OPINION OF THE COURT
Angiolillo, J.The petitioner, the director of a state facility, Creedmoor Psychiatric Center (hereinafter Creedmoor), commenced this proceeding for permission to administer antipsychotic medication to an involuntarily committed patient over his objection. At the hearing on the petition, Creedmoor chose not to present the testimony of the patient’s treating psychiatrist, and called a single expert witness who had interviewed the patient on one occasion and reviewed the patient’s records. The Supreme Court, applying “the missing witness rule,” denied the petition after making an adverse inference against Creedmoor that the treating psychiatrist, had he been called, would have given testimony unfavorable to Creedmoor. This appeal raises the is*171sues of whether the missing witness rule was properly applied under the circumstances of this case and, if so, whether Creed-moor satisfied its burden of establishing the grounds for its petition by clear and convincing evidence. Here, the missing witness rule was properly applied and Creedmoor failed to meet its burden by clear and convincing evidence.
The Regulatory Framework
The procedures for administering treatment over the objection of an involuntarily committed patient are set forth in detailed regulations promulgated by the Commissioner of the New York State Office of Mental Health, pursuant to Mental Hygiene Law § 7.09 (b) (see 14 NYCRR 501.1 [a]; 501.2 [b]). A facility must follow stringent procedures prior to filing a petition seeking court authorization to administer the treatment (see 14 NYCRR 527.8 [c] [4]). The process requires a series of clinical evaluations of the patient, all of which must be completed within 24 hours (see 14 NYCRR 527.8 [c] [4] [ii]).
First, the patient’s treating physician must determine that the treatment is in the patient’s best interests in light of all relevant circumstances, including the risks, benefits, and alternatives to treatment, and that the patient lacks the capacity to make a reasoned decision concerning treatment. The treating physician must forward the evaluation and findings to the clinical director with a request for further review, and notify, in writing, the patient, Mental Hygiene Legal Service (hereinafter MHLS), and any other representative of the patient (see 14 NYCRR 527.8 [c] [4] [ii] [a]).
Second, the clinical director must appoint a physician to review the patient’s record, and personally examine the patient, to evaluate whether the proposed treatment is in the patient’s best interests and whether the patient has the capacity to make a reasoned decision concerning treatment. If the reviewing physician determines that treatment over objection is appropriate, the physician must personally inform the patient of that determination (see 14 NYCRR 527.8 [c] [4] [ii] [b] [1]). Alternatively, if there is a substantial discrepancy between the opinions of the treating physician and the reviewing physician regarding the patient’s capacity or best interests, the clinical director may appoint a third physician to conduct an evaluation (see 14 NYCRR 527.8 [c] [4] [ii] [b] [2]).
Finally, if, after completion of the evaluation by the reviewing physician (or physicians), the patient continues to object to the *172proposed treatment, the clinical director must make a determination on behalf of the facility. If the director finds that the patient lacks capacity, and that treatment over objection is in the patient’s best interests, the director may apply for court authorization to administer the treatment and so notify the patient, MHLS, and any other patient representative. However, if the director makes the opposite determination, the patient’s objections must be honored (see 14 NYCRR 527.8 [c] [4] [ii] [b] [3]).
“[T]he due process clause of the New York State Constitution (art I, § 6) affords involuntarily committed mental patients a fundamental right to refuse antipsychotic medication,” and “due process requires that a court balance the individual’s liberty interest against the State’s asserted compelling need on the facts of each case to determine whether such medication may be forcibly administered” (Rivers v Katz, 67 NY2d 485, 492, 498 [1986]). Thus, in a proceeding for court authorization to administer medication over a patient’s objection pursuant to the State’s parens patriae power, the clinical director has the burden of demonstrating by clear and convincing evidence, first, that the patient “lacks the capacity to determine the course of his own treatment,” and second, that “the proposed treatment is narrowly tailored to give substantive effect to the patient’s liberty interest, taking into consideration all relevant circumstances, including the patient’s best interests, the benefits to be gained from the treatment, the adverse side effects associated with the treatment and any less intrusive alternative treatments” (id. at 497-498).
