Experiences of assessing mental capacity in England and Wales: A large-scale survey of professionals

Background: The Mental Capacity Act (2005) of England and Wales described in statute a test to determine whether a person lacked the “mental capacity” to make a particular decision. No large-scale survey has explored experiences of capacity assessment across professional groups. Methods: We administered an opportunistic self-report questionnaire survey of professionals who undertake capacity assessments in England and Wales (n= 611). Topics of interest included; how often and where capacity assessment took place, self-ratings of competency and challenges experienced in assessment, use of psychological testing and concerns about undue influence. We analysed the quantitative responses using a mixed-methods approach using regression methods for the quantitative ratings and a thematic analysis for qualitative data. Results: Our sample included 307/611 (50.2%), social workers, 89/611 (14.6%) psychiatrists, 62/611 (10.1%) nurses, 46/611 (7.5%) clinical psychologists, 30/611 (4.9%) doctors from other medical specialties, 12/611 (2.0%) speech and language therapists and 8/611 (1.3%) solicitors. 53% of these professionals undertook more than 25 capacity assessments per year, with psychiatrists, social workers and nurses undertaking them the most frequently. Most professionals reported high self-ratings of confidence in their assessment skills, although non-psychiatrist doctors rated themselves significantly lower than other groups (p< .005). Most professionals (77.1%) were at least moderately concerned about undue influence, with people with dementia and learning disabilities and older adults considered to be the most at risk. Qualitative themes for challenges in capacity assessment included inter-disciplinary working, complicated presentations and relational issues such as interpersonal influence. Requests for support mainly focused on practical issues. Conclusions: Most professionals feel confident in their ability to assess capacity but note substantial challenges around practical and relational issues. Undue influence is a particularly common concern amongst professionals when working with service users with dementia and learning disabilities which public services and policy makers need to be mindful of.


Introduction
The concept of mental capacity (or decision-making capacity) is fundamental to safeguarding the ethical principles of autonomy and informed consent. In England and Wales, the Mental Capacity Act 2005 (MCA 2005) and associated case law provide a legal definition of a test to determine whether a person aged 16 or over has the mental capacity to make a particular decision, and a framework to authorise an action to be taken in the best interests of a person who lacks capacity to take to a specific decision. Enshrined in the MCA 2005 are principles aiming to ensure that adults are supported to make informed decisions for themselves when they have the capacity to do so, and to have their views held at the core of the decision-making process even if they do not.
While the MCA 2005 has been heralded for its legal clarity, there are several ongoing debates over its implementation. Anyone can conduct or be subject to a capacity assessment at any time (Brown et al., 2015), so the challenges can vary considerably based on the person's needs and impairment, as well as the professional's skills and resources. To ensure high practice standards across the board, it is important to consider the full range of contexts in which these assessments can be applied. The recent National Institute for Care Excellence (NICE) guideline on decision making capacity outlined several research recommendations for health professionals who assess capacity (National Institute for Health and Care Excellence, 2018). Their key priorities included outlining training and support needs, the components of a capacity assessment and the potential use of tools to facilitate the assessment process. These should help to ensure the provision of high quality, legislation-compliant assessments (Jayes et al., 2020a;Jayes et al., 2020b) and to better engage with the most difficult aspects of assessing capacity, such as the 'use or weigh' criterion within the functional test (Keene et al., 2019).
Since the implementation of the MCA 2005 in 2007, only a small number of studies have explored the experiences of the professionals who carry out capacity assessments. These studies have each surveyed just one professional group (though some have included multiple sub-specialties of medicine), using opportunistic sampling, with fewer than 100 participants. Their findings have included: low confidence amongst doctors who assess capacity (Penn et al., 2020), concerns around underutilisation of speech and language therapists (McCormick et al., 2017) and difficulties assessing brain injury patients with fluctuating capacity (Norman et al., 2018). These studies suggest that experiences of capacity assessment may be context-specific, which could be explored within a larger and more varied sample.
Furthermore, to our knowledge, this literature has not considered professional views about how third-party social influences affect a person's mental capacity during assessments. For example, undue influence may result in a person being unable to exercise their capacity (Mandelstam, 2013 EWCA Civ 478, [2013] MHLO 61), but it is currently unknown to what extent they are reflected in everyday capacity assessments. A more comprehensive understanding of these issues will be key to ensuring effective implementation of the MCA 2005.
The main objectives of our study were to describe and compare the experiences of professionals who assess capacity in England and Wales. We sought to include professionals with a range of experience and responsibilities, across the health care, social care and legal disciplines. We were particularly interested in four questions: 1. When, where and how often do professionals assess capacity? 2. How effective do professionals feel they are at assessing mental capacity?
3. How do professionals consider undue influence when assessing mental capacity? 4. What are the challenges experienced when conducting any structured assessments?

