR = 1.80, 95% CI: 1.43, 1.28, P < 0.0001 for both). Likewise, urban location with teaching status of the hospital was associated with higher odds of longer LOS (aOR = 1.97, 95% CI: 1.36, 2.86 and aOR = 1.34, 95% CI: 1.12, 1.59, respectively, P < 0.0001 for both). However, higher cost (aOR = 1.9, 95% CI: 1.57, 2.31, P < 0.0001) was associated with teaching status only. In the sensitivity analysis (supplementary Table S2, available at Rheumatology online) for longer LOS after removing SSc patients who died in the hospital, PAH, urban status and teaching status of the hospital were no longer statistically significant. One reason for this could be the tendency of pooling of PAH patients in urban teaching hospitals by virtue of presence of highly specialized PAH centres who take care of these patients towards the end of their life. Of note, the median LOS was longer by 1 day (95% CI: 0.98, 1.03, P < 0.0001) among SSc patients who died in hospital compared with those who did not. Discussion With this study we have updated our understanding of current factors associated with in-hospital mortality, LOS and hospitalization costs among SSc patients. We have built on previous studies and analysed more than 9000 sample hospitalizations with SSc (representing an estimated 48 655 discharges) in the USA from the 2012 to 2013 NIS. We performed a detailed analysis of the factors associated with in-hospital mortality, LOS and cost of hospitalization. In previous studies reporting on inpatient mortality among SSc patients [6–10], SSc-related and cardiopulmonary causes were the most important determinants. In this study, in contrast to two previous studies using the same database, we have consolidated individual diagnostic codes into categories based on systemic/pathologic involvement to more comprehensively identify major diagnoses associated with hospitalization and mortality. We found that infections/septicaemia were the most common diagnoses in both overall SSc hospitalizations and SSc in-hospital mortality (17.4 and 32.7%, respectively). These results are supported by a study by Tyndall et al. [4] in which the authors longitudinally analysed 5860 patients from the EULAR Scleroderma Trials and Research cohort and reported infection to be attributed to 33% of non-SSc-related deaths. Similarly, in a more recent retrospective cross-sectional study of hospitalized SSc patients by Netwijitpan et al. [9] carried out in Thailand, infections were the most common non-SSc-related cause of hospitalization and lower respiratory tract infections were the most common (49.9%) overall cause of death. Another Iranian study by Shenavandeh and Naseri [10] looking at 446 admissions by 181 patients over 13 years also found infection to be the most common non-SSc-related cause for hospitalization. This finding could be explained by a more widespread utilization of immunosuppressive agents, as well as an increased comfort level among clinicians in utilizing such agents in SSc. With this in mind we further explored to determine whether a hospitalization with an infection diagnosis was associated with one of the collective codes representing possible use of chemotherapy, which can include both chemotherapy such as CYC or rituximab and any other immunosuppressive infusion during the hospitalization (supplementary Table S5, available at Rheumatology online). We found that, in comparison with non-SSc hospitalizations, SSc hospitalizations with infection/septicaemia had a significantly higher proportion of the possible use of chemotherapy among both patients who lived (9.2 vs 4.1%, P < 0.0001) and those who died during the hospitalization (17.4 vs 6.3%, P = 0.0005) (data not shown). Patients who died had significantly higher proportions of coding for PAH (43.3%), respiratory failure (72.4%) and CHF (38.4%), suggesting more aggressive SSc disease, which in turn may have required stronger immunosuppressive therapy predisposing to more infections. Similar to the study by Chung et al. [7] renal failure accounted for only a minor proportion of primary hospitalization diagnoses (2%) as well as inpatient mortality (1.7%), reflecting an early detection and management with angiotensin-converting enzyme inhibitors, which may have decreased both hospitalization and in-hospital mortality rates among this subset of SSc patients [5]. In multivariate analysis, however, a diagnosis of acute renal failure from any position of the discharge diagnoses was strongly associated with mortality. Most of these non-primary diagnoses likely do not represent scleroderma renal crisis—renal failure is common in patients with other major illnesses such as infection and heart failure, which are likely to be the primary drivers of mortality. But the evidence of renal failure’s independency in multivariate analysis suggests that it does amplify the odds of mortality and raises a concern that we might need to do more to prevent acute renal injury among hospitalizations with SSc. Interestingly, oesophageal dysfunction was associated with lower odds of mortality in both univariate and multivariate regression analysis. While it may seem contradictory, likely this is because oesophageal dysfunction is not coded as frequently in patients with other severe reasons for hospitalization and this might have falsely made oesophageal dysfunction look like it is protective. However, we noted a very strong association between mortality and aspiration, as in a study by Sehra et al. [6]. Also, the proportion of aspiration was almost double among patients with oesophageal dysfunction (5.95 vs 2.55%, P < 0.0001), indicating that oesophageal dysfunction is likely the index reason for aspiration. Similar to prior studies [7, 8], conditions such as pulmonary fibrosis, CHF, arrhythmias and transfer from another hospital all had higher odds of having a longer LOS. Additionally, as expected, PAH, acute renal failure, myositis, weight loss/cachexia and presence of any infection were some of the factors not studied or found insignificant in prior studies but which significantly increased the odds of longer LOS in our analysis. Overall in-hospital mortality among SSc patients was 5%, which is lower than the older database studies from the USA using the NIS (7.1% in 1995 and 6.3% in 2002–03) [7, 8] but similar to a more recent single-centre study looking at mortality among hospitalized SSc patients at an institution that has a dedicated scleroderma centre (5% in 2001–11) [6]. In our study, African-Americans consistently showed higher rates of mortality. It was particularly high as compared with the Other race groups, among older patients ≥65 years (11.1 vs 6.8% among Other race and 5% among Caucasians). This was reflected even in the multivariate analysis with an odds ratio of 1.46 (P = 0.01) for in-hospital death among African American SSc patients. This finding is in contrast to the two prior studies (Nietert et al. [8] and Chung et al. [7]) done with the same database, which found that race was not associated with higher in-hospital mortality. This may be explained by an ever changing and improving representation of racial information in the HCUP databases [16]. Curiously, higher income (quartile 4) was associated with higher odds (odds ratio = 1.41, P = 0.02) of dying in hospital. This might be explained by an access paradox, with wealthier patients being able to afford expensive immunosuppressive therapies and having more access to in-hospital care, or possible disease perception and management differences not clearly defined by our study. For comparison purpose, we had calculated the average LOS as 5.9 (±7.9) days with a median (interquartile range) of 4 (2, 7) days. This is lower than prior studies based on the same database (mean LOS of 6.6 days in the Chung study and 7.5 days in the Nietert study) with a chronological trend of decreasing overall LOS among SSc patients. This change of trend is most likely multifactorial and may represent advances in overall care, or shift in insurance regulations forcing shorter LOS. We saw a statistically significant relationship between longer LOS and Medicaid being the primary insurance during our multivariate analysis, supporting the hypothesis that SSc patients with lower socioeconomic status have longer LOS. SSc patients hospitalized in institutions labelled as urban and ones with a teaching status had longer LOS. This might be due to the tendency for sicker and more complex patients to present to these settings. The cost of hospitalization of SSc patients based on HCUP-NIS was studied by Nietert et al. [8] but not by Chung et al. [7]. In 1995 the estimated cost of healthcare expenditure in the inpatient setting among SSc patients was over $280 million. After adjusting for inflation [17], this translates to $422 million in 2012 and $429 million in 2013. We estimated a lower adjusted cost of $357 million and $361 million, respectively, for the years 2012 and 2013 in our study. While this could relate to better care and less hospitalizations, it could also relate to the insurance regulations forcing shorter LOS and thus incurring less cost. The findings in our report are subject to some limitations. There is the potential for misclassification of the diagnosis of SSc, although an inpatient diagnosis code for SSc has previously been validated. Also, being an administrative database created for billing purposes, the HCUP-NIS lacks the same rigor and diagnostic accuracy as compared with a database established for research purposes. The diagnosis of SSc per se does not differentiate between limited and diffuse variations of SSc due to the limitation of ICD-9 coding. The ICD-9 code used in this study (710.1) represents any of the following diagnoses: acrosclerosis, Calcinosis, Raynaud's, Esophageal dysmotility, Sclerodactyly, Telangiectasia syndrome, progressive SSc or scleroderma. We were unable to look at differences in hospitalization causes and LOS based on these variants. Similarly, the mortality data in our study are globally assessed disregarding the variants and we were unable to comment on mortality differences among SSc hospitalizations by subtypes. An absence of a diagnostic code does not mean the condition was absent, especially if it was not the primary diagnosis, which could make some comorbidities appear protective, like hypertension and diabetes in our study. As the sample is retrospective and cross-sectional it lacks any temporality in the conditions studied and cannot establish causality. Since multiple admissions of the same patient, if any, were counted as separate instances of hospitalization/discharge, readmissions are not accounted for, thus inflating the actual number of SSc patients utilizing the inpatient service. The cost estimation, which uses cost-to-charge ratios in the analysis, is based on aggregate hospital data and is not specific to a single diagnosis such as SSc, and might be different from the true amount. Despite these limitations, our study has several strengths. Being the largest inpatient and only database with all-payer information, it is the most representative database to reflect national estimates in the inpatient setting [11]. We have analysed a very large population of hospitalized adult SSc patients. Moreover, the redesign of the database since 2012, notably directly sampling discharges rather than hospitals like in previous years, has allowed wider and better representation of the US hospitals geographically and more complete information about race (fewer missing data) is present, which was lacking in the prior years’ databases [17]. With this study we have more clearly defined that the primary diagnosis for hospitalization and in-hospital mortality among SSc patients is infection and that in-hospital mortality and LOS have overtime declined. We believe our findings will serve to increase physician awareness on prevention (influenza and pneumococcal vaccinations) as well as early institution of antibiotherapy when facing a possible infective complication. Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: The authors have declared no conflicts of interest. References 1 Allanore Y , Simms R , Distler O et al. Systemic sclerosis . Nat Rev Dis Primers 2015 ; 1 : 1 – 21 . 2 Rubio-Rivas M , Royo C , Simeón CP , Corbella X , Fonollosa V. Mortality and survival in systemic sclerosis: systematic review and meta-analysis . Semin Arthritis Rheum 2014 ; 44 : 208 – 19 . Google Scholar CrossRef Search ADS PubMed 3 Bose N , Chiesa-Vottero A , Chatterjee S. Scleroderma renal crisis . Semin Arthritis Rheum 2015 ; 44 : 687 – 94 . Google Scholar CrossRef Search ADS PubMed 4 Tyndall AJ , Bannert B , Vonk M et al. Causes and risk factors for death in systemic sclerosis: a study from the EULAR Scleroderma Trials and Research (EUSTAR) database . Ann Rheum Dis 2010 ; 69 : 1809 – 15 . Google Scholar CrossRef Search ADS PubMed 5 Steen VD , Medsger TA. Changes in causes of death in systemic sclerosis, 1972-2002 . Ann Rheum Dis 2007 ; 66 : 940 – 4 . Google Scholar CrossRef Search ADS PubMed 6 Sehra ST , Kelly A , Baker JF , Derk CT. Predictors of inpatient mortality in patients with systemic sclerosis: a case control study . Clin Rheumatol 2016 ; 35 : 1631 – 5 . Google Scholar CrossRef Search ADS PubMed 7 Chung L , Krishnan E , Chakravarty EF. Hospitalizations and mortality in systemic sclerosis: results from the Nationwide Inpatient Sample . Rheumatology (Oxford) 2007 ; 46 : 1808 – 13 . Google Scholar CrossRef Search ADS PubMed 8 Nietert PJ , Silverstein MD , Silver RM. Hospital admissions, length of stay, charges, and in-hospital death among patients with systemic sclerosis . J Rheumatol 2001 ; 28 : 2031 – 7 . Google Scholar PubMed 9 Netwijitpan S , Foocharoen C , Mahakkanukrauh A , Suwannaroj S , Nanagara R. Indications for hospitalization and in-hospital mortality in Thai systemic sclerosis . Clin Rheumatol 2013 ; 32 : 361 – 7 . Google Scholar CrossRef Search ADS PubMed 10 Shenavandeh S , Naseri R. Assessment of hospitalization and mortality of scleroderma in-patients: a thirteen-year study . Reumatologia 2017 ; 55 : 163 – 8 . Google Scholar CrossRef Search ADS PubMed 11 HCUP-US Databases . Healthcare Cost and Utilization Project (HCUP). April 2017. Rockville, MD: Agency for Healthcare Research and Quality. https://www.hcup-us.ahrq.gov/databases.jsp (10 September 2017 , date last accessed). 12 Valenzuela A , Yaqub A , Fiorentino D , Krishnan E , Chung L. Validation of the ICD-9-CM code for systemic sclerosis using updated ACR/EULAR classification criteria . Scand J Rheumatol 2015 ; 44 : 253 – 5 . Google Scholar CrossRef Search ADS PubMed 13 HCUP CCS. Healthcare Cost and Utilization Project (HCUP) . Agency for Healthcare Research and Quality, Rockville, MD. HCUP Databases. Published March 2017. https://www.hcup-us.ahrq.gov/nisoverview.jsp (31 July 2017 , date last accessed). 14 Radensky PW , Berliner E , Archer JW , Dournaux SF. Inpatient costs of major cardiovascular events . Am J Cardiovasc Drugs 2001 ; 1 : 205 – 17 . Google Scholar CrossRef Search ADS PubMed 15 HCUP-US Tools & Software Page . Healthcare Cost and Utilization Project (HCUP). July 2017. Rockville, MD: Agency for Healthcare Research and Quality. https://www.hcup-us.ahrq.gov/tools_software.jsp (10 September 2017 , date last accessed). 16 Houchens R, Ross D, Elixhauser A, Jiang J. Nationwide Inpatient Sample (NIS) Redesign Final Report. 2014. HCUP Methods Series Report # 2014-04 ONLINE. April 4, 2014. U.S. Agency for Healthcare Research and Quality. http://www.hcup-us.ahrq.gov/reports/methods/methods.jsp (12 October 2015 , date last accessed). 17 Bureau of Labor Statistics. CPI Inflation Calculator . https://www.bls.gov/data/inflation_calculator.htm (2 November 2017 , date last accessed). © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Rheumatology Oxford University Press

Mortality, length of stay and cost of hospitalization among patients with systemic sclerosis: results from the National Inpatient Sample

Rheumatology , Volume 57 (9) – Sep 1, 2018

Loading next page...
 
/lp/ou_press/mortality-length-of-stay-and-cost-of-hospitalization-among-patients-C0SNyiR650
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
1462-0324
eISSN
1462-0332
D.O.I.
