Financial harm in the context of adult protection: the complexity of factors influencing joint decision-making

Abstract

Purpose

This research was conducted as part of a PhD study. The purpose of this paper is to explore the factors taken into consideration when multi-agency practitioners were considering financial harm in the context of adult protection and how this influenced their decision-making processes.

Design/methodology/approach

An adapted q sort methodology initially established the areas of financial harm considered to have additional factors, which led to complexity in adult protection decision making. These factors were further explored in individual interviews or focus groups.

Findings

The data identified that the decision-making process varied between thorough analysis, rationality and heuristics with evidence of cue recognition, factor weighting and causal thinking. This highlighted the relevance of Kahneman’s (2011) dual processing model in social work practice. Errors that occurred through an over reliance on System 1 thinking can be identified and rectified through the use of System 2 thinking and strengthen social work decision-making.

Originality/value

This paper considers the practice of multi-agency adult protection work in relation to financial harm and identifies the influences on decisions.

Citation

Durowse, M. and Fenton, J. (2024), "Financial harm in the context of adult protection: the complexity of factors influencing joint decision-making", The Journal of Adult Protection, Vol. 26 No. 1, pp. 36-45. https://doi.org/10.1108/JAP-08-2023-0022

Introduction

Financial harm is not new, arguably it has occurred for as long as people have had assets. However, the introduction of the Adult Support and Protection (Scotland) Act, 2007 (ASPA) was a means to protect adults who may be unable to safeguard themselves or their assets, are at risk of harm because they are affected by disability, mental disorder, illness or physical or mental infirmity and are more vulnerable to harm than those who are not so affected (Scottish Government, 2022). Under s5, this legislation requires multi-agency co-operation with the Local Authority making adult protection inquiries. These agencies include Local Authority services, Police and Health Services and each agency is required to have sufficient knowledge to make referrals of harm (Scottish Government, 2022). Although the terms “inter-agency” and “multi-agency” appear to be interchangeable, the Code of Practice refers to both, in respect of interagency referral discussions and multi-agency decisions (2022). However, “multi” refers to shared aims across partnerships with single agency guidance and processes (Dixon et al., 2022), whereas interagency refers to collaborative working with multi-disciplinary agencies (Doyle et al., 2023). This is significant for the delivery of safer and effective services (Rogers et al., 2020) to ensure a full understanding of the needs of adults at risk (Preston-Shoot, 2019).

Therefore, the outcome for the adult is dependent on the joint decision-making process, and the purpose of the research is not to explore the individual professional decisions but to understand this joint decision-making process. The recent overview report of adult protection services inspections highlighted the lack of routine interagency referral discussions (Care Inspectorate, 2023), which questions how decisions are made across the agencies, and if they are actually interagency decisions.

What is financial harm?

It is difficult to define financial harm due to the distinct types, the individuals affected by the harm and the varied terminology used. Financial harm can occur in the individual’s home, care settings and in the community. Financial harm can be committed online, over the telephone and in person, both within a relationship and from an unknown perpetrator. Although the Scottish legislation refers to “harm”, it can also be referred to as “financial abuse”, “financial exploitation”, “economic abuse” and “material abuse”. In addition, there is a lack of an agreed definition and way to measure financial harm to provide an accurate way to determine the scope of the problem (Yon et al., 2017).

As the research was conducted within one Adult Protection Committee area, it was appropriate to use their financial harm definition of Financial harm is caused by the illegal or improper use of an individual’s resources (both financial and property) by another person, without their informed consent or through the exercise of undue pressure (Fife Adult Support and Protection Committee, 2022).

Single and interagency decision-making in financial harm

As the outcome of a financial harm decision is made in partnership with multiple agencies, it is important all the professionals are able to recognise financial harm and make a collective judgement based on that knowledge. Single-agency professionals may have a good understanding of financial harm as it relates to their service, which may not be a matter for consideration under ASPA. For example, the Police and Trading Standards may have knowledge of cases involving financial scams but it is not automatic that the same cases involve an adult at risk of harm.

There are specific issues related to single agency involvement in recognition and reporting harm, with evidence of attitudes towards the police impacting on the likelihood of reporting harm, particularly if previous contact led to dissatisfaction with the outcome (Parti and Tahir, 2023). However, this could equally apply to any of the partner agencies. Police professionals are also considered to be more risk averse than social workers (Joseph et al., 2019) and nurses comment that they have a higher consideration of safety compared to social workers (Dingwall et al., 2015), which may have an impact on the way they make a decision or rank a good outcome. Similarly, there can be difficulties in information sharing from health professionals in adult protection, potentially due to the issue of consent (Joseph et al., 2019) as information is more frequently shared in situations of child abuse allegations. This highlights the lack of clarity for some occupational groups about the legislative requirements to share information where there are adult protection concerns.