Design
We developed a questionnaire for a cross-sectional survey (see extended data) (Ariyo et al., 2021) to explore the experiences of professionals assessing mental capacity. It was developed using insights from the literature and from the clinical experience of the research group. To maximise engagement from professionals, the survey was short and anonymous. It included a mixture of fixed-choice, Likert-scale and free text questions.
We then piloted the questionnaire within the research team, incorporating revisions where warranted.

Data collection
The web-based version of the survey was delivered via the academic survey provider onlinesurveys.ac.uk and was available from 07/02/2019 to 07/01/2020, in order to allow sufficient time to collect responses from a large sample and across different professions. Potential respondents were contacted using email and social networking sites, via professional networks and professionals with high profile social media presence. Hard paper copies of the same questionnaire were circulated at two academic conferences in London (in neuropsychiatry and in psychology). Partially completed questionnaires were included.

Data analysis
Quantitative analysis. Likert scale, yes/no responses and questions with numeric answers were analysed quantitatively, and K.A. performed the analyses using SPSS (version 27.0). We calculated descriptive statistics to determine means, standard deviations, normality of distributions (Shapiro-Wilk test and boxplots), homogeneity of variances (Levene's test) and missing data within the numerical responses. We considered 5% missingness as a reasonable cause for concern and 10% as significant cause for concern (Bennett, 2001;Schafer, 1999).
Following this, we used a series of one-way independent analyses of variance (ANOVAs) to explore differences between professionals to the following questions: how often they assessed mental capacity (Q4), how competent they felt in doing so (Q7), and how often they were concerned about undue influence (Q11). These were scored between one and five, with higher scores indicating higher values. As Q4 was originally scored on a four-point scale, we transformed each data point into a five-point scale for reporting purposes only, to ensure consistency.
The main effect of group differences is reported using eta squared (n 2 ), which is the most common effect size for ANOVA (Cohen, 1973). We also calculated 95% confidence intervals using the bias-corrected and accelerated (BCa) bootstrap method, as the assumption of univariate normality was violated. Following a significant result, we conducted a Tukey's post-hoc analysis to explore pairwise comparisons between professional groups (or the Games-Howell statistic when homogeneity of variances was violated).

Qualitative analysis.
Free text responses to one question, concerning the challenges experienced when assessing mental capacity, were analysed using a procedure based on the six steps of thematic analysis described by (Braun & Clarke, 2006), as this method has a track record of use in healthcare and social science research (Braun & Clarke, 2014).
The thematic analysis used a bottom-up, inductive, and experiential viewpoint (Clarke & Braun, 2013), comprising an iterative process of coding, derivation of thematic structure and comparison. Two researchers undertook the analysis (KA and AMcW).
We independently immersed ourselves in the responses, re-reading them while noting initial striking impressions, before each researcher independently applied to the responses codes which had emerged. Individual responses could be given more than one code, partial segments of responses could be coded, and every part of each response was placed within a code. We then compared the codes we had derived and, by referring back to the raw responses, agreed on a set of 51 codes. We then independently re-coded the responses and derived a prototype thematic structure. We continued an iterative process of coding and creating themes and subthemes until consensus was reached on both the coding and the thematic structure. We then chose the final names for the themes and subthemes they contained.
Although our survey generated free text in response to other questions, these generally were brief and somewhat factual responses, without sufficient elaboration to allow use of qualitative analysis. Instead, we outline the characteristics of the responses to portray the most salient features.