10.1093/rheumatology/key150
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives To evaluate the hospitalizations and define the factors associated with in-hospital mortality, longer length of stay (LOS) and higher hospital costs among SSc hospitalizations. Methods We used the National Inpatient Sample (2012–13) to identify adult hospitalizations with SSc, excluding patients with concomitant diagnosis of RA and systemic lupus. We calculated rates of hospitalization, in-hospital mortality, LOS and hospital costs. Factors associated with these outcomes were evaluated by univariate and backward stepwise multivariate logistic regression. Results There were 9731 hospitalizations in the sample representing an estimated 48 655 hospitalizations nationwide with SSc (0.09%), and the inpatient mortality rate was 5%. Patients were predominantly older (mean age 63.2 years), female (82.2%) and Caucasian (71.5%). Infections were the most common primary diagnoses among SSc hospitalizations (17.4%) and among those who died (32.7%). Acute renal failure [adjusted odds ratio (aOR) = 4.3, 95% CI: 3.3, 5.6] and aspiration (aOR= 3.5, 95% CI: 2.5, 4.9) were strongly associated with in-hospital mortality. The median (interquartile range) LOS was 4 days (−2, 7), and the median (interquartile range) cost was $8885 (−5169, 15921). While hospital from the West region, acute renal failure, acute bowel obstruction and aspiration (aOR > 2.0 with P < 0.0001 for all) seem to predict higher cost of hospitalization, pulmonary fibrosis, myositis and any type of infection in addition to the same factors, except the West region (aOR > 2.0 with P < 0.0001 for all), were associated with longer LOS. Conclusion Infections are currently the most common diagnoses among SSc hospitalizations and in-hospital deaths. This emphasizes the importance of being vigilant in prevention and early treatment of infections in SSc patients. systemic sclerosis, scleroderma, inpatient, in-hospital, mortality, hospitalization, national database, National Inpatient Sample, length of stay, cost Rheumatology key messages Infection is an important comorbid condition among hospitalized patients with SSc. Infection, acute renal failure and aspiration events are associated with significantly higher mortality in SSc. Focusing on infection prevention might make significant improvement in outcome among hospitalized SSc patients. Introduction SSc (scleroderma) is a chronic rheumatic CTD characterized by autoimmunity with predominantly vascular and fibrotic changes of the skin and visceral organs [1]. It is associated with significant morbidity and higher rates of mortality with standardized mortality rates ranging from 1.05 to 5.40 across various studies [2]. These rates vary by region as do clinical manifestations such as extent of skin and visceral organ involvement among geographically different populations. Although not much improvement has been made in discovering an effective disease-modifying agent for scleroderma, we have seen a decreasing trend in overall mortality and a change in the primary cause of SSc-related mortality from scleroderma renal crisis, in the pre-angiotensin-converting enzyme inhibitor era, to pulmonary-related complications currently [3–5]. Most of the studies looking at mortality or survival among SSc patients have done so among outpatient observational cohorts, which may not fully capture causes of mortality. Few have looked at the status and predictors of mortality among hospitalized patients [6–10]. The most recent comprehensive study of SSc hospitalizations by Chung et al. [7] a decade ago established CTD (SSc) as the principal cause of hospitalization and in-hospital mortality. The objective of our study was to identify current causes and factors associated with in-hospital mortality, length of stay (LOS) and hospitalization costs among patients with SSc by using a large database of hospitalized patients and comparing results to previous reports. Methods Database used We used the National Inpatient Sample (NIS), a part of the Healthcare Cost Utilization Project (HCUP), developed through a Federal State–Industry partnership sponsored by the Agency for Healthcare Research and Quality [11]. We utilized data for the years 2012 and 2013 to perform a cross-sectional analysis of discharge-level administrative data. NIS is the largest publicly available all-payer inpatient health care database in the USA. Each observation in this database represents a unique hospitalization. NIS provides weights for each observation in the sample that can be used to make national (weighted) estimates of hospital inpatient stays. It represents a 20% stratified sample of all discharges in the USA and contains data from >7 million unweighted and an estimated >35 million weighted hospital stays each year. Data for NIS 2012 and 2013 includes discharges from 44 participating states and represents >95% of the US population. NIS provides clinical, demographic and socio-economic information about each observation including up to 25 diagnosis codes (1 primary and 24 secondary), race, in-hospital death, LOS, hospital charges, hospital location, size, teaching status, primary insurance of the patient, transfer-in or -out status, etc. The NIS (HCUP) data are de-identified; thus, this study did not require institutional review board approval or patient consent. More details about this can be accessed from the HCUP website [11]. Selection of patients We identified adults (⩾18 years) with a discharge diagnosis of SSc (primary or secondary) based on the International Classification of Diseases version 9, Clinical Modification (ICD-9-CM) code 710.1 in any diagnosis position. An inpatient diagnosis code has been shown in previous work to have a positive predictive value of 76% for the diagnosis of SSc in administrative databases [12]. It identifies patients with limited SSc as well as dcSSc, but excludes morphea and other forms of localized scleroderma. To improve the specificity of the diagnosis code, we also excluded patients with a concomitant diagnosis code for RA (714.0, 714.1 and 714.2) or SLE (710.0). Hospitalization percentages and indications for hospitalization We used weighted estimates to calculate the proportion of all hospitalizations nationwide due to SSc during the years 2012–13. Indication for hospitalization was evaluated based on the primary diagnosis code and was examined among all patients with SSc and the subgroup of patients with SSc who died. The diagnoses were selected based on relevant ICD-9-CM codes and clinical classification software codes as supplied by HCUP (supplementary Table S1, available at Rheumatology online). Clinical classification software is a diagnosis and procedure categorization scheme where ICD-9-CM’s multitude of codes are collapsed into a smaller number of clinically meaningful categories that are more useful for presenting descriptive statistics than are individual ICD-9-CM codes [13]. While reporting the most common primary diagnoses associated with SSc hospitalization and deaths, we grouped the diagnoses into clinically relevant categories (beyond clinical classification software codes) based on pathologic as well as systemic involvement. Outcomes Among hospitalized patients with SSc, we evaluated the following outcomes: in-hospital mortality, LOS and cost. LOS and cost were treated as binomial variables and measurements greater than the 90th percentile for the overall SSc hospitalizations were considered as prolonged LOS or higher cost because of the highly skewed nature of these measures and clinical relevance of identifying the longest and most costly admissions [14]. Patients who were transferred out to a different acute care setting were excluded from these analyses as NIS does not enable identification of associated hospitalizations. Therefore, mortality, LOS and cost of the entire hospitalization episode cannot be determined by these high-risk patients. HCUP-NIS supplies hospital charges for each observation as well as the cost-to-charge ratio for each hospital in the database in a separate file. The product of the charge and corresponding cost-to-charge ratio gives the estimated cost (dollars spent) for each hospitalization. Factors associated with in-hospital mortality, longer LOS and higher cost of hospitalization Factors of interest included age (categorized as 18–44, 45–64, ≥65 years), sex, race (categorized as Caucasian, African American and Other), various relevant clinical factors, transfer status, primary payer, patient’s income quartile based on zip code, and hospital characteristic by region, locale, teaching status and size. When we identified infections as a leading cause of mortality in the population we were studying, we went back and also looked at the frequency of chemotherapy reception among these patients (considering ICD-9 codes for chemotherapy, immunotherapy infusion or central catheter placement as a surrogate of chemotherapy likely representing CYC or other infusions; supplementary Table S2, available at Rheumatology online). Comorbidities were defined based on Elixhauser Comorbidity Software as supplied by HCUP, which assigns variables that identify comorbidities in hospital discharge records using an algorithm based on ICD-9-CM codes [15]. These include 29 comorbidity measures encompassing all organ systems, and data on the presence of these conditions are supplied in a separate data file (severity file) by HCUP. Statistical reporting and modelling Continuous variables were expressed as mean (s.d.), or median with interquartile range for skewed data (LOS and cost). Categorical variables were expressed as percentages. Backward stepwise multivariate logistic regression excluding variables with P > 0.2 and forcing age and sex into the final model was used to identify independent factors associated with in-hospital mortality, prolonged LOS and high cost. Logistic regression estimates were reported as odds ratio in univariate analyses and as adjusted odds ratio (aOR) in multivariate analysis. Since in-hospital death necessarily impacts and could shorten LOS, a sensitivity analysis was done repeating multivariate regression after excluding patients who died in the hospital, although patients who died on average had longer LOS. All statistical analyses were conducted using statistical software STATA version 13.0 (College Station, TX, USA) and accounted for the stratified sampling design of NIS. A two-sided P < 0.05 was deemed to be statistically significant for all analyses. Results Overall hospitalizations with SSc We identified a total of 9731 hospitalizations with SSc diagnosis from the HCUP-NIS sample data for the years 2012–13, representing 48 655 hospitalizations (0.09%) nationwide. We estimated $719 million expenditure for those hospitalizations, which represented 0.12% of total cost for all hospitalizations within those years. SSc hospitalizations had a mean age of 63.3 years and almost 50% were ≥65 years of age. Female sex (82.2%) and Caucasian race (71.5%) were predominant. Most hospitalizations had Medicare (62.3%) as their primary insurance payer and the majority of the hospitalizations were emergent (83.3%). Distribution of other patient-, disease- and hospital-related characteristics are summarized in Table 1. Table 1 Baseline characteristics of hospitalized patients with diagnosis of SSc Characteristics Estimation or distribution P-valuea Overall SSc (sample n = 9731) Survived hospitalization (sample n = 9246) Died during hospitalization (sample n = 485) Age, mean( s.d.), years 63.3 (13.8) 63.1 (13.8) 66.0 (13.7) <0.001 Age category, %     18–44 years 9.4 9.5 7.8 0.19     45–64 years 41.4 41.7 35.1 0.003     ≥65 years 49.2 48.8 57.1 <0.001 Sex, male, % 17.8 17.7 19.8 0.24 Race, %     Caucasian/White 71.5 71.8 65.8 0.01     African-American 14.3 14.1 17.3 0.06     Othersb 14.2 14.1 16.9 0.13 Insurance type/status, %     Medicare 62.3 62.1 66.2 0.06     Medicaid 8.2 8.3 6.6 0.16     Private insurance 25.5 25.7 22.9 0.16     Self-pay 1.8 1.9 1.4 0.46     Uninsured 0.2 0.2 0.4 0.44     Otherc 2.0 2.0 2.5 0.49 Type of admission, %     Emergent