Exploring financial abuse involving people with dementia: an empirical legal study of nation-wide court rulings over a decade in Taiwan

Abstract

Previous literature attempted to gain insight into financial abuse involving people with dementia by analysing court cases, but these studies were limited in sample size or scope. This study collected 214 court rulings directly related to the financial decisions of people with dementia to identify characteristics of the financial abuse victim, perpetrators and the types of assets. The models of bystander intervention and routine activity theory were used as conceptual models to guide analysis regarding the role of bank staff as well as the court's decision in cases of financial abuse. The majority of financial abuse perpetrators were family members (73.8%), as opposed to outsiders (19.2%). Transfer of real estate was the most common legal issue, and land was the most common financial asset involved. Difficult intra-family relationships seem to pose a great risk of financial abuse involving people with dementia since adult children were found to be the most likely perpetrators (52.7%) but also plaintiffs accusing financial abuse (57.6%). In accordance with the bystander intervention model, bank staff were more likely to be suspicious of financial abuse when an outsider was regarded as the perpetrator. In accordance with the routine activity theory model, the court was more likely to acknowledge the case as an invalid financial decision when an outsider was regarded as the perpetrator in financial abuse cases. Since people with dementia suffer from greater losses due to their family members, future policies should establish guidelines for front-line bank staff to identify warning signs to reduce the risk of financial abuse involving people with dementia, not only to prevent fraud by outsiders but also exploitation by family members.

Background

Definition and prevalence of elder financial abuse

The Convention on the Rights of Persons with Disabilities (CRPD) emphasises protecting every person with disabilities from all forms of exploitation, violence and abuse (Department of Economic and Social Affairs, 2021). The World Health Organization (2021) states that elder abuse can be defined as ‘a single, or repeated act, or lack of appropriate action, occurring within any relationship where there is an expectation of trust which causes harm or distress to an older person’. Elder abuse can take various forms such as financial, physical, psychological and sexual. It can also be the result of intentional or unintentional neglect (World Health Organization, 2021). Although there is currently no standard terminology for elder financial abuse, similar terms such as financial mistreatment and financial exploitation all refer to a situation in which harm is perpetrated intentionally on victims by perpetrators to their financial benefit (Dalley et al., 2017a).

A systematic review and meta-analysis reported that the one-year prevalence rate of financial fraud and scams among older adults in the United States of America (USA) was 5.4 per cent (95% confidence interval (CI) = 3.2–7.6) in 2017 (Burnes et al., 2017). Another systematic review on elder abuse reviewed the prevalence figures of elder abuse globally and found the pooled prevalence rate of financial abuse in community settings was 6.8 per cent (95% CI = 5.0–9.2) (Yon et al., 2017). According to the statistics on the current situation of elder abuse in Taiwan, provided by the Ministry of Health and Welfare, the annual cases for older adults protection service reported in 2019 was 10,279. Among the different types of older adults protection cases, lack of support accounted for the highest proportion (52%), followed by physical and mental abuse (20%), and abandonment (13%), while financial abuse only accounted for 0.3 per cent of all elder protection cases (Department of Statistics, Ministry of Health and Welfare, 2021). However, previous literature on elder financial abuse has proposed the iceberg theory, which suggests that the figures reported to the older adults protection service only represent the ‘tip of the iceberg’ of all elder abuse cases that happen (National Centre on Elder Abuse, 1998). A recent news article in Taiwan also made the headlines with ‘the reporting rate of elder abuse was 0.3 per cent out of all cases happening’ (Taiwan News Group, 2021). The researcher interviewed in the news echoed the iceberg theory, saying that the majority of abuse cases were not reported because the victims often felt ashamed of being abused by their close family (Phelan, 2020).

