Prevalence and type of unmet needs experienced by carers of people living with dementia

Abstract

Objectives

Carers of people with dementia experience significant physical, emotional, and social burdens. Needs assessment can provide important information to assist services in providing support to carers to reduce these burdens. However, few studies have examined the prevalence of unmet needs experienced by carers of people with dementia using a quantitative instrument. This study aimed to examine the prevalence and type of unmet needs experienced by carers of people with dementia in Australia.

Methods

This was a cross-sectional survey of carers providing support to community-dwelling individuals living with dementia. Carers were recruited through geriatric clinics, aged care providers, support services and community organisations. Consenting carers completed a survey including an 80-item study-specific unmet needs instrument for carers (UNI-C) and sociodemographic characteristics. Descriptive statistics were used to identify the most prevalent unmet needs reported by carers.

Results

A total of 169 carers (response rate 47%) completed the survey. Most (87%) carers reported at least one unmet need. Carers reported a median of 16.0 (IQR = 24.0; max = 58) out of a possible 80 unmet needs. The most frequently endorsed unmet needs spanned a range of areas including emotional wellbeing, accessing health and aged care services, managing apathy, and finding information.

Conclusion

This study highlighted that a large proportion of carers experience unmet needs associated with caring for someone with dementia. Developing and rigorously testing interventions to meet carers’ commonly reported unmet needs are warranted.

Supplemental data for this article is available online at https://doi.org/10.1080/13607863.2022.2053833 .

Keywords:

• Support person
• carer
• Alzheimer’s disease
• measurement
• unmet needs

Introduction

Carers of people living with dementia experience significant burden

Approximately 50 million people are living with dementia worldwide, with an estimated 10 million new cases diagnosed per year (World Health Organisation, 2020). As dementia is incurable with symptoms worsening over time, carers play an important and ongoing role in supporting those with dementia. Carers may be family members or friends of the person with dementia and provide assistance with a range of practical tasks (e.g. activities of daily living, household tasks, coordinating health care), manage cognitive and behavioural symptoms, and provide emotional support (Brodaty & Donkin, 2009). These tasks are often sustained over a number of years and can take a physical, psychological, social and financial toll (Alzheimer’s Association, 2019). Population based data indicates that the proportion of carers that report at least one physical or emotional adverse effect associated with caregiving is higher for carers providing support to a person with dementia compared to other chronic conditions (57% vs 48%, respectively) (Australian Institute of Health and Welfare, 2012). Carers of people with dementia who report a greater impact are more likely to place the individual they support into full-time care (Schulz et al., 2004; Toot et al., 2017).

Needs assessment can inform service planning and tailoring of support for carers

A number of approaches are available to assess the impacts of providing care to a person with dementia, such as measuring quality of life (Horton et al., 2021), assessing burden experienced (Hébert et al., 2000) and measuring coping strategies (Cooper et al., 2008). While these approaches show the presence and magnitude of different types of problems experienced, they do not indicate whether carers need or want help addressing these issues. Needs assessment is an alternate paradigm which involves asking individuals about the extent to which they require help to address their needs (Wright et al., 1998). This approach can be used to guide the delivery of person-centred care by informing health care providers about the specific issues that individual carers themselves have identified as requiring support. Identifying unmet needs that are most prevalent among large samples of carers can also drive priority setting and resource allocation by governments and policymakers and inform the development of interventions to fulfil carers’ identified needs.

Few studies have examined the prevalence of unmet needs among carers of people with dementia

A recent review identified 31 studies that had explored needs of carers of people living with dementia (Queluz et al., 2020). This review found that the most commonly reported needs were related to the carer’s own health, and in particular their own emotional wellbeing. Information needs were also commonly reported. The review also highlighted a range of limitations in the existing literature. Just over one third of the included studies used a quantitative instrument to identify the prevalence of unmet needs. Quantitative studies enable an examination of the type, number and extent of needs experienced across different domains, allowing services to prioritise delivery of support to the areas of greatest identified need. Further, studies often did not distinguish between needs that were met and needs that remained unmet. Distinguishing needs that remain unmet allows the identification of areas where additional resources and support may be beneficial. The review also highlighted important gaps in the reporting of sociodemographic characteristics of study samples. For example, less than half of studies reported on living arrangements of the carer, which has been shown to be associated with the extent of carer stress (Brodaty & Donkin, 2009). Reporting on a comprehensive range of sociodemographic characteristics of the carer and person with dementia is essential for contextualising the findings and allowing for comparisons across studies. Finally, the majority of this research was undertaken in the United States, Finland, and the Netherlands, suggesting a paucity of information relevant to specific countries, different socio-cultural contexts, and different health care systems.

Limitations of existing instruments assessing unmet needs of carers of people with dementia

While there are existing instruments available to assess unmet needs of carers of people with dementia, they have significant limitations (Mansfield et al., 2017). Some measures (e.g. Johns Hopkins Dementia Care Needs Assessment) (Johnston et al., 2011), assess needs from a clinician’s perspective only. Given the large body of research showing inconsistencies between self and proxy-reported outcome measures (Jones et al., 2011; Perkins, 2007), this approach may introduce significant bias in identification of needs. Existing instruments are also limited in how comprehensively they cover the potential needs of carers. Some measures that have been used to assess needs for people with dementia such as the Camberwell Assessment of Need for the Elderly (Reynolds et al., 2000) and the Tayside Profile for Dementia Planning (Gordon et al., 1997) also include a sub-set of items to assess carers’ needs, however these items are limited in scope. Existing measures do not include important aspects of caregiving, such as dealing with healthcare systems, assisting the person with dementia to maintain their independence, and providing emotional support to the individual (Mansfield et al., 2017).

In this quantitative study we examined, using a study-specific measure, the unmet needs of carers of people with dementia in Australia, including:

1. The proportion of participants that reported any unmet need for help, and the median number of unmet needs identified;

2. The most prevalent items for which participants identified an unmet need.

Methods

Design

Cross-sectional survey of carers of people with dementia. The study was approved by the Human Research Ethics Committees of Hunter New England Health (17/05/17/4.07) and the University of Newcastle (H-2018-0308).

Participants

Carers were defined as people who have a significant personal relationship with the person with dementia and provide a main source of emotional and practical support. Eligible carers were aged 18 years or older, provided support to a community-dwelling person who had received a confirmed diagnosis of dementia from a medical professional, and were able to read and write English. Carers of people who were residing permanently in a residential aged care facility were ineligible.

Recruitment and data collection

Recruitment and data collection took place between July 2018 and June 2020. Five geriatric clinics, three respite centres for people with dementia and six carer support groups in New South Wales and an aged care provider in Victoria audited their client lists to identify and invite potentially eligible carers to participate in the study. A study information pack including a cover letter, information statement, hard copy of the survey, and reply-paid envelope was either mailed or provided to eligible carers when they visited the clinic. Participants were invited to complete the survey and return it in a reply-paid envelope provided. Completion and return of the survey was taken as implied consent to participate in the study. A reminder pack, including a second copy of the survey and reply-paid envelope was mailed after 4 weeks to all non-responders mailed the initial study pack. The geriatric clinics and aged care provider recorded the sex and age of the carer, if this information was available, in order to assess consent bias. Additional participants were recruited as part of a larger study of older adults’ wellbeing through five geriatrician clinics, 34 community groups (e.g. Country Women’s Associations, Men’s Sheds) and an aged care provider located across four Australian states. These participants completed an eligibility survey for the current study. Those who met inclusion criteria were sent the study pack and given the option of completing the survey by paper or online. A reminder was sent to non-responders four weeks later if required.

Measure

Carer unmet needs

An unmet needs instrument for carers of people with dementia, called the UNI-C, was developed using a mixed-methods approach. Initially, a list of possible items and domains was generated after a literature review to identify impacts, concerns and needs experienced by carers of people with dementia. Additional items were added based on the input of carers through qualitative interviews (n = 18 carers) and a focus group (n = 3 carers) exploring the type and frequency of needs they had experienced. A working group consisting of behavioural scientists (n = 5), a statistician (n = 1), health care professionals working in aged care (n = 2) and carers (n = 3) reviewed the resulting list of 147 items for relevance and completeness. Redundant and overlapping items were removed and item wording changed to ensure understanding. Draft versions of the instrument were sent to 14 carers, 10 health care professionals, three managers of community aged care services, 10 behavioural scientists, four representatives from relevant community organisations and one legal professional. These individuals provided feedback on item clarity, relevance and redundancy, and any items they perceived were missing. Cognitive interviewing was carried out with a sub-set of 10 carers to ensure adequate understanding of the items and response scale. A provisional version of the instrument was pilot tested with 27 carers. The proportion of missing data was low across all items (maximum = 7%). Some changes were made to the response scale to increase ease of understanding. Three items were removed due to floor effects and two items were added based on the feedback provided. The revised version was pilot tested with 36 carers.

The resulting unmet needs instrument, the UNI-C, consisted of 80 items across seven domains: Your own wellbeing (7 items); Your feelings and emotions (9 items); Your relationship with person you support (7 items); Managing practical needs of the person you support (18 items); Professional care of the person you support (15 items); Managing symptoms of dementia in the person you support (12 items); and Finding information (12 items). The overall reading age was Flesch-Kincaid grade 6. For each item, carers were asked to rate whether they needed help in the last month as a result of caring for a person with dementia. The four-point response scale included the options: ‘I didn’t need help’, ‘I had enough help’, ‘I could have used a little more help’, ‘I could have used a lot more help’. If an item was not relevant to the carer’s situation, they were advised to respond with ‘I didn’t need help’. See Supplementary File 1 for the complete list of items.

Carer sociodemographic characteristics

Carers were asked to self-report their sex, age, postcode, highest level of education, private health insurance status, concession card status (access to a range of medical services or medications in Australia for free or at a reduced rate), presence of other health conditions (heart disease, lung disease, back pain, cancer diabetes, arthritis, mental health condition, other), number of years spent caring for the person with dementia, number of hours per week spent providing care, relationship to the person with dementia, and whether they live with the person with dementia.

Sociodemographic, disease and functional characteristics of the person with dementia

Carers were asked about characteristics of the person they support, including their sex, age, month and year of diagnosis, type of dementia, other health conditions (heart disease, lung disease, back pain, cancer diabetes, arthritis, mental health condition, other), and an overall rating of severity of symptoms of dementia [on a scale of 1 (very mild) − 10 (severe)].

Statistical analysis

Analyses were conducted in R version 4.0.2 (2020-06-22) (R Core Team, 2020). Demographic, disease and functional characteristics are reported as counts and proportions for categorical variables and mean and standard deviation for continuous variables. Postcode was used to calculate an Accessibility/Remoteness Index of Australia Plus (ARIA+) score, which classifies individuals as living in one of five remoteness areas: major city, inner regional, outer regional, remote or very remote; and to determine the Socio-Economic Index for Areas (SEIFA) Index of Advantage and Disadvantage (2016), which provides an estimate of the degree of socio-economic advantage or disadvantage across geographical areas (Australian Bureau of Statistics, 2016). Possible response bias was assessed by comparing gender and age of carers who completed a survey with eligible carers who were sent a survey pack but did not respond. This data was available for those participants recruited through geriatric clinics and the aged care provider. The age, sex and remoteness status of the sample was also compared to national data on dementia and carers (Australian Institute of Health and Welfare, 2012) using Chi-squared tests.

As the proportion of missing data was low across items (only 5 items were missed by more than 10% of participants; Supplementary File 1) and participants (72% of participants had one or fewer missing responses), missing data for individual items were replaced with the response ‘I didn’t need help’. This is a conservative approach that underestimates prevalence of unmet needs. The analysis was also conducted as a complete case analysis with only minor differences.

Aim 1

Unmet needs were defined as items on which participants responded either ‘I could have used a little more help’ or ‘I could have used a lot more help’. The proportion of participants with at least one unmet need was calculated. The median, interquartile range (IQR) and maximum of the total number of unmet needs reported across participants was calculated.

Aim 2

Unmet needs were ranked in descending order and the top 25 items presented.

Results

Characteristics of the sample

Of 414 eligible carers sent an information pack, 196 returned a completed survey (response rate 47%). Of those who responded, a further 27 were removed as the person with dementia lived in a residential aged care facility (n = 26) or did not answer any of the unmet needs items (n = 1), leaving a final sample of 169 eligible carers included in the analysis. Based on the available data, there was no response bias for gender (n = 242; p = 0.886), however younger carers were less likely to respond (n = 147; p < 0.001). These trends were also observed when comparing our sample to national demographic data for carers of people with dementia (Australian Institute of Health and Welfare, 2012) (sex p = 0.635; age p < 0.001). Compared to national data, our sample contained more carers who were the partner of the person with dementia (p < 0.001) and more residing in major cities, though not significantly so (p = 0.083) (Australian Institute of Health and Welfare, 2012).

Characteristics of the sample are shown in Table 1. The majority of carers were female (74%, n = 123) with a mean age of 71.8 years (SD = 10.9) and were the partner of the person they support (83%, n = 139). A majority (84%, n = 139) lived in major cities, and 39% (n = 65) lived in areas of greater disadvantage according to the SEIFA Index of Advantage and Disadvantage.

Table 1. Sociodemographic characteristics of carers and the person with dementia they support (n = 169).

Variable		N^a (%)
Carer
Age	<60	25 (15%)
	60–74	62 (37%)
	75+	80 (48%)
Gender	Male	44 (26%)
	Female	123 (74%)
ARIA + score	Major city	139 (84%)
	Inner regional	22 (13%)
	Outer regional	5 (3.0%)
SEIFA	Most disadvantaged (decile ≤ 4)	65 (39%)
	Not disadvantaged (decile > 4)	101 (61%)
Education	High School or below	73 (45%)
	Vocational training, University or Other	91 (56%)
Relationship to person with dementia	Partner	139 (83%)
	Parent	22 (13%)
	Other	7 (4%)
Living arrangement	Live with person they support	156 (93%)
	Live separately	12 (7%)
Private health insurance	Yes	125 (75%)
	No	42 (25%)
Concession card	Yes	136 (81%)
	No	31 (19%)
Health conditions	None	54 (32%)
	At least one	114 (68%)
Person with dementia
Age	<60	3 (2%)
	60–74	32 (19%)
	75+	131 (79%)
Gender	Male	107 (64%)
	Female	60 (36%)
Time since diagnosis	Up to 2 years ago	40 (25%)
	>2 years ago	118 (75%)
Type of dementia	Alzheimer’s Disease	89 (53%)
	More than one type or Other	47 (28%)
	Don’t know	32 (19%)
Symptom severity rating	≤ 3 (mild)	14 (9%)
	4-7 (moderate)	84 (54%)
	≥ 7 (severe)	58 (37%)
Other health conditions	None	46 (27%)
	At least one	122 (73%)

