Skip to main content

Table 3 Variables significantly associated with screening performance

From: Ranked determinants of telemedicine diabetic retinopathy screening performance in the United States primary care safety-net setting: an exploratory CART analysis

Variable Label (n)

LS Group n = 51 (%)

HS Group n = 63 (%)

OR

95% CI

P-value#

Experience with TDRS?

    

0.008

  > 2 years (40)

11 (27.5)

29 (72.5)

3.1

1.4—7.1

 

 2 or fewer years (74)

40 (54.0)

34 (46.0)

REF

  

Frequency of use?

    

<.0001

 At least weekly (72)

20 (27.8)

52 (72.2)

16.5

4.4—61.8

 

 Monthly (19)

11 (57.9)

8 (42.1)

4.6

1.0—21.1

 

 Less than once per month (22)

19 (86.4)

3 (13.6)

REF

  

Why not: Patient Objection?

    

0.0003

 No (50)

32 (64.0)

18 (36.0)

0.2

0.1—0.5

 

 Yes (64)

19 (29.7)

45 (70.3)

REF

  

Champion: present, effective?

    

0.0545

 Yes, very effective (29)

8 (27.6)

21 (72.4)

2.0

0.8—5.3

 

 Yes, moderately effective or ineffective (16)

11 (68.8)

5 (31.2)

0.3

0.1—1.0

 

 Not present (64)

28 (43.8)

36 (56.2)

REF

  

Are patients compliant with screening guidelines?

    

<.0001

 0-25% (29)

23 (79.3)

6 (20.7)

0.1

0.0—0.2

 

 26-50% (29)

13 (44.8)

16 (55.2)

0.3

0.1—0.9

 

  > 50% (50)

11 (22.0)

39 (78.0)

REF

  

Effective reminder/alert in place?

    

0.0101

 Yes (59)

19 (32.2)

40 (67.8)

2.8

1.3—6.0

 

 No (51)

29 (56.9)

22 (43.1)

REF

  

Standing order?

    

0.0024

 Unsure (21)

17 (81.0)

4 (19.0)

0.2

0.1—0.7

 

 Yes (63)

21 (33.3)

42 (66.7)

1.5

0.6—3.8

 

 No (26)

11 (42.3)

15 (57.7)

REF

REF

 

Workflow present?

    

0.0014

 Yes (77)

26 (33.8)

51 (66.2)

3.9

1.7—9.1

 

 No (36)

24 (66.7)

12 (33.3)

REF

  

aExplicit positive instructions?

    

0.0554

 Always/Very frequently (58)

23 (39.7)

35 (60.3)

6.8

1.4—34.6

 

 Rarely/Never (22)

15 (68.2)

7 (31.8)

REF

  

aStaff autonomy?

    

0.0485

 Yes (51)

20 (39.2)

31 (60.8)

2.6

1.0—6.9

 

 No (27)

17 (63.0)

10 (37.0)

REF

  

bWhy not: Running behind?

    

0.0053

 No (11)

3 (27.3)

8 (72.7)

7.9

1.2—51.8

 

 Yes (21)

13 (61.9)

8 (38.1)

REF

  
  1. Abbreviations: TDRS telemedicine diabetic retinopathy screening, LS lower screening, HS higher screening, OR estimated odds ratio, CI confidence interval
  2. aResponses restricted to staff only (n = 82)
  3. bResponses restricted to providers only (n = 32)
  4. #P-values were obtained from univariate logistic regression model