Identification of Substance-involved Emergency Department Visits Using Data From the National Hospital Care Survey

Questions for Amy M. Brown, Health Statistician and Lead Author of “Identification of Substance-involved Emergency Department Visits Using Data From the National Hospital Care Survey

Q: Why is this National Health Statistics Report (NHSR) important?

AB: The use of substances containing drugs or alcohol continues to be an important national health concern.  According to data from the Substance Abuse and Mental Health Services Administration (SAMHSA), in 2011, an estimated 2.5 million emergency department (ED) visits resulted from medical emergencies involving drug misuse or abuse.  This paper presents two approaches (algorithms) to identify substance-involved ED visits using administrative claims data submitted to the National Hospital Care Survey. The ability to identify substance-involved ED visits will allow the National Center for Health Statistics (NCHS) and researchers to track and characterize these visits, including services provided, demographics, and co-morbidities.  


Q: What are the differences between these algorithms?  

AB: The two algorithms are termed ‘general’ and ‘enhanced.’ Both use selected diagnoses and external cause of injury codes. The general algorithm can be used to monitor trends in the number of ED patients with any record of substance use (either recent or past history). The enhanced algorithm adds codes for substance use-related symptoms and procedures and was designed to meet a more specific case definition to identify ED visits involving recent substance use that was related to the reason for visit.


Q: Which substances can be identified by the algorithms? 

AB: The general and enhanced algorithms can be used to identify 10 substance categories:  alcohol (under age 21); antidepressants; antipsychotics; benzodiazepines or sedatives; cannabinoids; cocaine; hallucinogens; heroin; opiates or opioids; and pharmaceutical central nervous system stimulants. 


Q: What was found when these algorithms were applied to survey data?

AB: For demonstration purposes, both algorithms were applied to unweighted data from the 2013 National Hospital Care Survey. Overall, the general algorithm identified 81% more ED visits involving at least one of the priority substance categories compared with the enhanced algorithm.  However, the relative percent difference in the number of ED visits identified between the general and enhanced algorithms varied widely depending on the type of substance involved, ranging from 28% for antidepressants to 120% for cannabinoids.

The percent distributions of patient sex, age, and expected source of payment across all substances were similar between the general and enhanced algorithms.  In contrast, there were differences in discharge status distributions between both algorithms across all substances.


Q: What is the take home message of this report?

AB: Two algorithms are described that search for selected standard medical codes in administrative claims to identify ED visits involving the use of selected substances. NCHS plans to continue refining the algorithms to incorporate additional data elements available in the growing volume of submitted electronic health record (EHR) data, such as clinical notes capturing patient statements regarding events leading up to an ED visit, positive blood or urine tests for specific substances, and types of medication administered or prescribed during the encounter. Once refined and formally validated to ensure accuracy, they can be used with National Hospital Care Survey data to eventually generate national estimates of substance-involved ED visits.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: