Which instruments are used to measure shared, supported and assisted healthcare decision-making between patients who have limited, impaired or fluctuating capacity, their family carers and healthcare professionals? A systematic review protocol

Background: Shared decision-making (SDM) is a dialogical relationship where the physician and the patient define the problem, discuss the available options according to the patient’s values and preferences, and co-construct the treatment plan. Undertaking SDM in a clinical setting with patients who have limited, impaired or fluctuating cognitive capacity may prove challenging. Supported (defined “Assisted” in the Irish context) decision-making describes how people with impaired or fluctuating capacity remain in control of their healthcare-related choices through mechanisms which build and maximise capacity. Supported and assisted decision-making (ADM) within healthcare settings is theoretically and practically novel. Therefore, there is a knowledge gap about the validity of psychometric instruments used to assess ADM and its components within clinical settings. This systematic review aims to identify and characterise instruments currently used to assess shared, supported and assisted healthcare decision-making between patients with limited, impaired or fluctuating capacity, their family carers and healthcare professionals. Methods: A systematic review and narrative synthesis will be performed using a search strategy involving the following databases (PubMed, Cinahl, Embase, Web of Science, Scopus and PsycINFO). Quantitative studies published in the last decade and describing psychometric instruments measuring SDM, supported decision-making and ADM with people having limited or fluctuating capacity will be considered eligible for inclusion. Title and abstract screening will be followed by full-text eligibility screening, data extraction, synthesis and analysis. This review will be structured and reported according to the PRISMA checklist. The COSMIN Risk of bias checklist will be used to assess the quality of the instruments. Discussion: The results will inform and be useful to HCPs and policymakers interested in having updated knowledge of the available instruments to assess SDM, supported and assisted healthcare decision-making between patients who have impaired or fluctuating capacity, their family carers and healthcare professionals. Registration: PROSPERO CRD42018105360; registered on 10/08/2018.


Introduction
Person-centred healthcare promotes the autonomy of persons about their treatment choices and places patients at the centre of care planning, considering them as partners in the decision-making process (Kusnanto, 2018;Tullo et al., 2018). Considering patients' will and preferences in the development of care plans and decision-making related to medical treatment choices is central to person-centred healthcare (Mulley et al., 2012). Research has highlighted that informed patients and families, receptive healthcare professionals, as well as coordinated and supportive healthcare environments, are crucial elements in the implementation of a patient-centred approach to health service planning and delivery (Epstein et al., 2010). Patient participation in care planning and treatment decision-making is the result of a cultural shift in historically paternalistic healthcare settings (Weston, 2001). This shift recognises the importance of considering patients' will and preferences in the development of care plans and decision-making related to medical treatment choices (Mulley et al., 2012). The research evidence has highlighted that informed patients and families, receptive healthcare professionals, as well as coordinated and supportive healthcare environments, are crucial elements in the implementation of a patient-centred approach to health service planning and delivery (Epstein et al., 2010).
This drive towards developing a culture of person-centredness and patient participation in healthcare policy, research, and service delivery has instigated a growing body of literature which is exploring patient engagement in care planning (Angel & Frederiksen, 2015). This literature focuses on the dynamic relationships between healthcare practitioners (HCPs), patients and their family caregivers as well as themes such as time, knowledge, the patient's situation and HCPs attitudes (ibidem; Davies et al., personal communication). Consequently, several new concepts have arisen in the last decade that describe the different roles that patients may assume within the healthcare system: patient participation, patient activation, patient engagement, shared decision-making (SDM), supported decision-making and patient involvement among the others (Barello et al., 2016;Browning et al., 2014;O'Donnell et al., 2018).

Shared decision-making and its recent development
Shared decision-making (SDM) is characterised by a dialogical relationship where the physician and the patient define the problem, discuss the available options according to the patient's values and preferences, and co-construct the treatment plan (Makoul & Clayman, 2006). The dynamic nature and complexity of the concept of shared decision-making has challenged scholars seeking to define it (Makoul & Clayman, 2006). SDM with patients who have limited, impaired or fluctuating cognitive capacity is defined supported decision-making (i.e. in Canada, Australia and US) or assisted decision-making (in Ireland) and may prove challenging. In this instance, specific support and assistance is required to build a patient's capacity and to enable shared decision-making. Supported decision-making describes how people with impaired or fluctuating capacity remain in control of their healthcare-related choices through mechanisms which build and maximise capacity (Browning et al., 2014;Davidson et al., 2015). These supportive mechanisms may include nominated decision-supporters, decision-making aids or assistive communication technologies.
Since 2006, successive states have ratified the United Nations Convention on the Rights of Persons with Disabilities (UNCRPD) (2006). This is a human rights instrument which aims to protect the autonomy and promote the full participation of people with disabilities within international human rights laws. The convention protects the rights of people with disabilities to autonomy and participation in all decisions which affect their lives, including healthcare decisions. Furthermore, it enshrines the right for people to have their decision-making capacity supported through an informed decision-making process (Davies et al., in publication).
Both supported and assisted decision-making are challenging constructs in terms of definition and implementation (Browning et al., 2014;O'Donnell et al., 2018). They can both be considered and defined as a process, a legal framework, a mechanism and a system, making its implementation and translation into practice harder (Browning et al., 2014;Kohn & Blumenthal, 2014 ). Assisted and Supported decision-making describe the process leading people with impaired or fluctuating capacity to remain in control of their healthcare-related choices through mechanisms which build and maximise capacity. These supportive mechanisms may include nominated decision-supporters, decision-making aids or assistive communication technologies (insert citation).
Several different instruments have been developed to assess the complex and multifaceted phenomenon of shared decision-making in the healthcare setting (Bouniols et al., 2016;Makoul & Clayman, 2006;Scholl et al., 2011;Simon et al., 2007). These scales measure different components or phases in the SDM process between physicians and patients having cognitive capacity, such as antecedents of the decision-making process, the process itself and the decision outcomes (Perestelo-Perez et al., 2017;Scholl et al., 2011).
Several systematic reviews have been undertaken within the last 20 years, which have mainly focused on the retrieval and analysis of instruments assessing the different components of SDM within healthcare settings between patients with mental capacity and physicians. Elwyn et al. (2001) could not retrieve any study evaluating the involvement of patients with assumed or established limited cognitive capacity in shared decision-making. Six years later, Simon et al. (2007) were able to identify 18

Amendments from Version 1
We have reviewed the parts that were less clear according to the reviewers' notes.
There are no major differences, just a few clarifications about terms and their definitions� .
Any further responses from the reviewers can be found at the end of the article instruments which measured the patients' perspective, preferences for information and participation, decisional conflict, self-efficacy as well as the evaluation of the decision-making process and outcomes. Scholl et al. (2011) contributed to this growing area retrieving 28 scales, underlining their development and validation in languages other than English, recognising an increasing internationalisation of SDM. Recently, 19 studies have been included by Bouniols et al. (2016), highlighting the evolution of instruments which take into account points of view of patients, HCPs and external observers. Recently, Perestelo-Perez et al. (2017) focused on SDM measures within the mental health area and reported 48 instruments, mainly assessing the SDM process from the patients' perspective.
Because of the novelty of concepts such as supported and assisted decision-making, there is no reference within the scientific literature about instruments able to assess these processes in the clinical practice. Accordingly, for this review, we consider shared decision-making instruments used with cohorts of people with a physical, age-related or mental health condition that may lead to limited, impaired and fluctuating capacity as plausible measures of supported and assisted decision-making. This systematic review aims to identify and synthesise the instruments used to measure shared, supported and assisted healthcare decision-making between patients who have limited, impaired or fluctuating capacity, their family carers and healthcare professionals.

Review question
Which instruments are used to assess shared, supported and assisted healthcare decision-making between patients who have limited, impaired or fluctuating capacity, their family carers and healthcare professionals?

