Nonfatal Drug Overdoses

CDC’s Enhanced State Opioid Overdose Surveillance (ESOOS) Program:* Trends† in Emergency Department Visits for Suspected Drug Overdose^§ for Selected States Providing Data, Q4 2016 (October 1, 2016-December 31, 2016) to Q4 2017 (October 1, 2017-December 31, 2017), ^¶ by State

* Data come from states participating in CDC’s Enhanced State Opioid Overdose Surveillance (ESOOS) program (https://www.cdc.gov/drugoverdose/foa/state-opioid-mm.html). Every four months, states share overdose data from emergency department (ED) visits to CDC, including syndromic or hospital billing data to identify all drug, opioid, and/or heroin overdoses that presented in the ED and demographic characteristics of those who overdosed, such as sex, age, and county of patient residence. States have several options for how they relay their ED data to CDC. States choose to share ED visits for suspected overdoses (e.g., all drug, opioid, and heroin) either directly with CDC using a secure server or they can allow CDC to have access to their states’ data in the National Syndromic Surveillance Program’s (NSSP) BioSense platform ( https://www.cdc.gov/nssp/overview.html ). The number of states included in the calculations of quarterly and yearly change will vary and will increase over time as additional states share data with CDC. Comparisons between states should not be made due to variations in data quality, completeness, and reporting across states.

† To account for changes occurring across time, quarterly and yearly trends for the rate of ED visits involving suspected drug overdoses (e.g., ED visits involving drug overdoses divided by total ED visits and multiplied by 10,000) were analyzed by U.S. state. Yearly change, controlling for seasonal effects, was estimated as the change from the final quarter of previous year to the final quarter of the current year (e.g., fourth quarter 2016 to fourth quarter 2017). Quarterly rate changes were calculated for all quarters. Significance testing was conducted using chi-square tests. Data table provides the yearly and quarterly rate changes by state. Bolded estimates indicate statistically significant results between quarters.

§ The case definitions used by states draw from multiple fields within emergency department (ED) data. For states with ED billing or discharge data visits assigned International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes, International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes, or SNOMED (Systematized Nomenclature of Medicine — Clinical Terms; https://www.nlm.nih.gov/healthit/snomedct/ ) codes consistent with any drug, opioid, or heroin overdose were included. In addition, states using syndromic ED data could use the ICD-9-CM, ICD-10-CM, or SNOMED codes to identify overdoses along with free text fields such as chief complaint, clinical impression, or triage notes. Free text searches included text strings indicating an overdose or poisoning and text string indicating the involvement of a drug (e.g., “drug overdose”). Common misspellings of key search terms were also included to increase specificity and sensitivity. States either developed their own definitions to capture the specific text and diagnoses used in their facilities or relied on pre-set queries developed by CDC (see March Vital Signs: https://www.cdc.gov/mmwr/volumes/67/wr/mm6709e1.htm )

¶ The following are several important caveats to consider when interpreting the data presented: (1) Data sent from facilities to health departments may be delayed or may stop for a period of time. When facilities begin sharing data again, information about visits during the lapse may never be shared; (2) For syndromic data, information from ~70% of visits arrive within 48 hours as the chief complaint of the visit. However, the chief complaint field may be incomplete. As updates to visits arrive weeks later, relevant overdose discharge diagnosis codes or revised chief complaint text may be received. Therefore, rates may change over time as the visit records are completed and new drug overdose visits are identified; (3) Because these data are not finalized based on toxicological results, they are not considered confirmed cases, but “suspected” overdoses. Data collected from syndromic surveillance should not be interpreted or represented as exact counts; and (4) Data likely represent an undercount, given inaccuracies in coding and missing chief complaint information.

** The funded ESOOS state did not provide CDC enough quarters of data to calculate yearly percent change. Some states provided enough data to calculate some quarterly changes.

†† The funded ESOOS state does not provide CDC estimates for emergency department visits for suspected drug overdose.

