Another person in a different facility said,
every day, facility staff told me with tears that they decided to evacuate.
One or two people a day.
I believe that it was a very hard decision to evacuate for
them after consulting with their families.
So, this facility member wanted to keep running the facility in the affected area,
but every day, was losing staff,
who had to concern themselves with their own families and children.
Another person said, after the accident, some families told me on the phone
that they would leave all care of their father or mother to us, and
then, they evacuated and left their parents behind.
Then some of us came to wonder why we had to take care of the residents whom their
family abandoned.
So here we begin to see immediately the breakdown of the kind of standard
public order that we expect in an ordinary society due to an earthquake and tsunami.
So, once we began our field work in Minamisoma, we realized that this had been
a big concern for many of the elderly care homes in the town.
We conducted a research project to understand exactly what happened.
So, we analyzed the evacuation and
mortality records from five of the eight elderly care homes in Minamisoma.
All of these homes evacuated after the Fukushima dai-ichi nuclear accident, but
they evacuated at different times and in different ways.
Our objective was to estimate the change in mortality due to evacuation
after the earthquake and nuclear accident, and also to try and
identify some facility-specific factors affecting mortality.
And this work was published in PLOS ONE in 2013.
So for this study, we collected facility entry, exit, and mortality records for all
of the five facilities from the period of 11th of March, 2006 to 10th of March 2011.
After the earthquake and the nuclear accident,
all of these facilities evacuated.
And they kept evacuation records.
So they kept records on where they went, when they went, and
the mortality of their residents during the period that they were evacuating.
So we collected those evacuation mortality records from the 11th of March 2011
to the 10th of March 2012.
Some facilities had shorter follow-up, but we used a method called Cox proportional
hazards regression to understand the mortality rates in this period of time.
This enabled us to compare pre- and post-earthquake time periods, and also to
incorporate the length of time that people were at risk of death due to evacuation.