How We Analyzed Where Older Californians Are at Increased Risk for Wildfire

The average age of those who died in the 2018 Camp Fire was 72. The average age of those who died in the fires that burned across Sonoma and Napa a year before: 73.

As wildfires grow more devastating in California, some of the most vulnerable are older people who live independently and who live in long-term care homes.

We wanted to examine the risk fire posed to older people because we know that California projects its population to grow grayer faster than the rest of the country.

There’s no comprehensive map of fire risk in California. Working with experts we created one for this analysis.

Risk analysis takes into account three qualities: a hazard, exposure to a hazard and a vulnerability to a hazard. To identify these qualities, we chose to analyze the geographic hazard of fire using two sets of shapefiles:

• The Fire Hazard Severity Zones show the physical conditions, like steepness or vegetation, that make it more likely for an area to burn. These are created by the California Department of Forestry and Fire Protection.
• Wildland Urban Interface areas include where development, geography and natural land mix to create fire risks. These areas were created by a team of scientists at the SILVIS lab at the University of Wisconsin.

We used three data sets to map the exposure of older and disabled people to wildfire hazard in certain regulated facilities:

• California Department of Social Services (CDSS) regulated facilities that house older people, specifically residential care facilities.
• California Department of Public Health (CDPH) regulated long-term-care facilities, specifically skilled nursing homes, intermediate care facilities and congregate living health facilities.
• Census tract level data from the 5-year 2018 American Community Survey showing how many people 65 and older live in each area.

To calculate the percentage of “risky” zones in each county, we used the “union” method to split the “risky” zones by county shapes and establish the percentage for each zone using the new shape’s area.

We also wanted to estimate how many people lived in those zones. To obtain the percentage of people 65 and older living in what we had established as “risky” zones, we filtered down both the Fire Hazard Severity Zones and Wildland Urban Interface to the categories experts associated with heightened risk and then combined the two shapes into one polygon, covering most of California. We then used the “union” method on Census tracts containing the ACS 5-year estimates and the counties’ shapes, essentially breaking down the tracts per county. We then used the resulting shapefile and again used “union” to get the parts of the tracts that fall into the “risky” zone. We then used the area of each shape to weight the population.

We used these same “risk zones” to calculate when long-term care facilities are situated in fire-prone areas.

To analyze how care facilities are prepared to handle fires or other disasters we requested data for fire and emergency violations over a two-year period from CDPH and CDSS, as well as other public records.

We picked the time period of November 15, 2017 through December 31, 2019 to account for federal regulations and a state law that expanded emergency preparedness requirements for long-term care facilities regulated by both agencies.

In November 2019, federal auditors from the Office of the Inspector General at the Department of Health and Human Services concluded that California should improve its oversight of skilled nursing facilities for wildfire emergency preparedness by looking at certain life safety code and emergency preparedness deficiencies.

Using the same “k-tags” and “e-tags” that federal auditors did for 20 facilities (including one that burned down in Paradise), we asked CDPH for deficiency records for more than 2,339 intermediate care and skilled nursing facilities. We made other public records requests for records related to that audit and for records related to individual facilities that have closed because they were destroyed or in areas that burned catastrophically.

For assisted living facilities, we requested records from CDSS related to new requirements for planning, training, and preparedness for disasters — including wildfire — as well as existing disaster and mass casualty, and training requirements.

Because we wanted to examine future risk in California, we analyzed data only for long-term care facilities that are currently open. That excludes several facilities that have burned down, some of which were found deficient, including in Paradise.

Meanwhile, the novel coronavirus is tearing through long-term care facilities throughout California. CDPH erratically provides data publicly on confirmed COVID-19 cases among staff and patients in skilled nursing homes. We included this data in our analysis so we could see how many facilities are already affected. Because CDSS masks data for assisted living outbreaks, we were unable to include them in this analysis, though their daily updated total deaths, available as a document, informs analysis about the percentage of COVID-related deaths in long-term care homes.

Older and Overlooked is an investigation by reporters in KQED Science and KQED News. KQED is the NPR and PBS member station for Northern California, serving audiences across the state on air and online with compelling stories that inform, inspire and involve. 

CalMatters’ Lo Bénichou contributed data analysis, editing and graphic design to this project. CalMatters.org is a nonprofit, nonpartisan media venture explaining California policies and politics. 

Capital Public Radio contributed photos to this series. As the NPR member station based in Sacrament, CapRadio serves California’s Capital region, Central Valley and Sierra Nevada.