Some stereotypical thinking continues to exist around the roles of occupational groups, such as functional understanding (Satiel and Lakey, 2019), clarity of role (Trainor, 2015) and responsibilities (Stevens, 2013). Police have a primary consideration of being in the interest of the public (Shearlock and Cambridge, 2009), whereas housing professionals have an overarching consideration for the property and ability to pay rent. These differences in professional role characteristics may have an impact on the ultimate decision made.

Decision-making

It is obvious from the above that the background to practitioners’ decision-making is extremely complex and saturated with subjective and role-related factors and understandings. To try to bring some clarity to the decision-making process in this context, we can turn to several decision-making models to tease out the process.

Rational choice decision-making, where practitioners must consider the alternatives and ascribe consequences to each alternate in a sequence of importance (Burns and Roszkowska, 2016) and bounded rationality decision-making where rational choice is limited by time, resources or available information (Taylor, 2012) are processes based on rational analysis. Alternatively, heuristic decision-making has the goal of making judgements quicker and more frugally and trades-off accuracy for less effort (Gigerenzer and Gaissmaier, 2011).

Kahneman (2011) brought these types of theories together in his “dual processing model” and described these different ways of decision-making as Systems 1 and 2. System 1 decisions are quick, effortless and automatic and reflect the heuristic models of decision-making (Gigerenzer and Gaissmaier, 2011). System 2 requires effort and attention and constructs thoughts into order or a series of steps. Therefore, it is slower, more deliberate and follows rules, reflecting rational choice and analysis.

These different models and perspectives will be applied to the findings in the discussion section.

Methodology

Although part of a wider study the focus of this paper is joint decision-making in adult protection in regard to financial harm. Q sort methodology combines the strengths of both qualitative and quantitative research methodologies (Watts and Stenner, 2012) combining statistical data (Barry and Proops, 1999) with an analysis of the narrative (Watts and Stenner, 2012). Q methodology can be used to establish commonly held beliefs or perspectives (Sneegas, 2019) and encourages empowerment and participation (Ellingsen et al., 2010) through ranking a series of statements on a scale (Robson, 2011). It is the ranking process that makes this methodology useful in identifying the statements the participants can neither agree nor disagree with and which attract a neutral score. The neutral score indicated the participants were unable to make a judgement and needed some further information, or that there were other factors that would influence their agreement with the statement. Q methodology, is therefore, considered to be an effective way to analyse perceptions and attitudes. The ranking process also negates the need for tests of validity as the view of the participant is interpreted by themselves (Lee et al., 2021).

Participants

The initial stage of the research focused on 96 practitioners from social work, health, police, housing, care providers, trading standards, banking, advocacy and office of the public guardian who had voluntarily attended an interagency conference about financial harm within one Adult Protection Committee area. All the practitioners had a role in identifying and reporting financial harm and representing their agency in the joint adult protection decision-making processes. Data regarding the occupational roles within their agencies were not gathered, in part because it would be easy to identify the participants in some of the smaller agencies and compromise anonymity. In the balance of collecting specific data about the participants, there was also consideration about the joint role of decision-making as the outcome of adult protection decisions are based on their collaborative knowledge.

Developing the statements

The 47 statements were developed from a wide-ranging literature search about all forms of financial harm or abuse and narrowed into eight categories related to financial harm. Drawing on the work of Wilber and Reynolds (1997), the first four sets of statements were on the themes of cost and benefit to the adult, characteristics of the victim, influence used and relationship between the adult and the perpetrator of harm. In addition, a further three sets of statements were developed in relation to risk: risk in the home, in the community and within services. Given the earlier findings relating to different understandings around risk between the professional groups, it was considered important to explore this area in more detail. The final category of statements had no explicit basis in the literature but was based on common misconceptions or myths to ascertain the extent to which professionals relied on them.

Developing the vignettes

From the scored statements, the two significant areas of interest about decision-making factors were where both matched statements scored in the neutral area, which indicated that these were the most difficult or ambiguous decisions (Table 1). The second area of interest were the matched statements that had scored differently, so the minor change made between the statements seemed to be an influential factor in their decision-making (Table 2). These statements are detailed below:

It is worth noting that the statements that scored neutrally predominately related to relationships and known perpetrators rather than financial harm situations where the perpetrator is unknown to the adult, such as scams. This would suggest that the participants were able to make a clear decision about harm where there was not a known perpetrator, and there were other considerations for situations that involved some form of relationship within the financial harm situation.