Financial abuse and dementia

The financial abuse of people lacking mental capacity is a complex phenomenon that involves a variety of victims and abusers. Moreover, it is also frequently hidden from the public's scrutiny and, therefore, results in minimal protection or redress for the abuse victims (Dalley et al., 2017a). According to the United Kingdom (UK) report on this issue of financial abuse of people lacking mental capacity, the perpetrators of financial abuse could be family members (between or within generations), friends, neighbours, acquaintances, paid workers or strangers, in person, by phone or online (Dalley et al., 2017a). It is well recognised that people with dementia can be subject to abuse in all care settings, and a previous report has discussed several factors that could make someone with dementia more vulnerable to abuse (Alzheimer's Society, 2011). First, having dementia can make someone an easy target, as they may not have the mental capacity to understand what they are being asked or forced to do. Other studies in the literature have also shown that people with mild cognitive impairment had poorer financial and health-care decision-making in real-world situations (James et al., 2014; Han et al., 2015). Secondly, victims of abuse generally find it hard to talk about their experiences, especially in cases where they were also dependent on others for household tasks (Alzheimer's Society, 2011). This can be exacerbated for a person with dementia, who may not be believed even if he or she discloses the abuse experienced due to communication problems. According to statistics reported to the adult safeguarding council in England, financial abuse was the third most frequent in the list of categories for abuse-related referrals between 2014 and 2016 at around 16–17 per cent of referrals, and it was ranked closely in proportion to the first two categories: (1) neglect and omission and (2) physical abuse (Dalley et al., 2017c). In fact, all three categories of abuse were often interlinked (Dalley et al., 2017a).

Research gap

Due to the under-reporting of cases of elder abuse, it is difficult to get a full picture of elder abuse through an analysis of elder protection data, especially due to the small case number of elder financial abuse as suggested by the iceberg theory. Therefore, we chose to use court rulings to be a large and open data source to examine elder financial abuse in detail. The advantages of using court rulings are that a judicial decision is an official solution to resolve disputes associated with financial abuse, the content in the court's judgement is subject to rigorous investigation and all testimonies must be held accountable.

Our review of previous literature of empirical legal studies related to people with dementia only found three articles (Doron et al., 2017; Dalley et al., 2017b; DeLiema, 2018). The study by Doron et al. (2017) reviewed court rulings in Israel to map out different legal contexts involving people with dementia and found that the majority of legal cases related to people with dementia were associated with legal capacity determination or legal issues in the field of criminal law. However, since that study aimed to provide an overview of the legal terrain, it contained no detailed information on financial abuse involving people with dementia. The second study looked at the range of misbehaviour conducted by the power of attorney of a person with dementia in the UK, but the limitation of that study was that its sample included only cases from the Court of Protection, so it was limited in scope (Dalley et al., 2017b). The third study used rich qualitative data at an elder abuse forensic centre in the USA to analyse the differences between elder fraud and financial exploitation, but its small sample size of 53 cases also implies that the findings of that study have limited generalisability (DeLiema, 2018).

Theoretical considerations

Previous literature on financial abuse of older adults or those lacking mental capacity has considered different theories in explaining this phenomenon. One study used the social exchange theory to describe intra-family abuse, which viewed the family as a ‘site of intergenerational stress, conflict and violence’ (Parrott and Bengtson, 1999). Another study used the routine activity theory as a contextual model to analyse the differences between elder fraud and financial exploitation (DeLiema, 2018). The study defined fraud perpetrators as those who used deception to obtain the victims' money or property, while financial exploiters were ‘trusted others’ who had pre-existing relationships with their victims (DeLiema, 2018). The routine activity theory predicts that criminal opportunities arise when a motivated offender and suitable target meet in the absence of capable guardians (Cohen and Felson, 1979). Since fraud victimisation occurs when an elder has cognitive impairments combined with a lack of financial oversight from capable guardians, the study found that financial losses were more than three times greater among fraud cases than financial exploitation (DeLiema, 2018). The other study adopted the bystander intervention model to understand decision-making in relation to the detection and prevention of elder financial abuse (Gilhooly et al., 2013). Banking-sector professionals who had prior experience in detecting incidents of financial abuse reported that they were more likely to become suspicious when a third party attempted to take control of an older customer's bank account, when requests for a large cash withdrawal were made, or when financial anomalies in accounts or bills were identified (Gilhooly et al., 2013). Following our review of the different theories discussed above, the current study compares the differences in financial losses and the validity of legal procedure by the type of perpetrators using the routine activity theory. In addition, the professional bystander intervention model was used as a conceptual model to guide our analysis regarding the role of bank staff in detecting cases of financial abuse involving people with dementia.

