Skip to main content

Table 3 Relationship between bilirubin intensities and respiratory health outcomes. Estimates represent the change in the parameter of interest for a 1 standard deviation change in bilirubin intensity

From: Bilirubin-associated single nucleotide polymorphism (SNP) and respiratory health outcomes: a mendelian randomization study

Ā 

Non-Hispanic White

African American

Estimate

95% Confidence Interval

Estimate

95% Confidence Interval

Acute Respiratory Events - IRR

0.85

0.75 to 0.96

-

-

Mortality - Odds Ratio

0.931

0.575 to 1.419

-

-

Mortality - Hazard Ratio

0.933

0.690 to 1.264

-

-

FEV 1 at baseline

0.064

0.016 to 0.111

0.044

-0.101 to 0.189

FVC at baseline

0.049

0.001 to 0.096

0.056

-0.110 to 0.221

FEV1/FVC at baseline

0.010

0.001 to 0.020

-0.003

-0.029 to 0.023

Change in FEV1, mL/yr

2.323

-0.998 to 5.644

-2.910

-13.864 to 8.044

Change in FVC, mL/yr

3.989

-1.358 to 9.335

3.736

-12.086 to 19.559

Change in Emphysema, % per yr

-0.019

-0.070 to 0.032

0.012

-0.096 to 0.120

Change in Gas Trapping, % per yr

-0.041

-0.135 to 9.335

0.291

-0.059 to 0.640

Change in Wall Area, mm per yr

-0.035

-0.086 to 0.032

-0.070

-0.310 to 0.169

Change in Pi10 mm per yr

-0.001

-0.004 to 0.002

-0.005

-0.019 to 0.009

Change in AWT mm per yr

-0.001

-0.002 to 0.002

-0.002

-0.008 to 0.004

Change in Walking Distance,ft/yr

8.42

-11.93 to 28.76

58.37

-17.98 to 134.71

5-Year Change in MMRC

0.026

-0.046 to 0.098

-0.074

-0.405 to 0.257

5-Year Change in Total SGRQ

-0.064

-0.892 to 0.764

0.197

-3.781 to 4.174

  1. AWT, airway wall thickness; FEV1, forced expiratory volume in 1-second; FVC, forced vital capacity; MMRC, Modified Medical Research Council Dyspnea Scale; Pi10, square root of the wall area for a theoretical airway with an internal perimeter of 10Ā mm;SGRQ, Saint Georgeā€™s Respiratory Questionnaire
  2. Models evaluating mortality as a binary and survival outcome were logistic and Cox proportional hazards models. All other regression models used linear regression. All models were adjusted for genetic ancestry, sex, age, and packs smoking at baseline. Models for all-cause mortality were additionally adjusted for clinical site at baseline. When applicable, an offset for months followed was used
  3. No mortality events occurred for African-Americans for this subset of data