Drugs, Alcohol, and Disposition
By Kevin Klauer, DO, EJD, Chief Medical Officer, Emergency Medicine Physicians, Canton, OH
Financial Disclosure: The following individuals disclose that they have no consultant, stockholder, speaker’s bureau, research, or other financial relationships with companies having ties to this field of study: Arthur R. Derse, MD, JD, FACEP (Physician Editor), Director and Professor, Center for Bioethics and Medical Humanities, Institute for Health and Society, Medical College of Wisconsin, Milwaukee, WI; Kevin Klauer, MD (Writer), Stacey Kusterbeck (Contributing Editor); Shelly Morrow Mark (Executive Editor); and Leslie Hamlin (Managing Editor). Kay Ball RN, PhD, CNOR, FAAN, Consultant/ Educator, K&D Medical Inc., Lewis Center, OH (Nurse Planner) is a speaker for AORN and a stockholder for STERIS, Inc.
There is much myth, controversy, and misunderstanding surrounding the safe and medical legally sound disposition of patients who are under the influence of drugs or alcohol. Notwithstanding the importance of defining what "under the influence" truly means, there are several clinical scenarios associated with drugs, alcohol, and the emergency department that are commonly encountered. There are many opinions on these matters. Most are worth exactly what was paid for them.
I’ll attempt to bring some clarity to these issues and dispel the myths regarding discharging patients after receiving opioid analgesics, refusals of care while intoxicated, and discharging intoxicated patients.
Background: Intoxication and Intent
Whether in a criminal or civil (tort) liability context, intoxication has been described in two contexts: voluntary and involuntary. Although our discussion focuses strictly on the liability of the negligence tort, it may be of value to recognize the similarities in definition with criminal law. Voluntary intoxication is just how it sounds. When people intentionally use an intoxicating substance, they are responsible for their actions. When people are given an intoxicating substance unknowingly, or are not informed of the side effects of a substance (i.e., opioid analgesics, sedative-hypnotics, etc.), they may claim involuntary intoxication as a defense to a general-intent crime.
The complex issues surrounding criminal law are rooted in the accused person’s ability to formulate intent and what intent is required for such a crime. Although not universally utilized in all jurisdictions, most distinguish between specific-intent and general-intent crimes. The tortious (tort liability) corollary will become fairly clear. So, to be convicted of a specific-intent crime, such as first-degree murder, a person must meet all of the elements of the crime, beyond just the general intent to perform the act. For example, if you intended to kill someone with your vehicle, you could meet the statutory requirements of first-degree murder (depending on the jurisdiction and specific required elements). However, if you killed someone with your car, but had no specific intent to do so, it would be unlikely that the prosecution could meet the burden for a first-degree murder charge, but the general intent to drive the car, and with reckless disregard due to voluntary intoxication, would certainly meet the burden of a lesser crime, requiring only general intent (i.e., reckless homicide, vehicular manslaughter.)1
Although criminal charges could be filed against an intoxicated patient who commits a crime, it is unlikely for criminal charges to be filed against the health care providers and institutions that cared for the accused. However, the exposure to civil (tortious) liability for negligence in these cases is very possible. Certainly, a defendant’s claim of involuntary intoxication due to a lack of information/warning or instruction by a health care provider or institution could prove problematic in a civil case.
Dram shop statutes have been adopted in many states. Such statutes are invoked as a matter of public policy to avoid unnecessary risk to the unnamed third party (those who may be injured by an intoxicated patron if that person drives). It is important to note that these statutes are narrowly crafted and are not utilized in medical negligence cases in which a patient received a sedating medication or the intoxicated patient was discharged and injured someone. These statutes traditionally are limited in scope to alcohol and vendors selling alcohol to minors or intoxicated patrons.
Although traditionally alcohol has been the primary intoxicant noted in legal definitions, challenges have been made that other mind-altering substances should be treated similarly. The Model Penal Code § 2.08 defines intoxication as, "A diminished ability to act with full mental and physical capabilities because of alcohol or drug consumption;" drunkenness. Thus, the medications we provide in the emergency department may be easily included in the legal definition of intoxicating substances. This inclusion could come into play when patients are discharged and deemed to be intoxicated, based on medications provided while under the care of medical personnel.