The present study

This study seeks to utilise the large sample size and open nature of the written judgements to have a more complete and broader picture of financial abuse cases involving people with dementia in Taiwan using an empirical legal study design (Epstein and Martin, 2010). More specifically, this study analyses court rulings to identify the characteristics of the financial abuse victim as well as the perpetrator, the types of legal issues and the types of assets involved in the cases; it then compares the differences in the amounts of financial losses across case types. The descriptive results from this study not only enhance our understanding of this often hidden form of abuse detrimental to people with dementia, but also provide an essential step to identifying future areas for targeted prevention, as we examined the role of bank staff in detecting cases of financial abuse.

Methods

Case selection

The judgement data were collected through the LawBank database (www.lawbank.com.tw), which is one of the major, privately operated, fee-based databases covering all court cases. The timeframe for our search was limited to the most recent decade (i.e. from 1 January 2010 to 31 December 2019). The search was conducted using relevant keywords, and the search term was defined in Chinese. By utilising both computerised searching and human screening, we obtained civil and criminal cases of first-instance rulings decided by district courts directly related to the financial abuse of people with dementia. Details on the search terms respectively used for criminal and civil cases are described in Table S1 in the online supplementary material.

For validity purposes, the screening procedure was performed independently by two authors, and in cases of disagreement, a third researcher was consulted to make the final determination. Cases that addressed dementia indirectly or incidentally were excluded. For example, some torts or criminal defendants claimed that family members suffering from dementia were in need of the defendants' support so that the defendants could pay fewer damages or receive lighter sentencing. Since neither the defendant nor the plaintiff was a person with dementia, we considered this kind of case to be ‘addressing dementia indirectly or incidentally’. Moreover, in torts cases there were victims of car accidents who would become people with dementia. However, as these rulings did not relate to the financial abuse of people with dementia, they were also excluded.

After applying the above-mentioned exclusion criteria, our sample narrowed to 393 cases (including 319 civil cases and 74 criminal cases). As this number is not large, we did not proceed with further case selection. That is to say, our sample includes all of the first-instance judgements directly related to the financial abuse of a person with dementia. Meanwhile, there were cases involving the same person with dementia in both civil and criminal cases. Since civil rulings reveal more details of people with dementia than criminal ones – because the focus of civil cases is the remedy for a person with dementia, while criminal ones emphasise the punishment for the perpetrator – we decided to keep the civil cases and delete the criminal ones in terms of duplicate cases from our sample set. This resulted in the reduction of criminal rulings to 56 cases. Finally, there were seven cases involving two people with dementia at the same time, so we broke it down to one person with dementia per observation. As a result, our final sample size consists of 382 observations, including 323 civil and 59 criminal observations.

Data abstraction

All decisions made by the court were analysed according to the different main characteristics of the cases. For the purposes of this study, we developed a coding schema with the goal of capturing the following groups of variables: (a) the demographic characteristics of the financial abuse perpetrator, (b) the demographic characteristics of the case plaintiff and the person with dementia, (c) the variables on legal issues and assets involved in cases of financial abuse, (d) the outcomes of the legal procedure (i.e. whether the financial behaviour of the person with dementia is acknowledged as void or revocable, or their financial losses are allowed to be recovered in civil cases, or whether the accused perpetrator in the criminal cases is sentenced guilty), (e) the amount of financial losses as a result of the court's rulings, and (f) the role of a banker (when it exists).

