Validation of the STARS risk assessment tool for age-related macular degeneration in an Algerian population

2016 
Purpose A risk assessment tool for age-related macular degeneration (AMD) has been previously developed and validated in two European countries (Italy, France). The objective of the present study was to validate its use in an Algerian population. Methods The STARS (Simplified Thea AMD Risk-assessment Scale) questionnaire is a simple 13-item, self-administered questionnaire, including questions on socio-demographic characteristics, family history of AMD, systemic and ocular risk factors. We included 1,183 patients over 55 years of age from 23 Algerian ophthalmologists. Patients completed the questionnaire, and ophthalmologists performed a fundus examination for AMD diagnosis (early AMD if soft drusen and/or pigmentary abnormalities; late AMD if atrophic and/or neovascular AMD). The validity of the final score was determined using ROC curves and Hosmer and Lemeshow testing. Patients were classified as low (score 0–9), moderate (10–19) or high risk (scores of 20+). Results Complete data was obtained for 942 (79.6%) patients, among whom 53 (5.6%) had late AMD and 288 (24.1%) had early AMD. The area under the ROC curve was 0.80, indicating good discrimination. The Hosmer and Lemeshow test was not statistically significant (p = 0.60), showing a good adaptation of the model to the data. Only 26 of the 662 (3.9%) patients without AMD were categorized as high risk and only 8 of the 53 patients (15%) with late AMD were categorized as low risk. Conclusions This study confirmed the validity of this short risk assessment tool, which shows good discrimination and high feasibility also in a North-African country. Early diagnosis of AMD is key to improve disease outcomes in patients. STARS is a promising questionnaire that allows early, simple and fast identification of patients at risk of AMD.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []