Skip to main content

Table 3 Factors associated with attrition in the ERMIES study (Cox regression analysis)

From: Attrition and social vulnerability during 2-year-long structured care in type 2 diabetes, the ERMIES randomized controlled trial

 

Univariate

Multivariate

Factor

Category/Class

HR

95% Cl

P value

HR

95% Cl

P value

Sociodemographic characteristics

 Age

   

0.23

   
 

18-60 years

Reference

     

≥ 60 years

0.61

[0.28 to 1.35]

    

 Sex

   

0.13

  

0.02

 

male

Reference

  

Reference

  
 

female

0.55

[0.25 to 1.19]

 

0.30

[0.11 to 0.81]

 

 Universal health coverage "CMU"

  

0.69

   
 

no

Reference

     
 

yes

1.17

[0.54 to 2.55]

    

 Education

  

0.24

   
 

Primary school (or less)

Reference

     
 

Middle school

2.13

[0.88 to 5.15]

    
 

Secondary school (or more)

1.53

[0.54 to 4.31]

    

 Cohabitation status

  

0.10

  

0.02

 

Living alone

Reference

  

Reference

  
 

Living with 1person

3.91

[1.09 to 14.05]

 

7.09

[1.84 to 27.37]

 
 

Living with 2 or more

2.43

[0.69 to 8.63]

 

3.32

[0.88 to 12.56]

 

 Gross monthly household income

  

0.08

  

0.01

 

<800 euros

Reference

  

Reference

  
 

800-1999 euros

0.34

[0.13 to 0.90]

 

0.17

[0.06 to 0.54]

 
 

>1999 euros

0.85

[0.34 to 2.16]

 

0.62

[0.22 to 1.72]

 
 

unknown

      

Clinical features

 BMI

   

0.70

   
 

<30

Reference

     
 

≥30

0.86

[0.39 to 1.87]

    

 Antidiabetic treatment

  

0.89

   
 

oral treatment

Reference

     
 

insulin +/- oral treatment

0.94

[0.42 to 2.14]

    

 Increased waist circumferencea

  

0.50

   
 

No

Reference

     
 

Yes

0.75

[0.33 to 1.72]

    

 Hba1c class

  

0.54

   
 

≤8.2

Reference

     
 

8.2-8.8

1.75

[0.56 to 5.53]

    
 

8.8-9.6

2.30

[0.75 to 7.04]

    
 

>9.6

1.60

[0.49 to 5.24]

    

Lifestyle 

 Dietary imbalanceb

  

0.15

   
 

No

Reference

  

-

-

-

 

Yes

0.57

[0.26 to 1.23]

    

 Physical activity at homec

  

0.34

   
 

Sedentary

Reference

     
 

Active

1.48

[0.67 to 3.27]

    

 Physical activity at workd

  

0.37

   
 

Unemployed

Reference

     
 

sedentary work

1.58

[0.62 to 4.02]

    
 

active work

0.53

[0.12 to 2.29]

    

 Smoking

  

0.54

   
 

Non-smoker

Reference

     
 

Smoker

1.67

[0.48 to 5.83]

    
 

former smoker

1.50

[0.65 to 3.47]

    

Health care access and pathway

 NEMT in past year

0.08

   
 

No

   

-

  
 

Yes

2.07

[0.93 to 4.6]

    

 Medical leave or hospitalization in past year

 

0.05

  

0.03

 

No

Reference

  

Reference

  
 

Yes

2.19

[1.00 to 4.79]

 

2.75

[1.13 to 6.72]

 

 Treatment compliancee

  

0.98

   
 

Good

Reference

     
 

Partially

1.11

[0.36 to 3.36]

    
 

noncompliant

1.05

[0.30 to 3.66]

    

 GP and diabetologist follow-up

  

0.82

   
 

<3consults/year

Reference

     
 

3-4 consults/year

1.05

[0.34 to 3.23]

    
 

> 4consults/year

1.38

[0.40 to 4.80]

    

Psychometric scales

 DQOL-BCI

    

 Satisfaction with diabetes control (6 items)

0.27

   
 

low f

Reference

     
 

High g

1.56

[0.71 to 3.44]

    

 Reported adherence with self-care regimen (6 items)

0.11

   
 

low f

Reference

  

-

-

-

 

high g

1.99

[0.85 to 4.69]

    

 CES-D: centre for epidemiologic studies depression scale

0.04

  

0.01

 

normal score

Reference

  

Reference

  
 

at risk for clinical depression

2.52

[1.05 to 6.06]

 

4.04

[1.35 to 12.02]

 

 HADS: Hospital Anxiety and Depression scale

0.80

   
 

normal score

Reference

     
 

probable mood disorder

0.90

[0.41 to 1.98]

    

MDQ Multidimensional diabetes questionnaire

 Outcome expectancies and social support in the self-care behaviours

0.18

   
 

low f

Reference

  

-

-

-

 

high g

1.83

[0.76 to 4.38

    

 Self-efficacy (7 items)

  

0.70

   
 

low f

Reference

     
 

high g

1.17

[0.53 to 2.57]

    

 Misguided support behaviours (4 items)

 

0.18

   
 

low f

Reference

  

-

  
 

high g

1.74

[0.78 to 3.85]

    

 Positive reinforcing behaviours (8 items)

 

0.67

   
 

low f

Reference

     
 

high g

1.19

(0.54 to 2.62]

    

Intervention status

 Randomization group

  

0.07

   
 

regular visits

Reference

  

-

  
 

intensive educational program

0.48

[0.21 to 1.07]

    
  1. Data are hazard ratios (HRs) with 95% confidence intervals (95% CIs) and P values for the global effect of the factor. All results were obtained after multiple imputation for missing baseline data.
  2. Abbreviations: HR Hazard ratios, 95% CI—95% Confidence interval, NEMT Nonemergency medical transportation, GP General practitioner, BMI Body mass index, IQR Interquartile range, DQOL-BCI Diabetes quality of life brief clinical inventor
  3. aWaist circumference > 88 cm (female) or > 102 cm (male)
  4. bdietary score (0–6), dietary imbalance if > 3
  5. cPhysical activity at home score (5–20), sedentary if < 13/20
  6. dPhysical activity at work score (5–20), sedentary if < 13/20
  7. eTreatment compliance score (0–6), 0 = good, 1–2 partially, >  = 3 noncompliant
  8. f< 50th percentile
  9. g> 50th percentile
Back to article page

BMC Endocrine Disorders

ISSN: 1472-6823

Contact us