Two authors followed a structured procedure for legal data extraction, under which each court ruling was analysed and transformed into a quantitative dataset. For example, two authors began the coding procedure by reading all of the cases and creating a variable of dementia confirmed by the court, which was applied if the verdict explicitly acknowledged that the victim had dementia. This variable was used as our last inclusion criteria to ensure the validity of our case selection (i.e. through this variable, we excluded cases where the plaintiff simply claimed to be a person with dementia without reasonable grounds). The severity of dementia was determined based on the Clinical Dementia Rating (CDR) score, indicating mild, moderate or severe dementia (CDR 1, 2 or 3). The authors also grouped main types of legal issues involved in cases of financial abuse, and this induction process led to a summary of nine types of legal issues including transfer of real estate, as well as the last category of others, which consists of all other non-specified types. Next, we created ten dummy variables, with each variable indicating one type of legal issue involved in cases of financial abuse.

Analysis

Data analyses were performed on our main variables of interests by examining descriptive statistics (mean, standard deviation and percentages) using SAS 9.4 software (SAS Institute Inc., Cary, NC, USA). SAS Viya: Visual Analytics was the software used to draw a box-and-whisker plot to compare the differences in financial losses by the type of financial abuse perpetrators. Since this study was based on publicly available court rulings that were obtained from an open-source website, the study's protocol was approved by the university's Institutional Review Board with a waiver for informed consent.

Results

The trend in the number of civil cases and criminal cases related to the financial abuse of people with dementia over a decade from 2010 to 2019 is displayed in Figure 1. There seemed to be a general upward trend in the number of legal cases in which dementia was claimed by the victim (the full line), but among our total of 382 cases, the cases in which dementia was confirmed by the court is shown by the lower dotted line. The number of cases confirmed by the court accounted for 56 per cent of the cases claimed by victims. Among the 214 legal cases comprising our sample related to the financial abuse of people with dementia confirmed by the court over the decade, 81 per cent are civil cases and 19 per cent are criminal cases.

Figure 1. The number of civil cases and criminal cases related to the financial abuse of people with dementia over a decade since 2010.

Note: N = 382 for dementia claimed by the victim; N = 214 for dementia claimed by the court.

The characteristics of the financial abuse perpetrators are shown in Table 1. Since some legal cases could have more than one perpetrator, the total number of perpetrators is 258, greater than the sample size of 214. We grouped the perpetrators by their relationships with the victims and found that the majority were family members of the people with dementia: 52.7 per cent were adult children, followed by grandson/granddaughter (12.4%) and spouse (8.5%).

Table 1. Financial abuse perpetrators

Notes: 1. After excluding double counting cases, perpetrators can be divided into two main types: family members (73.8%) and outsiders (19.2%). 2. The total number of perpetrators is more than the case number (N = 214) because there may be more than one perpetrator in a case.

Perpetrators who were outsiders to the people with dementia included 15.9 per cent of friends and 6.2 per cent of strangers. The identity of the party who claimed the incapacity of the person with dementia (hereinafter, the plaintiff) is listed in Table 2. Since some legal cases could have more than one plaintiff, the denominator used for the percentage calculation was 290 instead of 214. The people with dementia (or his/her guardian who brought a lawsuit on behalf of the person with dementia) accounted for 25.9 per cent of all plaintiffs who brought charges of financial abuse, which is the second largest group next to the adult child group which accounted for 57.6 per cent of the plaintiff population. Compared to people with dementia and their family members, the percentage of strangers or friends who served as a plaintiff to claim for the incapacity of the person with dementia was quite low at 3.5 per cent.

Table 2. Plaintiffs in legal cases of financial abuse

Notes: 1. The person who claimed for incapacity of the person with dementia in legal cases of financial abuse. 2. The total number of plaintiffs is more than the case number (N = 214) because there may have been more than one plaintiff in a case.

