On legal guardianship: An exploratory assessment of knowledge, attitudes and practices of resident physicians

Abstract Background Clinicians encounter patients under legal guardianship. We aimed to assess the knowledge, attitudes and practices (KAP) on legal guardianship in residents. Methods A KAP pilot survey about legal guardianship was developed by an interdisciplinary medicine-law-public health team and was distributed via institutional email to internal medicine, psychiatry, and neurology residents in a single academic institution. Results Of the 172 invited residents, 105 (61%) responded and 102 surveys were included in the final analysis. Most respondents (58% women; internal medicine 73%, neurology 15%, psychiatry 12%) had attended 42 medical schools from 16 countries and had heard about guardianship (88%), but only 23% reported having received training on guardianship during medical school or residency. The vast majority (97%) understood the intended benefit of guardianship, but only 22.5% reported knowing that guardianship removed an individual’s decision-making rights. Nearly half (47%) of respondents reported never having asked for documentation to prove that an individual was a patient’s guardian, and only 15% expected to see a court order as proof of guardianship status. Conclusions Although most residents intuitively understood the intended benefit of guardianship, they did not understand its full implications for clinical practice. Training interventions are warranted.


Introduction
In the United States, there are more than 1.5 million adults under legal guardianship or conservatorship (hereafter 'guardianship'), and that figure is likely to increase as the population continues to age (Blanck and Martinis 2015;Cohen et al. 2015).Guardianship over an adult age 18 or over is awarded when the court determines an individual incapacitated and unable to make some or all decisions; upon such a finding, the court appoints an individual or organization (i.e.guardian) to make decisions on the individual's behalf.Guardianship is a type of substitute-decision making.In most cases, guardians are family members or friends, but in approximately 25% of all cases, the guardian is an organization or official with no knowledge of the impaired individual prior to appointment, sometimes referred to as a 'professional guardian' (Cohen et al. 2015).State law governs the guardianship process (Whitton and Frolik 2012;State of New Jersey 2022;Zietlow et al. 2022) and the State decides whether complete decision-making authority and responsibility over the individual is needed (i.e.plenary or full guardianship) or whether only certain decision-making rights, for example finances or health care decisions, should be vested in another (i.e.limited guardianship).Full guardianship is the most prevalent type of guardianship awarded (Raley et al. 2020).
Physicians and residents must comply with laws and regulations governing and relating to physicians' practice

Practice points
Physicians should be able to identify and help address patient's unmet legal needs that have an adverse effect on their health and well-being.In the US, there are more than 1.5 million adults under legal guardianship and this number is likely to increase as the population ages.Physicians will likely encounter patients under legal guardianship in their practice.There is a paucity of data regarding resident physicians' knowledge of legal guardianship and its related processes.Our findings indicate that resident physicians may benefit from training in guardianship's legal definition, requirements, and effects on patients, as well as in how to interact with patients under guardianship to ensure their legal rights are respected.Further research on clinical trainees' education on guardianship is warranted.(Kollas & Frey 1999;McAbee et al. 2006;Evans and Refrow-Rutala 2010;Preston-Shoot et al. 2011;Preston-Shoot and McKimm 2013;Garagarza-Mariscal et al. 2016).In fact, the Accreditation Council for Graduate Medical Education (ACGME) Common Program Requirements for residency programs state that 'medical practice occurs in the context of an increasingly complex clinical care environment where optimal patient care requires attention to compliance with external and internal administrative and regulatory requirements' (Accreditation Council for Graduate Medical Education 2022).Ideally, physicians should be able to identify and help address patient's unmet legal needs that have an adverse effect on their health and well-being (Paul et al. 2009;Regenstein et al. 2018).Despite this, a recent narrative review points out that the ACGME does not consider education in some issues such as legal guardianship as a required competency in internal medicine (Zietlow et al. 2022).
In clinical practice, physicians at all stages of training will encounter patients who require substitute decisionmaking, including patients under guardianship or who will undergo the guardianship process (Cohen et al. 2015;Foley et al. 2017;Farrell et al. 2021).Since resident physicians in training will provide care to these patients, they need to understand the implications and legal processes related to caring for someone under guardianship.However, the evidence of physicians' knowledge about guardianship is scarce.Of the 1,115 titles retrieved over the last 102 years in a PubMed search using the terms ("guardianship" OR "legal guardianship" OR "conservatorship"), few are related to attending physicians, and even less to physicians in training (i.e.residents); this may reflect physicians' suboptimal understanding of guardianship-related processes (Leatherman and Goethe 2009;Moye and Naik 2011;Comer et al. 2017;Morphew et al. 2021;Zietlow et al. 2022).Basic information about what residents know and perceive about legal guardianship is needed to identify gaps that require educational approaches.In this study, we developed a proof-of-concept survey to explore the knowledge, attitudes, and practices of full guardianship in internal medicine, psychiatry, and neurology residents.