Discharging Patients after Receiving Opioids and the "Duty to Warn"
Is there liability associated with discharging a patient after he or she receives opioid analgesics, for instance? Yes, but the alleged negligence is ordinary negligence, as opposed to medical malpractice. However, if certain steps are taken, the liability exposure is minimal. What seems evident in many EDs is that some have misinterpreted and inflated the concerns of liability in these cases, and in order to protect themselves, have adopted defense strategies that are both unreasonable and inefficient. Some behaviors that seem both unreasonable and unwarranted are refusing to provide patients with opioid analgesics until after their ride can present to the emergency department, as well as attempting to detain the patient if it appears that person may be getting into a vehicle as the driver, despite showing no signs of impairment. Although there is certainly no prohibition to doing these things, they are unnecessary from a legal perspective. If a hospital or provider would like to notify the police, they should do so. However, there is no requirement of obligation to do so. Also, pain management shouldn’t be delayed until the patient’s ride is present. The discharge can be delayed for their arrival, but withholding treatment may be deemed unethical.
The common law roots of "duty to warn" by health care providers is defined in Tarasoff v. Regents of the University of California. In this case, Mr. Poddar, a student at the University of California, was romantically interested in Ms. Tarasoff. She didn’t share the same interest, and he became depressed and sought counseling. He made statements to his counselor that he had thoughts of hurting Ms. Tarasoff. However, the therapist didn’t initially report these statements. When Mr. Poddar failed to return to counseling, the counselor notified campus police, who questioned Mr. Poddar. No further action was taken. Mr. Poddar killed Ms. Tarasoff two months later. The Supreme Court of California held that providers have a duty to warn potential victims. This could be accomplished by notifying the police of the threat or by admitting the patient. This mental health case has direct implications for emergency medicine. The duty owed to a named third party, someone that the patient states he or she intends to harm (i.e., "when I leave here I am going to kill so and so"), can be fulfilled by notifying law enforcement and by hospitalizing the patient, if mental illness is the cause.
To avoid liability regarding the discharge of a patient who has received potential intoxicating substances from the ED, a duty is owed to the patient and to the unnamed third party. The duty owed to the patient is to adequately and reasonably inform the patient about the side effects of the treatment provided and any limitations on his or her activities to prevent unintentional self-harm. The duty owed to the unnamed third party is very similar, but the goal is to prevent harm of the unnamed third party by the actions of the patient. Patients must be told not to drive or operate machinery, etc. This should be clearly documented in the after care instructions and the nursing documentation. For the sake of completeness, this duty is usually owed to the unnamed third party when the patient poses risks to others for different reasons, such as driving while at risk of having a seizure or driving with an eye patch.
In general, once the duty to third parties has been met, the health care provider’s responsibility ends. However, it is still possible that someone could get in the car, drive away, and injure someone. It’s even possible that a lawsuit could be filed against the institution or providers. A Massachusetts case from 20092 established a strong legal basis that health care providers have no obligation to detain or otherwise control a patient to avoid potential harm to unnamed third parties. An additional holding was that medicated patients are not deemed inherently dangerous purely because they have been medicated.
Documentation that the patient has been adequately informed of the effects of medications provided and of any restrictions (e.g., driving or operating machinery) is essential to a good legal defense and to avoiding a lawsuit altogether.
Intoxication and AMAs
Despite the patient’s disinterest in seeking medical care, intoxicated patients are frequently brought to the emergency department by family, friends, or law enforcement. Also, alcohol intoxication may result in behavioral issues in patients who initially do want and consent for treatment, but subsequently change their mind.
So, when may an intoxicated patient refuse medical care? This is another complex question. First, it is imperative to recognize that most courts will give great deference to a physician attempting to provide reasonable care to a patient. Furthermore, stories of physicians being sued for torts such as false imprisonment, assault and battery, and criminal charges filed for kidnapping are nothing more than urban legends. Think this through. In order to file a lawsuit, a prospective plaintiff will need an attorney. Quite honestly, individuals who report being mistreated in an emergency department by health care professionals attempting to care for the patient while they were intoxicated, spitting, being verbally abusive, etc., just don’t make good witnesses or clients. Most attorneys see no benefit in taking on such cases. Having said that, it is important to respect a patient’s right to refuse care, and intoxication does not absolve a provider from allowing a patient to make his or her own decisions when capable of doing so.