The characteristics of the people with dementia involved in cases of financial abuse are presented in Table 3. Their mean age is 84.4 years, and males account for 56.1 per cent of the sample. Less than half of the subjects with dementia confirmed by the court also had their severity of dementia confirmed, so the distribution of severity among our sample is that 23.4 per cent of people with dementia had moderate dementia, followed by 13.6 per cent with severe dementia and 8.4 per cent with mild dementia. In addition to the status of dementia, we examined another variable of mental capacity, but that information was only available in civil cases and not in criminal cases, which account for 19.2 per cent. The percentage of the cases in which people with dementia were confirmed with impaired mental capacity was 38.3 per cent, a number lower than 42.5 per cent of not impaired. Furthermore, only 1.4 per cent of all people with dementia had a legal guardian or assistant to help with their decisions, and that percentage was much smaller than the percentage of those with impaired mental capacity. One reason may be that if a person with dementia has already been protected under guardianship or assistance, rarely will there be lawsuits in our sample because his/her financial legal capacity has already been determined by the law, leaving no place to dispute. We also found that a majority (57.5%) of the people with dementia died upon the filing of lawsuits or prosecution of financial abuse in our sample. We examined the outcome of the legal procedure acknowledged by the court and found that overall, in 52.8 per cent of the financial abuse cases, the court considered the financial decision conducted by the person with dementia to be invalid, so the transaction was void.

Table 3. The characteristics of people with dementia involved in cases of financial abuse

Note: SD: standard deviation.

Table 4 lists the various types of legal issues involved in cases of financial abuse, and since some legal cases could have more than one type of issue, the denominator used for the percentage calculation was 261. After reviewing the details of each case related to the financial abuse of people with dementia, the most common three types of legal issues can be summarised as follows: transfer of real estate (39.1%), withdrawal from bank account without permission while the person with dementia is alive (19.5%) and guarantee (8.8%). The remaining legal issues in cases of financial abuse include inheritance, fraud and transfer of other property, but each of those types accounts for around only 5 per cent of the total number of cases.

Table 4. The types of legal issues involved in cases of financial abuse

Note: 1. The total number of the types of legal issues are more than the case number (N = 214) because there may be more than one type of legal issue in a case.

Table 5 shows the types of assets involved in cases of financial abuse, and similar to the findings shown in Table 4, two of the top three assets involved with people with dementia were land and houses since both are related to the transfer of real estate. Cases with the second highest proportion were financial abuse involving cash/bank savings, but there were only a very few cases about other types of assets such as stocks.

Table 5. The types of assets involved in cases of financial abuse

Note: 1. The total number of the types of assets is more than the case number (N = 214) because there may be more than one type of asset in a case.

The differences in financial losses by the type of perpetrators according to the routine activity theory are shown in Figure 2. There were 144 cases with an amount of financial losses recorded by the court, and comparing the box and whiskers of those three groups suggests that the amount of financial losses associated with an outsider was more dispersed than the amount associated with a family member. Financial losses were about 1.5 times more among cases where the financial abuse perpetrator was a family member rather than an outsider. The far left box plot in Figure 2 shows that of the 96 cases in which the perpetrator was a family member, the median value (the horizontal line inside the box) of the financial losses was US $137,907 per victim, ranging from US $217 (the lower whisker) to US$ 665,916 (the upper whisker). The middle box plot in Figure 2 shows that of the 35 cases in which the perpetrator was an outsider, the median value of the financial losses was US $92,333 per victim, ranging from US $1,667 to 776,667. The far right box plot in Figure 2 shows that of the 13 cases in which the case perpetrators included both a family member and a stranger, the median value of the financial losses was US $329,333 per victim, ranging from US $1,400 to 617,294.

Figure 2. Differences in financial losses according to perpetrator types.

Note: USD: US dollars. ◆ indicate mean value (while the black lines inside each boxplot represent the median value).