From the findings of the initial Q sort, an unfolding vignette was developed to provide a continuous narrative (Hughes and Huby, 2004) explored in either individual interviews or focus groups involving 17 practitioners who had participated in the initial ranking exercise at the conference, and were employed in Health, Social Work, Advocacy, Housing, Police and Trading Standards. The purpose of the interviews and focus groups was to explore the factors involved in the thinking of professionals when confronted with the situations highlighted by the Q sort that were agree as the most difficult, ambiguous or complex.

Main results

There was considerable discussion around the term grooming, which differentiated between error or opportunistic behaviours and intent to harm. If it was a coffee and a biscuit I would think differently, if the cost of the meal was higher, consider it was grooming (Group 2), which suggests the amount of the spend was a factor in determining financial harm. However, this was not explicitly balanced against the benefit to the adult.

The participants sought concrete evidence, such as discussions around financial harm indicators considering the lifestyle of the adult not matching the income used. They used this information to inform their risk assessment. The participants also looked for evidence of potential control, such as the alleged perpetrator’s name on the bank account or access to passwords. The consistency of this response across the focus groups and interviews demonstrated that the participants were looking for the same types of information to inform their decisions.

Some of the responses indicated a collaborative response but highlighted some potential decision-making errors, particularly subjectivity. They used memories to identify links between the current circumstances and previous events I had a similar situation, but the person was doing their best (Group 3). They considered the nature of reciprocity (Fehr et al., 2002), which led to speculation about the motive of the alleged perpetrator, ordinarily no one would help someone out of the blue (Group 4), suggesting the participants had a picture of what the norms of relationships were and whether the relationship was considered trustworthy. One response referring to an informal hierarchy of harm in our heads (Group 3) suggests there is some prior ranking of the type of harm or impact of harm informing protection decisions.

Although the joint decision-making process was being explored, there was evidence of single agency thinking that impacted on the way particular issues were being considered. Police don’t have a view, Police consider the risk of harm when asked about Power of Attorney and capacity, indicated a clear perception of their role in joint decision-making. Similarly, the comment Eviction case, 70-year-old man says his daughter isn’t paying the rent, daughter doesn’t have the money, need to involve other agencies. Would it make a difference if it was or wasn’t financial harm re action that housing would take? (Interview 3) demonstrates the consideration of agency priorities rather than joint decision about financial harm. There were other comments that also demonstrated a lack of commitment to the process and a desire to be solution focussed. Remove the risk, it’s risk managed from a Police participant, combined single agency thinking with a solution focus, which negated the benefit of an interagency consideration of the wider factors. There was some acknowledgement that a solution focus was not the most appropriate response with the comment I find other agencies have the expectation of housing to resolve the situation by giving them a different house (Group 2). However, this understanding was based on direct experience and single agency perspective, rather than a commitment and understanding of interagency working.

Similarly, there was evidence of a lack of commitment to joint decision-making. Comments such as need Social Work to take the lead and direct, to get other services involved (Interview 3) and would expect social work to take the lead and look at the wider situation (Group 1) indicated the participants did not see themselves as equal in the decision-making process. This may also be related to the comment Can’t really do much else to protect the person as the traditional things we do to protect people we can’t do here (Group 4) suggesting there would be actions a single agency could take but the joint decision-making processes inhibits their action.

A key remark, can appear that people use the capacity decision, (the adult can) have capacity and rule out (adult protection) and not make any further inquiries (Interview 3) suggests that decisions are made prior to engaging with partner agencies, making the decision neither multi nor interagency.

Discussion

There was clear evidence of bounded rationality decision making (Gigerenzer and Selten, 2002) occurring with most of the participants. Rational choice decision-making (Burns and Roszkowska, 2016) based on data, facts and direct observation (Rutter and Brown, 2015) was, of course, hampered by the limits on information available, thus leading to the participants making bounded rational decisions. Rational choice (Burns and Roszkowska, 2016) has the key principle that all actors will behave rationally and full data is available. However, when all the data is not available, bounded rationality operates with the same principle of rationality but simplifies the decision-making process identifying the most important outcome or decision required, simplifying the search for information and identifying the point where the search for information stops (Gigerenzer and Selten, 2002). So, for example, the deliberate and conscious consideration of whether the adult was assessed as having capacity, whether there were concrete changes to the adult’s circumstances and whether concrete indicators of control were in evidence, are all examples of bounded rational choices as to whether financial harm was taking place. In each of these three examples, the most important outcome had been identified by the participants and indicated where they would search for evidence.