admission 83.3 82.8 92.8 <0.001     Transfer statusd, %     Transferred in 8.7 8.2 17.4 <0.001 Hospital region, %     South 34.3 34.3 34.2 0.15     Northeast 22.1 22.0 24.7 0.001     Midwest 23.8 24.2 17.7 0.97     West 19.7 19.5 23.3 0.07 Hospital teaching status, %     Teaching 57.9 57.8 60.6 0.21 Hospital location, %     Urban 90.7 90.6 92.2 0.21     Small 12.5 12.5 12.4 0.93 Hospital bed size, %     Medium 24.6 24.5 26.2 0.44     Large 62.9 62.9 61.4 0.52 Household income quartile, %     Quartile 1( $1–38 999) 25.8 25.8 23.8 0.32     Quartile 2( $39 000–47 999) 24.5 24.7 19.8 0.009     Quartile 3( $48 000–62 999) 24.5 24.4 25.5 0.58     Quartile 4( $63 000+) 25.3 25.0 30.8 0.006 Number of comorbid conditions, median( IQR) 3 (2–4) 3 (2–4) 4 (2–5) Hypertension 55.4 55.7 Fluid and electrolytes disorders (55.3) – Fluid and electrolytes disorders 33.7 32.6 Hypertension (50.1) – Top 5 comorbid conditionse, % Deficiency anaemias 28.7 28.6 Pulmonary circulation disorders (36.9) – Hypothyroidism 24.1 24.3 Deficiency anemias (30.7) – Chronic pulmonary disease 23.3 23.4 Congestive heart failure (29.5) – Number of diagnoses codes mentioned, median (IQR) 14 (10–18) 14 (9–18) 18 (15–24) <0.001 Number of procedure codes mentioned, median (IQR) 1 (0–3) 1 (0–3) 3 (1–6) <0.001 Relevant clinical conditions, %     Oesophageal dysfunction 47.7 48.0 41.7 0.005     Hypertension 57.8 58.2 51.1 0.002     Overall infections 32.7 31.0 63.9 <0.001         Opportunistic infections 0.9 0.9 0.6 0.48         Other infections 31.8 30.2 63.3 <0.001     Pulmonary arterial hypertension 26.8 25.9 43.3 <0.001     Renal failure (acute and unspecified) 26.5 25.1 53.2 <0.001     Digital ulcer/gangrene/Raynaud’s 25.6 25.7 23.5 0.29     Congestive heart failure 23.7 23 38.4 <0.001     Coronary artery disease 22.2 21.9 27.8 0.004     Chronic kidney disease 18.3 17.9 26.8 <0.001     Diabetes mellitus 17.9 18.1 13.8 0.008     Respiratory failure 17.3 17.3 72.4 <0.001     Pulmonary fibrosis/interstitial lung disease 15.9 15.3 27.4 <0.001     Cachexia/weight loss/FTT 8.7 8.1 20 <0.001     PEMC 8.4 8 16.1 <0.001     Arrhythmia 4.7 4.5 8.7 0.001     Acute bowel obstruction 4.2 4.2 4.5 0.71     Aspiration 4.2 3.6 15.1 <0.001     Stroke/transient ischaemic attack 4 3.9 6 0.08     Liver disease 2.7 2.7 4.1 0.12     Myositis 0.8 0.8 0.4 0.21 Characteristics Estimation or distribution P-valuea Overall SSc (sample n = 9731) Survived hospitalization (sample n = 9246) Died during hospitalization (sample n = 485) Age, mean( s.d.), years 63.3 (13.8) 63.1 (13.8) 66.0 (13.7) <0.001 Age category, %     18–44 years 9.4 9.5 7.8 0.19     45–64 years 41.4 41.7 35.1 0.003     ≥65 years 49.2 48.8 57.1 <0.001 Sex, male, % 17.8 17.7 19.8 0.24 Race, %     Caucasian/White 71.5 71.8 65.8 0.01     African-American 14.3 14.1 17.3 0.06     Othersb 14.2 14.1 16.9 0.13 Insurance type/status, %     Medicare 62.3 62.1 66.2 0.06     Medicaid 8.2 8.3 6.6 0.16     Private insurance 25.5 25.7 22.9 0.16     Self-pay 1.8 1.9 1.4 0.46     Uninsured 0.2 0.2 0.4 0.44     Otherc 2.0 2.0 2.5 0.49 Type of admission, %     Emergent admission 83.3 82.8 92.8 <0.001     Transfer statusd, %     Transferred in 8.7 8.2 17.4 <0.001 Hospital region, %     South 34.3 34.3 34.2 0.15     Northeast 22.1 22.0 24.7 0.001     Midwest 23.8 24.2 17.7 0.97     West 19.7 19.5 23.3 0.07 Hospital teaching status, %     Teaching 57.9 57.8 60.6 0.21 Hospital location, %     Urban 90.7 90.6 92.2 0.21     Small 12.5 12.5 12.4 0.93 Hospital bed size, %     Medium 24.6 24.5 26.2 0.44     Large 62.9 62.9 61.4 0.52 Household income quartile, %     Quartile 1( $1–38 999) 25.8 25.8 23.8 0.32     Quartile 2( $39 000–47 999) 24.5 24.7 19.8 0.009     Quartile 3( $48 000–62 999) 24.5 24.4 25.5 0.58     Quartile 4( $63 000+) 25.3 25.0 30.8 0.006 Number of comorbid conditions, median( IQR) 3 (2–4) 3 (2–4) 4 (2–5) Hypertension 55.4 55.7 Fluid and electrolytes disorders (55.3) – Fluid and electrolytes disorders 33.7 32.6 Hypertension (50.1) – Top 5 comorbid conditionse, % Deficiency anaemias 28.7 28.6 Pulmonary circulation disorders (36.9) – Hypothyroidism 24.1 24.3 Deficiency anemias (30.7) – Chronic pulmonary disease 23.3 23.4 Congestive heart failure (29.5) – Number of diagnoses codes mentioned, median (IQR) 14 (10–18) 14 (9–18) 18 (15–24) <0.001 Number of procedure codes mentioned, median (IQR) 1 (0–3) 1 (0–3) 3 (1–6) <0.001 Relevant clinical conditions, %     Oesophageal dysfunction 47.7 48.0 41.7 0.005     Hypertension 57.8 58.2 51.1 0.002     Overall infections 32.7 31.0 63.9 <0.001         Opportunistic infections 0.9 0.9 0.6 0.48         Other infections 31.8 30.2 63.3 <0.001     Pulmonary arterial hypertension 26.8 25.9 43.3 <0.001     Renal failure (acute and unspecified) 26.5 25.1 53.2 <0.001     Digital ulcer/gangrene/Raynaud’s 25.6 25.7 23.5 0.29     Congestive heart failure 23.7 23 38.4 <0.001     Coronary artery disease 22.2 21.9 27.8 0.004     Chronic kidney disease 18.3 17.9 26.8 <0.001     Diabetes mellitus 17.9 18.1 13.8 0.008     Respiratory failure 17.3 17.3 72.4 <0.001     Pulmonary fibrosis/interstitial lung disease 15.9 15.3 27.4 <0.001     Cachexia/weight loss/FTT 8.7 8.1 20 <0.001     PEMC 8.4 8 16.1 <0.001     Arrhythmia 4.7 4.5 8.7 0.001     Acute bowel obstruction 4.2 4.2 4.5 0.71     Aspiration 4.2 3.6 15.1 <0.001     Stroke/transient ischaemic attack 4 3.9 6 0.08     Liver disease 2.7 2.7 4.1 0.12     Myositis 0.8 0.8 0.4 0.21 a P-value represents the difference between those who survived and died; statistically significant values are in bold. b Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. c Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programs. d Transfer from other emergency, hospital, office or nursing home. e Comorbid conditions are counted as defined by Elixhauser Comorbidity Software algorithm provided by Healthcare Cost and Utilization Project. IQR: interquartile range; FTT: failure to thrive; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy. Table 1 Baseline characteristics of hospitalized patients with diagnosis of SSc Characteristics Estimation or distribution P-valuea Overall SSc (sample n = 9731) Survived hospitalization (sample n = 9246) Died during hospitalization (sample n = 485) Age, mean( s.d.), years 63.3 (13.8) 63.1 (13.8) 66.0 (13.7) <0.001 Age category, %     18–44 years 9.4 9.5 7.8 0.19     45–64 years 41.4 41.7 35.1 0.003     ≥65 years 49.2 48.8 57.1 <0.001 Sex, male, % 17.8 17.7 19.8 0.24 Race, %     Caucasian/White 71.5 71.8 65.8 0.01     African-American 14.3 14.1 17.3 0.06     Othersb 14.2 14.1 16.9 0.13 Insurance type/status, %     Medicare 62.3 62.1 66.2 0.06     Medicaid 8.2 8.3 6.6 0.16     Private insurance 25.5 25.7 22.9 0.16     Self-pay 1.8 1.9 1.4 0.46     Uninsured 0.2 0.2 0.4 0.44     Otherc 2.0 2.0 2.5 0.49 Type of admission, %     Emergent admission 83.3 82.8 92.8 <0.001     Transfer statusd, %     Transferred in 8.7 8.2 17.4 <0.001 Hospital region, %     South 34.3 34.3 34.2 0.15     Northeast 22.1 22.0 24.7 0.001     Midwest 23.8 24.2 17.7 0.97     West 19.7 19.5 23.3 0.07 Hospital teaching status, %     Teaching 57.9 57.8 60.6 0.21 Hospital location, %     Urban 90.7 90.6 92.2 0.21     Small 12.5 12.5 12.4 0.93 Hospital bed size, %     Medium 24.6 24.5 26.2 0.44     Large 62.9 62.9 61.4 0.52 Household income quartile, %     Quartile 1( $1–38 999) 25.8 25.8 23.8 0.32     Quartile 2( $39 000–47 999) 24.5 24.7 19.8 0.009     Quartile 3( $48 000–62 999) 24.5 24.4 25.5 0.58     Quartile 4( $63 000+) 25.3 25.0 30.8 0.006 Number of comorbid conditions, median( IQR) 3 (2–4) 3 (2–4) 4 (2–5) Hypertension 55.4 55.7 Fluid and electrolytes disorders (55.3) – Fluid and electrolytes disorders 33.7 32.6 Hypertension (50.1) – Top 5 comorbid conditionse, % Deficiency anaemias 28.7 28.6 Pulmonary circulation disorders (36.9) – Hypothyroidism 24.1 24.3 Deficiency anemias (30.7) – Chronic pulmonary disease 23.3 23.4 Congestive heart failure (29.5) – Number of diagnoses codes mentioned, median (IQR) 14 (10–18) 14 (9–18) 18 (15–24) <0.001 Number of procedure codes mentioned, median (IQR) 1 (0–3) 1 (0–3) 3 (1–6) <0.001 Relevant clinical conditions, %     Oesophageal dysfunction 47.7 48.0 41.7 0.005     Hypertension 57.8 58.2 51.1 0.002     Overall infections 32.7 31.0 63.9 <0.001         Opportunistic infections 0.9 0.9 0.6 0.48         Other infections 31.8 30.2 63.3 <0.001     Pulmonary arterial hypertension 26.8 25.9 43.3 <0.001     Renal failure (acute and unspecified) 26.5 25.1 53.2 <0.001     Digital ulcer/gangrene/Raynaud’s 25.6 25.7 23.5 0.29     Congestive heart failure 23.7 23 38.4 <0.001     Coronary artery disease 22.2 21.9 27.8 0.004     Chronic kidney disease 18.3 17.9 26.8 <0.001     Diabetes mellitus 17.9 18.1 13.8 0.008     Respiratory failure 17.3 17.3 72.4 <0.001     Pulmonary fibrosis/interstitial lung disease 15.9 15.3 27.4 <0.001     Cachexia/weight loss/FTT 8.7 8.1 20 <0.001     PEMC 8.4 8 16.1 <0.001     Arrhythmia 4.7 4.5 8.7 0.001     Acute bowel obstruction 4.2 4.2 4.5 0.71     Aspiration 4.2 3.6 15.1 <0.001     Stroke/transient ischaemic attack 4 3.9 6 0.08     Liver disease 2.7 2.7 4.1 0.12     Myositis 0.8 0.8 0.4 0.21 Characteristics Estimation or distribution P-valuea Overall SSc (sample n = 9731) Survived hospitalization (sample n = 9246) Died during hospitalization (sample n = 485) Age, mean( s.d.), years 63.3 (13.8) 63.1 (13.8) 66.0 (13.7) <0.001 Age category, %     18–44 years 9.4 9.5 7.8 0.19     45–64 years 41.4 41.7 35.1 0.003     ≥65 years 49.2 48.8 57.1 <0.001 Sex, male, % 17.8 17.7 19.8 0.24 Race, %     Caucasian/White 71.5 71.8 65.8 0.01     African-American 14.3 14.1 17.3 0.06     Othersb 14.2 14.1 16.9 0.13 Insurance type/status, %     Medicare 62.3 62.1 66.2 0.06     Medicaid 8.2 8.3 6.6 0.16     Private insurance 25.5 25.7 22.9 0.16     Self-pay 1.8 1.9 1.4 0.46     Uninsured 0.2 0.2 0.4 0.44     Otherc 2.0 2.0 2.5 0.49 Type of admission, %     Emergent admission 83.3 82.8 92.8 <0.001     Transfer statusd, %     Transferred in 8.7 8.2 17.4 <0.001 Hospital region, %     South 34.3 34.3 34.2 0.15     Northeast 22.1 22.0 24.7 0.001     Midwest 23.8 24.2 17.7 0.97     West 19.7 19.5 23.3 0.07 Hospital teaching status, %     Teaching 57.9 57.8 60.6 0.21 Hospital location, %     Urban 90.7 90.6 92.2 0.21     Small 12.5 12.5 12.4 0.93 Hospital bed size, %     Medium 24.6 24.5 26.2 0.44     Large 62.9 62.9 61.4 0.52 Household income quartile, %     Quartile 1( $1–38 999) 25.8 25.8 23.8 0.32     Quartile 2( $39 000–47 999) 24.5 24.7 19.8 0.009     Quartile 3( $48 000–62 999) 24.5 24.4 25.5 0.58     Quartile 4( $63 000+) 25.3 25.0 30.8 0.006 Number of comorbid conditions, median( IQR) 3 (2–4) 3 (2–4) 4 (2–5) Hypertension 55.4 55.7 Fluid and electrolytes disorders (55.3) – Fluid and electrolytes disorders 33.7 32.6 Hypertension (50.1) – Top 5 comorbid conditionse, % Deficiency anaemias 28.7 28.6 Pulmonary circulation disorders (36.9) – Hypothyroidism 24.1 24.3 Deficiency anemias (30.7) – Chronic pulmonary disease 23.3 23.4 Congestive heart failure (29.5) – Number of diagnoses codes mentioned, median (IQR) 14 (10–18) 14 (9–18) 18 (15–24) <0.001 Number of procedure codes mentioned, median (IQR) 1 (0–3) 1 (0–3) 3 (1–6) <0.001 Relevant clinical conditions, %     Oesophageal dysfunction 47.7 48.0 41.7 0.005     Hypertension 57.8 58.2 51.1 0.002     Overall infections 32.7 31.0 63.9 <0.001         Opportunistic infections 0.9 0.9 0.6 0.48         Other infections 31.8 30.2 63.3 <0.001     Pulmonary arterial hypertension 26.8 25.9 43.3 <0.001     Renal failure (acute and unspecified) 26.5 25.1 53.2 <0.001     Digital ulcer/gangrene/Raynaud’s 25.6 25.7 23.5 0.29     Congestive heart failure 23.7 23 38.4 <0.001     Coronary artery disease 22.2 21.9 27.8 0.004     Chronic kidney disease 18.3 17.9 26.8 <0.001     Diabetes mellitus 17.9 18.1 13.8 0.008     Respiratory failure 17.3 17.3 72.4 <0.001     Pulmonary fibrosis/interstitial lung disease 15.9 15.3 27.4 <0.001     Cachexia/weight loss/FTT 8.7 8.1 20 <0.001     PEMC 8.4 8 16.1 <0.001     Arrhythmia 4.7 4.5 8.7 0.001     Acute bowel obstruction 4.2 4.2 4.5 0.71     Aspiration 4.2 3.6 15.1 <0.001     Stroke/transient ischaemic