§§ The funded ESOOS state does not provide CDC emergency department data.

CDC’s Enhanced State Opioid Overdose Surveillance (ESOOS) Program:* Percent changes † in All Drug Overdoses ^§ for Selected States Providing Data, ^¶ Q4 2016 (October 1, 2016-December 31, 2016) to Q4 2017 (October 1, 2017-December 31, 2017), by Sex and Age Group

* Data come from states participating in CDC’s Enhanced State Opioid Overdose Surveillance (ESOOS) program (https://www.cdc.gov/drugoverdose/foa/state-opioid-mm.html). Every four months, states share overdose data from emergency department (ED) visits to CDC, including syndromic or hospital billing data to identify all drug, opioid, and/or heroin overdoses that presented in the ED and demographic characteristics of those who overdosed, such as sex, age, and county of patient residence. States have several options for how they relay their ED data to CDC. States choose to share ED visits for suspected overdoses (e.g., all drug, opioid, and heroin) either directly with CDC using a secure server or they can allow CDC to have access to their states’ data in the National Syndromic Surveillance Program’s (NSSP) BioSense platform ( https://www.cdc.gov/nssp/overview.html ). The number of states included in the calculations of quarterly and yearly change will vary and will increase over time as additional states share data with CDC. Comparisons between states should not be made due to variations in data quality, completeness, and reporting across states.

^† To account for changes occurring across time, quarterly and yearly trends for the rate of ED visits involving suspected drug overdoses (e.g., ED visits involving drug overdoses divided by total ED visits and multiplied by 10,000) were analyzed overall and by sex, age group, and U.S. state. Quarterly rate changes were calculated for all quarters. Yearly change, controlling for seasonal effects, was estimated as the change from the final quarter of previous year to the final quarter of the current year (e.g., fourth quarter 2016 to fourth quarter 2017). Significance testing was conducted using chi-square tests. Data table provides quarterly and yearly estimates of change for all ESOOS states with available data overall, and by sex and age. Bolded estimates indicate statistically significant results between quarters.

^§ The case definitions used by states draw from multiple fields within emergency department (ED) data. For states with ED billing or discharge data visits assignedInternational Classification of Diseases, Ninth Revision,Clinical Modification (ICD-9-CM) codes,International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes, or SNOMED (Systematized Nomenclature of Medicine — Clinical Terms; https://www.nlm.nih.gov/healthit/snomedct/ ) codes consistent with any drug, opioid, or heroin overdose were included. In addition, states using syndromic ED data could use the ICD-9-CM, ICD-10-CM, or SNOMED codes to identify overdoses along with free text fields such as chief complaint, clinical impression, or triage notes. Free text searches included text strings indicating an overdose or poisoning and text string indicating the involvement of a drug (e.g., “drug overdose”). Common misspellings of key search terms were also included to increase specificity and sensitivity. States either developed their own definitions to capture the specific text and diagnoses used in their facilities or relied on pre-set queries developed by CDC (see March Vital Signs: https://www.cdc.gov/mmwr/volumes/67/wr/mm6709e1.htm )

^¶ The following are several important caveats to consider when interpreting the data presented: (1) Data sent from facilities to health departments may be delayed or may stop for a period of time. When facilities begin sharing data again, information about visits during the lapse may never be shared; (2) For syndromic data, information from ~70% of visits arrive within 48 hours as the chief complaint of the visit. However, the chief complaint field may be incomplete. As updates to visits arrive weeks later, relevant overdose discharge diagnosis codes or revised chief complaint text may be received. Therefore, rates may change over time as the visit records are completed and new drug overdose visits are identified; (3) Because these data are not finalized based on toxicological results, they are not considered confirmed cases, but “suspected” overdoses. Data collected from syndromic surveillance should not be interpreted or represented as exact counts; and (4) Data likely represent an undercount, given inaccuracies in coding and missing chief complaint information.