Creedmoor’s Petition
By petition dated August 31, 2011, Creedmoor sought an order permitting the administration of certain antipsychotic medications to Adam K. (hereinafter the patient) over his objection. In the petition, Clinical Director William A. Fisher averred that the patient, who had no family involvement, suffered from a mental illness requiring treatment, refused medication, and was not competent to make an informed and reasoned treatment decision. Creedmoor sought authorization to administer certain medications and, as a last resort, in the event that the patient continued to refuse medication, sought to dispense the medications via nasogastric tube. In support of the petition, Creed-moor submitted the affidavits of two physicians, Thulasi R. Reddy and Robert Mathew, who opined with reasonable medical *173certainty that the patient was not competent to make an informed, reasoned decision concerning treatment and that it would be in the best interests of the patient to be treated as proposed. Both doctors relied on two documents prepared by Mathew: a “Clinical Summary” and an “Evaluation for Treatment over Objection” (hereinafter the Evaluation).
In his Clinical Summary, Mathew noted the following: the patient, 63 years old, homeless, and Polish-speaking, had been continuously hospitalized since 2010 and had been involuntarily admitted to Creedmoor in April 2011. The patient had a medical history which included traumatic brain injury, cognitive deficiencies, and alcohol abuse, and he was diagnosed with a cognitive disorder, psychosis secondary to the traumatic brain injury, and bilateral frontal encephalomalacia. The patient was very loud and disruptive and had a history of assaultive behavior at other facilities. Mathew stated that the patient was “refusing his medications most of the time” and his response to medication had been “poor.” He did not attend group therapy and was not making progress. Mathew further stated that when he or other staff members attempted to converse with the patient, he became angry and started shouting in Polish. A Polish-speaking nurse and a therapy aide would translate, and the patient also spoke some English. Mathew concluded that the patient was “very psychotic, disorganized, noncompliant with treatment, [and] unable to care for himself” and, thus, court authorization was needed to treat him over his objection.
In his Evaluation, Mathew stated that the patient had experienced a “partial to fair response” to the antipsychotic medications haldol and depakote benztropine. When the patient objected to treatment, he would state, “I am not sick.” The recommended treatment included certain doses of haldol decanoate, benztropine, divalproex sodium, and seroquel, and the recommended “reasonable alternative” to those medications included certain doses of prolixin or prolixin decanoate and olanzapine. The reasonably foreseeable adverse effects of these medications included sedation, muscle spasm, extra pyramidal side effects, metabolic side effects, risk of tardive dyskinesia, and neuroleptic malignant syndrome. In Mathew’s opinion, without treatment, the patient would “remain psychotic, labile, inappropriate, agitated, aggressive, assaultive and unable to care for himself,” as well as a danger to himself and others.
*174The Hearing
On September 20, 2011, the Supreme Court held a hearing on the petition at Creedmoor’s facility. The patient did not attend the hearing, but was represented by an attorney from MHLS. Creedmoor called a single witness, Dr. Ella Brodsky, who testified that she interviewed the patient on August 31, 2011, reviewed his clinical record, and spoke with his treating psychiatrist, whom she identified as Mathew, a full-time psychiatrist at Creedmoor. The parties stipulated to Brodsky’s qualifications in the field of psychiatry and to the admission into evidence of the clinical record, which included Mathew’s Clinical Summary and Evaluation and hundreds of pages of clinical notes.
At the outset, after the Supreme Court determined that Mathew was available to testify, Brodsky testified that Creed-moor had chosen not to call him as a witness to avoid interfering with the relationship between treating psychiatrist and patient. The court then inquired why, in other recent cases, Creedmoor had called treating psychiatrists to testify. Brodsky explained that she had been on vacation and unable to testify in those cases, and, thus, there had been no other choice. The court opined that there could be no professional or ethical rule preventing a treating psychiatrist from testifying because “every other hospital” that appeared before the court relied on the testimony of a treating psychiatrist. Brodsky explained that Creedmoor was a long-term facility for chronic patients as compared to other hospitals with short-term stays. Further, although “not forbidden,” and on occasion unavoidable, Creed-moor “preferred not to spoil the therapeutic relationship” between the treating psychiatrist and patient. At that point, prior to the continuation of Brodsky’s testimony, the court, sua sponte, held that it would be applying “the missing witness rule” in weighing whether Creedmoor satisfied its burden of proof.
