North Carolina Injury Data User Toolkit

Coming soon

What are Provisional Data?

Provisional data are early numbers or datasets with the information that is available now.

• These data have not yet been fully checked.
• They don't have all the information they normally would have before being considered final.

Why Does IVPB Use Provisional Data?

Provisional data helps us to predict trends, so we know when to act and how.

Having access to the most up-to-date information is critical for responding to health problems, managing resources and planning public health prevention activities.

• Final data are the most accurate and complete, but it can take a long time to get all the information needed for the data to be considered final.
• Provisional data give helpful insights into what's going on so far.

What are Final Data?

Final data are data that have been reviewed and have been confirmed to be correct and complete.

They have all or most of the information needed for the data to be used in public health surveillance, analysis and reporting. This can refer to a full year of data (year-end data) or individual records that make up the data for a given year.

• Final Record - All available information on a single death or health event is collected.
  • A single record or group of records in the data can be considered final when the full year of data is still provisional.
• Final Year-End Data - Most of the information has been collected and processed for all records reported to a data system for a single year.
  • Sometimes there is information that is still unavailable after the normal data processing time has passed.
  • The year-end data are reviewed and then considered final and complete, even if some information is missing, so that the data can be used for analysis and reporting.
  • The data usually are not updated after the year-end data have been finalized.

Considerations for Using Provisional Data?

Provisional data should be used carefully because there may be missing information or delays in reporting.

• Because data are always being collected and updated, provisional data can change as new information is added.
• How complete provisional data are varies by data source, data system and when the data are accessed.
• The quality and completeness of the information can also be different for different groups of people in the same data source.

Whenever provisional data are used, a note should be included to share with audiences that the data are provisional. 

• This allows people using the data to know that the data can change so they can use the data appropriately.
• For example:
  • Report Title: North Carolina Emergency Department (ED) Visits for Opioid Overdose: April 2025* Update
  • Report Footnote: *Data for 2024-2025 ED visits are provisional.

Provisional death certificates may have a pending cause of death code (R99) listed for a record while the death is still being investigated. This cause of death code is usually updated to a code for a specific cause of death before the data are finalized.

Example Provisional Death Certificate Record

Example Final Death Certificate Record

Benefits and Drawbacks to Using Provisional Data

Using provisional data gives us information more quickly 

We can use provisional data to:

• Spot changes early, before the final data are completely ready.
• See patterns that help us know when to take action, even if numbers might not be exact.

Provisional data can change and might not be complete

The quality and completeness of the information can also vary for different groups of people and for different types of injuries.

• It takes longer to finalize data about violent deaths and overdose deaths because they need more time for investigation.
• To learn more about provisional overdose data, visit Using Provisional Overdose Data.

The quality and speed of provisional data is different across data sources and systems.

• Each data source or data system checks and finalizes data in its own way.
• This affects how complete provisional data are and how long it takes for a full year of data to be considered final.
• IVPB allows time to pass (lag-time) before using provisional data to make sure as much information is available for each record as possible.
• Provisional data are used until the year-end data are considered final.

Estimated Data Lag-Time in Months by Data Source

Availability of Data for Injury Surveillance by Data Source

Death Data

• Death Certificate Data
  • There is a longer wait for final death certificate data than other data sources because it takes time to investigate deaths, finish death certificates and enter updated data into the system.
  • Coding of death certificates currently takes longer in NC than in most other states. That is because of system changes as well as delays in autopsy reporting and pathology reviews for legal purposes.
    • It can take as long as 14 months from the end of the calendar year before a full year of data are considered final.
      • For deaths that occur in January, that is as long as 26 months before they are included in a final year-end data set.
    • IVPB waits about six months after a death occurs before using provisional data to make sure most records have been assigned to a specific cause of death code.
      • IVPB gets provisional death certificate data from the State Center for Health Statistics (SCHS) every month.
      • Provisional datasets include all deaths that have been reported so far.
      • Although a death may be reported in the provisional data set, important details like the cause of death code may not be available yet.
• NC Violent Death Data Reporting System (NC-VDRS)
  • NC-VDRS uses information from several different sources to help us understand violent deaths. It takes time to gather and review all the details.
    • There is a delay of about 16 months from the end of the calendar year before NC-VDRS data are ready.
  • Provisional data can be used once the most important information is added to the system, like manner of death, weapon type and victim demographics.
    • This timeline depends on when data become available from the data sources that are used to populate NC-VDRS.
    • Information on the circumstance surrounding the death takes the longest to collect since these fields are based on both medical examiner and law enforcement reports.
    • Toxicology data and information on firearm type also take longer to finalize.
• NC State Unintentional Drug Overdose Reporting System (NC-SUDORS)
  • NC-SUDORS uses information from several different sources to help us understand unintentional and undetermined overdose deaths. It takes time to gather and review all the details.
    • There is a delay of about 12 months from the end of the calendar year before these data are ready.
    • Provisional data can be added once the most important information is added to the system.
      • This timeline depends on when data become available from the data sources that are used to populate NC-SUDORS.

Non-Fatal Data

• Hospital Discharge Data
  • It takes about six to eight months from the end of the calendar year before a full year of hospital discharge data is considered final.
    • For hospitalizations early in the year, that could be a delay of as long as 20 months before records are available in a final year-end dataset.
  • Provisional data are used after about six months to be sure the information is as complete as possible for most of the provisional hospitalization records.
    • IVPB gets updated provisional data on hospital discharges from the NC Healthcare Association every three months (quarterly) via the NC SCHS.
    • These updates include all reported information for hospitalizations that have occurred so far for that year.
• Emergency Department (ED) Visit Data from North Carolina Disease Event Tracking and Epidemiologic Collection Tool (NC DETECT)
  • ED visit data from NC DETECT are among the timeliest data available for injury surveillance.
    • It usually takes about six months after a full year of ED visits occur before ED data are considered final.
    • Provisional ED visit data are used after about one month to be sure the information is as complete as possible.
      • ED visit data are also the most complete provisional data among the data sources IVPB uses.
      • Provisional data are available to authorized users within the NC DETECT data portal in near real-time.
      • IVPB gets updated provisional data on ED visits from NC DETECT each month.
      • Once all of the information for a record is entered into the system, it is unlikely the provisional ED visit data will change.
• Emergency Medical Services (EMS) Data from NC DETECT
  • Like ED visit data, EMS data are available much faster from other data sources for injury surveillance.
    • It usually takes about six months after a full year of EMS responses before EMS data are considered final.
    • Provisional EMS visit data are used after about one month to be sure the information is as complete as possible.
      • IVPB gets updated provisional data from NC DETECT each month.
      • Provisional data are also available to authorized users within the NC DETECT data portal in near real-time.

Step 1: IVPB receives data from partners to identify and monitor injuries across the state.

Data sources differ in availability and timeliness. Depending on the data source, IVPB can receive data on a monthly, quarterly or annual basis.

Below are the main data sources that IVPB uses.

• Data on injury deaths:
  • Death certificate data from the NC State Center for Health Statistics (SCHS)
  • Medical examiner reports from the Office of the Chief Medical Examiner (OCME)
  • NC Violent Death Reporting System (NC-VDRS)
  • NC State Unintentional Drug Overdose Reporting System (NC-SUDORS)
• Data on non-fatal injuries:
  • Emergency Department (ED) visit data from NC Disease Event Tracking and Epidemiologic Collection Tool (NC DETECT)
  • Emergency Medical Services (EMS) records from the Office of EMS via NC DETECT
  • Hospital discharge data (HDD) from the NC Healthcare Association via SCHS
• Survey data:
  • Youth Risk Behavior Survey (YRBS)
  • Behavioral Risk Factor Surveillance System (BRFSS)

Other data sources are also used to provide more context to injuries that occur in NC and related social drivers of health. To learn more about each source, visit Data Sources IVPB Uses for Injury Surveillance.

Step 2: IVPB processes the data to identify specific types of injuries.

Each data file IVPB receives is processed to customize the data for use in injury surveillance and is then stored to keep the information private and secure.

• The data are checked for the important details that help to identify and describe injuries.
• IVPB applies case definitions to the data to identify specific types of injuries.
  • Cases are identified and grouped by injury mechanism and intent.
  • IVPB takes time to understand how national case definitions work when used with NC data and has developed additional NC definitions.
    • To learn more about surveillance case definitions, visit Understanding Injury Surveillance Case Definitions.
    • To review the specific case definitions IVPB uses, visit North Carolina Injury and Violence Prevention Branch Injury Case Definitions.
• The data are organized into groups based on things like sex, age, race/ethnicity, location and other important factors to prepare the data for analyses.
  • IVPB explores how best to define and identify specific groups, locations and lived experiences in the data.
  • IVPB makes efforts to use categories to group people that are fair and accurate.
    • Fair – Assigned without bias
    • Accurate – Match the true characteristics or experiences of the groups

Step 3: Once the data are in a format that can be used for injury surveillance, they are analyzed.

Data are analyzed to understand who is impacted by injuries in NC, what types of injuries are happening, when and where injuries are happening, and why.

• IVPB uses data to monitor injury trends over time and identify groups and counties or regions of the state that are most impacted by injuries.
  • IVPB analyzes data by demographic factors like sex, age, racial/ethnic group, and by place (region, county or census tract).
  • Injury and violence data are also reported to the Centers for Disease Control and Prevention (CDC) to be used in national injury surveillance for all states.
• IVPB often incorporates data from additional data sources to understand risk and protective factors that can contribute to or help to prevent injuries.
• IVPB links data to other sources to help answer questions about injuries that cannot be answered from a single dataset.

Step 4: Data are summarized and shared to help inform response and prevention activities. 

 IVPB creates a variety of data resources including interactive data dashboards, fact sheets, slide decks and data reports.

• These resources summarize injury trends in NC to help explain who is being impacted and where injuries are occurring.
• These data are shared with public health leadership and with partners across the state to focus prevention efforts and guide public health decision making.
  • Resources are often developed in collaboration with state and local government, nonprofit, and university partners to be sure they include the information and prevention messaging most helpful for public health planning.
• Some resources are updated monthly or quarterly with provisional data, while other resources are updated annually after final data are available.
  • For more information visit How IVPB Uses Provisional Data.
• Data resources are also posted to the IVPB website for public use.

More Resources

• Visit the Interactive Data Inventory for a full list of IVPB data resources. This tool can be used to search for the data resources that include the type of data you need.
• IVPB also processes custom data requests from the media, elected officials, other agencies and the public for information that might not be found in a resource listed in the Data Inventory. For more information on how to request injury data and the timeline for the data request process, visit the Data Request Policy and Procedure page.
• For more information on considerations for analyzing and sharing public health data, visit Council of State and Territorial Epidemiologist Data Release Reference Document.

Coming soon

Coming soon

Injury mechanism and intent are concepts used to understand how and why injuries occur. These concepts help us to design programs that respond to injury problems and prevent future injuries.

Injury Mechanism

IVPB groups the codes used to identify injury deaths (ICD-10 codes in death certificate data) and non-fatal injuries (ICD-10-CM codes in morbidity data) into injury mechanism categories to help understand the type of injuries that occur.

Injury mechanism helps us:

• Explain how the injury happened.
• Describe what created the force that caused the injury.

Sub-causes of Injury

• Allow certain mechanism categories to be broken out to more specific causes of injury.
• Provide more details about how the injury happened. 

Injury Intent

Injury intent helps us:

• Describe why an injury happened
• Explain if the injury happened on purpose (intentional) or not (unintentional)

¹ CDC WISQARS

Using Mechanism and Intent Together 

Together, injury mechanism and intent help us to monitor different types of injuries and understand their causes to inform injury-specific prevention strategies.

For more information on injury mechanism and intent and the codes used to identify injuries, visit NCHS - ICD Injury Codes and Matrices and NCHS - Injury Tools and Frameworks.

What Is a Case Definition?

A case definition is a set of rules used to determine what counts when identifying cases of a certain health problem, like injuries.

Case definitions:

• Provide consistency in data by outlining what should and should not be included.
• Ensure injuries or other health events are counted the same way in different places and overtime.
  • This allows data for different groups or places to be compared.
• Allow us to know when there are meaningful changes in the data.

National Case Definitions

The IVPB applies national case definitions to NC data to identify injuries.

• National definitions are established through a formal process with the Council of State and Territorial Epidemiologists (CSTE) with input from subject matter experts from across the country.
• Data on injuries identified with national case definitions are often reported to the Centers for Disease Control and Prevention (CDC) to be used alongside data from other states and jurisdictions.
• National case definitions need to work for every state. At times this means better or more specific case definition criteria that are available to some states will not be part of national case definitions.

NC Case Definitions

IVPB has adapted national case definitions and has worked with partners across the state to develop new case definitions that better match our state’s data.