Ethics statement
Our project was approved following a review by King's College London's College Research Ethics Committee (MRA-18/19-10500). An information sheet about the study describing how data would be used was provided to all participants. Participation in the survey was taken as implying consent.

Sample characteristics
We received questionnaires from a total of 621 participants (573 online and 48 paper). Ten participants (nine online, one paper) were excluded for the following reasons: not a health, social care or legal professional (n=4), did not report their profession (n=3), only assessed capacity outside of England and Wales (n=2), only assessed capacity as part of a research study (n=1) (Ariyo et al., 2021). Table 1 outlines all of the included participants by profession. The 30 non-psychiatrist doctors within the sample included general practitioners (n=8), obstetricians or gynaecologists (n=7), neurologists (n=6), geriatricians (n=4) and a neurorehabilitation specialist. Four doctors didn't state their specialty. We also created a composite 'other' category for professional groups that were poorly represented in the sample. These included midwives (n=7), social care managers (n=7), best interest assessors (n=5), assistant psychologists (n=4), assistant social care workers/practitioners (n=4), physiotherapists (n=3), legal case workers (n=3) and other roles (n=8). We excluded these participants from the significance testing to avoid introducing additional heterogeneity.
Of the 611 participants, 59 (9.57%) reported having been an expert witness for capacity issues in court, including 32.6% of the clinical psychologists and 21.4% of the psychiatrists in the sample. No occupational therapists or speech and language therapists reported having been expert witnesses for capacity issues.
Professionals most commonly reported having conducted capacity assessments in hospital/inpatient services (39.97%), care homes (33.88%), community/outpatient services (31.25%) or the person's own home (27.47%). Only 1.48% of the sample most commonly assessed capacity in court settings. The two most common decisions that needed to be assessed were for accommodation (65.63%) and financial management (47.95%).

Quantitative analyses
How often do professionals assess capacity? The assumptions of univariate normality and homogeneity of variances were both violated (p< .05). A one-way Welch ANOVA with bootstrapping revealed that self-reported frequency in conducting capacity assessments significantly varied between professional groups, Welch's F(7, 57.12)= 11.73, n 2 = .140, p< .001, 95% BCa CI [.08, .18]. Table 3, there were significant differences between certain professional groups. Psychiatrists and nurses reported the highest frequency of assessments, whereas clinical psychologists, solicitors and non-psychiatrist doctors reported the lowest. 70.9% of psychiatrists and 68.9% of nurses reported having conducted at least 25 capacity assessments within the previous year.

How effectively do professionals feel they assess capacity?
The assumptions of univariate normality and homogeneity of variances were both violated (p< .05). A one-way Welch ANOVA with bootstrapping revealed that self-reported competence  As shown in Table 3, speech and language therapists, nurses, psychiatrists and social workers rated themselves highest in terms of assessing mental capacity, whereas psychologists rated themselves lowest (p< .05). 41.7% of speech and language therapists, 29.5% of nurses, 25.3% of psychiatrists and 23.6% of social workers rated themselves as being able to assess capacity 'very well'. Only 19.57% of clinical psychologists rated themselves as such. The average score for self-rated competence was reasonably high overall (M= 3.94, SD= .74). Self-rated competence was also positively associated with the number of capacity assessments the professional undertook per year (p< .001).

