Skip to main content

Table 3 Prediction of sepsis knowledge

From: Most patients with an increased risk for sepsis-related morbidity or death do not recognize sepsis as a medical emergency: results of a survey study using case vignettes

Predictors

Model 1

Model 2

Model 3

Model 4

β [95% CI]

Demographics-value

β [95% CI]

Demographics-value

β [95% CI]

Demographics-value

β [95% CI]

Demographics-value

Demographics

 Chronic disease: yes

2.428* [1.147, 3.709]

0.000

1.345* [0.111, 2.578]

0.033

2.659* [1.365, 3.953]

0.000

1.479* [0.216, 2.742]

0.022

 Age (per 10 years)

0.993* [0.477, 1.509]

0.000

0.496 [− 0.016, 1.008]

0.057

1.018* [0.500, 1.535]

0.000

0.520* [0.010, 1.030]

0.047

 Gender: female

2.219* [1.185, 3.253]

0.000

1.403* [0.402, 2.405]

0.006

2.170* [1.138, 3.201]

0.000

1.407* [0.406, 2.409]

0.006

Education level

Ref.: low

 Intermediate

0.586 [− 0.811, 1.982]

0.411

0.542 [− 0.782, 1.865]

0.423

0.608 [− 0.773, 1.990]

0.388

0.551 [− 0.769, 1.871]

0.413

 High

2.391* [0.984, 3.798]

0.001

2.388* [1.061, 3.715]

0.000

2.388* [0.995, 3.781]

0.001

2.385* [1.062, 3.708]

0.000

 Employment state: employed

0.844 [− 0.340, 2.027]

0.162

0.823 [− 0.330, 1.976]

0.162

0.755 [− 0.438, 1.948]

0.215

0.788 [− 0.370, 1.946]

0.182

Health information seeking behaviour

 Frequency of health information seeking behavioura

Ref.: < 1 × per month

 1 × per month

  

0.273 [− 1.033, 1.579]

0.682

  

0.294 [− 1.008, 1.597]

0.658

 1 × per week

  

1.505 [0.002, 3.008]

0.050

  

1.532* [0.026, 3.038]

0.046

 > 1 × per week

  

3.237* [1.674, 4.800]

0.000

  

3.268* [1.721, 4.814]

0.000

 Variety of sources of health information

  

0.587* [0.305, 0.868]

0.000

  

0.553* [0.268, 0.838]

0.000

Health literacy

    

0.779* [0.105, 1.452]

0.023

0.341 [− 0.309, 0.991]

0.303

 R2

0.063*

0.000

0.139*

0.000

0.071*

0.000

0.141*

0.000

 Incremental R2

–

 

0.076*

0.000

0.008*

0.023

0.078*

0.000

  1. Presented are the results of linear regression models to predict sepsis knowledge. β: Unstandardized Regression Coefficients. CI: Confidence Intervals. Analyses were based on n = 740 cases and applied full information maximum likelihood information to adjust for missing values. Four regression models were estimated to disentangle the interrelatedness of different sets of predictors. Incremental R2 measures the proportion of variance in sepsis knowledge, which is explained by the predictors in models 2 to 4 in addition to the variance explained by demographics (model 1)
  2. aFor frequency of health information seeking the categories “daily” and “1–3 times per week” were collapsed to the category “ > 1 × per week”
Back to article page

Critical Care

ISSN: 1364-8535