Alcohol detox: To admit or not?

Protocols can help simplify decision on admission.

More than 30% of American adults have abused alcohol or been dependent on alcohol at some point in their lives, according to the National Council on Alcoholism and Drug Dependence.

It's no wonder, then, that hospitalists frequently get calls from the emergency department about alcohol detoxification. Depending on a hospital's policies, the final determination about admission may rest on the hospitalist's shoulders, and it's not always an easy decision to make.

Photo from Thinkstock
Photo from Thinkstock.

“Some patients will clearly benefit most from outpatient treatment, while others are so far into withdrawal that it's obvious they need to be admitted. But it's the cases in the middle that are difficult and require a collaborative effort among emergency department physicians and hospitalists, keeping in mind safety and quality of care,” said Anil A. Thomas, MD, clinical director in the Division of Alcohol and Drug Abuse and unit chief of the Detoxification Clinical Service at Bellevue Hospital Center in New York City.

The benefits of protocols

A protocol for making this decision can smooth the process and provide hospitalists with the tools they need to justify their decisions, said Brett Kronenberger, MD, FACP, chief hospitalist at St. James Healthcare in Butte, Mont. “Having a protocol cuts back on disagreements between emergency department physicians and hospitalists. That means less work for chiefs, who will be getting fewer complaints from emergency department physicians and hospitalists about each other.”

Protocols may also help physicians present a united front to the patient or family, Dr. Kronenberger added. “Sometimes the emergency physician will call the hospitalist because the family or patient is being demanding [without the patient meeting the criteria for admission]. Having a written protocol sets some ground rules,” she said.

This can be particularly important for smaller hospitals with fewer beds and limited resources. “If we admitted everyone who came in requesting detox, we'd have a problem,” Dr. Kronenberger said.

Yet for all the benefits a protocol can provide, there is a dearth of guidelines specifically designed to help hospitalists decide whether to admit a patient for detoxfication. The most common alcohol assessment, the Clinical Institute Withdrawal Assessment of Alcohol (CIWA) scale, provides a snapshot of withdrawal symptoms, but it was developed with young, otherwise healthy patients in mind and does not account for comorbidities or frailty. Hospitals with mental health experts and ready access to both inpatient and outpatient addiction programs, such as Bellevue, have the resources to develop their own protocols for decision making, but smaller and community hospitals often do not.

Clinicians at the University of North Carolina (UNC) School of Medicine may have a solution: They have developed a protocol to standardize care for patients who come to their emergency department requesting alcohol detoxification, and the entire protocol is summed up in a 1-page flow chart. The team published the protocol and related research in the April 2014 Journal of General Internal Medicine, and the flow chart is available through the UNC website (see sidebar) .

“When we did a literature search, we found only one paper that described a protocol for evaluating patients for alcohol detoxification [published in the July 2004 Journal of Family Practice] and one study that directly compared inpatient to outpatient treatment for alcohol detoxification [published in the New England Journal of Medicine in 1989]; the New England Journal study used an outpatient program that was pretty intensive and not available in many institutions, including ours,” said John R. Stephens, MD, associate professor of hospital medicine. He noted that while the UNC protocol is grounded in evidence, it was also prompted by clinicians' experiences with many potential detox admissions and readmissions.

The UNC protocol accounts for the patient's current physical and psychiatric condition, level of intoxication, and history of alcohol use or withdrawal. Emergency department physicians perform the initial evaluation and consult hospitalists if they feel a patient should be admitted.

“We decided to involve the emergency department because that is where the referrals come through, and we wanted them to use the protocol as well,” said Dr. Stephens.

In the 21 months after protocol implementation, the team found that the number of alcohol-related admissions dropped from an average of 18.9 per month to an average of 15.9 per month. The average length of stay was slightly higher, at 3.4 days versus 2.7 days before implementation, which the researchers did not consider statistically significant. There was no change in readmission rate.

“We did not demonstrate any change in readmission rate as we had hoped, but we may have been selecting for more serious drinkers,” Dr. Stephens said. “We also don't know all of the outcomes for patients not admitted, as it wasn't possible for us to go through every emergency department patient seen with alcohol-related issues.”