How often are professionals concerned of undue influence?
The assumption of univariate normality was violated (p< .05) but equal variances were assumed (p> .05). A one-way ANOVA with bootstrapping revealed that professional groups differed in the extent to which they considered undue influence, F(7)= 3.85, p< .001, n 2 = .46 95% BCa CI [.01, .07].
As shown in Table 3, solicitors reported suspecting undue influence the most often, and to a significantly greater extent than all other professions except for speech and language therapists (p< .05). Clinical psychologists also reported significantly greater suspicion than psychiatrists and non-psychiatrist doctors, who reported the least concern. The average level of suspicion was moderate overall (M= 3.02, SD= .83).
What are the challenges experienced when conducting any structured assessments? 210/611 (34.4%) of respondents indicated use of some kind of structured tool or psychological testing when assessing capacity. Figure 1 shows rates reported by the four largest professional groups, categorised by type of measure. The most common (61/611; 10.0%) type of tool used was a proforma of some kind to record capacity assessments, such as that made by the employer or third sector organisations, although no psychologist reported their use.  to "IQ" (4/611; 0.65%). Quantification of executive functioning was described in 9/611 (1.5%), standalone tests of language (receptive or expressive) in 8/611 (1.3%), and of memory in 5/611 (0.82%).
Use of a tool designed with the specific purpose of assessing mental capacity was made by 9/611 (1.5%) respondents.
9/611 (1.5%) of respondents specified that structured tests must only be used to assess impairment of brain or mind (MCA 2005 s.2) and not to assess impairment of decisional function (MCA 2005 s.3). Sceptical or disparaging comments about the usefulness of structured and psychological tests for assessing mental capacity were made by 14/611 (2.4%).

Qualitative analysis
Thematic analysis was performed separately on responses from two survey questions. The first ("What are the main challenges to capacity assessment?" n= 595) yielded nine themes relating to practices of capacity assessment. Analysis of the second question ("What would help you assess capacity better?" n=541) revealed responses which were generally much shorter, suggesting that respondents had attached less importance to this question. Responses generally repeated on those already given and generated no additional themes.
What are the main challenges to capacity assessment? Table 4 shows the thematic structure of nine themes and the subthemes they contained, which we then describe, using illustrative quotations, each originating from different, distinct individuals.
Theme 1: Practical issues A wide range of logistical issues with conducting assessments were reported consistently through the responses. Finding sufficient time for the assessment was a prime concern and shortages affected the possibility of developing a clear understanding of the decision before undertaking the assessment, obtaining background information on the person and triangulating that information when necessary: "Time to prepare to ensure I am completely familiar with P's circumstances to ensure the assessment is specifically tailored to P's current situation and their concrete options." Social worker "The main challenge is that organisational pressure means there is often not enough time. Time is often needed to make bespoke resources e.g. images to represent relevant information." Speech and Language Therapist Time to get to know the person was also crucial for some respondents: "Having the time to get to know the person being assessed, how they think and how they communicate." Social worker There was a lack of suitable private environments for encounters and inadequate documentation systems for assessment reports. Some participants stated it was impractical to make the multiple attendances they needed to complete an adequate assessment: If additional specialist support was required, either to give support during an assessment or to perform an additional one, this was not always available in a timely manner: "Organisational issues such as limited time and difficulties getting IMCA's as they have huge caseloads in my area" Social worker Some professionals wanted the assessment to be undertaken by a team rather than an individual, with responsibility for the outcome dissipated across the group. Some participants expressed concerns that they were asked to provide second opinion assessments, but as they were not members of the team working with the person, it was difficult to understand fully the decision in question and to get to know the person sufficiently to allow a quality assessment: "…often asked to engage in these assessments but we may not know the individual best-other staff groups are likely to be in a more informed position to make a judgement but lack the confidence to do so".

Clinical psychologist
There was also concern about where the responsibility for deciding issues of capacity lay when second opinion assessments has been requested: "Persuading my medical colleagues that the MCA COP[code of practice] is relevant to their practice Gathering collateral information from family or friends was experienced as hard, including learning more about the person. Some participants had difficulty choosing the best wording for asking questions during assessment.

Theme 4: Engagement and influence
Challenges were noted in building and maintaining effective relationships between people involved in assessment, be this with the person being assessed, people in their social environment and with navigating coercion and undue influence from these third parties.

"Enabling an environment where patients can be open (i.e. stopping family members/carers from influencing patient's responses)" Speech and Language Therapist
"Also, family members wanting to be present during the assessment can present difficulties. While this can put the person at ease and sometimes assist in the process, it is difficult to know how much the person is being "led" to provide certain answers by the family member." Social worker "If family members are present it can be difficult encouraging them to allow the person to answer for themselves / not correcting the person if they "get the answer wrong" / to prevent the individual from looking to their family for the answers" Social worker Others cited particular groups as needing special considerations during assessment (in particular intellectual disability, dementia, alcohol use disorders and executive function problems)… "lack of specialist knowledge ie. acquired brain injury" Social worker … and with more severe symptoms making assessment more complicated for some respondents.