Bounded rational choice theory is also congruent with Kahneman’s (2011) System 2 thinking. System 2 works in an orderly fashion going through a system of steps, it is thoughtful and applies weight to the factors in the decision to apply a balance to the information. This was evidenced when the discussion returned to the adult’s capacity, indicating the participants were taking a more systematic approach (Kahneman, 2011) and withholding their judgement to rethink the situation (Tay et al., 2016).

Less straightforward were the decisions, which referred to similar previous situations the participants had come across, such as assessing the nature of the relationship between the adult and alleged perpetrator and subjective assessments of the amount of money involved. In these instances, we saw the use of heuristics, for example “I had a similar situation”, which exemplified a good number of those types of responses. Drawing on prior knowledge, training and experience (Taylor, 2012), heuristic decision-making involves simple short cuts to progress to a logical conclusion. The use of a similar memory allowed the participant to identify similarities, select fewer pieces of information and consider fewer alternatives (Shah and Oppenheimer, 2008), allowing for a quick decision to be made. In practice, this has great benefits in making quick decisions but can fail to identify non similar information.

As well as heuristically based judgements, other issues were affected by an emotional reaction, as in the case of the word “grooming”, and thoughts about the authenticity of the relationship. The word “grooming” appeared to be a priming word (Grill-Spector, 2008) for some participants, and these heuristically based and more subtle and subjective assessments are congruent with Kahneman’s System 1 thinking (Kahneman, 2011). System 1 uses skilled intuition but also generates feelings and emotions that can become beliefs, which may influence the weight given to particular factors. Perception errors are more likely with familiar stimuli (Mezias and Starbuck, 2008) as perceivers tend to see the data that reinforces their current perception. In this research, the emotional response prompted a more authoritative stance on power and control. Kahneman (2011) indicates the creation of a coherent account based on limited information is a feature of System 1 thinking, where information is sought to support the story, rather than to consider other possibilities. For example, there was considerable speculation about the motive of the alleged perpetrator, which created a causal link (Kahneman, 2011). With prior knowledge that new friendships can be forged by perpetrators of financial harm to manipulative the adult into trusting the perpetrator (Wilber and Reynolds, 1997), the participants questioned the nature of the relationship and some practitioners assumed it was a new relationship on the basis there was a lack of contra information. The assumption of a new friendship and the belief the new friend was a potential perpetrator led to a causal link where a coherent story was created based on the assumption the alleged perpetrator was deliberately defrauding the adult. Once this was part of the thinking process, it was not re-examined when further information was provided. Kahneman (2011) suggests that this quick System 1 thinking can reject alternative accounts and is more likely to focus on existing evidence with little attention paid to missing evidence, so the participants who created the coherent account did not review if there were gaps in the information.

“An informal hierarchy in our heads” (Group 3) denoted some factors were considered more important than others (Taylor, 2012) based on the interpretation that unequal friendships would be exploitative. Therefore, the participants applied higher weight to a piece of information, which anchors the value of a particular factor in the judgement (Kahneman, 2011). In this vignette, the higher weight was given to the perceived unequal relationship, which led the participants to anchoring this at the centre of their decision-making based on their previous assumption. This thinking supported their coherent story that the alleged perpetrator was exploitative and ignored the alternative view that people in receipt of services may want to give gifts to balance the power of their relationship (Quinn, 2008). This was also observed in the comment Ordinarily no one would help someone out of the blue (Group 4), where the participants had anchored the basis of the relationships between adult and alleged perpetrator on their perception of an equal relationship.

Kahneman (2011) identifies that system 1 responds more strongly to risk of losses than it does to a potential gain. In the statement if the cost of the meal was higher, consider it was grooming (Group 2), the participants responded to the cost of a meal as a loss. They demonstrated aversion to the value of a meal based on a subjective assumption about the cost and affordability, rather than the potential positive social gains of having a meal. System 1 “invents causes and intentions” (p. 105, Kahneman, 2011), and in this case the participants have applied an arbitrary value to determine financial harm. Kahneman (2011) also observed that, when cognitively busy, System 2 can default to System 1 and can substitute a simple decision for the true, more complex one. So rather than tangle with the complexities of spending money on a meal but gaining a social and potentially enriching experience, some participants based their decision on a simple calculation of the cost of the meal.

