EpiCenter

EpiCenter is designed to be a flexible tool for viewing California data on injuries that result in death, hospitalization, or a visit to the emergency department (ED). Follow these three steps below to start exploring the data.
1. Select
Select the injury outcome from the blue panel at the top of the page. Outcomes include deaths, non-fatal hospitalizations, and non-fatal ED visits. Non-fatal hospitalizations tend to be for more severe injuries that require greater medical treatment compared to ED visits.

1. Filter: Filter the injury data using the drop-down menu on the left of the page (2-A). Filters are organized under the categories of person, place, time, injury, and treatment (for non-fatal injuries only). Click on a category to view/hide the filters it contains. Filters will be immediately applied.
2. Filter Description: A description of the current filters and the number of injuries displayed appears near the top of the page, under the blue panel (2-B).
3. Clear filters: To clear all filters, click on the “Reset Filters” button at the bottom of the filter panel on the left of the page (2-C).

Display the data using the three outputs listed as tabs under the description of the filtered data (Custom Table, Time Series, and Bar Chart).

1. Grouping: The data can be grouped by up to four variables using the drop-down menus above the table (3-A).
2. Reset groups: Click the "Reset groups" button to reset all grouping drop-down menus (3-B).
3. Show Rates: The table will display injury counts for each grouping level. Check the “show rates” box to also display incidence rates per 100,000 person-years (3-C).
4. Download Data: To download the table as an Excel file, click on the “Excel” download button below the table (3-D).

1. Group by: Group the data by variables from the drop-down menu above the graph (please note the graph is limited to displaying 10 or fewer groups) (4-A).
2. Injury Measure: Choose to display injury counts or incidence rates from the next drop-down menu (4-B).
3. Time Unit: Change the time unit from years to months or weeks using the third drop-down menu (4-C).
4. View Details: Hover over the graph to see values for each data point (4-D).
5. Download Image: To download the graph as an image (.png file), hover over the table and a camera icon will appear in the top right corner – click on it (4-E).
6. Download Data: To download the graph’s source data as a table (Excel file), click on the “Excel” download button below the graph (4-F).

1. Group by: Group the data by variables from the drop-down menu above the chart (please note the chart is limited to displaying 120 or fewer groups) (5-A).
2. Injury Measure: Choose to display injury counts or incidence rates from the next drop-down menu (5-B).
3. Time Unit: Change the bar order from the grouping variable (groups) order to the bar height (injury measure) order using the third drop-down menu (5-C).
4. View Details: Hover over the chart to see values for each bar (5-D).
5. Download Image: To download the chart as an image (.png file), hover over the table and a camera icon will appear in the top right corner – click on it (5-E).
6. Download Data: To download the chart’s source data as a table (Excel file), click on the “Excel” download button below the graph (5-F).

EpiCenter is designed to be flexible and support a wide a variety of injury data analyses. Here are three suggestions to get started.

Those working to improve the health of a specific population may benefit from knowing the causes of injury that contribute to the most morbidity or mortality among that specific population. The Custom Table display can provide that information. First, select the injury outcome of interest and filter the data to the population of interest. The example below looks at injury hospitalizations among children under the age of 18 for the year 2020 (6).

1. Select the injury outcome of interest and filter the data to the population of interest. The example below looks at injury hospitalizations among children under the age of 18 for the year 2020.
2. In the custom table display, choose to group the data by injury intent then by injury mechanism (6-A).
3. Next, sort the table by injury counts in descending order (6-B).

Intent by mechanism: Each intent-by-mechanism cause group consists of ICD-10 or ICD-10-CM codes that correspond to specific injuries. The example below builds on the leading cause of injury query above and looks at specific ICD-10-CM injury cause codes and descriptions for unintentional falls among children under the age of 18 for the year 2022.

4. To identify the specific codes/injuries within an intent-by-mechanism cause group, further filter the data by the injury intent (e.g., Unintentional) and mechanism (e.g., Falls) (7-A).
5. Then, group the table by Injury Cause ICD-10-CM Code and Injury Cause Description (7-B).
6. Sort the table by injury counts in descending order to see which specific codes/injuries contribute most to a given intent-by-mechanism cause group (7-C).

