Reduction of occupational accidents: Evidence based prevention and the Prevention Index (PI TOP )

Monday, 18 August 2014
Exhibit hall (Dena'ina Center)
Frank Bochmann, PhD , Institute for Occupational Safety and Health of German Social Accident Insurance (IFA), Santk Augustin, Germany
Martin Arning, PhD , German Social Accident Insurance Institution for the Woodworking and Metalworking Industrie, Hannover, Germany
Yi Sun, MD , Institute for Occupational Safety and Health of German Social Accident Insurance (IFA), Sankt Augustin, Germany
Annette Nold, MS , Institute for Occupational Safety and Health of German Social Accident Insurance (IFA), Sankt Augustin, Germany
INTRODUCTION:  

Prevention Index (PITOP) is a practical measurement tool which is currently used by the German Social Accident Insurance Institution for the Woodworking and Metalworking Industries to monitor injury prevention activities at workplaces. This 12-item scoring system has 3 subscales rating technical, organizational and personnel-related internal safety conditions of a company.  Each item of the score has a rating value ranging from 1 – 9, with higher values indicating higher level of safety conditions. The reliability and validity of this instrument were evaluated in a cross-sectional survey in the German metal industry during the time between December 2011 and May 2012.   

METHODS:  

In the first step the inter-rater-reliability of this instrument was examined by 2 trained supervisors of the German Social Accident Insurance Institution for the Woodworking and Metalworking Industries in 128 companies. The agreement of the double ratings was quantified by interclass correlation coefficient (ICC) and absolute agreement of the rating values.

 In the next step the construct validity of the score was examined by principle component analysis (factor analysis) while content validity was evaluated by quantifying the association between PITOP-values and 5-years injury rates of 35,000 companies. Poisson regression analysis was performed to assess the strength of the association adjusted for company size and related industrial sectors.

RESULTS:  

Our analysis indicate a moderate to good inter-rater-reliability (ICC=0.46 – 0.75) of PITOP-values with an absolute agreement between 72% and 81%.

Factor analysis identified three component subscales which meet exactly the structural measure theory behind the score.

The Poisson regression analysis demonstrates that PITOP-values ≥7 in the T-, O- and P-subscales are associated with a decrease of injury rates.      

CONCLUSIONS:  

Because this analysis indicates that PITOP is a valid and reliable instrument, it will be used to monitor safety conditions at workplaces in a longitudinal practical approach.