Abstract
INTRODUCTION: Currently, age-related hearing loss has become prevalent, awareness and screening rates remain dismally low. Duing to several barriers, as time, personnel training and equipment costs, available hearing screening tools do not adequately meet the need for large-scale hearing detection in community-dwelling older adults. Therefore, an accurate, convenient, and inexpensive hearing screening tool is needed to detect hearing loss, intervene early and reduce the negative consequences and burden of untreated hearing loss on individuals, families and society. OBJECTIVES: The study harnessed "medical big data" and "intelligent medical management" to develop a multi-dimensional screening tool of age-related hearing loss based on WeChat platform. METHODS: The assessment of risk factors was carried out by cross-sectional survey, logistic regression model and receiver operating characteristic (ROC) curve analysis. Combining risk factor assessment, Hearing handicap inventory for the elderly screening version and analog audiometry, the screening software was been developed by JavaScript language and been evaluated and verified. RESULTS: A total of 401 older adults were included in the cross-sectional study. Logistic regression model (univariate, multivariate) and reference to literature mention rate of risk factors, 18 variables (male, overweight/obesity, living alone, widowed/divorced, history of noise, family history of deafness, non-light diet, no exercising habit, smoking, drinking, headset wearer habit, hypertension, diabetes, hyperlipidemia, cardiovascular and cerebrovascular diseases, hyperuricemia, hypothyroidism, history of ototoxic drug use) were defined as risk factors. The area under the ROC curve (AUC) of the cumulative score of risk factors for early prediction of age-related hearing loss was 0.777 [95% CI (0.721, 0.833)]. The cumulative score threshold of risk factors was defined as 4, to classify the older adults into low-risk (< 4) and high-risk (≥ 4) hearing loss groups. The sensitivity, specificity, positive predictive value, and negative predictive value of the screen tool were 100%, 65.5%, 71.8%, and 100.0%, respectively. The Kappa index was 0.6. CONCLUSIONS: The screening software enabled the closed loop management of real-time data transmission, early warning, management, whole process supervision of the hearing loss and improve self-health belief in it. The software has huge prospects for application as a screening approach for age-related hearing loss.