The COVID-19 Pandemic contributed to drastic changes in health outcomes beyond Coronavirus deaths and illness. The Time Series display can provide that information. First, select the injury outcome of interest and filter the data to the population of interest.

1. Select the injury outcome of interest and filter the data to the population of interest (but be sure to include 2020 data). The example below looks at injury deaths among all Californians.
2. In the time series display, choose month as the time unit (8-A).
3. The graph will show counts of injuries by month. Hover over the graph’s data points to see exact time periods and injury counts. You may notice considerable changes to injuries during the late winter and early spring of 2020 (8-B).

The Bar Chart display can provide insight on how injuries vary geographically across California. Comparing injury rates across counties can help determine if a specific county has a relatively high or low injury rate compared to other locations.

1. Select the injury outcome of interest (for this example, MVT-pedal cyclist and pedal cyclist, other) and filter the data to the population of interest (but be sure to include all counties).
2. In the bar chart tab (9-A)
   • Group by county of residence
   • Choose rate as the injury measure
   • Choose injury measure as the bar order
3. The chart will now rank California counties from those with the highest to lowest injury rates.

EpiCenter was developed by the California Department of Public Health (CDPH) Injury and Violence Prevention Branch (IVPB) with funding from the Centers for Disease Control and Prevention (CDC).

For additional injury data help, please email us at EpiCenter@cdph.ca.gov .

Injury deaths are identified from the California Department of Public Health (CDPH) California Comprehensive Master Death File (CCMDF). The CCMDF includes in-state deaths and deaths of California residents that occurred in other states or jurisdictions. It is derived from the death registration process and includes information from the death certificate, which is provided by informants and clinicians. More information can be found on the CDPH Vital Records Data and Statistics web page.

Injury hospitalizations and emergency department (ED) visits are identified from the California Department of Health Care Access and Information (HCAI) Patient Discharge Data (PDD) and ED Data, respectively. The PDD include records of inpatient discharge from California-licensed hospitals, including general acute care, acute psychiatric, chemical dependency recovery, and psychiatric health facilities. The ED data include records of patient face-to-face encounters with providers at hospitals licensed to provide emergency medical services. If an ED encounter resulted in a same-hospital admission, the ED encounter would be combined with the PDD record and only appear as a hospitalization. More information can be found on the HCAI Data Documentation web page.

Injury deaths are identified using International Classification of Diseases, 10th Revision (ICD-10) codes appearing in the underlying cause of death field. ICD-10 is a coding system used for classifying cause of deaths. The underlying cause of death field contains the ICD-10 code describing “the disease or injury which initiated the causal chain of morbid events leading directly to death or the circumstances of the accident or violence which produced the fatal injury.” In the case of an injury death, the underlying cause of death field must be coded with an external cause-of-injury code (Table 1). The EpiCenter case definition for injury deaths are deaths with any of the ICD-10 external cause of injury codes appearing in the underlying cause of death field. The EpiCenter injury death case definition is consistent with the Centers for Disease Control and Prevention (CDC) recommended surveillance case definition, which is described in detail in its State Injury Indicators Report.

Injury hospitalizations and ED visits are identified using ICD-10, Clinical Modification (ICD-10-CM) codes appearing in any of 25 diagnosis fields and 12 external cause of morbidity fields. The ICD-10-CM is a coding system used for classifying diagnoses and reason for visits in all health care settings. Diagnoses fields include a principal diagnosis field, which describes the condition established to be the chief reason for care, as well as up to 24 other diagnoses that describe conditions that coexist or develop during care and affect the treatment received. External cause of morbidity fields describe mechanisms that resulted in the most severe conditions.

The Epicenter case definition for injury ED visits are patient encounters with any of the ICD-10-CM injury diagnosis codes listed in Table 2 appearing in any of the 25 diagnosis or 12 external cause of morbidity fields -or- any of the ICD-10-CM external cause of injury codes listed in Table 3 appearing in any of the 25 diagnosis fields or 12 external cause of morbidity fields. The EpiCenter injury ED visit case definition is consistent with the CDC recommended surveillance case definition, which is described in detail in National Health Statistics Report Number 164.

The case definition for injury hospitalizations are patient discharges with any of the ICD-10-CM diagnosis codes listed in Table 2 appearing in the principal diagnosis field or any of the ICD-10-CM codes with external cause of injury information listed in Table 4 appearing in any of the 25 diagnosis fields or 12 external cause of morbidity fields.

The EpiCenter injury hospitalization case definition differs from the CDC recommended injury surveillance case definition, which is described in detail in National Health Statistics Report Number 125. The CDC case definition limits injury hospitalizations to those with an injury diagnosis code listed as the principal diagnosis only. This is intended to limit cases to injuries of sufficient severity to warrant hospitalization. Authors of the CDC case definition acknowledges that, “focus on particular types, mechanisms, or intents of injury might require the use of different selection criteria”. To facilitate focus on specific injuries, the EpiCenter case definition is more comprehensive. In addition to hospital visits with an injury diagnosis code listed as the principal diagnosis, it also includes those with an external cause of injury code appearing in any diagnosis or external cause of morbidity field. This expanded EpiCenter case definition is intended to serve those interested in understanding causes of injury, which can inform prevention. For example, many patients hospitalized with a self-harm injury receive a principal diagnosis of a mental health condition. Overlooking these injuries may lead to an underestimate of the role that mental health plays in self-harm injuries.

The EpiCenter case definition is the default, and is called “Comprehensive”. To apply the more conservative CDC case definition, select “Surveillance” under the hospitalization case definition filter.

Injury deaths are identified from the California Department of Public Health (CDPH) California Comprehensive Master Death File (CCMDF). The CCMDF includes in-state deaths and deaths of California residents that occurred in other states or jurisdictions. It is derived from the death registration process and includes information from the death certificate, which is provided by informants and clinicians. More information can be found on the CDPH Vital Records Data and Statistics web page.

• Age unknown
• Age greater than 119
• Iatrogenic (has ICD-10 code of Y40-Y84, or Y88)
• Nonresidents - deaths where state of residence is not CA

Injury hospitalizations and emergency department (ED) visits are identified from the California Department of Health Care Access and Information (HCAI) Patient Discharge Data (PDD) and ED Data, respectively. The PDD include records of inpatient discharge from California-licensed hospitals, including general acute care, acute psychiatric, chemical dependency recovery, and psychiatric health facilities. The ED data include records of patient face-to-face encounters with providers at hospitals licensed to provide emergency medical services. If an ED encounter resulted in a same-hospital admission, the ED encounter would be combined with the PDD record and only appear as a hospitalization. More information can be found on the HCAI Data Documentation web page.

• Age unknown
• Age greater than 119
• Encounters that were fatal
• Encounters for newborn
• Nonresidents - encounters where state of residence is not CA

Injuries are classified based on the external-cause-of-injury matrix, which classifies each ICD-10 or ICD-10-CM external cause of injury code by intent (column) and mechanism (row). The specific ICD-10 and ICD-10-CM codes assigned to each intent-by-mechanism cell within the matrix can be found on the CDC's Tools for Categorizing Injuries using ICD Codes Web Page. Table 5 illustrates the structure of the matrix and identifies injury intent-by-mechanism cells with one or more ICD-10 or ICD-10-CM codes.

The following variables are available for filtering and grouping injury data. Click on the variable to view the definition.

Deaths: For injury deaths, counts provided on EpiCenter represent individual California residents. Each injury death is assigned to one mutually exclusive injury intent-by-mechanism cause category.

ED Visits and Hospitalizations: For non-fatal injury hospitalizations and ED visits, counts provided on EpiCenter are not equal to individual California residents. The same person could experience two or more separate injury events that required a hospitalization or ED visit. Each hospitalization or ED visit for that same person would be counted as a separate injury. For example, a person who was hospitalized for a fall, was discharged to go home, and then fell again two weeks later and was hospitalized would be counted as two separate hospitalization injuries.

Additionally, injury ICD-10-CM codes may appear in 37 fields (25 diagnosis fields and 12 external cause of morbidity fields). This means a single hospitalization or ED visit could include multiple ICD-10-CM external cause of injury codes that correspond to different injury intents and mechanisms. For example, a person could be treated at the ED for an assault cut/pierce injury and an unintentional fall injury during the same visit. EpiCenter will count this single visit as multiple injuries when grouped by intent or mechanism. If injuries are grouped by intent, this single ED visit would be counted twice: once as an assault injury ED visit and once as an unintentional injury ED visit. Likewise, if injuries are grouped by mechanism, this single ED visit would also be counted twice: once as a cut/pierce injury ED visit and once as a fall injury ED visit. In all other instances it would be counted as a single injury ED visit.