Methods
The present study employed a cross-sectional survey on full guardianship, developed by a multidisciplinary medicine-law-public health team based on the knowledge attitudes practice (KAP) framework (World Health Organization & Stop TB Partnership 2008; M edecins du Monde 2011; World Health Organization 2014); Survey reporting guidelines were followed (Artino et al. 2022) .

Survey design
The survey design began with the operationalization of variables, followed by the initial drafting of the items, and finalized with seven reviews and editions over a period of 4 months.A legal content-expert on guardianship (J.R.V.), a health services researcher (T.I.M.), and a resident physician from the participating academic institution who had taken care of patients under guardianship (A.B.O) participated in the survey design.The survey (Supplementary Material 1) was composed of sections related to the demographic characteristics of the participants, and the dimensions of knowledge, attitudes, and practices on guardianship.For each dimension, the variables and testing domains were defined a priori with operational definitions grounded in extant literature (Moye et al. 2007;Then et al. 2018;Chamberlain et al. 2018;State of New Jersey 2022;Raley et al. 2020;Werner & Holler 2020).The questions were administered in multiple-choice and Likert formats, with some questions allowing multiple responses per question.Some questions had a branching nature and only appeared based on the answer to the previous question.The survey was not piloted before implementation.
Since guardianship is regulated at the state level and the guardianship processes may vary among states (Zietlow et al. 2022), we assessed content areas relevant to physicians that were unlikely to vary by state.Content areas for the 'Knowledge' domain included the definition of guardianship, awareness of the topic, understanding the implications of guardianship on the individual's rights, and familiarity with the role a physician may play in the guardianship process.Content areas for the 'Attitudes' domain included the perceived value for clinical practice of knowing about guardianship, willingness to share information about or to suggest the need for guardianship to patients and/or caregivers, the position about educating patients and/or caregivers on guardianship, and the position on pursuing other options of substitute decision-making before guardianship.Content areas for the 'Practices' domain were the experience caring for patients under guardianship and experience participating in the guardianship process.The survey started with demographics, followed by an assessment of the knowledge and practices domains.The attitudes domain came last; to focus the respondents' attitudes on a unified concept of guardianship, a formal definition of guardianship was provided at the start of that domain (State of New Jersey 2022).

Survey distribution
In October 2021, surveys were distributed via institutional email to all residents of internal medicine, psychiatry, and neurology from a single academic institution (Rutgers New Jersey Medical School, USA) using Qualtrics (Qualtrics, Provo, Utah).We targeted residents in these specialities because they are most likely to encounter patients who lack decision-making capacity due to mental illness or medical conditions and to be asked to perform a capacity assessment for guardianship purposes.Participants received an individualized access link that anonymized responses and prevented duplicate responses from individual accounts.Survey participants were able to access the survey at their time of preference, and the survey could be completed at a later time after it had been started.Following the initial email, a reminder text through an institutional text platform was sent on day 4, followed by reminder emails on days 7 and 10.The survey remained accessible for 20 days, and unsubmitted surveys (i.e.those in which the respondent did not click the 'finish' button after the last question) at the time of closure were eliminated.Response to every question was not mandated.Participants were not offered any incentives.

Survey analysis
Given the non-mandatory nature of the items, surveys missing >4% of responses on the KAP items were excluded from the analysis.Study results were analyzed using descriptive statistics.Continuous variables are presented as means and standard deviations or medians and interquartile ranges, as appropriate.Categorical variables are expressed as absolute numbers and percentages; the percentages were calculated based on the number of each item's respondents unless otherwise specified.For reporting the five-point Likert scale data, the 'strongly disagree' and 'disagree' options were grouped and expressed as 'disagree,' and the 'strongly agree' and 'agree' options were grouped as expressed as 'agree.' No attempts to test for associations were made for the Likert-type items, as suggested (Harpe 2015).The exploratory nature of this survey and the expected differences in speciality sample sizes were limited using inferential statistics, but Chi-square for multiple groups was used when appropriate.Statistical significance was established at 0.05.Data was analyzed using IBM SPSS Statistics for Windows, version 28.0 (IBM Corp., Armonk, N.Y., USA).The study was approved by the Institutional Review Board at Rutgers University.