Brodsky proceeded to give her opinion, to a reasonable degree of medical certainty, that the patient was afflicted with a mental illness to such a degree that he lacked the capacity to understand his illness or the need for medication, and he lacked the capacity to give or withhold his consent to treatment. Brodsky answered “yes” when asked if the proposed treatment plan was narrowly tailored to the patient’s specific needs, and answered “no” when asked if there was any less intrusive alternative treatment available. Brodsky opined that if the patient did not take the proposed medications, he would “decompensate,” *175becoming aggressive and violent, his condition would continue to deteriorate, and he could not be placed in a supervised nursing facility. Further, Brodsky acknowledged that there were some possible side effects of the proposed medication such as “tardive dyskinesia and neuromalignant syndrome.” However, Brodsky noted that the patient had been taking the proposed medications intermittently without manifesting side effects, and the patient would be closely monitored for side effects.
On cross-examination, when asked about Mathew’s comment in the Evaluation that the patient had experienced a “partial to fair response” to medication, Brodsky opined that, during periods when he took his medication, the patient continued to exhibit some symptoms which were manageable, and therefore, a “partial to fair response is actually [a] good response.” Brodsky acknowledged the possibility that some of the patient’s behavioral issues were the result of frustration at his inability to express himself in English. However, Brodsky had never personally witnessed the episodes when the patient lashed out.
The Supreme Court’s Determination
At the conclusion of the hearing, the Supreme Court did not credit Creedmoor’s reason for choosing not to call Mathew, the treating psychiatrist. The court noted that Mathew was unquestionably a full-time employee under the control of Creedmoor, that Creedmoor had failed to explain his nonappearance at this hearing being held in Creedmoor’s facility, and that Creedmoor had called treating psychiatrists in other cases, indicating that the alleged interest in protecting the relationship between the patient and the treating psychiatrist was not of great importance. Thus, the court, sua sponte, applied the “missing witness rule” and presumed that Mathew’s testimony would have been adverse to Creedmoor’s interests had he been called to testify.
Upon drawing an adverse inference, the court determined that Creedmoor had met its burden of establishing the patient’s lack of capacity to determine the course of his own treatment, but had failed to satisfy its burden of establishing by clear and convincing evidence that the proposed treatment would benefit the patient and, thus, failed to establish that the treatment was narrowly tailored to give substantive effect to the patient’s liberty interest. Accordingly, over Creedmoor’s objection to the court’s application of the missing witness rule, the court determined that the petition should be denied. In its order entered September 28, 2011, the court, in effect, denied the petition and dismissed the proceeding.
*176The Adverse Inference for Failure to Call a Witness
We begin our analysis by addressing Creedmoor’s contention that the Supreme Court erred in applying the missing witness rule because that rule has no place in a proceeding for court authorization to administer treatment to an involuntarily committed patient. This issue is properly raised for our review because the Supreme Court applied the missing witness rule over Creed-moor’s objection. Further, we deem it necessary to address this issue in light of Creedmoor’s express intention to shield treating psychiatrists from testifying in treatment-over-objection proceedings generally, and the Supreme Court’s heavy reliance on the adverse inference specifically in this case.
In support of its contention that the missing witness rule has no application here, Creedmoor argues that the governing regulations do not require the treating psychiatrist to testify, that there are compelling therapeutic reasons why a treating psychiatrist should not be required to testify, and that the rationale underlying the missing witness rule is inconsistent with the regulatory framework, which generates multiple medical judgments, thus ensuring that the treating psychiatrist’s testimony would be favorable to the petitioning facility and cumulative to the professional judgments of the other evaluating doctors.
In addressing these contentions, we examine first the well-established principles underlying the missing witness concept in other contexts. The case law most frequently involves situations in which appellate courts review a trial court’s ruling on a party’s application to submit a missing witness instruction to the jury. “A missing witness instruction . . . tells a jury that it may draw an unfavorable inference based on a party’s failure to call a witness who would normally be expected to support that party’s version of events” (People v Hall, 18 NY3d 122, 131 [2011] [citation and internal quotation marks omitted]; see People v Savinon, 100 NY2d 192, 196 [2003]). The rationale underlying the rule “derives from the commonsense notion that the nonproduction of evidence that would naturally have been produced by an honest and therefore fearless claimant permits the inference that its tenor is unfavorable to the party’s cause” (People v Gonzalez, 68 NY2d 424, 427 [1986] [internal quotation marks and emphasis omitted]; see Graves v United States, 150 US 118, 121 [1893]; People v Savinon, 100 NY2d at 196; People v Hovey, 92 NY 554, 559-560 [1883]).
