Skip to Content

Hospital’s Duty to Intoxicated Patient; Settlements by Email

August 12, 2013 in  News

New York Law Journal

This article discusses two recent rulings in New York, the Kowalski decision by the Court of Appeals and the Forcelli decision by the Appellate Division, Second Department. The first determines whether a hospital, via its emergency room personnel, has a duty to retain intoxicated or compromised patients who wish to leave. The second answers whether a litigation settlement is binding when an email containing the material terms lacks a signature by the party to be charged (or his agent) but does furnish the sender’s name at the end of the message.


An intoxicated man is brought by a friend to a hospital’s emergency room and seeks admission to the detoxification facility. His blood alcohol content is extremely high but he is alert and able to walk. He is seen by an emergency room doctor and admitted to the hospital’s detox program. Four hours after arriving and while awaiting transport to the detox, the patient removes his IV and tells a nurse he plans to go home in a taxi.

The nurse urges him to call a friend to pick him up. He agrees. The nurse goes to tell the emergency room doctor the patient wants to leave. When she returns the patient is gone. The nurse asks the doctor whether she should call the police. The doctor says “no” but notifies hospital security. Having left unescorted, the patient is struck by a car an hour or two later and severely injured. Is the hospital liable to the patient for negligence and malpractice?

Would your answer be different if you knew that the patient had been admitted to the hospital’s detox program a month earlier with suicidal thoughts, had been placed on a “one-to-one watch,” had received medication, improved and had been discharged? And, what if no one at the hospital consulted the record of that previous visit when he returned? Is there a duty to prevent patients from leaving the emergency room? If the patient insists on leaving, must he be restrained? And, if he were restrained against his will, could the hospital be exposed to liability for false imprisonment?

Sounds like a fascinating essay question on a law school “Torts” exam, doesn’t it? Except that the scenario actually occurred and wound up being the subject of a New York Court of Appeals decision published on June 26. The case, Kowalski v. St. Francis Hospital and Health Centers,1  resulted in a divided court with five judges voting to affirm a Second Department ruling granting the hospital summary judgment. Two dissenting judges, however, said that “serious issues” of fact regarding defendant’s conduct and proximate cause warranted a trial. Kowalskireminds us that “duty” and proximate cause policy factors are not mere law school discussion topics. They may inhere in real-life dramas unfolding every day.

The court viewed the gist of plaintiff’s claim to be that the hospital personnel should have prevented plaintiff from leaving the emergency room. But, said the court, defendant “had no right, and therefore could have had no duty, to do so.” There are “surely few principles more basic than that members of a free society may, with limited exceptions, come and go as they please.” Although there are people so mentally impaired that they must be denied the right to “go as they please,” that category is “a narrow one” and does not include everyone who would be safer in a detoxification facility than on the street.

The court observed that New York’s Mental Hygiene Law (§22.09) specifies when a “substantially impaired” person may be involuntarily retained for emergency treatment. The law distinguishes between those who come voluntarily (or are brought in without objection) as opposed to those who are brought “with objection.” In the latter case, the patient may be retained for emergency treatment if incapacitated to such a degree that “there is a likelihood to result in harm to the person or others.” That likelihood must be “manifested by threats or attempts at suicide or serious bodily harm or other conduct” that demonstrates a danger of self-injury. The Mental Hygiene Law makes no provision for involuntary retention of people who, like plaintiff, come to the hospital voluntarily.

Plaintiff argued that the duty to restrain him flowed from the hospital’s and doctor’s common law duty of care. But, said the court: “there can be no duty to do that which the law forbids. To restrain plaintiff on these facts would have exposed defendants to liability for false imprisonment.” But, what about the hospital’s failure to consult the record of plaintiff’s prior hospitalization when he was contemplating suicide? And, what about the emergency room’s doctor rejecting a nurse’s suggestion to call the police?

“Neither fact changes the result,” said the court. A patient cannot be confined simply because he was having suicidal thoughts a month ago. And the doctor had no duty to call the police. The police could not, on the facts known to the emergency doctor, “have forced plaintiff to return.”

The dissenting judges, however, regarded the Mental Hygiene Law as “tangential” and inapposite to a “general hospital.” Rather, they viewed the issue as one regarding a common law duty to an intoxicated patient. They pointed to the hospital’s own protocols and policies that “one-to-one surveillance” may be needed in patients with a history of psychiatric hospitalization and, further, that “potentially unstable patients by history will not be left unattended while in the emergency department.”