attack 4 3.9 6 0.08     Liver disease 2.7 2.7 4.1 0.12     Myositis 0.8 0.8 0.4 0.21 a P-value represents the difference between those who survived and died; statistically significant values are in bold. b Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. c Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programs. d Transfer from other emergency, hospital, office or nursing home. e Comorbid conditions are counted as defined by Elixhauser Comorbidity Software algorithm provided by Healthcare Cost and Utilization Project. IQR: interquartile range; FTT: failure to thrive; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy. Principal diagnoses associated with SSc hospitalizations in patients with and without inpatient mortality The most common reasons for hospitalization based on primary diagnosis codes (collapsed into categories) were infection/septicaemia (17.4%) followed by involvement of cardiovascular (15.9%), gastrointestinal (13.3%), musculoskeletal (includes CTDs) (12.2%), pulmonary (10.9%) and nervous (3.8%) systems among the SSc patients. Among SSc patients who died in the hospital, infection/septicaemia remained the most common primary diagnosis (32.7%) followed by pulmonary (20.0%), cardiovascular (15.7%), musculoskeletal (12.0%) and gastrointestinal (4.8%) system involvement (Fig. 1). When the primary diagnoses are not collapsed into categories, the most common reasons for hospitalization were CTD (includes CTDs other than RA and SLE) (6.35%) followed by respiratory infections (5.96%) and septicaemia (5.44%), while the most common primary diagnoses among SSc patients who died in hospital were septicaemia (23.35%) followed by respiratory failure/insufficiency/arrest (11.98%) followed by CTDs (10.33%) (supplementary Tables S3 and S4, available at Rheumatology online). Fig. 1 View largeDownload slide Primary diagnosis category of SSc patients: overall and among those who died Fig. 1 View largeDownload slide Primary diagnosis category of SSc patients: overall and among those who died Outcomes: in-hospital mortality, hospital LOS and cost of hospitalization Table 2 summarizes the above outcomes among overall hospitalizations by age, race and sex subgroups. Overall mortality was 5%. Mortality was 5.8% among older patients ≥65 years, 5.6% among males and 6.1% among African-Americans. A lower mortality was found among Caucasians (4.6%, P < 0.05). Median LOS was similar across all age, race and sex categories [4 days, interquartile range (2–7)] other than slightly longer LOS among African-Americans [5 days (3–8)]. Cost of hospitalization was also similar in the different subgroups [overall median (interquartile range) cost $8993, (5259–16 420)]. Overall cost of SSc hospitalizations for the years 2012 and 2013 were $357 million and $361 million, respectively. Table 2 Hospitalization proportion, mortality rate, length of stay and hospitalization cost by age category, race and sex Hospitalization composition Total SSc hospitalizations Age group (years) Sexa Race 18–44 45–64 ≥65 Male Female Caucasians AA Othersb Hospitalization counts, n (N = weighted counts) 9731 (48 655) 916 (4580) 4030 (20 150) 4785 (23 925) 1729 (8645) 8001 (40 004) 6586 (32 930) 1313 (65 650) 1312 (6560) Hospitalization percentage 0.093 0.048 0.11 0.095 0.035 0.14 0.095 0.093 0.094 Race, % of hospitalization (% race)     Caucasian/White 0.095 (71.5) 0.042 (50.1) 0.113 (65.1) 0.1 (81.1) 0.035 (69.2) 0.151 (72.0) – – –     AA 0.093 (14.3) 0.068 (28.6) 0.126 (19.3) 0.07 (7.1) 0.04 (15.8) 0.14 (14.0) – – –     Others 0.094 (14.2) 0.051 (21.3) 0.123 (15.5) 0.10 (11.8) 0.04 (14.9) 0.151 (14.0) – – –     Mortality rate, % 5.0 4.1 4.2 5.8 5.6 4.9 4.62 6.10 5.95     LOS, median (IQR), days 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 5 (3–8) 4 (2–7)     Cost per hospitalization, median (IQR), $ 8885 (5169– 15 921) 8567 (4883– 15 629) 9341 (5363– 17 097) 8791 (5236– 15 969) 9435 (5552– 17 619) 8916 (5201– 16 501) 8842 (5194– 5933) 8932 (5165– 16 514) 9839 (5634– 18 281) Hospitalization composition Total SSc hospitalizations Age group (years) Sexa Race 18–44 45–64 ≥65 Male Female Caucasians AA Othersb Hospitalization counts, n (N = weighted counts) 9731 (48 655) 916 (4580) 4030 (20 150) 4785 (23 925) 1729 (8645) 8001 (40 004) 6586 (32 930) 1313 (65 650) 1312 (6560) Hospitalization percentage 0.093 0.048 0.11 0.095 0.035 0.14 0.095 0.093 0.094 Race, % of hospitalization (% race)     Caucasian/White 0.095 (71.5) 0.042 (50.1) 0.113 (65.1) 0.1 (81.1) 0.035 (69.2) 0.151 (72.0) – – –     AA 0.093 (14.3) 0.068 (28.6) 0.126 (19.3) 0.07 (7.1) 0.04 (15.8) 0.14 (14.0) – – –     Others 0.094 (14.2) 0.051 (21.3) 0.123 (15.5) 0.10 (11.8) 0.04 (14.9) 0.151 (14.0) – – –     Mortality rate, % 5.0 4.1 4.2 5.8 5.6 4.9 4.62 6.10 5.95     LOS, median (IQR), days 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 5 (3–8) 4 (2–7)     Cost per hospitalization, median (IQR), $ 8885 (5169– 15 921) 8567 (4883– 15 629) 9341 (5363– 17 097) 8791 (5236– 15 969) 9435 (5552– 17 619) 8916 (5201– 16 501) 8842 (5194– 5933) 8932 (5165– 16 514) 9839 (5634– 18 281) a One observation in the sample had missing data for sex. b Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. AA: African American; LOS: length of stay; IQR: interquartile range. Table 2 Hospitalization proportion, mortality rate, length of stay and hospitalization cost by age category, race and sex Hospitalization composition Total SSc hospitalizations Age group (years) Sexa Race 18–44 45–64 ≥65 Male Female Caucasians AA Othersb Hospitalization counts, n (N = weighted counts) 9731 (48 655) 916 (4580) 4030 (20 150) 4785 (23 925) 1729 (8645) 8001 (40 004) 6586 (32 930) 1313 (65 650) 1312 (6560) Hospitalization percentage 0.093 0.048 0.11 0.095 0.035 0.14 0.095 0.093 0.094 Race, % of hospitalization (% race)     Caucasian/White 0.095 (71.5) 0.042 (50.1) 0.113 (65.1) 0.1 (81.1) 0.035 (69.2) 0.151 (72.0) – – –     AA 0.093 (14.3) 0.068 (28.6) 0.126 (19.3) 0.07 (7.1) 0.04 (15.8) 0.14 (14.0) – – –     Others 0.094 (14.2) 0.051 (21.3) 0.123 (15.5) 0.10 (11.8) 0.04 (14.9) 0.151 (14.0) – – –     Mortality rate, % 5.0 4.1 4.2 5.8 5.6 4.9 4.62 6.10 5.95     LOS, median (IQR), days 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 5 (3–8) 4 (2–7)     Cost per hospitalization, median (IQR), $ 8885 (5169– 15 921) 8567 (4883– 15 629) 9341 (5363– 17 097) 8791 (5236– 15 969) 9435 (5552– 17 619) 8916 (5201– 16 501) 8842 (5194– 5933) 8932 (5165– 16 514) 9839 (5634– 18 281) Hospitalization composition Total SSc hospitalizations Age group (years) Sexa Race 18–44 45–64 ≥65 Male Female Caucasians AA Othersb Hospitalization counts, n (N = weighted counts) 9731 (48 655) 916 (4580) 4030 (20 150) 4785 (23 925) 1729 (8645) 8001 (40 004) 6586 (32 930) 1313 (65 650) 1312 (6560) Hospitalization percentage 0.093 0.048 0.11 0.095 0.035 0.14 0.095 0.093 0.094 Race, % of hospitalization (% race)     Caucasian/White 0.095 (71.5) 0.042 (50.1) 0.113 (65.1) 0.1 (81.1) 0.035 (69.2) 0.151 (72.0) – – –     AA 0.093 (14.3) 0.068 (28.6) 0.126 (19.3) 0.07 (7.1) 0.04 (15.8) 0.14 (14.0) – – –     Others 0.094 (14.2) 0.051 (21.3) 0.123 (15.5) 0.10 (11.8) 0.04 (14.9) 0.151 (14.0) – – –     Mortality rate, % 5.0 4.1 4.2 5.8 5.6 4.9 4.62 6.10 5.95     LOS, median (IQR), days 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 4 (2–7) 5 (3–8) 4 (2–7)     Cost per hospitalization, median (IQR), $ 8885 (5169– 15 921) 8567 (4883– 15 629) 9341 (5363– 17 097) 8791 (5236– 15 969) 9435 (5552– 17 619) 8916 (5201– 16 501) 8842 (5194– 5933) 8932 (5165– 16 514) 9839 (5634– 18 281) a One observation in the sample had missing data for sex. b Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. AA: African American; LOS: length of stay; IQR: interquartile range. Factors associated with in-hospital mortality, longer hospital LOS and higher hospitalization cost Tables 3–5 summarize the findings of univariate and backward stepwise multivariate logistic regressions to identify factors associated with higher likelihood of inpatient mortality, longer hospital LOS and higher hospitalization cost, respectively. Age and sex categories were not statistically associated with any of the outcomes studies. While African American race was associated with higher odds of mortality (aOR = 1.46, 95% CI: 1.08, 1.98, P = 0.01), races other than Caucasian and African American had higher cost of hospitalization (aOR = 1.53, 95% CI: 1.24, 1.89, P < 0.0001). Many of the relevant clinical factors such as pulmonary fibrosis, pulmonary arterial hypertension (PAH), acute renal failure, pericarditis/endocarditis/myocarditis/cardiomyopathy and aspiration were positively associated with all of the outcomes. Of these, acute renal failure (although it might not always be a result of SSc process) had the strongest association with inpatient mortality (aOR = 4.31, 95% CI: 3.32, 5.60) and aspiration had strongest association with both longer LOS (aOR = 2.88, 95% CI: 2.20, 3.77, P < 0.05) and higher cost (aOR = 2.29, 95% CI: 1.72, 3.04, P < 0.0001). Acute bowel obstruction, on the other hand, was associated with longer LOS only (aOR = 2.66, 95% CI: 1.98, 3.57). Interestingly, diabetes mellitus (aOR = 0.62, 95% CI: 0.46, 0.84), hypertension (aOR = 0.71, 95% CI: 0.57, 0.89) and oesophageal dysfunction (aOR = 0.72, 95% CI: 0.59, 0.89) were significantly (P < 0.05) associated with lower odds of inpatient mortality in both univariate as well as multivariate analyses. Diagnosis of chronic kidney disease, however, had higher odds of mortality during univariate analysis (aOR = 1.7, 95% CI: 1.39, 2.08, P < 0.001) but lower in multivariate analysis (aOR = 0.52, 95% CI: 0.39, 0.69, P < 0.0001). Patients who died had significantly higher proportions of PAH (43.3%), respiratory failure (72.4%) and congestive heart failure (CHF) (38.4%) (Table 1). Table 3 Factors associated with in-hospital mortality among patients with systemic sclerosis: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.01 (0.7, 1.46) 0.95 1.10 (0.73, 1.66) 0.64     ≥65 years 1.42 (0.99, 2.02) 0.054 1.35 (0.89, 2.06) 0.16 Male sex 1.16 (0.92, 1.45) 0.21 1.07 (0.81, 1.4) 0.64 Race (ref. Caucasian)     African-American 1.33 (1.04, 1.7) 0.02 1.46 (1.08, 1.98) 0.01     Othersa 1.30 (0.99, 1.69) 0.06 1.20 (0.89, 1.61) 0.24     Transfer status (ref. not transferred in) 2.41 <0.0001 2.03 (1.52, 2.71) <0.0001 Primary payer status (ref. Medicare)     Medicaid 0.75 (0.52, 1.1) 0.14 – –     Private insurance 0.84 (0.67, 1.05) 0.12 – –     Self-pay 0.73 (0.34, 1.58) 0.43 – –     No charge 2.04 (0.46, 9.0) 0.34 – –     Otherb 1.19 (0.66, 2.17) 0.57 – – Income quartile (ref. quartile 1)     Quartile 2 0.87 (0.66, 1.15) 0.33 0.86 (0.63, 1.18) 0.36     Quartile 3 1.14 (0.87, 1.48) 0.34 1.10 (0.82, 1.49) 0.52     Quartile 4 1.35 (1.05, 1.73) 0.02 1.41 (1.07, 1.87) 0.02 Hospital region (ref. south)     Northeast 1.14 (0.9, 1.44) 0.29 – –     Midwest 0.74 (0.56, 0.98) 0.03 – –     West 1.20 (0.92, 1.56) 0.17 – – Urban hospital (ref. rural) 1.18 (0.84, 1.66) 0.33 – – Teaching hospital (ref. non-teaching) 1.10 (0.91, 1.33) 0.31 – – Hospital size (ref. small)     Medium 1.06 (0.77, 1.47) 0.71 – –     Large 0.96 (0.72, 1.28) 0.79 – – Relevant clinical conditions     Aspiration 4.92 (3.73, 6.51) <0.0001 3.52 (2.51, 4.94) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.33 (0.42, 4.21) 0.62 0.91 (0.22, 3.76) 0.90         Other infections 4.07 (3.35, 4.94) <0.0001 3.36 (2.73, 4.14) <0.0001     Acute renal failure 3.47 (2.88, 4.17) <0.0001 4.31 (3.32, 5.6) <0.0001     Pulmonary fibrosis 2.95 (1.61, 5.42) <0.0001 2.23 (1.15, 4.3) 0.02     Weight loss/cachexia 2.88 (2.27, 3.66) <0.0001 2.31 (1.74, 3.05) <0.0001     PEMC 2.26 (1.75, 2.91) <0.0001 1.6 (1.19, 2.16) <0.0001     Pulmonary arterial hypertension 2.22 (1.85, 2.66) <0.0001 1.82 (1.47, 2.25) <0.0001     Congestive heart failure 2.12 (1.76, 2.55) <0.0001 – –     Arrhythmia 2.05 (1.48, 2.84) <0.0001 1.53 (1.04, 2.25) 0.03     Chronic kidney disease 1.70 (1.39, 2.08) <0.0001 0.52 (0.39, 0.69) <0.0001     Chronic liver disease 1.60 (1, 2.58) 0.05 – –     Stroke/TIA 1.56 (1.04, 2.35) 0.03 1.86 (1.17, 2.95) 0.01     Coronary artery disease 1.38 (1.12, 