At a criminal trial, a party seeking a missing witness instruction bears the initial burden of “showing that the uncalled wit*177ness could be expected to have knowledge about a material issue and to testify favorably to the opposing party” (People v Kitching, 78 NY2d 532, 536-537 [1991]). The expectation that the uncalled witness would testify favorably to the opposing party “has been referred to as the ‘control’ element, which requires the court to evaluate the relationship between the witness and the party to whom the witness is expected to be faithful” (People v Savinon, 100 NY2d at 197). Once the moving party meets this initial burden, the opposing party, to defeat the instruction, must
“account for the witnesses] absence or otherwise demonstrate that the charge would not be appropriate ... by demonstrating that the witness is not knowledgeable about the issue, that the issue is not material or relevant, that . . . the testimony would be cumulative to other evidence, that the witness is not ‘available’, or that the witness is not under the party’s ‘control’ such that [the witness] would not be expected to testify in [that party’s] favor” (People v Gonzalez, 68 NY2d at 428; see People v Smith, 71 AD3d 1174,1175 [2010]; People v Marsalis, 22 AD3d 866, 868 [2005]).
The same principles generally apply in a civil jury trial. “A party is entitled to a missing witness charge when the party establishes that an uncalled witness possessing information on a material issue would be expected to provide noncumulative testimony in favor of the opposing party and is under the control of and available to that party” (Zito v City of New York, 49 AD3d 872, 874 [2008] [internal quotation marks omitted]; see Jackson v County of Sullivan, 232 AD2d 954, 955 [1996]; Kupfer v Dalton, 169 AD2d 819 [1991]). Where one or more of these elements is absent, the movant is not entitled to the charge (see Pope v 818 Jeffco Corp., 74 AD3d 1163, 1164 [2010] [the witness was not under the opposing party’s control]; Holbrook v Pruiksma, 43 AD3d 603, 605-606 [2007] [the witness was not expected to offer testimony favorable to the opposing party]; Pasquaretto v Cohen, 37 AD3d 440, 441 [2007] [the witness was not available and the testimony would have been cumulative]; Cohen v Lukacs, 272 AD2d 501, 501-502 [2000] [same]). Moreover, the missing witness rule may be applied in a nonjury civil trial, where the trial court, as finder of fact, is permitted to draw a negative inference against a party failing to call a witness (see Coliseum Towers Assoc. v County of Nassau, 2 AD3d *178562, 565 [2003] [upon recognizing the applicability of the rule, holding that the trial court improperly drew a negative inference against the defendant County for its failure to call former County employees as witnesses because they were not under the County’s direction or control]).
The missing witness rule is related to the broader principle that “[a] trier of fact may draw the strongest inference that the opposing evidence permits against a witness who fails to testify in a civil proceeding” (Matter of Nassau County Dept. of Social Servs. v Denise J., 87 NY2d 73, 79 [1995]; see Matter of Commissioner of Social Servs. v Philip De G., 59 NY2d 137, 141 [1983]; Noce v Kaufman, 2 NY2d 347, 353 [1957]; Dowling v Hastings, 211 NY 199, 202 [1914]; Crowder v Wells & Wells Equip., Inc., 11 AD3d 360, 361 [2004]). This formulation of the broader principle is generally applied in cases where the missing witness is a party (see Matter of Nassau County Dept. of Social Servs. v Denise J., 87 NY2d at 79-80; Matter of Commissioner of Social Servs. v Philip De G., 59 NY2d at 141; Matter of Renee R. [Tonya D.], 98 AD3d 1048 [2012]; Matter of Clarissa S.P. [Jaris S], 91 AD3d 785, 786 [2012]; Katz v Gangemi, 60 AD3d 819 [2009]; Brown v City of New York, 50 AD3d 937, 938 [2008]; Matter of Cantina B., 26 AD3d 327, 328 [2006]; Crowder v Wells & Wells Equip., Inc., 11 AD 3d at 361-362). In these cases, which include jury trials, nonjury trials, and hearings, the negative inference is generally applied without reference to the prerequisites for the missing witness rule, since those factors would be either irrelevant (the party’s control over himself or herself) or deemed satisfied (the party’s availability and personal knowledge of noncumulative, material facts) (but see Crowder v Wells & Wells Equip., Inc., 11 AD3d at 361-362 [analyzing, in a multiple defendant case, the availability and noncumulative, personal knowledge of the missing defendant]).
