Skip to main content

Table 4 Factors associated with missed assessments over 6-, 12, and 24-month follow-up after ARDSa

From: Factors associated with missed assessments in a 2-year longitudinal study of acute respiratory distress syndrome survivors

 

Univariable ORa (95% CI)

P-Valuea

Multivariable ORa (95% CI)

P-Valuea

Age

0.99 (0.97, 1.01)

0.231

  

Male

0.80 (0.47, 1.35)

0.406

  

Race, White

0.71 (0.42, 1.21)

0.210

  

No more than high school education

1.20 (0.69, 2.11)

0.517

  

Comorbidity

 Charlson Comorbidity Index

1.02 (0.91, 1.14)

0.715

  

 Functional Comorbidity Index

0.97 (0.80, 1.17)

0.744

  

 Alcohol abuse

0.48 (0.25, 0.93)

0.030

0.53 (0.28, 0.97)

0.040

 Drug abuse

1.35 (0.76, 2.40)

0.307

  

 Psychiatric comorbidity

1.43 (0.81, 2.54)

0.217

  

Baseline status prior to ARDS

 Unable to walk for at least 5 min

0.46 (0.21, 0.99)

0.048

0.46 (0.23, 0.91)

0.027

 Residing at Home without services

1.21 (0.51, 2.87)

0.674

  

 Unemployed due to health

0.59 (0.32, 1.09)

0.093

0.47 (0.23, 0.96)

0.039

 Number of dependent ADLsc

1.06 (0.87, 1.28)

0.571

  

 Dependent in any ADLsc

1.55 (0.76, 3.14)

0.226

  

 Number of dependent IADLsc

1.03 (0.94, 1.14)

0.512

  

 Dependent in ≥2 IADLsc

1.49 (0.85, 2.61)

0.162

1.96 (1.08, 3.52)

0.026

 EQ-5D-Visual Analogue Scale

1.00 (0.99, 1.01)

0.776

  

 EQ-5D-Utility score

0.82 (0.24, 2.79)

0.754

  

 SF-36 Physical Component Score

1.00 (0.98, 1.03)

0.915

  

 SF-36 Mental Component Score

1.00 (0.97, 1.02)

0.753

  

From Prior Follow-up Visit

 Visit Incomplete

2.07 (1.33, 3.23)

0.001

2.26 (1.35, 3.79)

0.002

 Entire visit conducted in clinic

0.55 (0.33, 0.91)

0.019

0.54 (0.31, 0.96)

0.035

 Living at home without services

1.35 (0.77, 2.37)

0.299

  

 Number of dependent ADLsc

1.04 (0.93, 1.17)

0.507

  

 Dependent in any ADLsc

1.05 (0.63, 1.75)

0.850

  

 Number of dependent IADLsc

0.95 (0.87, 1.04)

0.297

  

 Dependent in ≥2 IADLsc

0.74 (0.47, 1.16)

0.194

0.80 (0.48, 1.33)

0.388

 Shortness of breath

1.06 (0.66, 1.72)

0.810

  

 HHIA-S Score

0.98 (0.95, 1.02)

0.343

  

 Unemployed due to Health

1.01 (0.62, 1.64)

0.973

  

 EQ-5D-Visual Analogue Scale

1.00 (0.99, 1.01)

0.441

  

 EQ-5D-Utility score, per .01 increase

1.00 (0.99, 1.01)

0.772

  

 SF-36 Physical Component Score

1.01 (0.99, 1.03)

0.341

  

 SF-36 Mental Component Score

1.00 (0.98, 1.02)

0.887

  

 HADS Anxiety Score

0.99 (0.95, 1.04)

0.761

  

 HADS Anxiety Score ≥ 8

1.08 (0.66, 1.76)

0.765

  

 HADS Depression Score

1.01 (0.95, 1.08)

0.759

  

 HADS Depression Score ≥ 8

0.91 (0.54, 1.53)

0.721

  

 IES-R Total Score

0.92 (0.71, 1.20)

0.556

  

 IES-R Total Score ≥ 1.6b

0.76 (0.42, 1.40)

0.385

  

 6-min Walk Test, % Predicted

1.00 (0.99, 1.01)

0.490

  

 MMT Score

1.01 (0.97, 1.04)

0.722

  

 Grip, % Predicted

1.00 (0.99, 1.01)

0.588

  

 MIP, % Predicted

1.00 (0.99, 1.01)

0.982

  
  1. Abbreviations: ADL Activities of daily living, EQ-5D EuroQOL-5D, IADL Instrumental activities of daily living, IES-R Impact of event scale-revised, HADS Hospital anxiety and depression scale, HHIA-S Hearing handicap inventory for adults-screening, ICU Intensive care unit, LOS Length of stay, SF-36 Short-form 36, MCS Mental component score, MIP Maximal inspiratory pressure, MMT Manual muscle testing, PCS Physical component score, VAS Visual analog scale
  2. aEstimates and p-values were calculated with binomial regression model using generalized estimating equations with an exchangeable correlation structure. Bivariable models included the exposure presented in each row and indicator variables for time. Variables significant at the p < 0.20 level were selected for the multivariable model. Multivariable models included all factors presented in the column and indicator variables for time. If a model contained data from a prior visit, the model also included a missing data indicator for that variable
  3. bVariable had a significantly different association with missed assessments over time. However, when the variable and interactions with time were added to the final multivariable model, results remained consistent. Therefore, these variables are excluded from the multivariable model
  4. cIf two definitions of the same variables were significant a p < 0.20 (example: > = 1 ADL and No. of ADLS), the one with the stronger association (i.e. lower p-value) was used in the multivariable model
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us