In September 1971, the U.S. House of Representatives Subcommittee on Investigations and Oversight met in Covington, Kentucky, to discuss the accident experience and remedial procedures taken to alleviate the frequency of accidents occurring on approximately five miles of I 75 just south of Cincinnati. This section of highway had been and was continuing to be the subject of several safety improvements. The objective of this study was to conduct before-and-after investigations of the effectiveness of the following safety improvement projects: (1) five variable message signs within a two-mile section, (2) a New Jersey-type median barrier wall extending approximately four miles, and (3) a general safety improvement project throughout the study section which included upgrading all guardrail to current safety standards, extension of existing guardrail to fill in gaps, installation of buried end treatments for guardrail, attachment of guardrail to concrete bridge end railings, flattening of side slopes, leveling of gores where feasible, installation of breakaway bases on exposed lighting standards, elimination of butterfly sign supports in gore areas by replacement with new overhead trusses spanning the roadway, and installation of median guardrail at twin bridges.

The before-and-after analyses are based on accident statistics. The "before" statistics are for the 1969 calendar year: the "after" period began after installation of the variable message signs and also encompassed a full year (May 1, 1971, to May 1, 1972). The statistics do not indicate which of the features, singly or collectively, effectively reduced accident rates; the most obvious statistic was the elimination of crossing-the-median head-on collisions. This was the intent and purpose of the concrete barrier.

The variable message signs were used to warn motorists of impending hazardous driving conditions ahead. Displayed messages generally related to accidents, congestion, and poor driving conditions due to adverse weather. Benefits associated with the variable message signs appear to be significant: in the northbound direction (the direction where signs were used), there was a 16.1 percent decrease in the accident rates as compared with only a 1.7 percent reduction in the southbound direction.

Accident rates for the entire study section decreased significantly (95-percent confidence level) between the before and after periods. This is an indication that the combination of safety improvements and corrective design procedures were effective in reducing the accident rate.

Report Date


Report Number

No. 391

Digital Object Identifier



Prepared for Presentation to the 44th Annual Meeting of the Institute of Traffic Engineers (September 15-20, 1974: Detroit, Michigan).