SAMHSA’s DAWN Increases Data Options

By Rebecca A. Clay

Since 1972, the Drug Abuse Warning Network (DAWN) has monitored trends in drug-related emergency room visits and drug-related deaths. Now SAMHSA’s national surveillance system has had an extreme makeover. “Basically, only DAWN’s name has remained the same,” said Judy K. Ball, Ph.D., M.P.A., DAWN Team Leader in SAMHSA’s Office of Applied Studies.

To reflect the dramatic changes over the last three decades in both the Nation’s demographics and its health care system, DAWN recently underwent a multi-year process of evaluation and redesign. The result is a system that’s more useful than ever before to the Federal agencies, state and local governments, pharmaceutical companies, hospitals, and others that rely on these data.

Back to Top

Redesigning DAWN

Initially a project of the Drug Enforcement Administration, DAWN became SAMHSA’s responsibility in the early 1980s. For a long time, DAWN didn’t change much. But by the mid-1990s, the Agency took a close look at whether DAWN was still reflecting the Nation’s substance abuse issues accurately.

“The health care system had changed a lot, and so had our country’s population,” explained Dr. Ball, noting that Americans were continuing to migrate south and west. “It was clear we needed to take a look at DAWN and see how it could be improved to reflect those and other changes.”

To find out, SAMHSA convened a panel of experts in 1997 to consider the question. DAWN was worth continuing, the panel concluded, but needed major changes to stay relevant. A 2-year assessment of every aspect of DAWN followed. Based on the resulting recommendations, a totally new protocol was introduced in January 2003.

The new DAWN differs from the old network in three major ways.

First, there’s a new sample of hospitals that better reflects the Nation’s changing demographics and health care system. Because it would be too expensive to collect data from every emergency room across the country, since the mid-1980s DAWN has relied on a scientific sample of emergency rooms across the Nation and in selected metropolitan areas.

The new sample better reflects the Nation’s shifts in population. It includes cities such as Houston, for example, one of the Nation’s largest. And the sample now covers the entire United States (previously, Alaska and Hawaii were excluded). Because DAWN doesn’t use a sample for medical examiners and coroners, it is now expanding its recruitment efforts to cover the entire area where there are hospital samples. In addition, DAWN is adding medical examiners and coroners for a number of states.

Second, the new DAWN defines eligible cases more broadly than the old DAWN, which collected data about substance abuse cases only, and defined substance abuse in narrow terms. In other words, the only cases that counted were ones in which medical charts documented that patients had used drugs because of dependence, a suicide attempt, or the desire to achieve a “high.”

That also meant that DAWN was missing a lot of cases that should have been included. “That kind of specificity about why a patient used a drug is something that’s often missing from medical charts,” explained Dr. Ball. “Why the patient took the drug may not be clinically relevant. Also, in some states, insurers can deny payment for emergency department visits associated with substance abuse. That is a real disincentive for writing it down.”

Casting its net more broadly, DAWN now collects data about all kinds of drug-related emergency department visits, whether they’re due to illegal drugs, prescription or over-the-counter medications, dietary supplements, or non-pharmaceutical inhalants.

And third, DAWN now uses a new “case-finding” technique. DAWN “reporters”—the people who collect the data—used to rely on shortcuts to find the eligible cases. To avoid having to review every patient’s chart, they would check logs or use billing codes to find cases likely to be related to drug abuse. SAMHSA’s evaluation revealed that such shortcuts missed a substantial number of cases. Now DAWN reporters review the charts of every single patient who is treated in the emergency room.

“Drug-related cases in emergency departments are not terribly frequent relative to the total number of visits,” explained Dr. Ball. “Basically, we’re looking for a needle in a haystack. Under the old DAWN, we only looked through part of the hay. Now we know that if we’re going to find those needles, we have to go through the whole haystack.”

Back to Top

Putting the Data to Use

Although the Public Health Service Act requires SAMHSA to collect the data, there’s no law requiring emergency rooms, medical examiners, or coroners to participate in DAWN. Since January 2004, SAMHSA has met with more than 200 hospitals at 23 community meetings convened as part of a major recruitment drive designed to encourage participation.

The compensation hospitals receive for their participation—not enough to cover the full costs, Dr. Ball said—isn’t the only benefit they receive. One of the major advantages is immediate access to their data via a new feature called DAWN Live! Authorized users can log on to see how many drug-related cases their emergency room is treating, what drugs are involved, and where the patients ended up. “That information wasn’t necessarily available to them before,” she added.

One hospital used DAWN data to detect a sudden surge in the number of semi-comatose young people landing in the emergency room, identify the little-known drug they were using, and train emergency department staff how to respond. Other hospitals use the data to determine how many patients are admitted as inpatients and where they’re admitted. This information allows them to make more informed decisions about resource allocation. Others have used the data to demonstrate the need for specialized substance abuse services.

Take the emergency room at Boston University, for example. Faced with a cocaine and heroin epidemic in the mid-1990s, the department used DAWN data to win a grant from SAMHSA’s Center for Substance Abuse Treatment that funded a project to screen patients, provide brief interventions, and refer them to treatment. Once that grant ended, the department used the data again to convince the hospital to make the project a line item in its budget.

“The data are now much more sophisticated, and the information is right at your fingertips,” said Edward Bernstein, M.D., Professor and Vice Chair for Academic Affairs in the emergency medicine department at Boston University School of Medicine. “Before, for example, we could find ‘seeking detox.’ Now we have data showing whether someone’s actually referred or not.” Data also allow the department to track its progress compared to its own past performance or to emergency rooms in Boston or across the Nation.

Hospitals aren’t the only ones that use these data. Federal agencies are one of the primary users, said Dr. Ball. Community epidemiologists supported by the National Institute on Drug Abuse, for example, use the data to assess substance abuse problems in different parts of the country. The Food and Drug Administration uses DAWN data to track adverse reactions to prescription drugs and look at the potential for abuse. And municipalities involved in the Office of National Drug Control Policy’s 25-Cities Initiative are using DAWN data as a way of measuring progress.

The only downside to the new DAWN is that data collected under the new protocol cannot be compared to data collected under the old network. “It’s a short-term problem,” said Dr. Ball. “You don’t take a statistical series that’s been going on for many years and break it without a good reason. We decided the benefits of making the changes far outweighed the costs.”

For more information about SAMHSA’s Drug Abuse Warning Network, visit dawninfo.samhsa.gov.

Back to Top

Back to Graphic Version