Theme 7: Navigating the functional test of capacity
The assessment of functional decision-making inability (as part of the MCA 2005 section 3 test) was seen as generally challenging, with the using-and-weighing of information seen as particularly problematic: "The test is, inherently, entirely subjective when it comes to the 'weighing up' criteria in particular, which is the element of the test most relevant to most patients with psychiatric disorders." Psychiatrist Demonstrating that the functional inability was caused by any impairment of brain or mind (known as the 'causative nexus') was difficult.

"Causative nexus. Stipulating that it is as a result of the disorder" Nurse
Decisions which appeared unwise presented problems, and it was difficult to be sure if mental capacity in the office setting translated to decisions taken in real life situations (sometimes called the "frontal lobe paradox" in the context of brain injury): "For more complex assessments it is often a challenge when P seems to be able to understand, retain, use/weigh information and communicate a decision but they have compromised executive functioning and will do completely the opposite of what they stated they will do." Social worker "seeing the whole picture, and not simply person's presentation on a given day, in a structured environment, where they are not challenged to make decision in a 'live/hot', real life situation." Occupational therapist This extended to concerns about impulsivity: "assessing people with brain injury who have frontal lobe damage and may seem to be able to hold and make decisions, but who have impulsive behaviour which really makes one question if that is full capacity" Occupational therapist

Theme 8: Complicated presentations and pressing decisions
A range of features of cases were cited as producing particularly difficult assessments. Factors about the decision itself made some assessments more pressurised, including when the health status of the person required the assessment to be completed quickly, when the person was experiencing a mental health crisis, or when the decision involved life sustaining treatment.
Setting the threshold of functional ability at which the person (P) would be viewed as capacitous was challenging, as was capacity judged as fluctuant or borderline: "on the cusp of capacity but they're frightened to admit the gaps in their memory and understanding" Social worker