On occasion, the broad principle allowing “the strongest inference that the opposing evidence permits” against a nontestifying party has been relied upon in cases against a municipal defendant which has failed to call its employees as witnesses (Piquette v City of New York, 4 AD3d 402, 404 [2004] [police officers]; cf. Coliseum Towers Assoc. v County of Nassau, 2 AD3d at 565 [former County employees]). Significantly, in a civil involuntary commitment proceeding pursuant to Mental Hygiene Law article 15, the Appellate Division, Third Department, has upheld the Supreme Court’s application of “ ‘the strongest inference against [petitioner] that the opposing evidence in the rec*179ord permits’ ” where the petitioning facility failed to call its employee, the patient’s treating psychologist (Matter of Richard E., 12 AD3d 1019, 1021 [2004], quoting Matter of Commissioner of Social Servs. v Philip De G., 59 NY2d at 141). In Matter of Richard E., the Court concluded that the negative inference was appropriately made because the treating psychologist “was under petitioner’s control and could provide relevant, noncumulative testimony regarding whether respondent is mentally retarded” (Matter of Richard E., 12 AD3d at 1021).
The rule permits the strongest possible adverse inference as to any evidence which the missing party or witness “would be in a position to controvert,” but the rule “may not be used to draw any inferences beyond that” (Matter of Jane PP. v Paul QQ., 65 NY2d 994, 996 [1985]). Thus, in a paternity proceeding, the putative father’s failure to testify would permit an inference establishing the date of his sexual relations with the mother based on evidence in the record, but does not permit establishment of the ultimate conclusion that he was the father of the child (see id.; see also Matter of Cantina B., 26 AD3d at 328 [father’s failure to testify at neglect proceeding warranted an adverse inference that the father knew of the mother’s drug use during pregnancy, as established by other evidence in the case]).
Here, Creedmoor would have us carve out an exception to the well-established missing witness rule, prohibiting its application generally in every treatment-over-objection proceeding or, specifically, in the instant proceeding. However, Creedmoor failed to demonstrate any appropriate ground for defeating the adverse inference permitted by the rule.
First, as a general matter, a treating psychiatrist in the petitioner’s employ would normally qualify as an available witness within the petitioner’s control (see Zito v City of New York, 49 AD3d at 874; Piquette v City of New York, 4 AD3d at 404; Leven v Tallis Dept. Store, 178 AD2d 466 [1991]; cf. Pope v 818 Jeffco Corp., 74 AD3d at 1164; Coliseum Towers Assoc. v County of Nassau, 2 AD3d at 565; Zeeck v Melina Taxi Co., 177 AD2d 692, 694 [1991]). While we do not foreclose the possibility that, in another case, a petitioner might successfully demonstrate its lack of control over, or the unavailability of, its treating psychiatrist, Creedmoor failed to meet its burden in this regard.
Creedmoor made no showing that Mathew was physically unavailable on the date of the hearing, but argued only, in effect, that he should be deemed unavailable in light of Creedmoor’s policy to avoid interference with the patient-treating psychia*180trist relationship. However, Creedmoor’s witness, Brodsky, failed to offer any basis for the alleged policy, whether its genesis was in a professional or ethical rule, a practical consideration of facility management, or previous cases in which a treating psychiatrist’s testimony interfered with the therapeutic relationship. Moreover, as Brodsky conceded during her testimony, Creedmoor has called its treating psychiatrists to testify in other treatment-over-objection proceedings. Under these circumstances, Creedmoor failed to establish that its alleged policy precludes the application of the missing witness rule generally in all treatment-over-objection proceedings where the facility fails to call its treating psychiatrist.