To explore the differences between elder fraud and financial exploitation further, we examined the outcome of legal procedures acknowledged by the court by the types of perpetrators. While in Table 3 we showed that the overall proportion of invalid financial decisions among our sample was 52.8 per cent, the proportion of invalid financial decisions among cases in which the perpetrator was an outsider was much higher at 65.9 per cent, compared to the proportion of 50 per cent invalid cases in which the perpetrator was a family member. This implies that the court was more likely to regard a case of financial abuse as an invalid financial decision when an outsider was regarded as the perpetrator rather than a case involving only a family member as a perpetrator.

In order to examine our findings according to the bystander intervention model, we selected the legal cases that included specific mentioning of the bank staff or teller operation in the verdicts and have presented the results in Table 6. Among those 62 cases, the role of bank staff was categorised as positive in 29 per cent of the cases since the court mentioned that the bank staff took action to protect the assets of a person with dementia. On the other hand, the bank staff played a negative role against the client's asset safety in only 8 per cent of the cases, while in two-thirds of the cases, the bank staff took a neutral role in which no value judgement could be made about their actions toward the client's assets. We specifically examined the type of perpetrators in cases where the court ruling indicated that the bank staff played a positive role in protecting the assets of a person with dementia. Most of the cases involving family members were related to bank account withdrawals, and the bank staff intervened positively by confirming with the person with dementia about their motives before proceeding with such a transaction. If we divided the number of positive cases by the number of total cases in which the perpetrator was an outsider, the proportion of positive cases would be 71 per cent compared to the proportion of 22 per cent for cases in which the perpetrator was a family member. This implies that while there were six more family-related than outsider-related financial abuse cases involving a person with dementia in absolute number, bank staff were more likely to be suspicious of financial abuse when an outsider was regarded as the perpetrator. Content analysis of the written judgements also revealed that the bank staff would often become alert when the case involved a large sum of money, and sometimes they turned to the police to report the fraudsters.

Table 6. The role of bank staff in cases of financial abuse

Note: 1. If the bank staff took an action to protect the assets of the person with dementia, then the role would be positive.

Discussion

By using court rulings and a systematic search, our analysis has provided an overview of the financial abuse cases involving people with dementia and reached an interesting finding: the majority of financial abuse perpetrators were family members (73.8%) as opposed to outsiders (19.2%), with half of the perpetrators being their adult children. Our conclusion that family members seemed to pose a greater threat than strangers to people with dementia in terms of financial abuse echoed previous studies on financial abuse of people lacking mental capacity using Court of Protection cases in the UK (Dalley et al., 2017b, 2017c). Our finding was also supported by previous studies on financial elder abuse, which found that while professionals like care workers, financial advisors or solicitors can be responsible for such misconduct, the majority of cases involved close family members or relatives as the perpetrators (Setterlund et al., 2007; Stelma-Roorda, 2021). The statistics of abusers reported in the UK involved family members of the financial abuse victims in over 76 per cent of the cases analysed (Dalley et al., 2017c). Dalley et al. (2017b) proposed that difficult intra-family dynamics were a source of risk to individuals dependent on family members for support and protection, and then they concluded from their analysis that vulnerable people who lacked capacity were often at greater risk from their relatives than is generally assumed. Specifically, empirical legal analyses have revealed that financial misbehaviours correspond to the proprietorial attitudes and assumed entitlements of some adult children towards their parents' assets and their inheritance quoted in the case details (Dalley et al., 2017b). Another study on the perpetrator of financial abuse also noted the so-called ‘sense of entitlement’ to the older adult's assets was an underlying motive for conducting financial exploitation (Conrad et al., 2011). That sense of entitlement could lead to what has been referred to as ‘asset-stripping’ (Tilse et al., 2005), whereby the perpetrator takes actions to transfer an adult's asset to family members in order to reduce the adult's asset value below the threshold required to qualify for the age pension or to avoid high accommodation fees in residential care. Even worse, the perpetrator may deliberately manipulate the vulnerable adult into making gift arrangements solely for their own personal interests, i.e. not only abusing the adult financially, but also reducing the inheritance of other heirs (Stelma-Roorda, 2021). This scenario was also shown in our research, where the largest group of perpetrators are adult children of people with dementia (52.7%; see Table 1), while the largest group of plaintiffs are also adult children (57.6%; see Table 2). Since most of the transactions between people with dementia and his/her adult child (the accused perpetrator) are often contested by the perpetrator's siblings, the financial abuse itself is indeed related to inheritance disputes.