Results
Emails were sent to all one-hundred seventy-two residents in the selected specialities (124 from internal medicine, 28 from psychiatry, 20 from neurology).There were 105 complete and 16 unsubmitted surveys at closure time.We eliminated the sixteen unsubmitted surveys and excluded from analysis two completed surveys for which consent was not provided, and one survey with consent but no other answers.
One hundred and two complete surveys by resident physicians were used for the final analysis (response rate 59%).Table 1 shows the respondents' characteristics, composed of internal medicine residents (73%), neurology residents (15%), and psychiatry residents (12%).Most were first and second-year residents (31% and 30%, respectively).Psychiatry n ¼ 5/13 [38%]; X 2 p ¼ 0.13), and when asked to select each individual right removed, none identified that all rights were removed (Table 2).About 70% (n ¼ 71) of respondents agreed that the treating physician may play a role in the guardianship process, with sixty-two (87%) of these respondents indicating that the physician's most likely role would be to assess the decision-making capacity of the individual.

Attitudes domain
Eighty-five per cent (n ¼ 86) of respondents agreed that less restrictive support options to make decisions, such as a healthcare proxy or a power of attorney, should be considered before resorting to full guardianship.
Table 3 shows respondents' attitudes toward providing information and suggesting the pursuit of guardianship for each clinical vignette.For the clinical vignettes of patients with prolonged coma and severe autism, most residents leaned towards providing guardianship information (76% and 93%, respectively) and recommending the pursuit of guardianship (69% and 93%, respectively), which contrasted with the residents' attitudes for mild dementia, quadriplegia, and severe substance use disorder vignettes.
Ninety-one per cent (n ¼ 93) of respondents agreed that physicians in their specialities should know how to perform a capacity assessment for guardianship, while 89% (n ¼ 91) agreed that knowing about the guardianship process was necessary for their clinical practice.Furthermore, 92% (n ¼ 94) agreed that physicians should provide guardianship education to those patients for whom guardianship is being sought, and 90% (n ¼ 92) agreed that physicians also should educate caregivers pursuing guardianship about the rights that the individual will lose if guardianship is granted.Additionally, 85% (n ¼ 87) of respondents agreed that physicians should ask patients for whom guardianship is being sought if they want a legal guardian appointed to make decisions for them, and 78% (n ¼ 80) agreed that the patient's position about guardianship should be included in the capacity assessment report.
Ninety-six per cent (n ¼ 98) of respondents thought they would benefit from learning more about guardianship and their role in the process as physicians, with similar percentages among all specialties (Internal Medicine n ¼ 71/74

Practices domain
Fifty-six per cent (n ¼ 57) of respondents had cared for a patient with a guardian, while 22% (n ¼ 22) had not and 22% (n ¼ 23) could not recall.Of those who had treated patients under guardianship, 84% (n ¼ 48) stated having cared for 1 to 5 patients in the last 12 months and only 7% (n ¼ 4) stated having cared for 6 or more during that period.Additionally, 70% (n ¼ 71) of respondents stated having cared for a patient who would have benefited from having a guardian but did not have one, based on their professional opinion.Twenty-one percent (n ¼ 21) stated they had recommended guardianship to a patient and/or caregiver in the past.
When asked the question 'when a patient is accompanied by another person to the clinical visit, I ask if that person is their legal guardian,' 2% (n ¼ 2) answered 'always,' 47% (n ¼ 48) answered 'never,' and 41% (n ¼ 42) answered 'sometimes.'However, only thirty-eight (37%) respondents answered that they ask for a document to prove the guardianship status (17 'always' asked) and, of these, only 15 (15% of the total) indicated that the document they expect to see is a court order.When asked 'If the alleged legal guardian offers no proof/documentation of guardianship, I still take directives from them?' only 47.5% (n ¼ 47) answered 'never.' Only two of the respondents answered that they had completed a Physician Certification Form in support of a guardianship application, and only one stated to have assessed the patient's ability to understand and retain information to make decision and communicate agency over their own affairs.Both respondents were neutral on whether the form was straightforward to complete.

