Investigation of Occupational Health and Safety Levels in Genetic Disease Centres in Istanbul

VEDAT CANER ^*

• Beykent University, Büyükçekmece, Türkiye

Genetic disorders significantly impact public health and quality of life, necessitating precise and timely diagnosis for effective risk management and treatment. Genetic diagnostic centres (GDCs) play a critical role in this process but face numerous occupational health and safety (OHS) challenges. Existing literature primarily classifies GDCs based on biosafety levels, which is insufficient to assess their comprehensive OHS conditions. This study aims to systematically evaluate OHS practices in GDCs and propose a new classification approach based on multidimensional hazard dimensions. A cross-sectional study was conducted in 15 GDCs in Istanbul, including two public and 13 private facilities with 75 employees. Data were collected through a structured survey with 49 statements covering seven hazard dimensions: physical safety, chemical safety, biosafety, laboratory safety, personal protective equipment, waste management, and risk management. Regression and correlation analyses were used to assess the impact and interrelationships of these dimensions on risk management. Principal Component Analysis (PCA) was applied for dimensionality reduction, and the k-Nearest Neighbours (k-NN) algorithm classified laboratories into safety levels. The findings revealed that personal protective equipment had the highest impact on risk management (56.3%), while physical security had the lowest (34.8%). Among the 21 identified hazard relationships, 18 were very strong, and 3 were strong. PCA reduced the data into three primary components explaining 81.9% of variance, and the k-NN algorithm achieved a classification accuracy of 93.33%, consolidating six hazard dimensions into three and categorizing centres into three safety levels. This study demonstrates that genetic disease diagnostic centres should be classified not only by biosafety levels but also by multidimensional security factors. It proposes a new OHS classification model using PCA and k-NN algorithms, showing that personal protective equipment and chemical safety are the most critical factors in risk management. Additionally, it introduces a security rating system that is compatible with current biosafety levels but more comprehensive. These findings emphasize the need for an updated OHS classification for GDCs beyond biosafety levels.