CDC’s Enhanced State Opioid Overdose Surveillance (ESOOS) Program: * Trends † in Emergency Department Visits for Suspected Opioid Overdose ^§ for Selected States Providing Data, Q4 2016 (October 1, 2016-December 31, 2016) to Q4 2017 (October 1, 2017-December 31, 2017), ^¶ by State

* Data come from states participating in CDC’s Enhanced State Opioid Overdose Surveillance (ESOOS) program (https://www.cdc.gov/drugoverdose/foa/state-opioid-mm.html). Every four months, states share overdose data from emergency department (ED) visits to CDC, including syndromic or hospital billing data to identify all drug, opioid, and/or heroin overdoses that presented in the ED and demographic characteristics of those who overdosed, such as sex, age, and county of patient residence. States have several options for how they relay their ED data to CDC. States choose to share ED visits for suspected overdoses (e.g., all drug, opioid, and heroin) either directly with CDC using a secure server or they can allow CDC to have access to their states’ data in the National Syndromic Surveillance Program’s (NSSP) BioSense platform ( https://www.cdc.gov/nssp/overview.html ). The number of states included in the calculations of quarterly and yearly change will vary and will increase over time as additional states share data with CDC. Comparisons between states should not be made due to variations in data quality, completeness, and reporting across states.

^† To account for changes occurring across time, yearly and quarterly trends for the rate of ED visits involving suspected opioid overdoses (e.g., ED visits involving drug overdoses divided by total ED visits and multiplied by 10,000) were analyzed by U.S. state. Yearly change, controlling for seasonal effects, was estimated as the change from the final quarter of previous year to the final quarter of the current year (e.g., fourth quarter 2016 to fourth quarter 2017). Quarterly rate changes were calculated for all quarters. Significance testing was conducted using chi-square tests. Data table provides the yearly and quarterly rate changes by state. Bolded estimates indicate statistically significant results between quarters.

^§ The case definitions used by states draw from multiple fields within emergency department (ED) data. For states with ED billing or discharge data visits assignedInternational Classification of Diseases, Ninth Revision,Clinical Modification (ICD-9-CM) codes,International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes, or SNOMED (Systematized Nomenclature of Medicine — Clinical Terms; https://www.nlm.nih.gov/healthit/snomedct/ ) codes consistent with any drug, opioid, or heroin overdose were included. In addition, states using syndromic ED data could use the ICD-9-CM, ICD-10-CM, or SNOMED codes to identify overdoses along with free text fields such as chief complaint, clinical impression, or triage notes. Free text searches included text strings indicating an overdose or poisoning and text string indicating the involvement of a drug (e.g., “drug overdose”). Common misspellings of key search terms were also included to increase specificity and sensitivity. States either developed their own definitions to capture the specific text and diagnoses used in their facilities or relied on pre-set queries developed by CDC (see March Vital Signs: https://www.cdc.gov/mmwr/volumes/67/wr/mm6709e1.htm )

^¶ The following are several important caveats to consider when interpreting the data presented: (1) Data sent from facilities to health departments may be delayed or may stop for a period of time. When facilities begin sharing data again, information about visits during the lapse may never be shared; (2) For syndromic data, information from ~70% of visits arrive within 48 hours as the chief complaint of the visit. However, the chief complaint field may be incomplete. As updates to visits arrive weeks later, relevant overdose discharge diagnosis codes or revised chief complaint text may be received. Therefore, rates may change over time as the visit records are completed and new drug overdose visits are identified; (3) Because these data are not finalized based on toxicological results, they are not considered confirmed cases, but “suspected” overdoses. Data collected from syndromic surveillance should not be interpreted or represented as exact counts; and (4) Data likely represent an undercount, given inaccuracies in coding and missing chief complaint information.

** The funded ESOOS state did not provide CDC enough quarters of data to calculate yearly percent change. Some states provided enough data to calculate some quarterly change.

