Industry-Regulatory Reports


In 2000, 2002, 2004, 2006, 2007, and 2017 Saferparks submitted public records requests to U.S. state safety agencies that regulate amusement rides, asking for data on ride-related accidents and injuries. A few additional requests were submitted during that time, targeted to specific agencies (including the U.S. Consumer Product Safety Commission) for specific purposes. The resulting records have been harmonized and incorporated into the Saferparks Database.

The processes, agency responses, data formats, and harmonization strategies were all non-uniform, resulting in a hodge-podge of U.S. accident data pertaining to amusement rides and devices.

  • Note that reporting criteria, level of detail provided, equipment types included, and years covered vary widely by year, jurisdiction, industry sector, and a variety of other factors.
  • In almost all cases, the accident descriptions were provided by the owner/operator of the ride, and may lack detail. Some records include information from regulatory officials based on investigation or discussion with the owner/operator.
  • In aggregate, the regulatory records provide some insight into the ways in which patrons are hurt using amusement rides and attractions, but the data set cannot be used to estimate injury rates or compare states, parks, rides, or years.


The bulk of the data is clustered around the early to mid 2000s and the early to mid 2010s. The Saferparks Database splits the aggregated records into two datasets: accidents that occurred prior to January 1, 2010, and those that occurred afterward.

  • 2017 Dataset - This includes all the data from the 2017 records request, except for a few older accident records provided by responding agencies that occurred prior to 2010. 28 states and the U.S. Consumer Product Safety Commission provided more than 8000 accident reports, primarily from 2011-2016. In a few cases, media reports of serious accidents were included because they happened to show up in a web search while researching government records, and the incidents had not been included in the public records sent to Saferparks.
  • Legacy Dataset - The Saferparks legacy data set has approximately 15,000 amusement ride accident reports from 30 agencies. Dates range from 1986-2009 with most records from years 1999-2007.

Saferparks' Research Goals

  • Inventory public accident records available from state and federal agencies - Quantity, quality, and ease of access to public safety records on amusement rides and devices vary wildly from state to state. This variability is evident in the data set. The more visibility we have into safety failures, the better the odds of avoiding repeat tragedies. States that make even minor incidents public create incentive for industry to correct problems early, before someone is seriously hurt. When consumers can see the specific types of injury risk associated with specific rides or devices, and which groups of people are most vulnerable, they can make more informed choices about which amusement experiences are right for themselves and their children. This inventory highlights the safety transparency, or lack thereof, for each U.S. state.
  • Learn about injury patterns in newer industry sectors - The amusement industry has experienced rapid expansion and innovation over the past decade, particularly in the area of participatory attractions (e.g., trampoline courts, challenge courses, etc.). Some states are now regulating those types of attractions and public records provide some visibility into the ways in which patrons can be injured.
  • Compare injury patterns over time - Saferparks legacy data set, collected during the early-to-mid 2000s, illuminated areas of risk, such as falls/ejections of young children in certain types of rides with certain types of containment systems. Industry and regulators have worked to reduce those risks through evolution of safety standards, ride designs, and operational policies. One goal of this research project was to see whether the 2017 data set indicates that incident patterns have changed as a result, and to look for new patterns in this rapidly evolving product space.


2017 Data Set
Legacy Data Set