Skip to main content

Table 2 Multivariable analysis for each site of tuberculosis

From: Association of Underlying Comorbidities and Sites of tuberculosis: an analysis using surveillance data

 

Univariable

Multivariable

 

OR

95% CI

OR

95% CI

TB pleurisy

 Age

1.011

1.007–1.015

1.007

1.003–1.012

 Body mass index

1.047

1.027–1.068

1.054

1.032–1.076

 Chronic kidney disease

1.982

1.453–2.706

1.926

1.368–2.711

 Long-term steroid use

2.586

1.29505.162

–

–

 Cough/sputum

1.258

1.094–1.447

–

–

 Dyspnoea

5.857

5.050–6.793

5.134

4.366–6.036

 Chest pain

7.510

6.226–9.060

7.190

5.845–8.843

 Fever

1.760

1.464–2.117

1.598

1.304–1.958

TB lymphadenitis

 Age

0.968

0.963–0.972

0.967

0.962–0.972

 Female sex

3.115

2.542–3.817

3.407

2.765–4.197

 Body mass index

1.102

1.073–1.131

1.113

1.083–1.144

 Fever

0.498

0.337–0.735

0.462

0.310–0.688

 General weakness

0.288

0.128–0.648

–

–

 Weight loss

0.377

0.211–0.674

–

–

Abdominal TB

 Age

0.981

0.976–0.987

0.983

0.977–0.989

 Female sex

1.360

1.082–1.709

1.414

1.121–1.785

 Body mass index

1.062

1.029–1.096

1.066

1.032–1.100

 Diabetes

0.477

0.331–0.688

0.537

0.366–0.787

 Chronic liver disease

2.172

1.242–3.799

2.686

1.522–4.740

 Fever

0.593

0.386–0.913

0.593

0.385–0.914

 General weakness

0.417

0.185–0.942

–

–

Bone/joint TB

 Age

1.010

1.001–0.019

1.008

1.000–1.017

 Female sex

1.836

1.332–2.532

1.747

1.265–2.413

 Weight loss

0.273

0.087–0.858

0.305

0.097–0.961

Central nervous system TB

 Age

0.978

0.968–0.988

0.978

0.969–0.988

 Body mass index

1.128

1.073–1.186

1.127

1.072–1.186

 Malignant disease

0.206

0.051–0.836

  

 Long-term steroid use

3.925

0.935–16.47

–

–

 Fever

4.481

2.954–6.798

4.580

3.009–6.970

Urogenital TB

 Body mass index

1.123

1.050–1.201

1.119

1.047–1.196

 Chronic kidney disease

2.984

1.180–7.549

2.748

1.081–6.989

  1. Logistic regression was performed to calculate odds ratio and 95% confidence interval. Multivariable analysis included variables that were significant in univariable analysis, and final model was selected by backward elimination method