1.7) <0.0001 1.3 (1.02, 1.66) 0.03     Acute bowel obstruction 1.12 (0.72, 1.73) 0.62 – –     Raynaud’s/ulcer/gangrene 0.89 (0.71, 1.11) 0.29 – –     Oesophageal dysfunction 0.77 (0.64, 0.92) <0.0001 0.72 (0.59, 0.89) <0.0001     Hypertension 0.75 (0.63, 0.9) <0.0001 0.71 (0.57, 0.89) <0.0001     Diabetes mellitus 0.72 (0.56, 0.94) 0.02 0.62 (0.46, 0.84) <0.0001     Myositis 0.53 (0.13, 2.2) 0.39 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.01 (0.7, 1.46) 0.95 1.10 (0.73, 1.66) 0.64     ≥65 years 1.42 (0.99, 2.02) 0.054 1.35 (0.89, 2.06) 0.16 Male sex 1.16 (0.92, 1.45) 0.21 1.07 (0.81, 1.4) 0.64 Race (ref. Caucasian)     African-American 1.33 (1.04, 1.7) 0.02 1.46 (1.08, 1.98) 0.01     Othersa 1.30 (0.99, 1.69) 0.06 1.20 (0.89, 1.61) 0.24     Transfer status (ref. not transferred in) 2.41 <0.0001 2.03 (1.52, 2.71) <0.0001 Primary payer status (ref. Medicare)     Medicaid 0.75 (0.52, 1.1) 0.14 – –     Private insurance 0.84 (0.67, 1.05) 0.12 – –     Self-pay 0.73 (0.34, 1.58) 0.43 – –     No charge 2.04 (0.46, 9.0) 0.34 – –     Otherb 1.19 (0.66, 2.17) 0.57 – – Income quartile (ref. quartile 1)     Quartile 2 0.87 (0.66, 1.15) 0.33 0.86 (0.63, 1.18) 0.36     Quartile 3 1.14 (0.87, 1.48) 0.34 1.10 (0.82, 1.49) 0.52     Quartile 4 1.35 (1.05, 1.73) 0.02 1.41 (1.07, 1.87) 0.02 Hospital region (ref. south)     Northeast 1.14 (0.9, 1.44) 0.29 – –     Midwest 0.74 (0.56, 0.98) 0.03 – –     West 1.20 (0.92, 1.56) 0.17 – – Urban hospital (ref. rural) 1.18 (0.84, 1.66) 0.33 – – Teaching hospital (ref. non-teaching) 1.10 (0.91, 1.33) 0.31 – – Hospital size (ref. small)     Medium 1.06 (0.77, 1.47) 0.71 – –     Large 0.96 (0.72, 1.28) 0.79 – – Relevant clinical conditions     Aspiration 4.92 (3.73, 6.51) <0.0001 3.52 (2.51, 4.94) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.33 (0.42, 4.21) 0.62 0.91 (0.22, 3.76) 0.90         Other infections 4.07 (3.35, 4.94) <0.0001 3.36 (2.73, 4.14) <0.0001     Acute renal failure 3.47 (2.88, 4.17) <0.0001 4.31 (3.32, 5.6) <0.0001     Pulmonary fibrosis 2.95 (1.61, 5.42) <0.0001 2.23 (1.15, 4.3) 0.02     Weight loss/cachexia 2.88 (2.27, 3.66) <0.0001 2.31 (1.74, 3.05) <0.0001     PEMC 2.26 (1.75, 2.91) <0.0001 1.6 (1.19, 2.16) <0.0001     Pulmonary arterial hypertension 2.22 (1.85, 2.66) <0.0001 1.82 (1.47, 2.25) <0.0001     Congestive heart failure 2.12 (1.76, 2.55) <0.0001 – –     Arrhythmia 2.05 (1.48, 2.84) <0.0001 1.53 (1.04, 2.25) 0.03     Chronic kidney disease 1.70 (1.39, 2.08) <0.0001 0.52 (0.39, 0.69) <0.0001     Chronic liver disease 1.60 (1, 2.58) 0.05 – –     Stroke/TIA 1.56 (1.04, 2.35) 0.03 1.86 (1.17, 2.95) 0.01     Coronary artery disease 1.38 (1.12, 1.7) <0.0001 1.3 (1.02, 1.66) 0.03     Acute bowel obstruction 1.12 (0.72, 1.73) 0.62 – –     Raynaud’s/ulcer/gangrene 0.89 (0.71, 1.11) 0.29 – –     Oesophageal dysfunction 0.77 (0.64, 0.92) <0.0001 0.72 (0.59, 0.89) <0.0001     Hypertension 0.75 (0.63, 0.9) <0.0001 0.71 (0.57, 0.89) <0.0001     Diabetes mellitus 0.72 (0.56, 0.94) 0.02 0.62 (0.46, 0.84) <0.0001     Myositis 0.53 (0.13, 2.2) 0.39 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker's Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programmes. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; TIA: transient ischaemic attack. Table 3 Factors associated with in-hospital mortality among patients with systemic sclerosis: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.01 (0.7, 1.46) 0.95 1.10 (0.73, 1.66) 0.64     ≥65 years 1.42 (0.99, 2.02) 0.054 1.35 (0.89, 2.06) 0.16 Male sex 1.16 (0.92, 1.45) 0.21 1.07 (0.81, 1.4) 0.64 Race (ref. Caucasian)     African-American 1.33 (1.04, 1.7) 0.02 1.46 (1.08, 1.98) 0.01     Othersa 1.30 (0.99, 1.69) 0.06 1.20 (0.89, 1.61) 0.24     Transfer status (ref. not transferred in) 2.41 <0.0001 2.03 (1.52, 2.71) <0.0001 Primary payer status (ref. Medicare)     Medicaid 0.75 (0.52, 1.1) 0.14 – –     Private insurance 0.84 (0.67, 1.05) 0.12 – –     Self-pay 0.73 (0.34, 1.58) 0.43 – –     No charge 2.04 (0.46, 9.0) 0.34 – –     Otherb 1.19 (0.66, 2.17) 0.57 – – Income quartile (ref. quartile 1)     Quartile 2 0.87 (0.66, 1.15) 0.33 0.86 (0.63, 1.18) 0.36     Quartile 3 1.14 (0.87, 1.48) 0.34 1.10 (0.82, 1.49) 0.52     Quartile 4 1.35 (1.05, 1.73) 0.02 1.41 (1.07, 1.87) 0.02 Hospital region (ref. south)     Northeast 1.14 (0.9, 1.44) 0.29 – –     Midwest 0.74 (0.56, 0.98) 0.03 – –     West 1.20 (0.92, 1.56) 0.17 – – Urban hospital (ref. rural) 1.18 (0.84, 1.66) 0.33 – – Teaching hospital (ref. non-teaching) 1.10 (0.91, 1.33) 0.31 – – Hospital size (ref. small)     Medium 1.06 (0.77, 1.47) 0.71 – –     Large 0.96 (0.72, 1.28) 0.79 – – Relevant clinical conditions     Aspiration 4.92 (3.73, 6.51) <0.0001 3.52 (2.51, 4.94) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.33 (0.42, 4.21) 0.62 0.91 (0.22, 3.76) 0.90         Other infections 4.07 (3.35, 4.94) <0.0001 3.36 (2.73, 4.14) <0.0001     Acute renal failure 3.47 (2.88, 4.17) <0.0001 4.31 (3.32, 5.6) <0.0001     Pulmonary fibrosis 2.95 (1.61, 5.42) <0.0001 2.23 (1.15, 4.3) 0.02     Weight loss/cachexia 2.88 (2.27, 3.66) <0.0001 2.31 (1.74, 3.05) <0.0001     PEMC 2.26 (1.75, 2.91) <0.0001 1.6 (1.19, 2.16) <0.0001     Pulmonary arterial hypertension 2.22 (1.85, 2.66) <0.0001 1.82 (1.47, 2.25) <0.0001     Congestive heart failure 2.12 (1.76, 2.55) <0.0001 – –     Arrhythmia 2.05 (1.48, 2.84) <0.0001 1.53 (1.04, 2.25) 0.03     Chronic kidney disease 1.70 (1.39, 2.08) <0.0001 0.52 (0.39, 0.69) <0.0001     Chronic liver disease 1.60 (1, 2.58) 0.05 – –     Stroke/TIA 1.56 (1.04, 2.35) 0.03 1.86 (1.17, 2.95) 0.01     Coronary artery disease 1.38 (1.12, 1.7) <0.0001 1.3 (1.02, 1.66) 0.03     Acute bowel obstruction 1.12 (0.72, 1.73) 0.62 – –     Raynaud’s/ulcer/gangrene 0.89 (0.71, 1.11) 0.29 – –     Oesophageal dysfunction 0.77 (0.64, 0.92) <0.0001 0.72 (0.59, 0.89) <0.0001     Hypertension 0.75 (0.63, 0.9) <0.0001 0.71 (0.57, 0.89) <0.0001     Diabetes mellitus 0.72 (0.56, 0.94) 0.02 0.62 (0.46, 0.84) <0.0001     Myositis 0.53 (0.13, 2.2) 0.39 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.01 (0.7, 1.46) 0.95 1.10 (0.73, 1.66) 0.64     ≥65 years 1.42 (0.99, 2.02) 0.054 1.35 (0.89, 2.06) 0.16 Male sex 1.16 (0.92, 1.45) 0.21 1.07 (0.81, 1.4) 0.64 Race (ref. Caucasian)     African-American 1.33 (1.04, 1.7) 0.02 1.46 (1.08, 1.98) 0.01     Othersa 1.30 (0.99, 1.69) 0.06 1.20 (0.89, 1.61) 0.24     Transfer status (ref. not transferred in) 2.41 <0.0001 2.03 (1.52, 2.71) <0.0001 Primary payer status (ref. Medicare)     Medicaid 0.75 (0.52, 1.1) 0.14 – –     Private insurance 0.84 (0.67, 1.05) 0.12 – –     Self-pay 0.73 (0.34, 1.58) 0.43 – –     No charge 2.04 (0.46, 9.0) 0.34 – –     Otherb 1.19 (0.66, 2.17) 0.57 – – Income quartile (ref. quartile 1)     Quartile 2 0.87 (0.66, 1.15) 0.33 0.86 (0.63, 1.18) 0.36     Quartile 3 1.14 (0.87, 1.48) 0.34 1.10 (0.82, 1.49) 0.52     Quartile 4 1.35 (1.05, 1.73) 0.02 1.41 (1.07, 1.87) 0.02 Hospital region (ref. south)     Northeast 1.14 (0.9, 1.44) 0.29 – –     Midwest 0.74 (0.56, 0.98) 0.03 – –     West 1.20 (0.92, 1.56) 0.17 – – Urban hospital (ref. rural) 1.18 (0.84, 1.66) 0.33 – – Teaching hospital (ref. non-teaching) 1.10 (0.91, 1.33) 0.31 – – Hospital size (ref. small)     Medium 1.06 (0.77, 1.47) 0.71 – –     Large 0.96 (0.72, 1.28) 0.79 – – Relevant clinical conditions     Aspiration 4.92 (3.73, 6.51) <0.0001 3.52 (2.51, 4.94) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.33 (0.42, 4.21) 0.62 0.91 (0.22, 3.76) 0.90         Other infections 4.07 (3.35, 4.94) <0.0001 3.36 (2.73, 4.14) <0.0001     Acute renal failure 3.47 (2.88, 4.17) <0.0001 4.31 (3.32, 5.6) <0.0001     Pulmonary fibrosis 2.95 (1.61, 5.42) <0.0001 2.23 (1.15, 4.3) 0.02     Weight loss/cachexia 2.88 (2.27, 3.66) <0.0001 2.31 (1.74, 3.05) <0.0001     PEMC 2.26 (1.75, 2.91) <0.0001 1.6 (1.19, 2.16) <0.0001     Pulmonary arterial hypertension 2.22 (1.85, 2.66) <0.0001 1.82 (1.47, 2.25) <0.0001     Congestive heart failure 2.12 (1.76, 2.55) <0.0001 – –     Arrhythmia 2.05 (1.48, 2.84) <0.0001 1.53 (1.04, 2.25) 0.03     Chronic kidney disease 1.70 (1.39, 2.08) <0.0001 0.52 (0.39, 0.69) <0.0001     Chronic liver disease 1.60 (1, 2.58) 0.05 – –     Stroke/TIA 1.56 (1.04, 2.35) 0.03 1.86 (1.17, 2.95) 0.01     Coronary artery disease 1.38 (1.12, 1.7) <0.0001 1.3 (1.02, 1.66) 0.03     Acute bowel obstruction 1.12 (0.72, 1.73) 0.62 – –     Raynaud’s/ulcer/gangrene 0.89 (0.71, 1.11) 0.29 – –     Oesophageal dysfunction 0.77 (0.64, 0.92) <0.0001 0.72 (0.59, 0.89) <0.0001     Hypertension 0.75 (0.63, 0.9) <0.0001 0.71 (0.57, 0.89) <0.0001     Diabetes mellitus 0.72 (0.56, 0.94) 0.02 0.62 (0.46, 0.84) <0.0001     Myositis 0.53 (0.13, 2.2) 0.39 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker's Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programmes. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; TIA: transient ischaemic attack. Table 4 Factors associated with longer length of stay (>90th percentile) among patients with SSc: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.18 (0.91, 1.52) 0.21 1.23 (0.93, 1.61) 0.14     ≥65 years 0.92 (0.71, 1.19) 0.53 0.85 (0.64, 1.12) 0.25 Male sex 1.27 (1.07, 1.51) 0.01 1.11 (0.92, 1.35) 0.27 Race (ref. Caucasian)     African-American 1.33 (1.09, 1.62) 0.01 – –     Othersa 1.31 (1.07, 1.6) 0.01 – – Transfer status (ref. not transferred in) 2.55 (2.11, 3.09) <0.0001 2.09 (1.68, 2.59) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.31 (1.02, 1.69) 0.04 – –     Private insurance 1.02 (0.85, 1.21) 0.85 – –     Self-pay 1.17 (0.72, 1.9) 0.54 – –     No charge 2.68 (0.93, 7.72) 0.07 – –     Otherb 1.27 (0.8, 2.01) 0.30 – – Income quartile (ref. quartile 1)     Quartile 2 0.88 (0.72, 1.08) 0.22 – –     Quartile 3 1.07 (0.87, 1.31) 0.53 – –     Quartile 4 1.04 (0.85, 1.26) 0.70 – – Hospital region (ref. south)     Northeast 1.05 (0.86, 1.28) 0.62 – –     Midwest 0.82 (0.67, 1.02) 0.08 – –     West 0.92 (0.74, 1.13) 0.41 – – Urban hospital (ref. rural) 2.67 (1.88, 3.79) <0.0001 1.97 (1.36, 2.86) <0.0001 Teaching hospital (ref. non-teaching) 1.69 (1.45, 1.99) <0.0001 1.34 (1.12, 1.59) <0.0001 Hospital size (ref. small)     Medium 0.97 (0.73, 1.3) 0.85 – –     Large 1.32 (1.02, 1.7) 0.03 – – Relevant clinical conditions     Aspiration 3.79 (2.98, 4.81) <0.0001 2.88 (2.2, 3.77) <0.0001     Infection (ref. no infection)         Opportunistic infections 2.40 (1.27, 4.53) 0.01 2.21 (1.13, 4.33) 0.02         Other infections 3.11 (2.7, 3.59) <0.0001 2.8 (2.41, 3.25) <0.0001     Weight loss/cachexia 2.99 (2.48, 3.61) <0.0001 2.41 (1.95, 2.97) <0.0001     Myositis 2.61 (1.49, 4.58) <0.0001 2.17 (1.21, 3.9) 0.01     Pulmonary fibrosis 2.58 (1.55, 4.27) <0.0001 2.01 (1.14, 3.55) 0.02     Acute bowel obstruction 2.43 (1.87, 3.15) <0.0001 2.66 (1.98, 3.57) <0.0001     Acute renal failure 2.33 (2.01, 2.71) <0.0001 2.24 (1.8, 2.78) <0.0001     Arrhythmia 2.15 (1.67, 2.77) <0.0001 1.62 (1.22, 2.16) <0.0001     PEMC 1.98 (1.61, 2.44) <0.0001 1.42 (1.13, 1.78) <0.0001     Congestive heart failure 1.81 (1.56, 2.1) <0.0001 1.44 (1.22, 1.71) <0.0001     Chronic kidney disease 1.58 (1.34, 1.88) <0.0001 0.73 (0.57, 0.95) 0.02     Pulmonary arterial hypertension 1.56 (1.34, 1.81) <0.0001 1.23 (1.04, 1.45) 0.02     Chronic liver disease 1.45 (1.02, 2.08) 0.04 – –     Diabetes mellitus 1.12 (0.94, 1.34) 0.21 – –     Raynaud’s/ulcer/gangrene 1.10 (0.93, 1.28) 0.26 – –     Hypertension 1.10 (0.96, 1.28) 0.17 – –     Coronary artery disease 1.00 (0.85, 1.19) 0.99 – –     Stroke/TIA 0.98 (0.69, 1.4) 0.9 – –     Oesophageal dysfunction 0.95 (0.83, 1.09) 0.44 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.18 (0.91, 1.52) 0.21 1.23 (0.93, 1.61) 0.14     ≥65 years 0.92 (0.71, 1.19) 0.53 0.85 (0.64, 1.12) 0.25 Male sex 1.27 (1.07, 1.51) 0.01 1.11 (0.92, 1.35) 0.27 Race (ref. Caucasian)     African-American 1.33 (1.09, 1.62) 0.01 – –     Othersa 1.31 (1.07, 1.6) 0.01 – – Transfer status (ref. not transferred in) 2.55 (2.11, 3.09) <0.0001 2.09 (1.68, 2.59) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.31 (1.02, 1.69) 0.04 – –     Private insurance 1.02 (0.85, 1.21) 0.85 – –     Self-pay 1.17 (0.72, 1.9) 0.54 – –     No charge 2.68 (0.93, 7.72) 0.07 – –     Otherb 1.27 (0.8, 2.01) 0.30 – – Income quartile (ref. quartile 1)     Quartile 2 0.88 (0.72, 1.08) 0.22 – –     Quartile 3 1.07 (0.87, 1.31) 0.53 – –     Quartile 4 1.04 (0.85, 1.26) 0.70 – – Hospital region (ref. south)     Northeast 1.05 (0.86, 1.28) 0.62 – –     Midwest 0.82 (0.67, 1.02) 0.08 – –     West 0.92 (0.74, 1.13) 0.41 – – Urban hospital (ref. rural) 2.67 (1.88, 3.79) <0.0001 1.97 (1.36, 2.86) <0.0001 Teaching hospital (ref. non-teaching) 1.69 (1.45, 1.99) <0.0001 1.34 (1.12, 1.59) <0.0001 Hospital size (ref. small)     Medium 0.97 (0.73, 1.3) 0.85 – –     Large 1.32 (1.02, 1.7) 0.03 – – Relevant clinical conditions     Aspiration 3.79 (2.98, 4.81) <0.0001 2.88 (2.2, 3.77) <0.0001     Infection (ref. no infection)         Opportunistic infections 2.40 (1.27, 4.53) 0.01 2.21 (1.13, 4.33) 0.02         Other infections 3.11 (2.7, 3.59) <0.0001 2.8 (2.41, 3.25) <0.0001     Weight loss/cachexia 2.99 (2.48, 3.61) <0.0001 2.41 (1.95, 2.97) <0.0001     Myositis 2.61 (1.49, 4.58) <0.0001 2.17 (1.21, 3.9) 0.01     Pulmonary fibrosis 2.58 (1.55, 4.27) <0.0001 2.01 (1.14, 3.55) 0.02     Acute bowel obstruction 2.43 (1.87, 3.15) <0.0001 2.66 (1.98, 3.57) <0.0001     Acute renal failure 2.33 (2.01, 2.71) <0.0001 2.24 (1.8, 2.78) <0.0001     Arrhythmia 2.15 (1.67, 2.77) <0.0001 1.62 (1.22, 2.16) <0.0001     PEMC 1.98 (1.61, 2.44) <0.0001 1.42 (1.13, 1.78) <0.0001     Congestive heart failure 1.81 (1.56, 2.1) <0.0001 1.44 (1.22, 1.71) <0.0001     Chronic kidney disease 1.58 (1.34, 1.88) <0.0001 0.73 (0.57, 0.95) 0.02     Pulmonary arterial hypertension 1.56 (1.34, 1.81) <0.0001 1.23 (1.04, 1.45) 0.02     Chronic liver disease 1.45 (1.02, 2.08) 0.04 – –     Diabetes mellitus 1.12 (0.94, 1.34) 0.21 – –     Raynaud’s/ulcer/gangrene 1.10 (0.93, 1.28) 0.26 – –     Hypertension 1.10 (0.96, 1.28) 0.17 – –     Coronary artery disease 1.00 (0.85, 1.19) 0.99 – –     Stroke/TIA 0.98 (0.69, 1.4) 0.9 – –     Oesophageal dysfunction 0.95 (0.83, 1.09) 0.44 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programmes. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; ref.: Reference group; TIA: transient ischaemic attack. Table 4 Factors associated with longer length of stay (>90th percentile) among patients with SSc: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.18 (0.91, 1.52) 0.21 1.23 (0.93, 1.61) 0.14     ≥65 years 0.92 (0.71, 1.19) 0.53 0.85 (0.64, 1.12) 0.25 Male sex 1.27 (1.07, 1.51) 0.01 1.11 (0.92, 1.35) 0.27 Race (ref. Caucasian)     African-American 1.33 (1.09, 1.62) 0.01 – –     Othersa 1.31 (1.07, 1.6) 0.01 – – Transfer status (ref. not transferred in) 2.55 (2.11, 3.09) <0.0001 2.09 (1.68, 2.59) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.31 (1.02, 1.69) 0.04 – –     Private insurance 1.02 (0.85, 1.21) 0.85 – –     Self-pay 1.17 (0.72, 1.9) 0.54 – –     No charge 2.68 (0.93, 7.72) 0.07 – –     Otherb 1.27 (0.8, 2.01) 0.30 – – Income quartile (ref. quartile 1)     Quartile 2 0.88 (0.72, 1.08) 0.22 – –     Quartile 3 1.07 (0.87, 1.31) 0.53 – –     Quartile 4 1.04 (0.85, 1.26) 0.70 – – Hospital region (ref. south)     Northeast 1.05 (0.86, 1.28) 0.62 – –     Midwest 0.82 (0.67, 1.02) 0.08 – –     West 0.92 (0.74, 1.13) 0.41 – – Urban hospital (ref. rural) 2.67 (1.88, 3.79) <0.0001 1.97 (1.36, 2.86) <0.0001 Teaching hospital (ref. non-teaching) 1.69 (1.45, 1.99) <0.0001 1.34 (1.12, 1.59) <0.0001 Hospital size (ref. small)     Medium 0.97 (0.73, 1.3) 0.85 – –     Large 1.32 (1.02, 1.7) 0.03 – – Relevant clinical conditions     Aspiration 3.79 (2.98, 4.81) <0.0001 2.88 (2.2, 3.77) <0.0001     Infection (ref. no infection)         Opportunistic infections 2.40 (1.27, 4.53) 0.01 2.21 (1.13, 4.33) 0.02         Other infections 3.11 (2.7, 3.59) <0.0001 2.8 (2.41, 3.25) <0.0001     Weight loss/cachexia 2.99 (2.48, 3.61) <0.0001 2.41 (1.95, 2.97) <0.0001     Myositis 2.61 (1.49, 4.58) <0.0001 2.17 (1.21, 3.9) 0.01     Pulmonary fibrosis 2.58 (1.55, 4.27) <0.0001 2.01 (1.14, 3.55) 0.02     Acute bowel obstruction 2.43 (1.87, 3.15) <0.0001 2.66 (1.98, 3.57) <0.0001     Acute renal failure 2.33 (2.01, 2.71) <0.0001 2.24 (1.8, 2.78) <0.0001     Arrhythmia 2.15 (1.67, 2.77) <0.0001 1.62 (1.22, 2.16) <0.0001     PEMC 1.98 (1.61, 2.44) <0.0001 1.42 (1.13, 1.78) <0.0001     Congestive heart failure 1.81 (1.56, 2.1) <0.0001 1.44 (1.22, 1.71) <0.0001     Chronic kidney disease 1.58 (1.34, 1.88) <0.0001 0.73 (0.57, 0.95) 0.02     Pulmonary arterial hypertension 1.56 (1.34, 1.81) <0.0001 1.23 (1.04, 1.45) 0.02     Chronic liver disease 1.45 (1.02, 2.08) 0.04 – –     Diabetes mellitus 1.12 (0.94, 1.34) 0.21 – –     Raynaud’s/ulcer/gangrene 1.10 (0.93, 1.28) 0.26 – –     Hypertension 1.10 (0.96, 1.28) 0.17 – –     Coronary artery disease 1.00 (0.85, 1.19) 0.99 – –     Stroke/TIA 0.98 (0.69, 1.4) 0.9 – –     Oesophageal dysfunction 0.95 (0.83, 1.09) 0.44 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.18 (0.91, 1.52) 0.21 1.23 (0.93, 1.61) 0.14     ≥65 years 0.92 (0.71, 1.19) 0.53 0.85 (0.64, 1.12) 0.25 Male sex 1.27 (1.07, 1.51) 0.01 1.11 (0.92, 1.35) 0.27 Race (ref. Caucasian)     African-American 1.33 (1.09, 1.62) 0.01 – –     Othersa 1.31 (1.07, 1.6) 0.01 – – Transfer status (ref. not transferred in) 2.55 (2.11, 3.09) <0.0001 2.09 (1.68, 2.59) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.31 (1.02, 1.69) 0.04 – –     Private insurance 1.02 (0.85, 1.21) 0.85 – –     Self-pay 1.17 (0.72, 1.9) 0.54 – –     No charge 2.68 (0.93, 7.72) 0.07 – –     Otherb 1.27 (0.8, 2.01) 0.30 – – Income quartile (ref. quartile 1)     Quartile 2 0.88 (0.72, 1.08) 0.22 – –     Quartile 3 1.07 (0.87, 1.31) 0.53 – –     Quartile 4 1.04 (0.85, 1.26) 0.70 – – Hospital region (ref. south)     Northeast 1.05 (0.86, 1.28) 0.62 – –     Midwest 0.82 (0.67, 1.02) 0.08 – –     West 0.92 (0.74, 1.13) 0.41 – – Urban hospital (ref. rural) 2.67 (1.88, 3.79) <0.0001 1.97 (1.36, 2.86) <0.0001 Teaching hospital (ref. non-teaching) 1.69 (1.45, 1.99) <0.0001 1.34 (1.12, 1.59) <0.0001 Hospital size (ref. small)     Medium 0.97 (0.73, 1.3) 0.85 – –     Large 1.32 (1.02, 1.7) 0.03 – – Relevant clinical conditions     Aspiration 3.79 (2.98, 4.81) <0.0001 2.88 (2.2, 3.77) <0.0001     Infection (ref. no infection)         Opportunistic infections 2.40 (1.27, 4.53) 0.01 2.21 (1.13, 4.33) 0.02         Other infections 3.11 (2.7, 3.59) <0.0001 2.8 (2.41, 3.25) <0.0001     Weight loss/cachexia 2.99 (2.48, 3.61) <0.0001 2.41 (1.95, 2.97) <0.0001     Myositis 2.61 (1.49, 4.58) <0.0001 2.17 (1.21, 3.9) 0.01     Pulmonary fibrosis 2.58 (1.55, 4.27) <0.0001 2.01 (1.14, 3.55) 0.02     Acute bowel obstruction 2.43 (1.87, 3.15) <0.0001 2.66 (1.98, 3.57) <0.0001     Acute renal failure 2.33 (2.01, 2.71) <0.0001 2.24 (1.8, 2.78) <0.0001     Arrhythmia 2.15 (1.67, 2.77) <0.0001 1.62 (1.22, 2.16) <0.0001     PEMC 1.98 (1.61, 2.44) <0.0001 1.42 (1.13, 1.78) <0.0001     Congestive heart failure 1.81 (1.56, 2.1) <0.0001 1.44 (1.22, 1.71) <0.0001     Chronic kidney disease 1.58 (1.34, 1.88) <0.0001 0.73 (0.57, 0.95) 0.02     Pulmonary arterial hypertension 1.56 (1.34, 1.81) <0.0001 1.23 (1.04, 1.45) 0.02     Chronic liver disease 1.45 (1.02, 2.08) 0.04 – –     Diabetes mellitus 1.12 (0.94, 1.34) 0.21 – –     Raynaud’s/ulcer/gangrene 1.10 (0.93, 1.28) 0.26 – –     Hypertension 1.10 (0.96, 1.28) 0.17 – –     Coronary artery disease 1.00 (0.85, 1.19) 0.99 – –     Stroke/TIA 0.98 (0.69, 1.4) 0.9 – –     Oesophageal dysfunction 0.95 (0.83, 1.09) 0.44 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programmes. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; ref.: Reference group; TIA: transient ischaemic attack. Table 5 Factors associated with higher cost of hospitalization (>90th percentile) among patients with SSc: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.23 (0.96, 1.57) 0.10 1.21 (0.92, 1.6) 0.17     ≥65 years 0.96 (0.75, 1.24) 0.78 0.89 (0.67¸ 1.19) 0.43 Male sex 1.21 (1.03¸ 1.44) 0.02 1.11 (0.92¸ 1.34) 0.27 Race (ref. Caucasian)     African-American 1.24 (1.01, 1.52) 0.04 1.22 (0.97, 1.54) 0.09     Othersa 1.96 (1.62, 2.37) <0.0001 1.53 (1.24, 1.89) <0.0001 Transfer status (ref. not transferred in) 2.05 (1.68¸ 2.5) <0.0001 1.80 (1.44, 2.26) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.17 (0.92, 1.5) 0.20 – –     Private insurance 1.21 (1.04, 1.41) 0.02 – –     Self-pay 0.77 (0.46, 1.31) 0.34 – –     No charge 2.17 (0.76¸ 6.25) 0.15 – –     Otherb 0.87 (0.52, 1.44) 0.58 – – Income quartile (ref. quartile 1)     Quartile 2 1.10 (0.9, 1.34) 0.38 1.16 (0.93, 1.45) 0.18     Quartile 3 1.48 (1.21, 1.81) <0.0001 1.42 (1.13, 1.79) <0.0001     Quartile 4 1.78 (1.44, 2.19) <0.0001 1.80 (1.43, 2.28) <0.0001 Hospital region (ref. south)     Northeast 1.22 (0.97, 1.55) 0.09 1.03 (0.81, 1.31) 0.80     Midwest 1.22 (0.96, 1.55) 0.10 1.19 (0.92, 1.53) 0.20     West 3.00 (2.39, 3.77) <0.0001 2.98 (2.34, 3.78) <0.0001 Urban hospital (ref. rural) 3.19 (2.16, 4.71) <0.0001 – – Teaching hospital (ref. non-teaching) 1.91 (1.6, 2.27) <0.0001 1.90 (1.57, 2.31) <0.0001 Hospital size (ref. small)     Medium 1.06 (0.79, 1.43) 0.69 – –     Large 1.20 (0.92, 1.57) 0.18 – – Relevant clinical conditions     Myositis 2.83 (1.69, 4.77) <0.0001 – –     Aspiration 2.81 (2.19, 3.6) <0.0001 2.29 (1.72, 3.04) <0.0001     Pulmonary fibrosis 2.59 (1.61, 4.17) <0.0001 1.83 (1.06, 3.15) 0.03     Arrhythmia 2.42 (1.93, 3.05) <0.0001 1.95 (1.51, 2.52) <0.0001     Chronic liver disease 2.22 (1.6, 3.09) <0.0001 1.67 (1.18, 2.36) <0.0001     Acute renal failure 2.17 (1.89, 2.48) <0.0001 2.31 (1.85, 2.89) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.89 (1.01, 3.55) 0.047 1.39 (0.71, 2.71) 0.33         Other infections 2.08 (1.81, 2.39) <0.0001 1.87 (1.61, 2.18) <0.0001     PEMC 1.98 (1.63, 2.4) <0.0001 1.61 (1.29, 2) <0.0001     Acute bowel obstruction 1.93 (1.48, 2.51) <0.0001 2.32 (1.71, 3.15) <0.0001     Weight loss/cachexia 1.80 (1.47, 2.19) <0.0001 1.5 (1.2, 1.89) <0.0001     Chronic kidney disease 1.53 (1.31, 1.79) <0.0001 0.69 (0.54, 0.89) <0.0001     Pulmonary arterial hypertension 1.50 (1.29, 1.74) <0.0001 1.3 (1.1, 1.53) <0.0001     Congestive heart failure 1.49 (1.28, 1.72) <0.0001 – –     Stroke/TIA 1.40 (1.05, 1.87) 0.02 1.66 (1.21, 2.29) <0.0001     Diabetes mellitus 1.32 (1.12, 1.54) <0.0001 1.23 (1.03, 1.47) 0.02     Raynaud’s/ulcer/gangrene 1.21 (1.05, 1.4) 0.01 1.27 (1.08, 1.5) <0.0001     Coronary artery disease 1.19 (1.03, 1.38) 0.02 1.27 (1.07, 1.51) 0.01     Hypertension 1.17 (1.03, 1.33) 0.02 1.19 (1.03, 1.39) 0.02     Oesophageal dysfunction 0.96 (0.85, 1.09) 0.53 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.23 (0.96, 1.57) 0.10 1.21 (0.92, 1.6) 0.17     ≥65 years 0.96 (0.75, 1.24) 0.78 0.89 (0.67¸ 1.19) 0.43 Male sex 1.21 (1.03¸ 1.44) 0.02 1.11 (0.92¸ 1.34) 0.27 Race (ref. Caucasian)     African-American 1.24 (1.01, 1.52) 0.04 1.22 (0.97, 1.54) 0.09     Othersa 1.96 (1.62, 2.37) <0.0001 1.53 (1.24, 1.89) <0.0001 Transfer status (ref. not transferred in) 2.05 (1.68¸ 2.5) <0.0001 1.80 (1.44, 2.26) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.17 (0.92, 1.5) 0.20 – –     Private insurance 1.21 (1.04, 1.41) 0.02 – –     Self-pay 0.77 (0.46, 1.31) 0.34 – –     No charge 2.17 (0.76¸ 6.25) 0.15 – –     Otherb 0.87 (0.52, 1.44) 0.58 – – Income quartile (ref. quartile 1)     Quartile 2 1.10 (0.9, 1.34) 0.38 1.16 (0.93, 1.45) 0.18     Quartile 3 1.48 (1.21, 1.81) <0.0001 1.42 (1.13, 1.79) <0.0001     Quartile 4 1.78 (1.44, 2.19) <0.0001 1.80 (1.43, 2.28) <0.0001 Hospital region (ref. south)     Northeast 1.22 (0.97, 1.55) 0.09 1.03 (0.81, 1.31) 0.80     Midwest 1.22 (0.96, 1.55) 0.10 1.19 (0.92, 1.53) 0.20     West 3.00 (2.39, 3.77) <0.0001 2.98 (2.34, 3.78) <0.0001 Urban hospital (ref. rural) 3.19 (2.16, 4.71) <0.0001 – – Teaching hospital (ref. non-teaching) 1.91 (1.6, 2.27) <0.0001 1.90 (1.57, 2.31) <0.0001 Hospital size (ref. small)     Medium 1.06 (0.79, 1.43) 0.69 – –     Large 1.20 (0.92, 1.57) 0.18 – – Relevant clinical conditions     Myositis 2.83 (1.69, 4.77) <0.0001 – –     Aspiration 2.81 (2.19, 3.6) <0.0001 2.29 (1.72, 3.04) <0.0001     Pulmonary fibrosis 2.59 (1.61, 4.17) <0.0001 1.83 (1.06, 3.15) 0.03     Arrhythmia 2.42 (1.93, 3.05) <0.0001 1.95 (1.51, 2.52) <0.0001     Chronic liver disease 2.22 (1.6, 3.09) <0.0001 1.67 (1.18, 2.36) <0.0001     Acute renal failure 2.17 (1.89, 2.48) <0.0001 2.31 (1.85, 2.89) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.89 (1.01, 3.55) 0.047 1.39 (0.71, 2.71) 0.33         Other infections 2.08 (1.81, 2.39) <0.0001 1.87 (1.61, 2.18) <0.0001     PEMC 1.98 (1.63, 2.4) <0.0001 1.61 (1.29, 2) <0.0001     Acute bowel obstruction 1.93 (1.48, 2.51) <0.0001 2.32 (1.71, 3.15) <0.0001     Weight loss/cachexia 1.80 (1.47, 2.19) <0.0001 1.5 (1.2, 1.89) <0.0001     Chronic kidney disease 1.53 (1.31, 1.79) <0.0001 0.69 (0.54, 0.89) <0.0001     Pulmonary arterial hypertension 1.50 (1.29, 1.74) <0.0001 1.3 (1.1, 1.53) <0.0001     Congestive heart failure 1.49 (1.28, 1.72) <0.0001 – –     Stroke/TIA 1.40 (1.05, 1.87) 0.02 1.66 (1.21, 2.29) <0.0001     Diabetes mellitus 1.32 (1.12, 1.54) <0.0001 1.23 (1.03, 1.47) 0.02     Raynaud’s/ulcer/gangrene 1.21 (1.05, 1.4) 0.01 1.27 (1.08, 1.5) <0.0001     Coronary artery disease 1.19 (1.03, 1.38) 0.02 1.27 (1.07, 1.51) 0.01     Hypertension 1.17 (1.03, 1.33) 0.02 1.19 (1.03, 1.39) 0.02     Oesophageal dysfunction 0.96 (0.85, 1.09) 0.53 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programs. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; TIA: transient ischaemic attack. Table 5 Factors associated with higher cost of hospitalization (>90th percentile) among patients with SSc: results of logistic regression analyses Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.23 (0.96, 1.57) 0.10 1.21 (0.92, 1.6) 0.17     ≥65 years 0.96 (0.75, 1.24) 0.78 0.89 (0.67¸ 1.19) 0.43 Male sex 1.21 (1.03¸ 1.44) 0.02 1.11 (0.92¸ 1.34) 0.27 Race (ref. Caucasian)     African-American 1.24 (1.01, 1.52) 0.04 1.22 (0.97, 1.54) 0.09     Othersa 1.96 (1.62, 2.37) <0.0001 1.53 (1.24, 1.89) <0.0001 Transfer status (ref. not transferred in) 2.05 (1.68¸ 2.5) <0.0001 1.80 (1.44, 2.26) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.17 (0.92, 1.5) 0.20 – –     Private insurance 1.21 (1.04, 1.41) 0.02 – –     Self-pay 0.77 (0.46, 1.31) 0.34 – –     No charge 2.17 (0.76¸ 6.25) 0.15 – –     Otherb 0.87 (0.52, 1.44) 0.58 – – Income quartile (ref. quartile 1)     Quartile 2 1.10 (0.9, 1.34) 0.38 1.16 (0.93, 1.45) 0.18     Quartile 3 1.48 (1.21, 1.81) <0.0001 1.42 (1.13, 1.79) <0.0001     Quartile 4 1.78 (1.44, 2.19) <0.0001 1.80 (1.43, 2.28) <0.0001 Hospital region (ref. south)     Northeast 1.22 (0.97, 1.55) 0.09 1.03 (0.81, 1.31) 0.80     Midwest 1.22 (0.96, 1.55) 0.10 1.19 (0.92, 1.53) 0.20     West 3.00 (2.39, 3.77) <0.0001 2.98 (2.34, 3.78) <0.0001 Urban hospital (ref. rural) 3.19 (2.16, 4.71) <0.0001 – – Teaching hospital (ref. non-teaching) 1.91 (1.6, 2.27) <0.0001 1.90 (1.57, 2.31) <0.0001 Hospital size (ref. small)     Medium 1.06 (0.79, 1.43) 0.69 – –     Large 1.20 (0.92, 1.57) 0.18 – – Relevant clinical conditions     Myositis 2.83 (1.69, 4.77) <0.0001 – –     Aspiration 2.81 (2.19, 3.6) <0.0001 2.29 (1.72, 3.04) <0.0001     Pulmonary fibrosis 2.59 (1.61, 4.17) <0.0001 1.83 (1.06, 3.15) 0.03     Arrhythmia 2.42 (1.93, 3.05) <0.0001 1.95 (1.51, 2.52) <0.0001     Chronic liver disease 2.22 (1.6, 3.09) <0.0001 1.67 (1.18, 2.36) <0.0001     Acute renal failure 2.17 (1.89, 2.48) <0.0001 2.31 (1.85, 2.89) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.89 (1.01, 3.55) 0.047 1.39 (0.71, 2.71) 0.33         Other infections 2.08 (1.81, 2.39) <0.0001 1.87 (1.61, 2.18) <0.0001     PEMC 1.98 (1.63, 2.4) <0.0001 1.61 (1.29, 2) <0.0001     Acute bowel obstruction 1.93 (1.48, 2.51) <0.0001 2.32 (1.71, 3.15) <0.0001     Weight loss/cachexia 1.80 (1.47, 2.19) <0.0001 1.5 (1.2, 1.89) <0.0001     Chronic kidney disease 1.53 (1.31, 1.79) <0.0001 0.69 (0.54, 0.89) <0.0001     Pulmonary arterial hypertension 1.50 (1.29, 1.74) <0.0001 1.3 (1.1, 1.53) <0.0001     Congestive heart failure 1.49 (1.28, 1.72) <0.0001 – –     Stroke/TIA 1.40 (1.05, 1.87) 0.02 1.66 (1.21, 2.29) <0.0001     Diabetes mellitus 1.32 (1.12, 1.54) <0.0001 1.23 (1.03, 1.47) 0.02     Raynaud’s/ulcer/gangrene 1.21 (1.05, 1.4) 0.01 1.27 (1.08, 1.5) <0.0001     Coronary artery disease 1.19 (1.03, 1.38) 0.02 1.27 (1.07, 1.51) 0.01     Hypertension 1.17 (1.03, 1.33) 0.02 1.19 (1.03, 1.39) 0.02     Oesophageal dysfunction 0.96 (0.85, 1.09) 0.53 – – Factor Univariate regression Multivariate regression OR (95% CI) P-value aOR (95% CI) P-value Age category (ref. 18–44 years)     45–64 years 1.23 (0.96, 1.57) 0.10 1.21 (0.92, 1.6) 0.17     ≥65 years 0.96 (0.75, 1.24) 0.78 0.89 (0.67¸ 1.19) 0.43 Male sex 1.21 (1.03¸ 1.44) 0.02 1.11 (0.92¸ 1.34) 0.27 Race (ref. Caucasian)     African-American 1.24 (1.01, 1.52) 0.04 1.22 (0.97, 1.54) 0.09     Othersa 1.96 (1.62, 2.37) <0.0001 1.53 (1.24, 1.89) <0.0001 Transfer status (ref. not transferred in) 2.05 (1.68¸ 2.5) <0.0001 1.80 (1.44, 2.26) <0.0001 Primary payer status (ref. Medicare)     Medicaid 1.17 (0.92, 1.5) 0.20 – –     Private insurance 1.21 (1.04, 1.41) 0.02 – –     Self-pay 0.77 (0.46, 1.31) 0.34 – –     No charge 2.17 (0.76¸ 6.25) 0.15 – –     Otherb 0.87 (0.52, 1.44) 0.58 – – Income quartile (ref. quartile 1)     Quartile 2 1.10 (0.9, 1.34) 0.38 1.16 (0.93, 1.45) 0.18     Quartile 3 1.48 (1.21, 1.81) <0.0001 1.42 (1.13, 1.79) <0.0001     Quartile 4 1.78 (1.44, 2.19) <0.0001 1.80 (1.43, 2.28) <0.0001 Hospital region (ref. south)     Northeast 1.22 (0.97, 1.55) 0.09 1.03 (0.81, 1.31) 0.80     Midwest 1.22 (0.96, 1.55) 0.10 1.19 (0.92, 1.53) 0.20     West 3.00 (2.39, 3.77) <0.0001 2.98 (2.34, 3.78) <0.0001 Urban hospital (ref. rural) 3.19 (2.16, 4.71) <0.0001 – – Teaching hospital (ref. non-teaching) 1.91 (1.6, 2.27) <0.0001 1.90 (1.57, 2.31) <0.0001 Hospital size (ref. small)     Medium 1.06 (0.79, 1.43) 0.69 – –     Large 1.20 (0.92, 1.57) 0.18 – – Relevant clinical conditions     Myositis 2.83 (1.69, 4.77) <0.0001 – –     Aspiration 2.81 (2.19, 3.6) <0.0001 2.29 (1.72, 3.04) <0.0001     Pulmonary fibrosis 2.59 (1.61, 4.17) <0.0001 1.83 (1.06, 3.15) 0.03     Arrhythmia 2.42 (1.93, 3.05) <0.0001 1.95 (1.51, 2.52) <0.0001     Chronic liver disease 2.22 (1.6, 3.09) <0.0001 1.67 (1.18, 2.36) <0.0001     Acute renal failure 2.17 (1.89, 2.48) <0.0001 2.31 (1.85, 2.89) <0.0001     Infection (ref. no infection)         Opportunistic infections 1.89 (1.01, 3.55) 0.047 1.39 (0.71, 2.71) 0.33         Other infections 2.08 (1.81, 2.39) <0.0001 1.87 (1.61, 2.18) <0.0001     PEMC 1.98 (1.63, 2.4) <0.0001 1.61 (1.29, 2) <0.0001     Acute bowel obstruction 1.93 (1.48, 2.51) <0.0001 2.32 (1.71, 3.15) <0.0001     Weight loss/cachexia 1.80 (1.47, 2.19) <0.0001 1.5 (1.2, 1.89) <0.0001     Chronic kidney disease 1.53 (1.31, 1.79) <0.0001 0.69 (0.54, 0.89) <0.0001     Pulmonary arterial hypertension 1.50 (1.29, 1.74) <0.0001 1.3 (1.1, 1.53) <0.0001     Congestive heart failure 1.49 (1.28, 1.72) <0.0001 – –     Stroke/TIA 1.40 (1.05, 1.87) 0.02 1.66 (1.21, 2.29) <0.0001     Diabetes mellitus 1.32 (1.12, 1.54) <0.0001 1.23 (1.03, 1.47) 0.02     Raynaud’s/ulcer/gangrene 1.21 (1.05, 1.4) 0.01 1.27 (1.08, 1.5) <0.0001     Coronary artery disease 1.19 (1.03, 1.38) 0.02 1.27 (1.07, 1.51) 0.01     Hypertension 1.17 (1.03, 1.33) 0.02 1.19 (1.03, 1.39) 0.02     Oesophageal dysfunction 0.96 (0.85, 1.09) 0.53 – – Significant P-values are in bold. a Others includes: Hispanic, Asian or Pacific Islander, Native American and Other. b Other includes: Worker’s Compensation, Civilian Health and Medical Program of the Uniformed Services (CHAMPUS), Civilian Health and Medical Program of the Department of Veterans Affairs (CHAMPVA), Title V and other government programs. aOR: adjusted odds ratio; OR: odds ratio; PEMC: pericarditis, endocarditis, myocarditis or cardiomyopathy; TIA: transient ischaemic attack. Among other studied clinical factors, CHF was significantly associated with longer LOS (aOR = 1.44, 95% CI: 1.22, 1.71) but not with mortality or cost. Arrhythmia, weight loss/cachexia and transfer-in status had higher odds of mortality, longer LOS and higher cost. Infection other than opportunistic had a very strong association with higher odds of mortality (aOR = 3.36, 95% CI: 2.73, 4.41, P < 0.0001) and comparable to that of acute renal failure and aspiration. Interestingly, higher income quartiles were associated with higher odds of mortality (quartile 4, aOR = 1.41, 95% CI: 1.07, 1.87, P = 0.02) and higher cost (quartile 3 aOR = 1.48, 95% CI: 1.13, 1.79; quartile 4 aOR = 1.80, 95% CI: 1.43, 1.28, P < 0.0001 for both). Likewise, urban location with teaching status of the hospital was associated with higher odds of longer LOS (aOR = 1.97, 95% CI: 1.36, 2.86 and aOR = 1.34, 95% CI: 1.12, 1.59, respectively, P < 0.0001 for both). However, higher cost (aOR = 1.9, 95% CI: 1.57, 2.31, P < 0.0001) was associated with teaching status only. In the sensitivity analysis (supplementary Table S2, available at Rheumatology online) for longer LOS after removing SSc patients who died in the hospital, PAH, urban status and teaching status of the hospital were no longer statistically significant. One reason for this could be the tendency of pooling of PAH patients in urban teaching hospitals by virtue of presence of highly specialized PAH centres who take care of these patients towards the end of their life. Of note, the median LOS was longer by 1 day (95% CI: 0.98, 1.03, P < 0.0001) among SSc patients who died in hospital compared with those who did not. Discussion With this study we have updated our understanding of current factors associated with in-hospital mortality, LOS and hospitalization costs among SSc patients. We have built on previous studies and analysed more than 9000 sample hospitalizations with SSc (representing an estimated 48 655 discharges) in the USA from the 2012 to 2013 NIS. We performed a detailed analysis of the factors associated with in-hospital mortality, LOS and cost of hospitalization. In previous studies reporting on inpatient mortality among SSc patients [6–10], SSc-related and cardiopulmonary causes were the most important determinants. In this study, in contrast to two previous studies using the same database, we have consolidated individual diagnostic codes into categories based on systemic/pathologic involvement to more comprehensively identify major diagnoses associated with hospitalization and mortality. We found that infections/septicaemia were the most common diagnoses in both overall SSc hospitalizations and SSc in-hospital mortality (17.4 and 32.7%, respectively). These results are supported by a study by Tyndall et al. [4] in which the authors longitudinally analysed 5860 patients from the EULAR Scleroderma Trials and Research cohort and reported infection to be attributed to 33% of non-SSc-related deaths. Similarly, in a more recent retrospective cross-sectional study of hospitalized SSc patients by Netwijitpan et al. [9] carried out in Thailand, infections were the most common non-SSc-related cause of hospitalization and lower respiratory tract infections were the most common (49.9%) overall cause of death. Another Iranian study by Shenavandeh and Naseri [10] looking at 446 admissions by 181 patients over 13 years also found infection to be the most common non-SSc-related cause for hospitalization. This finding could be explained by a more widespread utilization of immunosuppressive agents, as well as an increased comfort level among clinicians in utilizing such agents in SSc. With this in mind we further explored to determine whether a hospitalization with an infection diagnosis was associated with one of the collective codes representing possible use of chemotherapy, which can include both chemotherapy such as CYC or rituximab and any other immunosuppressive infusion during the hospitalization (supplementary Table S5, available at Rheumatology online). We found that, in comparison with non-SSc hospitalizations, SSc hospitalizations with infection/septicaemia had a significantly higher proportion of the possible use of chemotherapy among both patients who lived (9.2 vs 4.1%, P < 0.0001) and those who died during the hospitalization (17.4 vs 6.3%, P = 0.0005) (data not shown). Patients who died had significantly higher proportions of coding for PAH (43.3%), respiratory