As with any other patient, the ability to consent and refuse is based on medical decision-making capacity. If you don’t have decision-making capacity, you can’t refuse. Competence is determined by a court and cannot be established by a physician or at the time of patient care delivery. Medical decision-making capacity can be established by making certain the patient is oriented to person, place, time, and situation, and that the patient understands the risks and benefits of the proposed treatment, and alternatives to treatment, including no treatment at all. It is very possible that someone under the influence of alcohol or other substances may still have medical decision-making capacity. However, the decision to allow a "drunk" patient to refuse care should be approached with great caution. When in doubt, err on the side of providing medical care. The chances of being sued for missing a subdural hematoma in an intoxicated patient you allowed to sign out against medical advice (AMA) are exponentially greater than being sued for detaining the patient. Furthermore, the exposure for third-party injury by an intoxicated patient who arguably didn’t understand the informed refusal is likely greater than the previously discussed scenario of the unimpaired patient who was discharged following the delivery of opioid analgesics. In both cases, the duty to the third party must be met. However, in the obviously intoxicated patient attempting to leave, some jurisdictions or courts may apply a higher standard on the health care worker for the duty to unnamed third parties than for the patient who received therapeutic doses of analgesics and exhibits no signs of impairment. Again, in the obviously intoxicated patient, it may be prudent to take extra steps, such as calling law enforcement, when a patient is clearly a risk to himself or others.
It is recommended that this be a team decision. If the physician and nurse caring for the patient are in agreement that this patient has decision-making capacity and the patient shows no signs of impairment, they should both document this in their respective records when allowing a refusal. Exercise great caution with intoxicated patients who may have pathology altering their decision-making capacity. For instance, if a patient who has consumed alcohol seems mildly intoxicated (i.e., slurred speech, mild somnolence, etc.), he or she may actually have traumatic brain injury or a diabetic complication. The physician’s evaluation and documentation should reflect attempts to identify any such medical etiologies that would preclude safe refusal and discharge. Of course, the best strategy is to convince to patient to consent to treatment. All such attempts to do so should be documented.
AMA forms are frequently utilized, but the protection they provide is often overrated. Just like a consent form, the AMA refusal form should outline what treatment is proposed, any alternatives, and the patient’s understanding of the informed discussion. Signing an AMA form, in the absence of documenting that a true informed refusal was obtained, is a recipe for disaster. It is evidence of a signature, but not true understanding of the risks of refusal. The most important documentation of the refusal is the assessment of decision-making capacity and the informed discussion. Documenting this in the medical record is much more valuable than an AMA form, particularly when AMA forms contain little more than a signature. Strategically, providers may want to avoid the use of AMA forms, as they frequently result in confrontation. "If you don’t want to then you have to sign this form accepting responsibility for your bad decision." The conflict created may not be worth the protection provided by such forms, particularly when a well-documented informed refusal in the medical record accomplishes so much more. All patients should be made aware that they are welcome to return anytime should they change their mind, and this notification should be noted in the record.
DWI (Discharged While Intoxicated)
The discharge of intoxicated patients can be safely accomplished after completing treatment if the patient is medically stable. If the patient shows signs of intoxication, then providers should confirm that the patient has a safe means of transportation and will not be left unattended. So, if a patient suffered a hand laceration from a broken glass and experienced no other injuries, but is intoxicated (i.e., slurred speech, mild disequilibrium, etc.), it is acceptable to discharge the patient if someone can drive him home and will be available to stay with him. Being intoxicated and alone may not be a medical emergency, but recommending that a drunk patient under your care stay home alone is a risky proposition. If the appropriate resources are not available to safely discharge the medically stable, intoxicated patient, then hold the patient until he or she is sober or when such resources become available. This topic leads us to a very important issue: How do you know when someone is intoxicated?
