Demographic and hospitalization-related characteristics of the study population are listed in Table 1. Discharges in the study population of 41,581 people were distributed by race as follows: white (78%), African-American (8%), and Asian (8%). The study population was divided ethnically into Hispanic (11%) and non-Hispanic (86%). More than one in five discharges (21.88%) died during their hospitalization. Half the study population (50%) was 80 years of age or older. We found no evidence that the mean lengths of stay differed by race (p=0.82). However, the differences in mean lengths of stay by ethnicity were statistically significant (p=0.03), with mean length of stay for Hispanics (11.6 days) higher than non-Hispanics (10.5 days).

Table 1 also lists the adjusted odds of hospital death associated with race and ethnicity. These odds are adjusted for differences in mortality risk by patient age group, sex, whether the hospitalization was an emergency admission, whether the patient was transferred from another hospital, and for the presence of 223 separate categories of comorbid disease and 36 separate categories of conditions related to aspiration pneumonia. Only categories of comorbid disease and conditions related to aspiration that occurred in 1% or more of the study population were included in the model.

Table 1. Frequency, Unadjusted Odds of Death, and Adjusted Odds of Death by Study Population Hospitalization Characteristics, Demographics, and Selected Comorbid Disease Categories:

Frequency Percent

Unadjusted Odds of Death (95% CI) Adjusted Odds of Death** (95% CI)
Race: White

77.99

reference group

reference group

Race: African-American

8.40

1.02 (0.94-1.11)

1.01 (0.91-1.11)

Race: Asian

7.87

0.95 (0.87-1.04)

0.83 (0.75-0.91)

Ethnicity: Non-Hispanic

86.02

reference group

reference group

Ethnicity: Hispanic

11.42

0.86 (0.80-0.93)

0.90 (0.82-0.98)

Died in hospital

21.88

Emergency admission

88.29

1.09 (1.01-1.17)

0.97 (0.89-1.05)

Transferred patient

4.70

1.05 (0.94-1.17)

1.20 (1.06-1.35)

Male

54.50

0.95 (0.91-0.99)

1.02 (0.96-1.08)

Female

45.50

reference group

reference group

Age up to 9

1.72

reference group

reference group

Age 10-19

0.93

1.73 (0.96-3.13)

1.50 (0.81-2.78)

Age 20-29

1.27

2.61 (1.57-4.35)

2.14 (1.25-3.67)

Age 30-39

2.33

3.02 (1.91-4.79)

2.04 (1.24-3.35)

Age 40-49

3.68

4.20 (2.72-6.49)

2.61 (1.62-4.18)

Age 50-59

4.75

5.28 (3.45-8.07)

3.02 (1.90-4.80)

Age 60-69

9.97

7.28 (4.816-11.017)

4.26 (2.71-6.70)

Age 70-79

25.32

8.22 (5.46-12.37)

5.36 (3.42-8.42)

Age 80-89

35.62

9.49 (6.31-14.27)

6.92 (4.41-10.86)

Age 90-99

13.79

10.05 (6.66-15.16)

7.50 (4.77-11.82)

Age 100 and older

0.60

15.28 (9.43-24.76)

10.56 (6.22-17.90)

Septicemia *

12.31

2.34 (2.20-2.49)

1.91 (1.78-2.06)

Thyroid disorders

8.20

0.87 (0.80-0.95)

0.87 (0.79-0.95)

Diabetes mellitus without complication

11.40

1.06 (0.98-1.14)

1.03 (0.95-1.11)

complications *

7.95

1.41 (1.30-1.53)

1.22 (1.10-1.34)

Nutritional deficiencies

14.30

1.22 (1.14-1.30)

1.12 (1.04-1.20)

Fluid and electrolyte disorders

39.40

1.46 (1.40-1.53)

1.26 (1.19-1.33)

Anemia

24.04

0.99 (0.94-1.05)

0.83 (0.78-0.89)

Hypertension with complications

7.35

1.54 (1.42-1.67)

1.27 (1.16-1.40)

Coronary atherosclerosis

19.54

1.25 (1.18-1.32)

1.05 (0.99-1.12)

Cardiac dysrhythmias

20.36

1.52 (1.44-1.61)

1.20 (1.12-1.27)

The statistical performance of the regression model (c-index=0.74, pseudo-R2=0.12) in the development population (1996-1998 discharges, numbering 41,581) declined only slightly (c-index=0.73, pseudo-R^O.l 1) in the validation population (1999 discharges, numbering 15,747).

