Table 2 Risk of 30-day, 90-day, and 180-day re-hospitalization among individuals with chronic obstructive pulmonary disease hospitalized with community-acquired pneumonia in the by tertiles of the health-related quality of life measured at admission.

From: Health-related quality of life predicts prognosis in individuals with COPD hospitalized with community-acquired pneumonia – a prospective cohort study

Re-hospitalization

HR (95% CI)

Unadjusted

(n = 140)

P

HR (95% CI)

Adjusted

Complete case (n = 81)

P

HR (95% CI)

Adjusted

Imputed data

(n = 140)

P

30-day

EQ utility score

0.51 (0.2; 1.2)

0.11

0.88 (0.2; 3.2)

0.84

0.52 (0.2; 1.2)

0.14

Lowest tertile

Ref.

 

Ref.

 

Ref.

 

Middle tertile

0.57 (0.3; 1.0)

0.06

0.62 (0.3; 1.4)

0.26

0.58 (0.3; 1.1)

0.08

Highest tertile

0.64 (0.4; 1.1)

0.11

0.65 (0.3; 1.4)

0.29

0.62 (0.4; 1.1)

0.09

90-day

EQ utility score

0.44 (0.2; 0.8)

0.01*

0.81 (0.3; 2.0)

0.65

0.50 (0.3; 0.9)

0.03*

Lowest tertile

Ref.

 

Ref.

 

Ref.

 

Middle tertile

0.45 (0.3; 0.7)

< 0.01*

0.57 (0.3; 1.2)

0.12

0.46 (0.3; 0.8)

< 0.01*

Highest tertile

0.59 (0.4; 0.9)

0.01*

0.72 (0.4; 1.3)

0.27

0.59 (0.4; 0.9)

0.01*

180-day

EQ utility score

0.55 (0.3; 0.9)

0.01*

0.86 (0.4; 1.7)

0.68

0.63 (0.4; 1.0)

0.06

Lowest tertile

Ref.

 

Ref.

 

Ref.

 

Middle tertile

0.71 (0.5; 1.0)

0.04*

0.77 (0.5; 1.3)

0.29

0.74 (0.5; 1.0)

0.06

Highest tertile

0.73 (0.5; 1.0)

0.052

0.90 (0.6; 1.4)

0.66

0.76 (0.6; 1.0)

0.09

  1. Health-related quality of life was measured as the total utility score derived from the EuroQoL (EQ) 5D-5L questionnaire. Cox regression was fitted for new hospital admission within 30, 90 and 180 days after discharge with mortality as competing event. Estimates shown are hazard ratio (HR) with corresponding 95% confidence intervals (CI) and p-value. The first column shows the unadjusted complete case model. The second column shows the adjusted complete case model. The third column shows the adjusted model with imputed variables. The adjusted models include the following variables: age, sex, FEV1 (pulmonary function), CURB-65 (initial severity of CAP), Charlson comorbidity index, and nutritional status (undernourished, well-nourished, overweight, and obese). Values were imputed in FEV1 (n=18), CURB-65 (n=16), and nutritional status (n=32). *Statistically significant (p < 0.05).