Discussion
This large-scale survey provides quantitative and qualitative insights into experiences of capacity assessment by members of different professions. To our knowledge, this is the largest survey of its kind in England and Wales and the first to explore the experiences of clinical psychologists and solicitors. We found differences between the professional groups for each of our main topics of interest: how often they assessed capacity, how competent they felt in assessing capacity, and how often they considered undue influence. Overall, the experiences of non-psychiatrist doctors and clinical psychologists seemed to have been the most divergent.
We found that non-psychiatrist doctors and clinical psychologists were less confident than other professionals in assessing capacity. Only 6.9% (2/30) of non-psychiatrist doctors rated themselves as assessing capacity 'very well', which was similar to the 4% (3/83) who rated themselves as 'extremely competent' within a previous vignette study of non-psychiatrist doctors (Penn et al., 2020). The finding for clinical psychologists is novel. Both findings may reflect the observation that non-psychiatrist doctors and clinical psychologists in our sample undertook fewer capacity assessments, relative to other professional groups. They may also reflect a relative lack of knowledge and training around the MCA 2005 (Penn et al., 2020;Schofield, 2008). This would be a concern if non-psychiatrist doctors and clinical psychologists felt less able to participate in the assessment process, given that our sample desired better access to psychological or speech and language therapist expertise, particularly for complex cases and more acknowledgement that the MCA is relevant to non-psychiatric decision making.
The responses also shed light on the notion of undue influence on capacity. While this has received scant attention within the empirical literature, professionals were most likely to report that they were 'sometimes' (56.9%; 347/610) or 'quite frequently' (18.0%; 110/610) concerned about undue influence. Solicitors were the most frequently concerned, which is perhaps unsurprising given that undue influence is most extensively scrutinised within legal cases. Psychiatrists and non-psychiatric doctors were least concerned, which may reflect less professional familiarity with the concept.
When we asked professionals who they thought were most vulnerable to undue influence, their responses were considerably varied. Older adults and people with learning disabilities, dementia and severe mental illness were the most frequently mentioned, but many emphasised that potentially anyone could be affected. Both findings suggest that undue influence is poorly researched, relative to its prevalence. This raises important concerns of potentially unmet needs which are important to decisional autonomy. Furthermore, a lack of guidance beyond the MCA code of practice was reported as a major challenge within our sample. Future research should therefore build on frameworks that can enable professionals to identify and support people who may be unduly influenced (Craigie, 2020; Kong & Keene, 2018) using data from both court and non-court settings. Literature on relational autonomy may also provide useful background (Ho, 2008;Series, 2015).
Use of some kind of structured or psychological tools was made by around a third of respondents with most being proformas. It is likely that many of the template assessment documents produced by employers and other organisations are little more than itemised proformas listing principles of the MCA 2005 and the assessment criteria, rather than providing specific assessment strategies.
A small number of respondents used some tools designed to measure decisional abilities themselves, with the challenge of matching the content of the tool to the actual decisions being faced by P. The relative absence of use of other capacityspecific measures suggests that professionals or organisations do not see them as useful in clinical settings -unpacking the reasons for this would allow barriers to be addressed when producing future tools.
Respondents used standard psychological tests, mainly use full cognitive testing, cognitive screening tests or tests of executive function. The most common use of the MoCA might reflect cognitive assessment practices in health services, or might have alternatively been chosen specifically for questions of mental capacity, for reasons such as its usefulness in providing a snapshot of executive function. Recent work (McWilliams et al., 2020) has found that the MMSE and "IQ" were the most commonly discussed measures in healthcare professional evidence in the Court of Protection when determining capacity (though not the ACE). It is difficult to say whether selection of the measurement by professionals is most closely related to considerations of their properties to measure decisional function, impairment of brain or mind, or whether these tests are simply available and easy to administer. There is an opportunity to develop panels of tests of function, tailored to the needs of different patient groups, decisional types and professional groups.
Our qualitative analysis of free text responses highlighted a wide range of challenging situations which professionals can experience. Many participants described careful consideration of their own practice and the practical and organisational barriers they faced, such as finding adequate time to carry out meaningful assessment. Capacity assessment was seen as sometimes complex, requiring training and access to resources.
Tailoring assessment to different presentations was challenging, as was being certain about the questions to ask and how to assess responses -especially concerning assessment of usingand-weighing information. Despite this, the levels of self-rated competence have weighted means that are fair.
Differences in experiences of capacity assessment between professional groups and settings emerged from the qualitative data, though the nature of our analysis was not designed to make formal comparisons. Examples included tussles over which professional held responsibility for a decision and how to resolve disagreement.
The main strength of the present study is the relatively large cross-sectional sample, with reasonable power to detect quantitative differences between the professional groups. Many of these professional groups (solicitors, clinical psychologists, social workers and nurses) have not been represented in a similar survey on mental capacity in England and Wales. Another strength is that we designed our free text questions to provide further context to the main three questions, which improved interpretability. We also had minimal missing data for the three main questions. Finally, approximately one in four survey respondents consented to be contacted about participation in follow-up interviews, to explore the themes in even more depth.

Limitations
We also note some important limitations. First, although most professional groups were reasonably well represented, our total sample is overrepresented by social work professionals (302/562) and underrepresented by solicitors (8/562). Uneven sample sizes are not uncommon in observational studies and the standard errors did not vary much between groups, suggesting that the analyses were reasonably valid. To avoid misinterpretations, we have reported either averages and frequencies for specific professional groups, or weighted averages within the results section. Any unweighted summary statistics should be interpreted with caution.
Secondly, as we recruited partly through social media and could not obtain national mailing lists, it would have been impossible to estimate the number of eligible people who did not respond. Therefore, we cannot rule out selection bias and our results may not be an accurate representation of each profession. It is likely that professionals who were most engaged with the issues around assessment of mental capacity will have responded. To mitigate against this risk, the survey was designed to be short -and therefore attractive to other professionals. We also deliberately targeted professionals in generic and non-specialist workplaces. It is important to note that the challenges described in our themes will not have been felt by all respondents across the sample, and so cannot be regarded as majority concerns.