Rates are calculated using population projections from the California Department of Finance’s Report P-3: Complete State and County Projections Dataset :

• Data years 2016 – 2019 use the Vintage 2020 release (2021.7.14).
• Data years 2020 – 2022 use the Vintage 2023 release (2023.7.19). Please note: prior to December 2023, rates for data years 2020 and 2021 were calculated using the Vintage 2020 release. Rates currently shown for these years may differ from those that appeared on EpiCenter prior to December 2023

The P-3 dataset includes population projections by year, age, sex, race/ethnicity, and county. Rates are not available for populations not included in this dataset, including education level, veteran status, expected payer (i.e., insurance type). Rates are also not available for other/unknown categories of populations included in the dataset (e.g., other/unknown sex, unhoused Californians). Additionally, rates are not available for hospital/ED facility county or region location because the source population (i.e., catchment area) is not clearly defined.

Due to differences in how race/ethnicity is reported in the population and injury data, caution should be used when interpreting rates involving race/ethnicity.

Because injuries are events that occur over time, incidence rates are used and expressed as injuries per 100,000 person-years. Person-years are the population multiplied by the number of years of observation. Incidence rates can be conceptualized as the number of events per 100,000 persons, per year. A thorough explanation of incidence rates can be found in Tenny & Boktor’s, "Incidence" StatPearls Article.

Asian and Pacific Islander race/ethnicity: Per Assembly Bill (AB) 1726, The Accounting of Health and Education in Asian and Pacific Islander (API) Demographics Act, CDPH is mandated to expand the number of API groups for which information is collected and reported. Detailed race/ethnicity data are available for death data starting in 2023. See the CDPH Asian and Pacific Islander Data Disaggregation page for more details. Due to small cell sizes and California Health and Human Services data de-identification guidelines data for some expanded API groups may not be available on EpiCenter.

Sex: California law requires that individuals completing death certificates accurately record a decedent’s sex to reflect the decedent's gender identity (“Respect After Death Act” AB 1577, 2014). In 2021, the law was expanded to provide a nonbinary option for gender identity on California death certificates (AB 439, 2021). Due to small cell sizes and data de-identification guidelines nonbinary is added to the unknown/other sex category on EpiCenter.

The injury data presented on EpiCenter are intended to provide a comprehensive count of injuries in California. There are, however, important limitations:
Deaths

• Injuries are identified through ICD-10 codes in the underlying cause of death field only. They do not include deaths where an injury was an intermediate or contributory cause of death. For example, if a person had a heart attack while driving and was subsequently fatally injured in a fall, the death would not be included because the heart attack would be considered the underlying cause of death.
• Data come from death certificate information provided by informants and clinicians and may be subject to error.
• Injury deaths are limited to California residents. They do not include injury deaths to non-residents that occurred in California.

• The data only capture injuries treated at state-licensed hospitals and emergency medical services facilities. They do not include injuries treated in the outpatient setting or at federally licensed facilities, such as Veterans Affairs (VA) hospitals. They also do not capture injuries that go untreated.
• Injuries are identified through ICD-10-CM code, and are therefore only as accurate as the medical coding used to populate the data. Injuries treated at state-licensed facilities may be missed or misclassified if not accurately coded.
• Beginning in 2019, the number of external cause of morbidity fields in the non-fatal hospitalization and ED visit data was expanded from 5 to 12. This may result in more injuries being recorded starting in 2019
• The data are limited to California residents treated in California. They do not include injuries to non-residents treated in California or injuries to California residents treated elsewhere.

Injury data are updated annually upon receipt of all source data files, usually in December of the following year.

Injury data prior to 2016 are still available on the legacy EpiCenter site . In the fourth quarter of 2015, the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) coding system replaced the ICD-9-CM coding system. These codes represent health conditions and are used to identify and classify injuries treated in the hospital or ED. The ICD-10-CM update resulted in changes to the way injuries were identified and classified that were not readily compatible with ICD-9-CM. Due to this coding system update, a new EpiCenter site was created and the original, legacy site was retired.

