New Keratoconus Risk Factors: A Cross-Sectional Case—Control Study
Table 4
Risk factors of keratoconus in the eye rubbing population and multivariate logistic regression.
Characteristic
Cases N = 164
Controls N = 66
Univariate analysis
Multivariate analysis
OR
95% CI
value
OR
95% CI
value
Rubbing frequency
Once/day maximum
97 (59)
17 (26)
—
—
—
—
Often, 1–10 times/day
32 (20)
48 (73)
8.12
4.13–16.6
<0.001
9.48
4.62–20.6
<0.001
Very often, >10 times/day
35 (21)
1 (1.5)
52.1
10.4–951
<0.001
57.3
11.0–1,059
<0.001
Rubbing duration
<15 sec
34 (22)
6 (9.2)
—
—
≥15 sec
121 (78)
59 (91)
2.74
1.16–7.57
0.032
Intensity of rubbing
Superficial
59 (36)
32 (48)
—
—
Moderate
62 (38)
27 (41)
1.15
0.61–2.18
0.67
Hard
43 (26)
7 (11)
3.06
1.28–8.20
0.017
Rubbing noise
No
92 (56)
48 (73)
—
—
Yes
71 (44)
18 (27)
1.97
1.06–3.77
0.036
Type of rubbing
Index fingertips
No
103 (63)
24 (36)
—
—
—
—
Yes
61 (37)
41 (63)
0.34
0.18–0.62
<0.001
0.27
0.13–0.55
<0.001
The palm of the hand
No
138 (84)
54 (82)
—
—
Yes
26 (16)
11 (17)
0.89
0.41–2.01
0.77
Knuckles
No
73 (45)
44 (67)
—
—
Yes
91 (55)
21 (32)
2.46
1.35–4.59
0.004
Thumb and index clamp
No
141 (86)
56 (85)
—
—
Yes
23 (14)
9 (14)
0.91
0.40–2.22
0.83
Back of the hand
No
132 (80)
59 (89)
—
—
Yes
32 (20)
6 (9)
2.97
1.19–9.05
0.032
OR = odds ratio and CI = confidence interval. Data are n (%) or odds ratio (OR) and 95% confidence interval (CI). Odds ratios were calculated by univariate and multivariate logistic regression. The multivariate model was created with the variables with in the univariate analysis with a method of eliminating variables by using the corrected Akaike information criterion. A generalized variance inflation factor was used to check for multicollinearity.