After comprehensively adjusting for baseline characteristics, including comorbid disease, Asian discharges were found to have a probability of death significantly different from that of white discharges in the study population. The odds of death for Asian discharges were lower [adjusted odds ratio (OR), 0.83; 95% confidence interval (CI), 0.75-0.911] compared to whites. The odds of death for Hispanics were lower (adjusted OR=0.90; CI, 0.816-0.981) compared to non-Hispanics. The numbers of discharges whose racial category was listed as “Native American,” “other,” or “unknown,” or whose ethnicity was recorded as “unknown” were too small to provide meaningful statistics about their odds of death. silagra uk

Table 2. Ten Comorbid Illnesses Common among Aspiration Pneumonia Discharges that Have Statistically Significant Differences in Prevalence by Race or by Ethnicity

Asian

Black

White

P Value Hispanic Non-Hispanic

P Value

Septicemia*

13.63

17.34

11.56

O.0001

13.24

12.14

0.0254

Thyroid disorders

3.94

4.32

9.25

O.0001

5.89

8.53

0.0001

Diabetes mellitus without complication

15.83

15.03

10.42

O.0001

14.08

11.11

0.0001

Diabetes mellitus with complications*

10.39

13.02

6.88

O.0001

13.39

7.24

0.0001

Nutritional deficiencies

17.14

17.89

13.82

O.0001

13.54

14.41

0.2651

Fluid and electrolyte disorders

39.78

42.70

39.22

O.0001

37.79

39.52

0.0068

Anemia

24.29

33.09

23.19

0.0001

23.94

24.07

0.8760

Hypertension with complications

12.19

13.62

6.11

0.0001

9.28

7.10

0.0001

Coronary atherosclerosis

17.97

15.17

20.37

0.0001

17.16

19.94

0.0001

Cardiac dysrhythmias

18.91

14.22

21.57

0.0001

14.36

21.16

0.0001

Table 2 lists the 10 most commonly occurring comorbid diseases represented in the complete multivariable logistic regression model for which statistically significant differences existed in prevalence across any racial or ethnic categories. Septicemia, diabetes, nutritional deficiencies, fluid and electrolyte disorders, anemia, and hypertension were all more prevalent among Asians and African Americans. Asians and African Americans both had lower frequencies of thyroid disorders, coronary atherosclerosis, and cardiac dysrhythmias. Similar differences in the prevalence of comorbidities were demonstrated for Hispanic versus non-Hispanic discharges, with the exception of nutritional deficiencies and anemia. Specific frequencies, unadjusted, and adjusted odds of death for these 10 categories of comorbid disease are included in Table 1. cialis canadian pharmacy

Figure 1. Adjusted odds

Figure 1. Adjusted odds of in-hospital death and their 95% confidence intervals for aspiration pneumonia discharges with 10 most common comorbid illnesses stratified by race. For each comorbid illness, the odds of in-hospital death is depicted for whites, blacks, and Asians.

While there were differences in prevalence, we did not find evidence that the effects of these categories of comorbid disease were different by race or ethnicity. The interaction terms measuring the race or ethnicity specific effects of comorbid disease on mortality risk were not statistically significant in any of the models we tested.

Figure 1 lists odds ratios for in-hospital death by racial category for each of the 10 comorbid diseases, adjusted for patient age, sex, hospitalization characteristics, other categories of comorbid disease, and conditions related to aspiration pneumonia. Differences by racial group are demonstrated for several categories of comorbid disease. Thyroid disorders are associated with a statistically significant increased risk of death among African Americans but not among either whites or Asians. In the total study population, thyroid disorders are associated with a statistically significant reduced risk of death. Diabetes mellitus with complications and hypertension with complications both represent a statistically significant increased risk of death for all discharges, except Asians. Anemia represents a statistically significant lower risk of death for all discharges, except for African Americans. buy tadacip

Figure 2. Adjusted odds

Figure 2. Adjusted odds of in-hospital death and their 95% confidence intervals for aspiration pneumonia discharges with 10 most common comorbid illnesses stratified by ethnicity. For each comorbid illness, the odds of in-hospital death is depicted for Hispanics and non-Hispanics.

Figure 2 lists adjusted odds ratios for the effects of the 10 comorbid disease categories on the risk of in-hospital death by ethnic category. Differences in the effects of some categories of comorbid disease by ethnic group were also demonstrated. While thyroid disorders and anemia were associated with decreased risks of death for all discharges, the decreased risk was not statistically significant for Hispanics. Statistically significant increased risks of death were associated with both nutritional deficiencies and diabetes mellitus, with complications for all discharges except Hispanics. tadalis sx