Table 4.
Adjusted ORs for Sensitivity Analyses of Low Gait Speed and Low Peak Torque by Serum Bicarbonate Level*
| CRP & dietary protein† |
eGFR ≥ 60 mL/min/1.73 m‡ |
No diuretic use§ |
Cystatin C-based eGFR¶ |
|
|---|---|---|---|---|
| Low Gait Speed# | ||||
| Number in analysis | 2631 | 2230 | 2122 | 2120 |
| Bicarbonate < vs ≥23 mEq/L |
1.43 (1.05 – 1.96) | 1.69 (1.21 – 2.36) | 1.50 (1.11 – 2.02) |
1.68 (1.13 – 2.50) |
| Low Peak Torque†† | ||||
| Number in analysis | 2095 | 1824 | 1739 | 1643 |
| Bicarbonate < vs ≥23 mEq/L |
||||
| Overall | 1.38 (1.08 – 1.74) | 1.68 (1.31 – 2.16) | 1.34 (1.06 – 1.68) |
1.44 (0.96 – 2.15) |
| White Women | 0.94 (0.47 – 1.88) | 1.30 (0.66 – 2.56) | 0.85 (0.43 – 1.68) |
0.63 (0.26 – 1.49) |
| Non-white women |
2.44 (1.23 – 4.83) | 2.77 (1.36 – 5.57) | 2.20 (1.14 – 4.27) |
2.07 (0.84 – 5.09) |
| Men | 1.49 (1.06 – 2.10) | 1.86 (1.22 – 2.84) | 1.53 (1.06 – 2.19) | 1.65 (0.96 – 2.84) |
Unless otherwise indicated, values shown are adjusted OR (95% confidence interval).
Abbreviations: OR, odds ratio; eGFR, estimated glomerular filtration rate; CRP, C-reactive protein.
Defined as the lowest sex-specific quartile of gait speed and peak torque, respectively.
Log-transformed C-reactive protein and dietary protein added to variables included in Model 3 from Table 3 (see below for adjustors)
Analysis (Model 3 from Table 3) restricted to participants with eGFR ≥ 60 mL/min/1.73 m2
Analysis (Model 3 from Table 3) restricted to participants reporting no diuretic use
Analysis (Model 3 from Table 3) performed with eGFR calculated using cystatin C, age, sex, and race
Models adjusted for age, sex, race/ethnicity, BMI, poverty, education, smoking, alcohol intake, activity level, performance of muscle strengthening activities, diuretic use by self-report, diagnosis of hypertension, arthritis by self-report, liver disease by self-report, eGFR <60 ml/min/1.73 m2, log-transformed urine albumin-creatinine ratio, serum albumin, serum calcium
Models adjusted for age, sex (for overall cohort), race/ethnicity (for overall cohort, non-white women and men), BMI, poverty, education, marital status, smoking, alcohol intake, activity level, diuretic use by self-report, diagnosis of hypertension, diagnosis of diabetes mellitus, cardiovascular disease by self-report, arthritis by self-report, thyroid disease by self-report, eGFR <60 ml/min/1.73 m2, log-transformed urine albumin-creatinine ratio, serum albumin, serum calcium, hemoglobin