It is worth noting that while all of our study samples had dementia confirmed by the court, only 38.3 per cent of the people with dementia were confirmed with impaired mental capacity, and more people with dementia (42.5%) were considered as not impaired by the court. These data show that even when a person is confirmed to be a person with dementia by the court, only in less than half of the cases was his/her financial decision-making denied (i.e. the mental capacity was recognised as impaired). The statistics reported in England show that the percentages of safeguarding adult referrals where the individual was assessed as lacking mental capacity accounted for 25–27 per cent in 2014–2016 (Dalley et al., 2017c). Doron et al. (2017), in their review of Israeli court rulings involving people with dementia found that the courts ruled in a majority of those cases that the people with dementia were legally capable of executing a will (61%), to stand trial in criminal cases (38%), and either fully or partially legally capable of entering into a contract or executing property transactions (57%). These findings from Taiwan, England and Israel are consistent in suggesting that having a diagnosis of dementia does not make the court automatically regard the person as lacking mental capacity. From the viewpoint of respecting the autonomy of the principal, this result may sound fair. Nevertheless, if these financial decisions essentially involve abuse, the low rate of impaired capacity means a low possibility of receiving a remedy.

Results from our analyses of the types of legal issues and assets involved in cases of financial abuse show that the transfer of real estate (land) was the predominant type, followed by withdrawal (cash) from a bank account without permission while the person with dementia is alive. The second type of cash withdrawal legal issue echoed the Court of Protection cases study in England, which found that the range of financial misbehaviour conducted by a power of attorney included co-mingling of funds and failure to keep and provide accounts of spending from the bank accounts of the person with dementia (Dalley et al., 2017b). While the English study did not mention real estate specifically, real estate was ranked seventh in the type of legal issues involved in the Israeli court rulings study and accounted for just 8 per cent of all cases (Doron et al., 2017). Our finding that the transfer of real estate and disputes of land and houses were among the most common types of assets involved in legal cases in Taiwan is different from other countries and may be due to the traditional preference of holding on to property rather than cash among Chinese older adults (Ministry of Finance, 2019). A typical Chinese custom is for parents to leave a house as an inheritance for the next generation (Jhang and Lu, 2012), so our finding that an adult child was the most likely perpetrator and that transfer of real estate was the number one legal issue is consistent with that tradition.

Our findings on comparing the financial losses by financial abuse perpetrator types revealed that the value of financial losses was about 1.5 times more among cases, and this result is quite different from the US study, which found that the financial losses were more than three times greater among fraud cases than financial exploitation, which that study defined as abuse committed by trusted others including family members (DeLiema, 2018). There are some plausible reasons for our differences in results. First, the data sources of the two studies are hardly comparable and have quite different sample sizes. As acknowledged by the author in the limitation section (DeLiema, 2018), the forensic centre cases were not representative of most reports to adult protective services in the USA, as they were the most serious cases, so the sample size with data available on financial losses was only 44. The current study, on the other hand, used the open data from civil and criminal cases including a large sample of 214 cases, among which 144 cases had amounts of financial losses reported in the verdict. Secondly, land and houses were among the top three assets involved in cases of financial abuse as they tend to have much higher values than cash, thus leading to greater financial losses among cases where the perpetrator was a family member rather than an outsider, someone who is less likely to have access to documents of real estate and less involvement in the transfer of such assets.