Selection and inclusion criteria
The following selection inclusion criteria will be considered: e) Papers describing psychometric instruments as objects of a creation and validation study or used as part of a battery within a broader study; f) Instruments assessing SDM, supported or assisted decisionmaking related antecedents, process and outcomes constructs; g) The population targeted by the instruments will include: People presenting limited, impaired or fluctuating capacity due to a physical or diagnosed mental health condition; healthcare professionals of any type (i.e. physicians, nurses, occupational therapist, physiotherapist and so on) working in primary, secondary and tertiary care such as hospitals, nursing homes, psychiatric hospitals and rehabilitation hospitals; family members and patient nominees acting as surrogate or decisionmaking supporters; h) The outcomes will be direct patient-reported outcome measures or family carer-reported outcome measures, clinician-reported outcome measures and objective observer-based outcome measures. c) We will exclude those instruments that the authors do not explicitly consider as measures of SDM, supported or assisted decision-making even if labelled otherwise (e.g. decisional conflict scale not used as a measure of SDM).

Information sources
Two authors developed and agreed on the search strategy (keywords, subject headings, limiters, and so on). Two authors (FF, BRM) will run the search independently on the following electronic databases Cinahl, PubMed, Embase, Web of Science, Scopus and PsycINFO. The results of the independent searches will be exported and uploaded on a reference management software (Zotero). After the elimination of the duplicates, and as highlighted previously, the above-cited limits will apply to the search (please refer to paragraph Selection criteria). The first search will be run in July 2019. As a secondary search, reference lists of the included full-texts will be used as a further retrieval source.

Search strategy
The keywords composing the search strings have been adapted from a previous realist review of research evidence concerning the mechanisms which support ADM in healthcare settings (Davies et al., 2019). This previous search strategy was modified according to the aims of the present review and will be used as the basis for the development of the database-specific (Mesh and Headings) search strategies as outlined here below in Table 1.
We will use a flow diagram to report the inclusion and eligibility process and a table to describe the main features of the studies included in the final review. The full search strategies for the three databases with relative Boolean rationales will be reported in the appendix.

Studies selection and screening criteria
Initially, the authors will retrieve the initial pool of studies from the databases following the inclusion criteria described above. We will use Zotero to remove duplicates. Then, we will proceed with the title and abstract screening. FF and BRM will conduct the screening and the full-texts review phase independently and will identify cases which require discussion and resolution by consensus. DOD will verify a sub-proportion of the texts, and the inter-rater agreement will be assessed. Again, disagreements will be solved by a discussion between FF and BRM; if a solution cannot be found, DOD will decide. Assessment of the quality of the studies and instruments FF and BRM will proceed to the assessment of the quality of both the instruments and the papers independently. Any disagreement will be referred to a third reviewer (DOD). The quality of the retrieved instruments will be assessed with the COSMIN Risk of bias checklist (Mokkink et al., 2018). This appraisal will evaluate the methodological rigour of the validation process and the psychometric properties of the instruments. The instruments will be presented in such a way as to align them with patient characteristics, healthcare setting and outcome.

Data extraction
Data synthesis Findings will be synthesised using a narrative synthesis approach (Popay et al., 2006). There is currently no method to empirically group measurement properties; synthesis is then recommended (Beattie et al., 2014). We will categorise the instruments according to the assessed variable, their format, and what feature of the ADM process investigate (antecedents, the process itself or the outcomes).
We will also discuss patterns of recurrences and differences between the instruments, uncovering and highlighting the communal components, identifying patterns that lead to the utilisation of effective tools assessing ADM with people with limited, impaired or fluctuating capacity.

Study registration
The systematic review is registered with the protocol number CRD42018105360 (10/08/2018) in the PROSPERO register.

Discussion
The systematic review aims to analyse the last ten years of development of psychometric instrument assessing shared decisionmaking with people with limited, impaired or fluctuating capacity, supported and assisted decision-making. Due to the novelty of supported and assisted decision-making constructs, we explicitly chose to address instruments assessing shared decision-making with people who have limited, impaired or fluctuating capacity. This will allow us to formulate plausible inferences about the validity of tools which measure supported and assisted decision-making.
By considering instruments assessing supported decisionmaking processes with samples of people with limited, impaired and fluctuating capacity, this review will inform practice development about building and supporting decision-making with this cohort of patients. Through the categorisation of the results, policy-makers and managers of healthcare settings may find useful insights on which instruments are more appropriate than others to assess different components of the complex process of supporting the decision-making of patients who have limited, impaired or fluctuating capacity. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Paul Webb
Praxis Care, Belfast, UK Thank you for inviting me to review a study protocol for a systematic review and narrative synthesis of the psychometric instruments which are used to measure shared, supported and assisted healthcare decision-making between patients who may have limited and impaired capacities or capacities which change through time and their family carers and healthcare professionals.
I think that this study could potentially be an important contribution to the literature as well as being an excellent resource for healthcare practitioners (HCPs). In general, the study protocol is well written although I'm not entirely sure that key constructs like SDM and ADM are described in sufficient detail and that the study design is wholly appropriate to the research question. My reasoning is as follows: Shared decision-making, supported decision-making and assisted decision-making are not clearly defined in the study protocol e.g. "SDM with patients who have limited, impaired or fluctuating cognitive capacity is defined supported decision-making (i.e. in Canada, Australia and US) or assisted decision-making (Ireland) and may prove challenging" (p. 3). A bit more clarity and detail may perhaps be needed so that the reader can understand the distinctions between shared decision-making, supported decision-making and assisted decision-making and thereby appreciate what the respective instruments are trying to measure.