Conclusions
In this study, we have explored experiences of capacity assessments within a large and diverse sample of professionals.
Professionals were generally confident in their own abilities and this was positively associated with how often they conducted assessments. However, the assessment process can include a wide range of challenges. Our findings suggest that clinical psychologists and non-psychiatrist doctors could particularly benefit from support to conduct capacity assessments. Structured tools or psychological tests do not find a high level of use and their use appeared inconsistent. This may reflect the limits of tests and the contextual and interpretative nature of capacity assessment. Guidelines on the use of psychological tests would be valuable to clarify limits and improve consistency of use. Finally, undue influence was a common concern amongst all of the groups. To mitigate against concerns of unmet need, this will need to be explored further in future research and policy.

Peter Bartlett
School of Law and Institute of Mental Health, University of Nottingham, Nottingham, UK This reports the findings of a questionnaire on professionals' experiences of assessing mental capacity in England and Wales. The questionnaire is short, and largely quantitative (with some short spaces to add in qualitative comments).
The paper points up the strengths and weaknesses of the methodology selected. On the plus side, there is a high number of responses -n=611 -allowing for a sensibly wide array of responses. That no doubt flows in part from the fact the questionnaire is short and to the point. That in turn, however, means that there is not much by way of nuance in the findings. It is interesting that significant numbers of assessors are concerned about coercion from others; one would sensibly wonder, however, whether this was consistent across types of decisions for which assessments were being made. Was there more concern about undue influence in financial decisions or consent to treatment decisions? A significant number of professionals used structured instruments. Did they always do so, or only sometimes? Did that depend on the sort of client they were assessing, or on the type of decision, or what? Does the confidence of the assessor in their ability depend on the sort of judgment (or sort of client) that is being assessed? The structure of the survey does not allow this to be explored. One of the advantages of a big number of responses is that sort of analysis should be statistically viable. The structure of the questionnaire means that this sort of analysis is not possible here.
Similarly, because of the questionnaire format, it has not been possible for the researchers to probe into the qualitative statements. There are the beginnings of some quite interesting themes that are identified, but I confess I am left wanting more detail and exploration from the responders.
To be clear -the researchers never seem to make claims on these matters that are not supported by their findings. The frustration is that the structure chosen is limiting -to which the researchers might well respond that a more detailed structure would have reduced the response rate, creating a different set of limitations on the data. their experiences of capacity assessment conducted under the MCA 2005. The study is soundlydesigned, properly-analysed and well-reported. It has clear aims. The methods are well-described, including the statistical analysis (although I am not qualified to assess the quality of that analysis). The study's limitations, including possible selection bias, are fairly discussed.
Particular strengths include the large sample (600+ respondents), drawn from several health disciplines, with sufficient numbers for their responses to be usefully compared; plus the inclusion of many respondents who conduct a large number of capacity assessments (25+ a year). This means that, between them, the respondents have a lot of experience in this field.
The study is particularly useful as a composite portrait of the current challenges facing UK health professionals involved in capacity assessment, especially the practical challenges. In a rather 'gritty' manner, it brings the process to life. It reaches important findings concerning: the frequency of concerns about undue influence; the relative infrequency of assessors being required to testify about their findings in court; the infrequent use of structured capacity assessment tools; and the distinct experiences of non-psychiatrist doctors and psychologists.
Overall, the study usefully complements the large body of more theoretical writing on the capacity assessment process and on the law established by the MCA. It should be useful in teaching and as reading material in professional courses.
Open access is provided to the study's quantitative data, while the qualitative data can be made available in appropriate cases, on request. Generally, the study provides a good example of the value of open access publishing and data sharing.
I have no specific changes to recommend.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? Yes

Are sufficient details of methods and analysis provided to allow replication by others? Yes
If applicable, is the statistical analysis and its interpretation appropriate? I cannot comment. A qualified statistician is required.
Are all the source data underlying the results available to ensure full reproducibility? Partly

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Mental health law; mental capacity law; socio-legal research methods and ethics.
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.