To these “theories” the court majority replied: “Nothing in this record…supports an inference that there was any causal connection between any of the alleged departures from protocol…and plaintiff’s injury.” This case is about whether defendants “had a duty to prevent plaintiff from leaving the hospital, and nothing else.”


An agreement between parties or their attorneys relating to any matter in an action—other than one made in open court—is “not binding” upon a party unless it is in a writing “subscribed by him or his attorney” or reduced to the form of an order and entered. So says CPLR 2104. Further, with respect to stipulations of settlement, the terms of such stipulation shall be filed by defendant with the county clerk. Thus, according to CPLR 2104, if the party wants an out-of-court stipulation of settlement to be “binding” it has to be in written form and signed by the party or his attorney.

The CPLR uses the word “subscribed.” Webster’s dictionary defines “subscribe” as “to sign (one’s name) at the end of a document.” Also, “to sign one’s name to, in attestation, testimony or consent.” Given the text of the rule and common definition of “subscribe,” the document must be signed by the party or his attorney.

Is that always true? Not according to the Second Department’s ruling issued on July 24 in Forcelli v. Gelco.2 There, the communication by an insurance company’s claims adjuster was an email message saying (in part), “Per our phone conversation today…you accepted my offer of $230,000 to settle this case.” The message also called for an attached Medicare form to be executed. It further stated, “you also agreed to prepare the release” and then specified the names to be included. The email closed: “Please forward the release and dismissal for my review. Thanks Brenda Greene.”

Was that email message compliant with CPLR 2104’s requirement of a writing “subscribed” by the party or his attorney? The Appellate Division concluded, “yes,” but not before quite some discussion and reasoning. Here, in a simplified nutshell, is why the issue became quite important. P was injured in a car accident. He sued D (there were multiple defendants but we label them collectively as D). Each side filed motions for summary judgment. On the same day, the motions have submitted the parties, their attorneys and Brenda Greene (the insurer’s adjuster) appeared for mediation. A settlement was not concluded, however.

A month later settlement negotiations revived. The adjuster first offered $200,000. Following further negotiations, Greene offered $230,000 and P’s counsel orally accepted. That same day (May 3), Greene sent P’s counsel the email message quoted above. The next day (May 4) P signed a notarized release. On May 11, P’s counsel sent to Greene by fax and certified mail the signed release and a stipulation of discontinuance.

The problem was that on May 10 the court had granted some of the defendants summary judgment and granted P summary judgment on liability against some of the defendants. Counsel for the winning defendant, on May 12, faxed and mailed a letter to P’s counsel advising that P’s release and stipulation of discontinuance “is hereby rejected.” The letter further stated that since there was no settlement consummated under CPLR 2104, “we considered this matter dismissed by the court’s decision…dated May 10.”

Two weeks later, P moved to vacate the court’s order of May 10 and to enforce the settlement agreement as set forth in Brenda Greene’s email message. P argued that there was a binding, written settlement agreement. D urged to the contrary. Two months later, the trial court vacated his order dismissing P’s complaint and enforced the settlement agreement, entering judgment in favor of P for $230,000.

The Appellate Division affirmed. Noting that stipulations of settlement are “judicially favored,” that the adjuster’s email “set forth the material terms of the agreement” and that the message “contained an expression of mutual assent,” the Second Department concluded that the email was a “writing made…with authority to bind the party.”

The question remained: Was the email “subscribed” under CPLR 2104 and, thus, capable of enforcement? Letters are usually signed at the end by the author. Email messages “cannot be signed in the traditional sense.” The court cited a decision from the First Department to the effect that “emails exchanged between counsel, which contain their printed names at the end, constitute signed writings…within the meaning of the statute of frauds.” And, in a Third Department case, the appellate panel said that “an exchange of emails may constitute an enforceable contract, even if a party subsequently fails to sign implementing documents when the communications are sufficiently clear and concrete to establish such an intent.”3

The court in Forcelli reasoned that, given the now-widespread use of email in personal and business affairs, it would be “unreasonable” to conclude that email messages are incapable of conforming to CPLR 2104 criteria “simply because they cannot be physically signed in a traditional fashion.” The panel’s conclusion was buttressed by reference to a “statement of legislative intent” accompanying the “Electronic Signatures and Records Act” enacted in 2002. The act was “intended to support and encourage electronic commerce…by allowing people to use electronic signatures and electronic records in lieu of handwritten signatures and paper documents.”