Because of the transition from the ICD-9-CM to ICD-10-CM coding system, comparisons of non-fatal injury hospitalizations and ED visits before and after 2015 are discouraged. However, there were no changes to the ICD-10 coding system used for fatal injury identification and classification. Injury deaths before and after 2015 can be compared.

For injury hospitalizations, EpiCenter gives the option of applying two different case definitions: comprehensive and surveillance.

• The comprehensive case definition is the default and includes hospitalizations with an injury diagnosis ICD-10-CM code listed as the principal diagnosis or an ICD-10-CM code with external cause of injury information appearing in any diagnosis or external cause of morbidity field. This comprehensive case definition should be used to capture hospitalizations with any mention of an injury cause. It is intended to inform prevention efforts that focus on understanding causes of injury.
• The surveillance case definition is limited to hospitalizations with an injury diagnosis code listed as the principal diagnosis only. This definition excludes cases where the injury alone may not warrant hospitalization. It should be used to exclude injuries that may not be treated in the hospital if not for some other health condition. Because it is consistent with CDC recommendations, it should also be used for comparison to other states or jurisdictions that use the CDC-recommended case definition.

For injury deaths and ED visits, only one case definition is available, which is consistent with CDC recommendations. For more information, see the injury case definitions section of the EpiCenter Methods page.

If an ED visit resulted in admission to the hospital at the same facility, the ED visit would be combined with the hospitalization record and only appear as a hospitalization. It may, however, be possible for a patient with an injury to be treated at an ED, then transferred to a different hospital. In this case, the same injury may appear in both the ED visit and hospitalization data.
Injuries treated in the ED or hospital with a fatal patient disposition are excluded and would only appear in the death data.

In an effort to use the most accurate population data to calculate injury rates, sometimes rates will be revised when new population projections become available. Rates for data years 2020 and 2021 were updated on EpiCenter in December 2023 using the latest population projections from the California Department of Finance, which were released in 2023 and are based on 2020 U.S. Census results.

The legacy EpiCenter site contains death injury data through 2020. The number of injury deaths that appear on the legacy EpiCenter site differ slightly from the new EpiCenter site for years 2016 through 2019. This is because upgrades to the process for applying exclusion criteria were applied retrospectively to those years of data when creating the new EpiCenter site. These upgrades better identified California residents and those with known ages <1 year (EpiCenter data exclude non-residents and those with an unknown age). The upgrades resulted in a net change of -32 injury deaths in 2016, -41 injury deaths in 2017, +3 injury deaths in 2018, and +1 injury deaths in 2019.

When non-fatal injury hospitalizations and ED visits are grouped by injury cause variables (intent, mechanism, ICD-10-CM code, or code description), the sum of the groups may be larger than the total number of unique visits shown at the top of the page along with the filter description. This is because injury ICD-10-CM codes, which are used for identifying and classifying injuries, may appear in up to 37 fields in the data. This means a single hospitalization or ED visit could include multiple ICD-10-CM codes that correspond to different injury causes. For example, a person could be treated at the ED for an assault cut/pierce injury and an unintentional fall injury during the same visit. EpiCenter will count this single visit twice when grouped by intent or mechanism. If injuries are grouped by intent, this single ED visit would be counted once as an assault injury ED visit and once as an unintentional injury ED visit. Likewise, if injuries are grouped by mechanism, this single ED visit would also be counted once as a cut/pierce injury ED visit and once as a fall injury ED visit. In all other instances it would be counted as a single injury ED visit

Person-years are the population multiplied by the number of years of observation. They are used to calculate incidence rates, which are the rate of new events that occur during a specified period for the population at risk for the event. The injury incident rates shown on EpiCenter can be conceptualized as the number of injuries per 100,000 persons, per year. A thorough explanation of incidence rates can be found in Tenny & Boktor’s, “Incidence” StatPearls Article.

• ED visit data from the California Department of Health Care Access and Information (HCAI).
• Patient Discharge data (hospitalizations) from the California Department of Health Care Access and Information (HCAI).
• Death certificate data from the CDPH Center for Health Statistics and Informatics (CHSI).