Moreover, Creedmoor failed to establish the importance of its alleged policy with respect to this specific proceeding. Mathew’s testimony would not have interfered with the therapeutic relationship, since the patient did not attend the hearing but appeared by counsel. Thus, Mathew’s appearance and testimony at the hearing would have intruded no more on the patient-psychiatrist relationship than his required evaluation with notice to the patient (see 14 NYCRR 527.8 [c] [4] [ii] [a]) and his affidavit filed in support of the instant proceeding. Accordingly, Creedmoor failed to satisfy its burden to “account for the witnesses] absence or otherwise demonstrate that the [adverse inference] would not be appropriate” (People v Gonzalez, 68 NY2d at 428).
Second, it goes without saying that the treating psychiatrist, as opposed to the reviewing doctors, possesses the greatest knowledge about the patient and “information on a material issue” raised in the proceeding (Zito v City of New York, 49 AD3d at 874). On appeal, Creedmoor does not contest the fact that Mathew had knowledge relevant to material issues.
Third, the treating psychiatrist “would be expected to provide . . . testimony in favor of’ Creedmoor (id. [internal quotation marks omitted]). In fact, Creedmoor concedes that “as a categorical matter, the treating psychiatrist cannot have an opinion adverse to the proposed treatment,” since that doctor is responsible for the initial evaluation and recommendation to commence the proceeding. Creedmoor unpersuasively argues that this fact militates against application of the missing witness rule. On the contrary, an expectation that the witness would testify favorably to the party failing to call the witness establishes a prerequisite for making the adverse inference permitted by the missing witness rule (see People v Gonzalez, 68 NY2d at 427).
*181Although Creedmoor emphasizes that the governing regulations do not require the treating psychiatrist to testify in a treatment-over-objection proceeding, this fact has no bearing on the issue of whether the missing witness rule was properly applied in this proceeding. The missing witness rule is not premised upon a party’s violation of a regulatory or statutory requirement, but on the commonsense notion that a party will normally call a witness who would be expected to provide testimony in the party’s favor. Moreover, in an analogous context, the Third Department applied the missing witness rule after rejecting, inter alia, the petitioner’s contention that there was “no authority requiring him to call respondent’s treating psychologist to testify at the retention hearing” (Matter of Richard E., 12 AD3d at 1021).
Further, we find no merit to Creedmoor’s contention that, in every treatment-over-objection proceeding, the treating psychiatrist’s testimony is necessarily cumulative because the regulatory framework is “designed to generate cumulative, confirmatory medical judgments separately indicating the need for the proposed treatment.” Creedmoor’s reliance on the multi-layered review procedure is misplaced. Manifestly, the regulatory procedure is designed to ensure that two or more doctors independently arrive at the same opinion as to the need to treat the patient over his or her objection. However, an opinion is not the same as the information upon which that opinion was based. The regulatory procedure does not ensure that the reviewing physicians spend a significant amount of time treating the patient or otherwise acquire an amount of knowledge about the patient comparable to that of the treating psychiatrist. In other contexts, most notably in personal injury actions, application of the missing witness rule for physician testimony is based on the recognition that the testimony of two physicians examining the same patient may be noncumulative, and it is the burden of the party failing to call the witness to demonstrate otherwise in order to defeat the adverse inference (see Lauro v City of New York, 67 AD3d 744, 746 [2009]; Hanlon v Campisi, 49 AD3d 603, 604 [2008]; Brooks v Judlau Contr, Inc., 39 AD3d 447, 449 [2007]). Accordingly, we reject Creedmoor’s contention that the governing regulations provide sufficient assurance of cumulative testimony as to preclude the application of the missing witness rule generally in all treatment-over-objection proceedings.
