More than 55 million adults in the United States use cannabis or other drugs, and approximately 7% meet DSM-5 diagnostic criteria for substance use disorder (SUD). However, the number of patients that receive such a diagnosis is much lower (0.8-4.6%). Low rates of diagnosis make it harder for patients to receive treatment and other supports to help them reduce or stop their use.
Brief, validated screening instruments help primary care providers assess SUD symptoms and patient treatment needs. They often ask about frequency of cannabis and other drug use and are effective for screening for SUDs. However, while these brief screeners have been validated for SUD diagnosis generally, they don’t provide information on DSM-5 SUD symptoms and they also typically haven’t been validated in general medical settings and electronic health record systems (EHRs).
This study aimed to evaluate the clinical use of a new screener, the Substance Use Symptom Checklist, using data collected as part of routine care after Kaiser Permanente Washington began using it to screen patients who reported daily cannabis or other drug use.
This study used EHR data from Kaiser Permanente Washington, a large integrated system providing primary care across 32 clinics in Washington State. Beginning in 2015, adult patients were asked to complete a 7-item behavioral health screen as part of routine care. The screen included separate questions for cannabis and other drug use. Staff entered responses in the EHR, and if patients reported “daily or almost daily” cannabis or other drug use, the EHR then prompted staff to ask patients to complete the Symptom Checklist as a follow-up screening.
The Symptom Checklist includes 11 items corresponding to SUD criteria in the DSM-5. Researchers examined data pulled from the EHR to see whether the checklist was effective at measuring substance use disorder symptoms and determining their severity. They also examined whether the checklist performed similarly across age, sex, race, and ethnicity.
After analyzing data from 23,304 screens, researchers found that for all cannabis and drug subsamples, the Symptom Checklist was able to identify symptoms of substance use disorder and discriminate between higher and lower levels of severity. The checklist also appeared to work consistently across all analyzed demographic groups.
Symptom Checklist results also revealed some interesting data about substance use and SUD prevalence: Of the 23,304 screens examined, 16,140 patients reported daily cannabis use only, 4,791 reported other drug use only, and 2,373 reported both daily and other drug use. Patients who endorsed 2 or more items on the Symptom Checklist (suggesting possible SUD) included: 4,242 (26.3%) of those who reported daily cannabis use only, 1,446 (30.2%) reporting other drug use only, and 1,229 (51.8%) reporting both daily and other drug use.
Measurement is essential for improving care for patients using cannabis and other drugs, and being able to effectively screen for substance use disorders in primary care settings could make a real difference in reducing barriers to treatment for people with SUD. This study found that the Substance Use Symptom Checklist performed well across patient demographics and can be used in primary care as an effective tool to help clinicians elicit patient symptoms, identify a spectrum of substance use disorder severity, and make informed clinical and treatment decisions.