failure (72.4%) and CHF (38.4%), suggesting more aggressive SSc disease, which in turn may have required stronger immunosuppressive therapy predisposing to more infections. Similar to the study by Chung et al. [7] renal failure accounted for only a minor proportion of primary hospitalization diagnoses (2%) as well as inpatient mortality (1.7%), reflecting an early detection and management with angiotensin-converting enzyme inhibitors, which may have decreased both hospitalization and in-hospital mortality rates among this subset of SSc patients [5]. In multivariate analysis, however, a diagnosis of acute renal failure from any position of the discharge diagnoses was strongly associated with mortality. Most of these non-primary diagnoses likely do not represent scleroderma renal crisis—renal failure is common in patients with other major illnesses such as infection and heart failure, which are likely to be the primary drivers of mortality. But the evidence of renal failure’s independency in multivariate analysis suggests that it does amplify the odds of mortality and raises a concern that we might need to do more to prevent acute renal injury among hospitalizations with SSc. Interestingly, oesophageal dysfunction was associated with lower odds of mortality in both univariate and multivariate regression analysis. While it may seem contradictory, likely this is because oesophageal dysfunction is not coded as frequently in patients with other severe reasons for hospitalization and this might have falsely made oesophageal dysfunction look like it is protective. However, we noted a very strong association between mortality and aspiration, as in a study by Sehra et al. [6]. Also, the proportion of aspiration was almost double among patients with oesophageal dysfunction (5.95 vs 2.55%, P < 0.0001), indicating that oesophageal dysfunction is likely the index reason for aspiration. Similar to prior studies [7, 8], conditions such as pulmonary fibrosis, CHF, arrhythmias and transfer from another hospital all had higher odds of having a longer LOS. Additionally, as expected, PAH, acute renal failure, myositis, weight loss/cachexia and presence of any infection were some of the factors not studied or found insignificant in prior studies but which significantly increased the odds of longer LOS in our analysis. Overall in-hospital mortality among SSc patients was 5%, which is lower than the older database studies from the USA using the NIS (7.1% in 1995 and 6.3% in 2002–03) [7, 8] but similar to a more recent single-centre study looking at mortality among hospitalized SSc patients at an institution that has a dedicated scleroderma centre (5% in 2001–11) [6]. In our study, African-Americans consistently showed higher rates of mortality. It was particularly high as compared with the Other race groups, among older patients ≥65 years (11.1 vs 6.8% among Other race and 5% among Caucasians). This was reflected even in the multivariate analysis with an odds ratio of 1.46 (P = 0.01) for in-hospital death among African American SSc patients. This finding is in contrast to the two prior studies (Nietert et al. [8] and Chung et al. [7]) done with the same database, which found that race was not associated with higher in-hospital mortality. This may be explained by an ever changing and improving representation of racial information in the HCUP databases [16]. Curiously, higher income (quartile 4) was associated with higher odds (odds ratio = 1.41, P = 0.02) of dying in hospital. This might be explained by an access paradox, with wealthier patients being able to afford expensive immunosuppressive therapies and having more access to in-hospital care, or possible disease perception and management differences not clearly defined by our study. For comparison purpose, we had calculated the average LOS as 5.9 (±7.9) days with a median (interquartile range) of 4 (2, 7) days. This is lower than prior studies based on the same database (mean LOS of 6.6 days in the Chung study and 7.5 days in the Nietert study) with a chronological trend of decreasing overall LOS among SSc patients. This change of trend is most likely multifactorial and may represent advances in overall care, or shift in insurance regulations forcing shorter LOS. We saw a statistically significant relationship between longer LOS and Medicaid being the primary insurance during our multivariate analysis, supporting the hypothesis that SSc patients with lower socioeconomic status have longer LOS. SSc patients hospitalized in institutions labelled as urban and ones with a teaching status had longer LOS. This might be due to the tendency for sicker and more complex patients to present to these settings. The cost of hospitalization of SSc patients based on HCUP-NIS was studied by Nietert et al. [8] but not by Chung et al. [7]. In 1995 the estimated cost of healthcare expenditure in the inpatient setting among SSc patients was over $280 million. After adjusting for inflation [17], this translates to $422 million in 2012 and $429 million in 2013. We estimated a lower adjusted cost of $357 million and $361 million, respectively, for the years 2012 and 2013 in our study. While this could relate to better care and less hospitalizations, it could also relate to the insurance regulations forcing shorter LOS and thus incurring less cost. The findings in our report are subject to some limitations. There is the potential for misclassification of the diagnosis of SSc, although an inpatient diagnosis code for SSc has previously been validated. Also, being an administrative database created for billing purposes, the HCUP-NIS lacks the same rigor and diagnostic accuracy as compared with a database established for research purposes. The diagnosis of SSc per se does not differentiate between limited and diffuse variations of SSc due to the limitation of ICD-9 coding. The ICD-9 code used in this study (710.1) represents any of the following diagnoses: acrosclerosis, Calcinosis, Raynaud's, Esophageal dysmotility, Sclerodactyly, Telangiectasia syndrome, progressive SSc or scleroderma. We were unable to look at differences in hospitalization causes and LOS based on these variants. Similarly, the mortality data in our study are globally assessed disregarding the variants and we were unable to comment on mortality differences among SSc hospitalizations by subtypes. An absence of a diagnostic code does not mean the condition was absent, especially if it was not the primary diagnosis, which could make some comorbidities appear protective, like hypertension and diabetes in our study. As the sample is retrospective and cross-sectional it lacks any temporality in the conditions studied and cannot establish causality. Since multiple admissions of the same patient, if any, were counted as separate instances of hospitalization/discharge, readmissions are not accounted for, thus inflating the actual number of SSc patients utilizing the inpatient service. The cost estimation, which uses cost-to-charge ratios in the analysis, is based on aggregate hospital data and is not specific to a single diagnosis such as SSc, and might be different from the true amount. Despite these limitations, our study has several strengths. Being the largest inpatient and only database with all-payer information, it is the most representative database to reflect national estimates in the inpatient setting [11]. We have analysed a very large population of hospitalized adult SSc patients. Moreover, the redesign of the database since 2012, notably directly sampling discharges rather than hospitals like in previous years, has allowed wider and better representation of the US hospitals geographically and more complete information about race (fewer missing data) is present, which was lacking in the prior years’ databases [17]. With this study we have more clearly defined that the primary diagnosis for hospitalization and in-hospital mortality among SSc patients is infection and that in-hospital mortality and LOS have overtime declined. We believe our findings will serve to increase physician awareness on prevention (influenza and pneumococcal vaccinations) as well as early institution of antibiotherapy when facing a possible infective complication. Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: The authors have declared no conflicts of interest. References 1 Allanore Y , Simms R , Distler O et al. Systemic sclerosis . Nat Rev Dis Primers 2015 ; 1 : 1 – 21 . 2 Rubio-Rivas M , Royo C , Simeón CP , Corbella X , Fonollosa V. Mortality and survival in systemic sclerosis: systematic review and meta-analysis . Semin Arthritis Rheum 2014 ; 44 : 208 – 19 . Google Scholar CrossRef Search ADS PubMed 3 Bose N , Chiesa-Vottero A , Chatterjee S. Scleroderma renal crisis . Semin Arthritis Rheum 2015 ; 44 : 687 – 94 . Google Scholar CrossRef Search ADS PubMed 4 Tyndall AJ , Bannert B , Vonk M et al. Causes and risk factors for death in systemic sclerosis: a study from the EULAR Scleroderma Trials and Research (EUSTAR) database . Ann Rheum Dis 2010 ; 69 : 1809 – 15 . Google Scholar CrossRef Search ADS PubMed 5 Steen VD , Medsger TA. Changes in causes of death in systemic sclerosis, 1972-2002 . Ann Rheum Dis 2007 ; 66 : 940 – 4 . Google Scholar CrossRef Search ADS PubMed 6 Sehra ST , Kelly A , Baker JF , Derk CT. Predictors of inpatient mortality in patients with systemic sclerosis: a case control study . Clin Rheumatol 2016 ; 35 : 1631 – 5 . Google Scholar CrossRef Search ADS PubMed 7 Chung L , Krishnan E , Chakravarty EF. Hospitalizations and mortality in systemic sclerosis: results from the Nationwide Inpatient Sample . Rheumatology (Oxford) 2007 ; 46 : 1808 – 13 . Google Scholar CrossRef Search ADS PubMed 8 Nietert PJ , Silverstein MD , Silver RM. Hospital admissions, length of stay, charges, and in-hospital death among patients with systemic sclerosis . J Rheumatol 2001 ; 28 : 2031 – 7 . Google Scholar PubMed 9 Netwijitpan S , Foocharoen C , Mahakkanukrauh A , Suwannaroj S , Nanagara R. Indications for hospitalization and in-hospital mortality in Thai systemic sclerosis . Clin Rheumatol 2013 ; 32 : 361 – 7 . Google Scholar CrossRef Search ADS PubMed 10 Shenavandeh S , Naseri R. Assessment of hospitalization and mortality of scleroderma in-patients: a thirteen-year study . Reumatologia 2017 ; 55 : 163 – 8 . Google Scholar CrossRef Search ADS PubMed 11 HCUP-US Databases . Healthcare Cost and Utilization Project (HCUP). April 2017. Rockville, MD: Agency for Healthcare Research and Quality. https://www.hcup-us.ahrq.gov/databases.jsp (10 September 2017 , date last accessed). 12 Valenzuela A , Yaqub A , Fiorentino D , Krishnan E , Chung L. Validation of the ICD-9-CM code for systemic sclerosis using updated ACR/EULAR classification criteria . Scand J Rheumatol 2015 ; 44 : 253 – 5 . Google Scholar CrossRef Search ADS PubMed 13 HCUP CCS. Healthcare Cost and Utilization Project (HCUP) . Agency for Healthcare Research and Quality, Rockville, MD. HCUP Databases. Published March 2017. https://www.hcup-us.ahrq.gov/nisoverview.jsp (31 July 2017 , date last accessed). 14 Radensky PW , Berliner E , Archer JW , Dournaux SF. Inpatient costs of major cardiovascular events . Am J Cardiovasc Drugs 2001 ; 1 : 205 – 17 . Google Scholar CrossRef Search ADS PubMed 15 HCUP-US Tools & Software Page . Healthcare Cost and Utilization Project (HCUP). July 2017. Rockville, MD: Agency for Healthcare Research and Quality. https://www.hcup-us.ahrq.gov/tools_software.jsp (10 September 2017 , date last accessed). 16 Houchens R, Ross D, Elixhauser A, Jiang J. Nationwide Inpatient Sample (NIS) Redesign Final Report. 2014. HCUP Methods Series Report # 2014-04 ONLINE. April 4, 2014. U.S. Agency for Healthcare Research and Quality. http://www.hcup-us.ahrq.gov/reports/methods/methods.jsp (12 October 2015 , date last accessed). 17 Bureau of Labor Statistics. CPI Inflation Calculator . https://www.bls.gov/data/inflation_calculator.htm (2 November 2017 , date last accessed). © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

RheumatologyOxford University Press

Published: Sep 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off