1.
It's unclear from the study protocol whether the scope of the systematic review and narrative synthesis includes psychometric instruments which measure SDM, supported decision-making and ADM or whether the scope is, in effect, restricted to instruments which measure SDM alone. The authors state that "this systematic review aims to identify and characterize instruments currently used to assess shared, supported and assisted healthcare decision-making" (p. 1) whilst writing further on in the study protocol that "there is no reference within the scientific literature" (p. 4) to instruments which can assess supported decision-making and ADM.

2.
The authors then suggest that shared decision-making instruments can be "plausible 3.
measures of supported and assisted decision-making" (p. 4) which seems to confirm that there are a lack of instruments assessing supported decision-making and ADM. In the absence of instruments which assess supported decision-making and ADM, what's the rationale for assuming that shared decision-making instruments can be used as measures of supported and assisted decision-making? If there isn't a convincing rationale, then the scope of the systematic review may be narrower than the study protocol suggests.
Moving on to the selection inclusion criteria, the target population on page 4 includes health professionals i.e. 'physician', 'nurse', 'occupational therapist' and 'physiotherapist'. However, 'occupational therapist' and 'physiotherapist' are not included in the search terms listed in Table 1 (p. 5). In a similar vein, 'psychologist' is omitted from the list of search terms too. Is there a rationale for this? Conversely, 'social worker' is not listed as an example of a healthcare professional in the selection inclusion criteria but appears as a search term in Table 1. Is there a rationale for this?

4.
I hope that this review is of assistance and thank you again for the invitation to review this study protocol.

Is the rationale for, and objectives of, the study clearly described? Partly
Is the study design appropriate for the research question? Partly decision-making, among those with limited or fluctuating decision-making capacity.
I found the protocol to be clearly written, and overall to describe a suitably rigorous approach to the review. I did have two reservations, which are described below. These relate to the clarity of the protocol description and a potential issue with the exclusion of relevant papers from the observational research literature.
Point 1: On p3 the authors write "Accordingly, for this review, we consider shared and assisted decision-making instruments used with cohorts of people with a physical or mental health condition that may lead to limited, impaired and fluctuating capacity as plausible measures of supported and assisted decision-making." While it becomes clear later in the manuscript that tools relevant for participants with age-related cognitive impairments are included in the review, this definition is potentially confusing in its implication that measures/instruments used among people with age-related cognitive impairments (which are not necessarily a mental health condition) are not included. This could be addressed relatively easily.
Point 2: On p4, the final search term string is defined as "randomised controlled trial OR controlled clinical trial OR randomised OR placebo OR randomly OR trial OR groups OR nonrandomised controlled trial OR cohort OR control". I am surprised that the terms here are so narrow, potentially excluding psychometric validation studies, epidemiological and/or observational studies, all of which might contribute useful tools for measurement of shared, supported or assisted decision-making. I would recommend the inclusion of terms "validation", "longitudinal" and "observational".
Thank you once again for the opportunity to contribute to this work.
Is the rationale for, and objectives of, the study clearly described? Yes

Are sufficient details of the methods provided to allow replication by others? Yes
Are the datasets clearly presented in a useable and accessible format? Not applicable Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Psychology, health services research, older adult decision-making, dementia, supported decision-making, advance care planning.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.