Independent riskfactors for AK from multiple logistic regression models using the main analysis sample (213 controls and 63 AK cases). ORs are adjusted for possible confounding
| Exposures | Adjusted OR* | P values | 95% CI for OR |
| Lens disinfectant: classified by principal active ingredient (PAI)† | |||
| Polyhexanide (0.00005–0.0001%) | (Referent) | ||
| PQ+ALDOX/PQ+Alex‡ | 2.32 | 0.184 | 0.67 to 8.04 |
| Hydrogen peroxide3% | 1.77 | 0.534 | 0.29 to 10.83 |
| Polyquad-1(0.0001%)+Polyaminopropyl biguanide (0.00013%) | 0.80 | 0.851 | 0.08 to 8.03 |
| Oxipol | 4.74 | 0.001 | 1.83 to 12.30 |
| Rigid CL solutions | 7.34 | 0.011 | 1.57 to 34.24 |
| CL materials classification: simplified from the ANSI classification§ | |||
| Groups I+II+III | 2.39 | 0.313 | 0.44 to 12.98 |
| Group IV | 6.71 | 0.022 | 1.31 to 34.29 |
| Group VA | (Referent) | ||
| Groups VB+VC | 2.3 | 0.107 | 0.84 to 6.32 |
| Hygiene score: categories split between the top (worst) quartile of the sample and the rest¶ | |||
| Good–moderate (1.75–5.08) | (Referent) | ||
| Poor (5.09–8.08) | 3.34 | 0.003 | 1.52 to 7.38 |
| Hand washing: | |||
| Yes | (Referent) | ||
| No/uncertain | 3.65 | 0.004 | 1.52 to 8.77 |
| Shower wearing CLs: | |||
| No | (Referent) | ||
| Yes | 1.81 | 0.150 | 0.81 to 4.07 |
| Water activities wearing CLs: | |||
| None | (Referent) | ||
| In ocean/sea/river/lake | 1.45 | 0.519 | 0.47 to 4.45 |
| In public pool/private pool/hot tub | 3.49 | 0.003 | 1.51 to 8.04 |
| Ethnic group: for categories** | |||
| Other | (Referent) | ||
| British (white) | 4.82 | 0.005 | 1.61 to 14.46 |
| Occupation: for categories†† | |||
| Class 1+2+3: professional/director/senior official | (Referent) | ||
| Classes 4–9: | 3.51 | 0.003 | 1.52 to 8.11 |
*In constructing each final MV model, one exposure variable was considered as the ‘exposure of main interest’ and all others as ‘auxiliary factors’ (potential confounders). The final MV model was thus optimised to estimate the adjusted OR for the exposure of main interest’ (see online supplementary table 6 covariate adjustment).
†Classification of CL solutions is given in online supplementary table 2.
‡PAI category PQ+ALDOX/PQ+Alex=Polyquad-1 (0.001%)+ALDOX (0.0005%) or Polyquad-1 (0.0003%)+Alexidine (0.00016%)—the latter for 4 controls and 1 case only·
§American National Standard Institute (ANSI) standard Z80.20-2016. American national standard for ophthalmics—contact lenses—standard terminology, tolerances, measurements and physicochemical properties (ophthalmic).
¶Hygiene scores were calculated for the 14 hygiene variables listed in online supplementary table 5 using the rules described in that table. The mean score for each patient was then used to divide the patient sample into a simple binary classification for MV analysis in which the worst quartile (the quartile with the highest score) of the sample was categorised as having “Poor” hygiene and compared with the rest categorised as having “Good-Moderate” hygiene. “Good-Moderate” mean scores were 1-75-5.08 and “Poor” mean scores were 5.09-8.08. Hand washing before handling CLs and showering while wearing CLs were kept as separate variables and and analysed as such.
**Ethnic categories (UK census categories): 1: Asian or Asian British (Bangladeshi, Indian, Pakistani); 2: black or black British (African, Caribbean, other); 4: British (white); 6: other.
††Occupation: 1: managers, directors and senior officials; 2: professional occupations; 3: associate professional and technical occupations; 4: administrative and secretarial occupations; 5: skilled trades occupations; 6: caring, leisure and other service occupations; 7: sales and customer service occupations; 8: process, plant and machine operatives; 9: occupations requiring no specific training or skills and student: categorised by parents’ occupation.
AK, Acanthamoeba keratitis; CL, contact lens; MV, multivariate.