^§§ The funded ESOOS state does not provide CDC emergency department data.f

CDC’s Enhanced State Opioid Overdose Surveillance (ESOOS) Program:* Percent Changes † in Opioid Overdoses ^§ for Selected States Providing Data, ^¶ Q4 2016 (October 1, 2016-December 31, 2016) to Q4 2017 (October 1, 2017-December 31, 2017), by Sex and Age Group

* Data come from states participating in CDC’s Enhanced State Opioid Overdose Surveillance (ESOOS) program (https://www.cdc.gov/drugoverdose/foa/state-opioid-mm.html). Every four months, states share overdose data from emergency department (ED) visits to CDC, including syndromic or hospital billing data to identify all drug, opioid, and/or heroin overdoses that presented in the ED and demographic characteristics of those who overdosed, such as sex, age, and county of patient residence. States have several options for how they relay their ED data to CDC. States choose to share ED visits for suspected overdoses (e.g., all drug, opioid, and heroin) either directly with CDC using a secure server or they can allow CDC to have access to their states’ data in the National Syndromic Surveillance Program’s (NSSP) BioSense platform ( https://www.cdc.gov/nssp/overview.html ). The number of states included in the calculations of quarterly and yearly change will vary and will increase over time as additional states share data with CDC. Comparisons between states should not be made due to variations in data quality, completeness, and reporting across states.

^† To account for changes occurring across time, quarterly trends for the rate of ED visits involving suspected opioid overdoses (e.g., ED visits involving drug overdoses divided by total ED visits and multiplied by 10,000) were analyzed overall and by sex, age group, and U.S. state. Quarterly rate changes were calculated for all quarters. Yearly change, controlling for seasonal effects, was estimated as the change from the final quarter of previous year to the final quarter of the current year (e.g., fourth quarter 2016 to fourth quarter 2017). Significance testing was conducted using chi-square tests. Data table provides quarterly and yearly estimates of change for all ESOOS states with available data overall, and by sex and age. Bolded estimates indicate statistically significant results between quarters.

^§ The case definitions used by states draw from multiple fields within emergency department (ED) data. For states with ED billing or discharge data visits assignedInternational Classification of Diseases, Ninth Revision,Clinical Modification (ICD-9-CM) codes,International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes, or SNOMED (Systematized Nomenclature of Medicine — Clinical Terms; https://www.nlm.nih.gov/healthit/snomedct/ ) codes consistent with any drug, opioid, or heroin overdose were included. In addition, states using syndromic ED data could use the ICD-9-CM, ICD-10-CM, or SNOMED codes to identify overdoses along with free text fields such as chief complaint, clinical impression, or triage notes. Free text searches included text strings indicating an overdose or poisoning and text string indicating the involvement of a drug (e.g., “drug overdose”). Common misspellings of key search terms were also included to increase specificity and sensitivity. States either developed their own definitions to capture the specific text and diagnoses used in their facilities or relied on pre-set queries developed by CDC (see March Vital Signs: https://www.cdc.gov/mmwr/volumes/67/wr/mm6709e1.htm )

^¶ The following are several important caveats to consider when interpreting the data presented: (1) Data sent from facilities to health departments may be delayed or may stop for a period of time. When facilities begin sharing data again, information about visits during the lapse may never be shared; (2) For syndromic data, information from ~70% of visits arrive within 48 hours as the chief complaint of the visit. However, the chief complaint field may be incomplete. As updates to visits arrive weeks later, relevant overdose discharge diagnosis codes or revised chief complaint text may be received. Therefore, rates may change over time as the visit records are completed and new drug overdose visits are identified; (3) Because these data are not finalized based on toxicological results, they are not considered confirmed cases, but “suspected” overdoses. Data collected from syndromic surveillance should not be interpreted or represented as exact counts; and (4) Data likely represent an undercount, given inaccuracies in coding and missing chief complaint information.