In addition, Creedmoor did not specifically establish that Mathew’s testimony would have been merely cumulative in the *182instant proceeding. In evaluating Creedmoor’s contention, we remain cognizant of the context in which it is made. In this proceeding, Creedmoor seeks to forcibly administer antipsychotic drugs over the patient’s objection, by nasogastric tube if necessary. The proposed medication has potentially serious side effects, and the patient’s liberty interest is at stake, requiring a clear and convincing standard of proof (see Rivers v Katz, 67 NY2d at 490 n 1, 493). In this context, and on this record, Creed-moor failed to establish that Mathew’s testimony would have been merely cumulative. Brodsky testified that she reviewed the medical records and interviewed the patient on one occasion, whereas the certified medical record establishes that Mathew treated the patient continuously over a period of months. The quantity and quality of the information possessed by Brodsky and Mathew were vastly different and cannot be considered equivalent. Further, while much of the certified medical record was consistent with Brodsky’s testimony, certain portions of the record on key issues with respect to the patient’s behavior and Mathew’s opinion as to the patient’s response to treatment were inconsistent with the testimony offered by Brodsky. In sum, Creedmoor failed to meet its burden of establishing that Mathew’s testimony would have been merely cumulative to Brodsky’s testimony.
Accordingly, under the particular circumstances of this case, and on this record, the Supreme Court properly drew an adverse inference against Creedmoor for its failure to call an available witness under its control who would be expected to offer noncumulative testimony on material issues in favor of Creed-moor’s position (see Matter of Richard E., 12 AD3d at 1021; see also Zito v City of New York, 49 AD3d at 874).
Whether Creedmoor Satisfied its Burden of Proof
In support of its petition, Creedmoor was required to establish by clear and convincing evidence that (1) “the patient lacks the capacity to determine the course of his own treatment,” and (2) “the proposed treatment is narrowly tailored to give substantive effect to the patient’s liberty interest, taking into consideration all relevant circumstances, including the patient’s best interests, the benefits to be gained from the treatment, the adverse side effects associated with the treatment and any less intrusive alternative treatments” (Rivers v Katz, 67 NY2d at 497-498; see Matter of Harvey S., 38 AD3d 906 [2007]). Here, after drawing an adverse inference against Creed-*183moor, the Supreme Court found that Creedmoor satisfied the first element, but failed to satisfy the second element by clear and convincing evidence. We agree with the Supreme Court that Creedmoor’s failure to prove the second element required denial of the petition.
Several entries in the certified medical record were inconsistent with Brodsky’s testimony. Most notably, while Brodsky opined that the patient actually had shown a “good response” to medication in the past, her opinion apparently conflicted with Mathew’s notations in the Clinical Summary that the patient had shown a “poor” response to medication, and in his Evaluation that the patient had a “partial to fair response” to medication. While Mathew’s notations may have reflected the interruptions in treatment occasioned by the patient’s intermittent refusals to accept it, Mathew was not in court to offer an explanation. Further, the certified medical record contained several clinical notes indicating that, during periods when the patient had refused his medication, he was calm and exhibited no aggressive, violent, or threatening behavior. Brodsky testified that the patient had previously taken all of the proposed medications without exhibiting any severe side effects, yet Mathew’s Evaluation contained a list of past medications which did not include all of the proposed medications. Thus, Brodsky’s testimony provided no assurance that none of the proposed medications would cause the patient severe side effects. Moreover, Brodsky’s testimony that the proposed treatment plan had been narrowly tailored to the patient’s specific needs, and that there existed no less intrusive alternative treatment, consisted of conclusory “yes” and “no” answers without elucidation. This, too, conflicted with Mathew’s Evaluation, which listed different medications as “reasonable alternatives.”
Given the Supreme Court’s proper application of the adverse inference permitted by the missing witness rule and the other circumstances presented, the court was entitled to infer that Mathew’s testimony with respect to the inconsistent matters in the record would not be favorable to Creedmoor’s position. Specifically, for example, the Supreme Court was entitled to deem established that the patient had previously experienced a “poor,” “fair,” or “partial” response to medication, taking Mathew’s notes at face value, in light of his failure to come to court and explain (see Matter of Jane PP. v Paul QQ., 65 NY2d at 996). Accordingly, the Supreme Court correctly determined that Creedmoor failed to meet its burden of establishing, by
*184clear and convincing evidence, that the proposed treatment is narrowly tailored to protect the patient’s liberty interest (see Rivers v Katz, 67 NY2d at 497-498; cf. Matter of William S., 31 AD3d 567, 568 [2006]), and thereupon, properly, in effect, denied the petition and dismissed the proceeding.
Creedmoor’s remaining contention is without merit.
Accordingly, the order is affirmed.