Section 302(3) of that statute says that “electronic signature” shall mean an “electronic sound, symbol or process, attached to or logically associated with an electronic record and executed or adopted by a person with the intent to sign the record.” Section 304(2) of the statute says that “an electronic signature may be used by a person in lieu of a signature affixed by hand [and] [t]he use of an electronic signature shall have the same validity and effect as the use of a signature affixed by hand.”4

‘Signature’ Intended

But the email message from the adjuster contained only her printed name at the end, not an “electronic signature” as defined by the act. “Nevertheless,” said the appellate court, the record supports the conclusion that Greene, “in effect, signed the email message.” The email ended with the simple expression, “Thanks Brenda Greene.” This indicates that the author “purposefully added her name to this particular email message” rather than one where software is programmed to automatically generate the name of the sender every time an email message is sent. Additionally, the adjuster’s participation in the face-to-face mediation and the telephone calls “support the conclusion that Greene intended to ‘subscribe’ the email settlement for purposes of CPLR 2104.”

The court expressed its dramatic ruling as follows: “…we hold that where, as here, an email message contains all material terms of a settlement and a manifestation of mutual accord, and the party to be charged, or his or her agent, types his or her name under circumstances manifesting an intent that the name be treated as a signature, such an email message may be deemed a subscribed writing within the meaning of CPLR 2104 so as to constitute an enforceable agreement.”

The court’s assessment of this particular adjuster’s likely “intent” to effect a settlement seems hardly debatable. So, too, the fact that, as a matter of policy, courts favor settlements and will make efforts to enforce them cannot be denied. Probably “justice” has been well served in this case. Nevertheless, a smidgen of disquieting thoughts may linger, especially for other cases and in general. CPLR 2104 clearly says “subscribed”—meaning a signature—not just insertion of a name at the end of a document. In common understanding, a name at the end of an email is not a signature.

Although modern email messaging customs and practices have introduced levels of informality in most communications, CPLR 2104 has not. That rule of procedure still calls for a “subscribed,” i.e., signed, writing. Maybe today’s legislature would agree that settlements by email with just a name inserted at the end suffice. But, then arguably, the legislature should tell us that. Modernizing CPLR 2104 via judicial means can raise questions.

Let us, at least, peek briefly at another arguable point of view. Settlement of a case is serious subject matter. Litigants are committing themselves to forgoing certain rights and, often, transferring large sums of money from one side to the other. Peace is being bought as well as protection against the same dispute reappearing. Requiring some ritualistic sense of solemnity and formality to accompany and attest to such momentous decisionmaking seems quite justified. Requiring a party’s (or agent’s) signature provides that solemn formality. The signature evidences aforethought and decisive intent to be bound.

We can see this operative principle well reflected in Galetta v. Galetta,5 a Court of Appeals decision issued on May 30 declaring invalid a prenuptial agreement due to a defective acknowledgment by a notary public. The court would not permit the notary’s acknowledgment to be “cured” post hoc via affidavits. The court observed that the acknowledgment requirement fulfills “two important purposes.” First, it serves to prove the identity of the person whose name appears on the instrument and to authenticate his signature. Second, “it necessarily imposes on the signer a measure of deliberation in the act of executing the document.” The notarial formality “underscores the weighty personal choices to relinquish significant property…rights or to resolve important issues…”

The same “important purposes” arguably might be served by an actual “subscribed” or “signature” standard for binding settlement agreements. Maybe, from a policy standpoint, there’s something to be said for not fretting too much about a delay between the parties’ stated intent to settle and later validating that intent with a signature. That interval of delay would, to paraphrase Galetta’s language, “impose on the signer a measure of deliberation” and “foster reflection and care” about the decision. Further, counsel in a real hurry to create a binding agreement could always shorten that interval on a “rush” basis to obviate second thoughts. Better yet, settlements can be concluded: “in open court.” Indeed, the entire problem in Forcelli seems to have stemmed from not advising the trial court about the settlement until after the summary judgment issued. So, perhaps, there is decent food for thought by the Legislature were it to reassess CPLR 2104’s requirement of signed writing.

Michael Hoenig is a member of Herzfeld & Rubin.


  1.  2013 N.Y. LEXIS 1677 (N.Y. Ct. App. June 26, 2013).
  2.  2013 NY Slip Op 05437 (2d Dept. July 24, 2013).
  3.  Forcelli v. Gelco, 2013 Slip Op 05437 (2d Dept. July 24, 2013) (citing, respectively, Williamson v. Delsener, 59 AD3d 291 (1st Dept. 2009); Brighton Inv. v. Har-Zvi, 88 AD3d 1220, 1222 (3d Dept. 2011)).
  4.  Forcelli, Id., Slip Op, at p. 8.
  5.  2013 N.Y. LEXIS 1339 (N.Y. Ct. App. May 30, 2013).