Our finding that the court was more likely to consider financial abuse committed by fraud perpetrators as invalid decisions as compared to financial abuse committed by financial exploiters was consistent with a model of elder financial victimisation adapted from the routine activity theory (DeLiema, 2018). The court was probably more vigilant to the behaviours of fraudsters in cases of financial abuse, or gave more lenient consideration to the financial decision of the person with dementia when the perpetrator was an outsider. On the other hand, it seemed more difficult for the court to rule against financial exploitation, since our data showed that the invalid proportion of financial abuse involving a family member was much lower. Given that family members are usually viewed as ‘trusted others’ or ‘capable guardians’ to protect the assets of the person with dementia, it would thus be much harder to identify or define a case of financial exploitation committed by family members. The privacy surrounding family interactions creates a confined environment where older individuals' personal finances can be easily accessed. Additionally, personal vulnerabilities such as cognitive declines make some older people more susceptible to abuse. These circumstances, combined with the low likelihood of a formal complaint being filed, further increase the potential for financial abuse to occur for people with dementia.

Previous research using the bystander intervention model to study the role of bank staff with regard to the financial abuse of an older customer as a bystander in emergencies has shown that they are more likely to become suspicious when a third party attempts to take control of the customer's bank account (Gilhooly et al., 2013). The findings of our analysis on the role of bank staff are consistent with previous research in showing that staff seemed more suspicious of financial abuse involving people with dementia, so they took preventive action to protect the assets when an outsider rather than a family member was regarded as the perpetrator. Similarly, our analyses on the proportion of invalid financial decisions by types of perpetrators also supported the routine activity theory in finding that the court was more likely to intervene when an outsider rather than a family member was regarded as the perpetrator. However, our results also revealed that people with dementia were at a higher risk from their family members than from outsiders in terms of greater financial losses. As suggested by research that applied the routine activities theory to study financial abuse in families, in order to prevent financial abuse more effectively, it is important to increase the probability of being held accountable for one's actions, even among family members (Setterlund et al., 2007). Thus, in addition to fraud prevention (which usually targets outsiders), future policies may include establishing guidelines for front-line bank staff on how to identify warning signs of abnormal transactions conducted by those trusted others of people with dementia. This would echo the description by Setterlund et al. (2007) that family members who assume the responsibility for assisting an older person with their finances must recognise that regardless of the familial context, they are still bound by legal and ethical obligations towards the older person.

Strengths and limitations

Considering the scarcity of literature identified in our review, the strength of our study lies in using a large dataset, including both civil and criminal cases identified through a systematic search; therefore, the people with dementia included in our analysis have generalisability. However, there are also limitations with this nationally representative data. The LawBank database we used did not contain information on medical records, so less than half of our sample had their dementia severity recorded in the written judgement. We also had little information on the health status of a person with dementia; therefore, we were unable to compare the effects of different levels of cognitive functioning on the outcomes of financial abuse.

Conclusions

Our study found that difficult intra-family relationships seemed to pose a greater risk of financial abuse involving people with dementia since adult children were found to be the most likely perpetrators. Transfer of real estate and disputes of land and houses were the predominant types of assets involved in legal cases, thus leading to a greater amount of financial losses in cases where the perpetrator was a family member rather than an outsider. In accordance with the bystander intervention model, bank staff were more likely to be suspicious of financial abuse involving a person with dementia when an outsider was regarded as the perpetrator. As such, staff played a positive role to intervene. The court was also more likely to intervene when an outsider rather than a family member was regarded as the perpetrator in financial abuse cases, and this finding was consistent with what the routine activity theory predicts about fraud perpetrators. However, as we have shown that people with dementia were often at a greater risk from their family members than from outsiders, future prevention strategies should also consider the training of bank staff and establishing policy guidelines to safeguard the assets of people with dementia from financial exploitation by trusted others.