Surviving collapsed structure entrapment after earthquakes: a "time-to-rescue" analysis

Prehosp Disaster Med. Jan-Feb 2006;21(1):4-17; discussion 18-9. doi: 10.1017/s1049023x00003253.


Introduction: Massive earthquakes often cause structures to collapse, trapping victims under dense rubble for long periods of time. Commonly, this spurs resource intensive, dangerous, and frustrating attempts to find and extricate live victims. The search and rescue phase usually is maintained for many days beyond the last "save," potentially diverting critical attention and resources away from the pressing needs of non-trapped survivors and the devastated community. This recurring phenomenon is driven by the often-unanswered question "Can anyone still be alive under there?" The maximum survival time in entrapment is an important issue for responders, yet little formal research has been conducted on this issue. Knowing the maximum survival time in entrapment helps responders: (1) decide whether or not they should continue to assign limited resources to search and rescue activities; (2) assess the safety risks versus the benefits; (3) determine when search and rescue activities no longer are indicated; and (4) time and pace the important transition to community recovery efforts.

Methods: The time period of 1985-2004 was selected for investigation. Medline and Lexis-Nexis databases were searched for earthquake events that occurred within this timeframe. Medical literature articles providing time-torescue data for victims of earthquakes were identified. Lexis-Nexis reports were scanned to select those with time-to-rescue data for victims of earthquakes. Reports from both databases were examined for information that might contribute to prolonged survival of entrapped individuals.

Results: A total of 34 different earthquake events met study criteria. Forty-eight medical articles containing time-to-rescue data were identified. Of these, the longest time to rescue was "13-19 days" post-event (secondhand data and the author is not specific). The second longest time to rescue in the medical articles was 8.7 days (209 hours). Twenty-five medical articles report multiple rescues that occurred after two days (48 hours). Media reports describe rescues occurring beyond Day 2 in 18 of 34 earthquakes. Of these, the longest reliably reported survival is 14 days after impact, with the next closest having survived 13 days. The average maximum times reported from these 18 earthquakes was 6.8 days (median = 5.75 days). The event with the most media reports of distinct rescue events was the 1999 Marmara, Turkey earthquake (43 victims). Times range from 0.5 days (12 hours) to 6.2 days (146 hours) for this event. Both databases provide little formal data to develop detailed insight into factors affecting survivability during entrapment.

Conclusions: A thorough search of the English-language medical literature and media accounts provides a provocative picture of numerous survivors beyond 48 hours of entrapment under rubble, with a few successfully enduring entrapment of 13-14 days. These data are not necessarily applicable to non-earthquake collapsed-structure events. For incident managers and their medical advisors, the study findings and discussion may be useful for post-impact decision-making and in establishing and/or revising incident priorities as the response evolves.

Publication types

  • Review

MeSH terms

  • Confined Spaces*
  • Disaster Planning
  • Disasters*
  • Humans
  • Rescue Work*
  • Resource Allocation
  • Survival*
  • Time and Motion Studies*