• State-developed definitions use special rules to consider nuances of NC data and more closely track injury cases.
• IVPB works to test these new definitions to make sure they correctly identify injury cases, but this can be a long and costly process.
  • Once tested, North Carolina’s definitions are given priority over national ones that haven’t been tested with NC data.
• IVPB uses both state and national case definitions to better understand injuries happening in NC.
  • Sometimes IVPB reports data differently on the same topic depending on their purpose.
  • For example, IVPB uses national definitions to complete reporting requirements for federal funders like CDC but uses state definitions for resources posted to the website and shared with NC partners.

To see a full list of the case definitions IVPB uses, visit North Carolina Injury and Violence Prevention Branch Injury Case Definitions.

Common Components of an Injury Case Definition

There are elements included in each case definition to outline what should be counted as a case.

Injury case definitions have different criteria depending on the data source being used and what kind of injury or health outcome is being counted.

Who

• Case definitions include criteria to identify events among specific groups of people and describe characteristics people have in common.
  • Most NC injury case definitions limit data to injury cases among people who are residents of NC.
    • This includes people who are NC residents but were injured somewhere else.
• Some injury case definitions limit who is included using criteria to identify populations at increased risk of injury. For example:
  • Limiting by age group – Self-inflicted injuries and suicides are limited to those ages 10 and older.
  • Limiting by sex – Some intimate partner violence (IPV) case definitions limit cases to females since they are statistically more likely to experience certain forms of IPV-related injuries.

What

Case definitions include clinical criteria to identify cases of a specific health event. For injuries, these criteria depend on the type of injury and the data source being used.

Below are ways IVPB identifies injuries in surveillance data.

• Cause of Death and Diagnosis Codes
  • These are codes used to group diseases and conditions within health care data sets. A particular code or ranges of codes are used to identify injury-related events in death certificate and morbidity data sets.
    • Death Certificate Data – International Classification of Diseases 10 (ICD-10) is a system used to classify and code the primary and underlying causes of death.
    • Morbidity Data – A coding structure based on ICD-10 coding with a clinical modification (ICD-10-CM) is used to classify diagnoses and reasons for health care visits.
      • ICD-10-CM codes are intended for administrative and billing purposes, not public health surveillance. This can affect which codes are or are not assigned to a record and therefore which events are included or excluded as injury cases.
  • For more information on the ICD-10 and ICD-10-CM codes used to classify injuries, visit ICD Injury Codes and Matrices (CDC) and Injury Tools and Frameworks.
• Keywords
  • Some case definitions search for keywords or phrases in free text data fields, like medical visit notes, to help identify cases. Keyword searches can include:
    • Root words, abbreviations, and misspellings to ensure the case definition captures as many relevant cases as possible
    • Words to exclude, or negations, to remove records that should not be considered as a case
  • Keywords are often included in syndromic surveillance case definitions.
    • Syndromic surveillance uses near-real-time health data to detect and monitor public health threats.
    • Syndromic surveillance case definitions search chief complaint fields and triage notes in ED visit data. Text fields like the narrative can be included in case definitions for EMS encounter data.
    • These definitions can include a combination of keywords and diagnosis codes.
    • Keywords are useful for identifying cases where ICD-10-CM codes may not have been assigned to the record yet or are not specific enough to define the case.

When

Case definitions limit cases to a specific timeframe.

• Most injury case definitions include cases that occurred within a calendar year.
• Case definitions can be applied to a shorter time frame (cases in the most recent month or quarter) or over multiple years.

Where

• Case definitions can include criteria to identify cases based on where they occurred.
  • Most injury case definitions are restricted to NC residents no matter where they were injured.
  • Some definitions identify cases among residents of a specific county or region of the state.
• Differences Between Occurrent and Resident Injuries and Deaths
  • Data Among Residents – Includes all resident cases, even if the injury or death occurred in another county or state in the US.
    • Limiting cases to NC residents allows us to calculate rates to better understand meaningful differences between populations.
  • Data by Occurrence – Includes all injuries in a specific location, regardless of a person’s residence status.
    • Counting cases based on where they happen helps us understand how they affect local hospital systems and services.
    • This can also be useful for areas with lots of people coming and going, including college towns, cities or counties that are common tourist destinations, or areas with large populations of people experiencing homelessness. 

Coming soon

Coming soon