Developing a protocol

Whether hospitalists adopt an existing protocol, such as UNC's, or develop their own to meet their hospital's needs and make the best use of available resources, experts agree that it should have several core elements.

Alcohol-related history. Assessment should begin with the patient's history, particularly with respect to alcohol, said Reid Finlayson, MD, medical director of the Vanderbilt Comprehensive Assessment Program and associate professor of clinical psychiatry at the Vanderbilt University School of Medicine in Nashville.

“The first thing is to find out how much the person has been drinking, whether the person has a history of previous withdrawals, and if the person has had seizures, hallucinations, or delirium tremens [DTs] before,” said Dr. Finlayson. “Those events are quite predictive of recurrence.”

Dr. Finlayson added that physicians should be aware of other substances the patient may be taking or abusing. “We're particularly concerned about benzodiazepines combined with alcohol because the effect is more than cumulative, and withdrawal can be severe.”

The UNC protocol notes that if a patient has a documented history of DTs or seizures related to alcohol withdrawal and is having at least moderate withdrawal (CIWA score ≥8) the patient should be admitted, but could be discharged if having minimal symptoms (CIWA score >8).

But hospitalists should also keep in mind the nature of alcoholism and addiction, said Dr. Kronenberger. “Denial is one of the biggest problems in this population, so take the patient's history from its source. They can lie to themselves, and to family.”

Dr. Thomas noted that elevated levels of the liver enzymes, mean corpuscular volume (MCV), gamma-glutamyl transpeptidase and the blood serum protein carbohydrate-deficient transferrin may indicate chronic alcohol abuse. “These can help sort out if the patient is truly alcohol dependent.”

Clinical findings. The patient's current symptoms and vital statistics are intrinsic to a decision to admit, said James B. Franko, MD, FACP, from the Carilion Clinic in Roanoke, Va.

“We admit patients based on clinical findings such as profound delirium, profound tachycardia, hallucinations, or acute infections,” said Dr. Franko. “Likewise, if they show up stuporous, that could indicate hospitalization to prevent withdrawal.”

Dr. Franko emphasized the need to consider the whole patient, not just the patient's alcohol use. “Most of the time these patients have other issues going on. For example, they might not have DTs, but dehydration that would prompt hospitalization in anyone. It just so happens that the dehydration is caused by alcohol abuse.”

Lab work. The UNC protocol calls for complete blood count, chemistry panel, liver function tests, urine toxicology screens, and blood alcohol level. Major electrolyte disturbances require admission.

Comorbidities. The experts agree that the presence of other medical conditions should tilt the decision toward admission, especially decompensated medical disease. “We found that acute medical disease seemed to add to the risk of complicated withdrawal, so that's where our first step is,” said Dr. Stephens. “Is the patient at baseline other than the issue with alcohol?”

Conditions that favor admission include liver disease, acute pancreatitis, congestive heart failure, heart arrhythmias, aspiration pneumonia, and uncontrolled hypertension because of the risk of stroke or heart attack, Dr. Thomas said. Also, “if the patient is vomiting blood, has stomach ulcers, or if there is anything rupturing, that calls for immediate admittance. Remember bleeds don't clot as quickly in this population,” said Dr. Thomas.

CIWA score. The CIWA score is a major component of UNC's protocol and Bellevue's internal guidelines, both of which recommend discharge if the patient's score is below 8 and there are no pressing medical needs. Both recommend admission if the score is higher than 15. “I like the CIWA assessment because it's fast to administer,” said Dr. Thomas. “It's just 10 questions, and it gives you a good picture of symptoms and severity.”

Finally, hospitalists should remember that patients with alcohol dependency have a high risk of relapse, so they may appear in the emergency department multiple times, said Dr. Kronenberger. “Even alcoholics who enter treatment have a high probability of relapse because of the other mental and emotional issues that go along with alcoholism,” she said. “But if you stick by your protocol or rules of engagement, you can maintain integrity about admitting.”