Some of the responses were seemingly dismissive of the need for joint decision-making and the respondents appeared to be defaulting to their professional or agency decision processes. Clearly, each agency professional has a responsibility to their agency, which may be a dominant consideration in decision-making. Thinking in a wider context and taking in the factors other professionals consider relevant to a decision may generate feelings of doubt or ambiguity, which are uncomfortable feelings in situations of risk. System 1 can suppress doubt to allow for quicker decisions to be made. Most professionals have spent considerable time learning and fine tuning their decision making within their sphere of their own professional knowledge, and when something does not fit into the perceived context, System 1 can treat it as a departure from established norms and try to suppress it to allow for quick thinking. Whilst system 1 is also responsible for uncritical acceptance of suggestions, prospect theory (Kahneman, 2011) indicates people are far more likely to avoid loss than look for a gain. In adult protection, any form of loss would mean the adult is potentially at risk of harm, which would be unacceptable to workers who have a protection focus. Whether it is the potential risk that may place an adult at risk of harm, or a default to thinking in line with single agency priorities, it would appear that decisions are not made in multi or interagency ways. Based on this observation, it may be more likely that practitioners have pre-formed their decision prior to sharing it in a multi-agency forum. Coupled with the observations of an expectation that one agency will have a lead role in the decision-making process would suggest that some multi agency decision forums are, in reality, more of an information sharing session, with each agency bringing their distinct professional view and the lead agency making the ultimate decision.

It can be seen from the above that it is difficult to bring coherence to the complexity of decision-making about potential financial harm. What the research has shown however, is that Kahneman’s dual processing model can usefully delineate and identify tendencies towards Systems 1 or 2 in participants’ thinking during the decision-making process. Engaging explicitly in System 2 thinking in the interagency discussion would also allow for departure from one professional, distinct view and a proper consideration of all factors and information in a more rational, less heuristic way.

Conclusion

The complexity of decision-making in this context demonstrates the participants were knowledgeable about the types of financial harm that would be relevant to their field of work and were able to draw on their heuristic knowledge, identifying cues and patterns to inform their decisions. There was also evidence of analytical decision-making and discussions in areas where further information was required.

By using Kahneman’s Systems 1 and 2 thinking, we identified dual processing occurred throughout decision making processes, but observed the participants to be unaware of the errors that occurred in creating coherent accounts through System 1 thinking and the subjective weight applied to particular information and particular professional priorities.

This is an extremely valuable piece of learning from the research because what has been uncovered is the unwitting, and very human, reliance on System 1 thinking with all the error, confidence and bias that entails. There may, therefore, be real value in helping practitioners learn about Kahneman’s dual processing model and in demonstrating how practitioners may be relying on System 1 thinking in their decisions. The objective would be to help practitioners identify consciously when they were reverting to System 1, to help them stop and think more rationally and slowly about assumptions they may be making and about information from different professional perspectives. In essence, Kahneman’s dual processing model can bring practitioners insight and self-awareness in relation to their decision-making in these most complex and difficult situations.

Table 1

Example of matched statements with a central score

Statement	Matched Q set no.	Reference
8. Financial harm and evidence of neglect are closely linked	35	Allen-Devlin and Freyne (2013)
35. Adults whose basic needs are not met could be the victims of financial harm	8	Allen-Devlin and Freyne (2013), Crosby et al. (2008)
5. The self-appointment of family members as carers is a factor in financial harm	46	Home Office (1999)
46. Where the adult has a ‘good faith’ relationship, it can be detrimental to allow the relationship to change to one where the friend makes decisions in the adult's best interest	5	Home Office (1999)

Source: Created by authors

Table 2

Example of matched statements with a different score

Statement	Matched Q set no.	Reference
10. The power dynamics between carers and adults make it difficult to tell if someone is being harmed	44	Faulkener and Sweeney (2011)
44. The power dynamics between paid care services and the service user make it difficult for service users to report financial harm	10	Faulkener and Sweeney (2011)
29. The cultural norms of exchanges such as gift giving have the potential to be exploited	20	Cropanzono and Mitchell (2005)
20. The cultural norms of exchanges, such as gift giving, need to be suspended when an adult no longer has the capacity to make reasoned decisions	29	Stoller (1985)
41. An adult is free to pay for their friends even if the arrangement is not reciprocal	6	Emerson (1976)
6. If and adult thinks they are not at risk from financial harm they should be allowed to spend their money any way they want	41	Emerson (1976)
3. Adults are less likely to be the victim of financial harm if they have family support	21	Dixon et al. (2013)
21. Those who are required to make decision in the best interest of an adult are also the same people who could apply undue influence on how funds are spent	3	Tilse and Wilson (2013)

Source: Created by authors