Who’s drunk and who isn’t? Traditionally, providers have relied on blood alcohol levels to determine a patient’s level of intoxication and to estimate when the person will no longer be intoxicated (e.g., hepatic clearance). This rationale for this approach is horribly flawed. First, ethanol intake produces different levels in every patient, which, in turn, result in different, and very individual, effects. Second, ethanol metabolism and clearance is variable and very individual. Traditional thinking holds that certain rates of clearance, such as 0.2-0.3 mg/dL per hour, are predictable and reproducible. In addition, different rates were reported for chronic versus occasional users. Although these numbers may have some validity when considering large populations, attempting to apply them to individual patients is nearly worthless. In a 2010 article published in the Journal of Medical Toxicology, the authors stated, "Attempting to relate observed signs of alcohol intoxication or impairment, or to evaluate sobriety, by quantifying blood alcohol levels can be misleading, if not impossible."3
The final ace in this house of cards is that managing blood alcohol levels is medically managing with a legal definition of intoxication, as opposed to using more appropriate clinical indicators of intoxication. We have all seen and cared for the patient with an alcohol level of 230 mg/dL who can carry on a normal conversation, ambulate without ataxia, and show no obvious signs of intoxication.
So, is the patient intoxicated? It will depend on whom you ask. If you ask the plaintiff’s attorney hired by an injured third party, a judge, law enforcement officer, etc., most will say that patient is intoxicated because his or her ethanol level is above the legal limit. However, this has nothing to do with the patient’s clinical state of intoxication. As a clinician, how can you refute their claim and assertion? The easiest way is to stop ordering blood alcohol levels on patients who don’t need them. From a clinical perspective, any test that is ordered to answer a question you already know the answer to is probably unnecessary. For instance, when a patient tells you he has been drinking and he appears mildly intoxicated, what useful clinical information does an ethanol level provide? In most cases, none. Of course, there may be regulatory/logistical reasons to obtain such a test (i.e., required by mental health professionals to screen for admission, court order following a motor vehicle accident, etc.). However, there is little to no medical utility in ordering these tests.
So, when is an ethanol level of value? As mentioned previously, when a test helps answer an unanswered question, it’s a useful test. When alcohol ingestion has not been confirmed, and its presence or absence will help guide the management of the patient, this test may be essential. For instance, a seemingly intoxicated patient is being treated for a scalp laceration and his ethanol level is "0," suspicion for other etiologies of somnolence or behavior, such as intracranial injury or other toxicological exposures, should be prompted. An ethanol level may also be helpful when a complete history is not available and the patient presents with a mental status change.
Impairment is a clinical diagnosis or assessment, not one made by the laboratory. To that end, a well-documented examination including mental status, speech, and motor function is very effective for determining whether or not the patient is impaired. Remember, our goal is not to determine if they have substances on board, but if they are impaired. This is akin to a roadside sobriety test. Law enforcement officers don’t use their Breathalyzer on every traffic stop. They perform their roadside testing, and if no signs of impairment are present, the DUI evaluation is usually complete. Why should we be held to a different standard? Fortunately, we aren’t. However, when an alcohol level is recorded in the chart, it opens Pandora’s box when the patient is deemed unimpaired clinically. If the only sobriety/impairment test in the record is a detailed and well-documented examination by the physician (or other primary provider) corroborated in the nursing documentation, then it is very difficult to refute that the patient wasn’t impaired. Using ethanol levels when not indicated can create serious legal exposure for clinicians.
When dealing with patients who are potentially impaired, either by their own choosing or by our treatments, great caution should be taken to make certain they are stable for discharge, and that such discharges can be carried out safely. It is critical to make certain that providers have met their duty to unnamed third parties, informing patients of their restrictions and documenting their understanding. Avoid the unnecessary use of alcohol levels and meticulously document your clinical evidence that your patient is unimpaired. When in doubt, err on the side of treating the intoxicated patient, even if detaining the patient is necessary.
- LaFave, Modern Criminal Law: 4th edition, Thomson/West, 2006:469-475.
- Leavitt v. Brockton Hospital, Inc.
- Roberts JR, Dollard D. Alcohol levels do not accurately predict physical or mental impairment in ethanol-tolerant subjects: Relevance to emergency medicine and dram shop laws. J Med Toxicol. 2010;6(4):438-442.