Discussion
This exploratory survey showed that encountering patients under guardianship was not uncommon for residents at a large and diverse academic institution.It also showed that there were multiple opportunity areas to improve the knowledge and practices when caring for patients under guardianship.There are two main findings.
First, most residents intuitively understood the intended benefit of guardianship but did not comprehend its full implications, including the most relevant ones.These include not knowing that guardianship results in the removal of an individual's decision-making rights and not asking for the document that proves that a person claiming to be a patient's guardian actually is the guardian.These findings indicate that residents may benefit from training on the legal definition of, requirements for, and effects of guardianship, and on how medical professionals should interact with a patient who has a guardian or who is undergoing the evaluation for guardianship to ensure their legal rights are respected.Knowing the law and regulations is important, as just guessing may lead to unwanted legal scenarios.The only other study that has assessed residents' attitudes toward guardianship, but from a different perspective, also shows that residents may benefit from education about guardianship (Morphew et al. 2021).
Physicians will also benefit from understanding the potential harms caused by depriving the person of their self-determination as a result of guardianship, which could lead to lowered self-esteem, lowered perceived self-efficacy, behavioral passivity, and the potential for abuse of privilege by appointed guardians (Jameson et al. 2015;Zietlow et al. 2022).
Second, almost all respondents expressed interest in learning more about guardianship and the role they may play in it as physicians.Although most believed that their likely role would be to assess the decision-making capacity of the individual over whom guardianship is sought, their actual role might be far broader.For instance, internal medicine, neurology, and psychiatry physicians will encounter hospitalized patients who have impaired decisionmaking capacity and yet have no one appointed to engage in decision-making for them (Anderson-Shaw 2019;Schweikart 2019;Farrell et al. 2021).In some of these cases, patients may require a temporary or permanent guardian to be appointed during hospitalization (Bandy et al. 2010;Ricotta et al. 2018).Physicians may volunteer or be subpoenaed to participate as expert witnesses in court cases regarding an individual's capacity and need for guardianship.Physicians should also know about the essential elements of a capacity evaluation or where to obtain more information (Moye and Naik 2011;Zietlow et al. 2022).
Physicians also encounter patients who already are under guardianship, and thus must verify that an individual is a patient's guardian by asking to see the court order awarding guardianship and the type of guardianship (full or limited) awarded (Leatherman and Goethe 2009).Physicians can assist guardians in making decisions while protecting the rights of the patients, but require knowledge of state laws and regulations that govern guardianship to provide guidance on the decisions a guardian can make independently and which may require further judicial review (Cohen et al. 2015(Cohen et al. , 2019(Cohen et al. , 2021;;Effiong & Harman 2014;Hastings 2014;Sandler 2014).For instance, in 2015 only five states had laws permitting guardians to make end-of-life decisions independently; eight prohibited these decisions without further judicial review; and 37 had no specific language about a guardian's authority to make such decisions (Cohen et al. 2015).Notably, most of the residents stated they had not received any training about guardianship during their medical school or residency training.
A strength of this study is that respondents included residents from 42 different medical schools from 16 countries, and 68% of respondents were from 24 U.S. medical schools.This might shed light on what happens in other academic centers given the diversity of our sample, although it also supports the need to study this topic with larger, multicenter samples that allow for more robust subgroup analyses.However, our study has several limitations.It is a cross-sectional survey using self-reported data; accordingly, the results may differ from the residents' actual practices.Also, we only evaluated residents from internal medicine, psychiatry, and neurology.However, we focused on these specialities as they encounter patients who lack the decision-making capacity due to mental illness or medical conditions.Furthermore, we evaluated only a portion of the residents from the selected specialities.Since the survey was voluntary, our results could be an underestimation if any of those who opted not to participate did so because of a deliberate reason, such as not having knowledge about the guardianship process.Additionally, this study is exploratory and uses Likert-type scales that preclude inferential statistics, but the data is relevant because it shows that there is a relevant gap in the information for the clinical practice of residents; it allows planning to solve this need for information and provides the platform for further hypothesis-testing studies.Although there were multiple iterations in the construction of the questionnaire, and we followed the recommendations to create KAP questionnaires, our questionnaire is not a validated tool, so the results may vary if the wording of the questions is changed.Finally, the laws regulating guardianship vary by US state.Although we focused the items on clinically relevant areas that are unlikely to vary by state, some items may need adjustments if used in other geographic regions.
In summary, our study provides information that guardianship-related KAP among residents from a single US academic center has opportunities to improve.Legal guardianship in the clinical care setting seems to be an understudied topic inside and outside of the US.Yet, physicians are likely to take care of patients on guardianship or under consideration for it, and a lack of understanding of the processes related to the care of these patients may result in adverse consequences for the patient or the physician (e.g.disregard of end-of-life preferences, legal conflicts).As more studies help to more precisely elucidate the generalizability and implications of our results, it would be valuable for clinicians to learn about the processes required to take care of those under guardianship and for medical education institutions to foster educational efforts on this topic.

Glossary
Assisted-decision making: Individuals with disabilities aged 18 and over may require help to make decisions.The legal system provides a range of approaches for assisted decision-making that vary in their restrictiveness, from supported decision-making to substitute decision-making (e.g.guardianship).

Guardianship:
A type of substitute-decision making in which a court determines an individual incapacitated and unable to make some or all decisions; upon such a finding, the court appoints an individual or organization (i.e.guardian) to make decisions on the individual's behalf.Also known as conservatorship.

Table 1 .
Demographic characteristics of the respondents.Other languages were Chinese, French, German, Russian, Serbian, Telugu, Turkish and Urdu.

Table 2 .
Decision-making rights that respondents selected as being removed from the patient when guardianship is granted.Respondents had attended 42 different medical schools from 16 countries, but 70% were English-speakers from the U.S. (68%).Only 58 respondents provided their age (average age 30 ± 3.9).

Table 3 .
Attitudes to provide information about guardianship or to recommend pursuing guardianship on the clinical vignettes.