a May not sum to n = 169 due to missing data.

Most of the carers provided support to a male (64%, n = 107) with a mean age of 79.4 years (SD = 8.6) and who had been diagnosed with dementia over 2 years ago (75%, n = 118). Most carers reported the severity of symptoms in the person with dementia was moderate (mean = 5.9; SD = 1.8).

Prevalence and number of unmet needs

Eighty-seven percent (n = 147) of all participants identified at least one unmet need on the UNI-C. The median number of unmet needs reported was 16.0 out of a possible 80 (IQR = 24.0). The full break down of responses for each item are provided in Supplementary File 1.

Most frequently reported unmet needs

The top 25 unmet needs from the 80 item UNI-C are presented in Table 2. The three most frequently endorsed unmet needs were related to the carer’s wellbeing, feelings and emotions, including ‘Finding the emotional energy that you need to care for the person you support’ (53%, n = 90), ‘Worrying about what will happen in the future’ (45%, n = 76), and ‘Feeling angry and frustrated’ (45%, n = 76). Other frequently reported unmet needs relating to the carers’ feelings and emotions included ‘Feeling sad’ (37%, n = 63), ‘Feeling as though you are the only one supporting the person with dementia’ (37%, n = 62) and ‘Feeling lonely’ (31%, n = 52). Items relating to impacts on carers’ wellbeing were also among the most frequently endorsed unmet needs and included ‘Managing stress’ (40%, n = 67), ‘Keeping up your social activities or hobbies’ (34%, n = 58), ‘Balancing your role as a carer with other responsibilities’ (33%, n = 56), and ‘Getting enough sleep’ (33%, n = 55).

Table 2. Items and their corresponding domains endorsed by at least 20% of respondents as needing at least a little more help, sorted in descending order (n = 169).

Rank	Item	Domain	N (%)
1	Finding the emotional energy that you need to care for the person support you	Your own wellbeing	90 (53%)
2	Worrying about what will happen in the future	Your feelings and emotions	76 (45%)
3	Feeling angry or frustrated	Your feelings and emotions	76 (45%)
4	Keeping the person you support motivated	Managing practical needs of the person you support	71 (42%)
5	Managing stress	Your own wellbeing	67 (40%)
6	Being aware of clinical trials and research that we could be involved in	Finding information	66 (39%)
7	Coping with thinking for the person you support as well as yourself	Your relationship with person you support	64 (38%)
8	Feeling sad	Your feelings and emotions	63 (37%)
9	Keeping the person you support physically active	Managing practical needs of the person you support	63 (37%)
10	Uncertainty about who will care for the person you support if something happens to you	Professional care of the person you support	63 (37%)
11	Feeling as though you are the only one supporting the person with dementia	Your feelings and emotions	62 (37%)
12	Finding meaningful and enjoyable activities for the person you support	Managing practical needs of the person you support	62 (37%)
13	Understanding which symptoms are caused by dementia and which are not	Finding information	62 (37%)
14	Keeping up your social activities or hobbies	Your own wellbeing	58 (34%)
15	Navigating the My Aged Care website	Professional care of the person you support	58 (34%)
16	Balancing your role as a carer with other responsibilities	Your own wellbeing	56 (33%)
17	Getting enough sleep	Your own wellbeing	55 (33%)
18	Feeling unsure about how the person you support would like to be treated	Your feelings and emotions	54 (32%)
19	Explaining to others what you are going through	Your relationship with person you support	53 (31%)
20	Feeling lonely	Your feelings and emotions	52 (31%)
21	Finding information on possible treatments for dementia	Finding information	51 (30%)
22	Knowing when the person you support should move to a nursing home	Finding information	51 (30%)
23	Understanding the perspective of the person you support	Your relationship with person you support	50 (30%)
24	Accessing a home care package with enough help for the person you support	Managing practical needs of the person you support	50 (30%)
25	Accessing dementia-specific care or services	Professional care of the person you support	49 (29%)

Unmet needs related to keeping the person with dementia active and engaged were frequently identified, including ‘Keeping the person you support motivated’ (42%, n = 71), ‘Keeping the person you support physically active’ (37%, n = 63), and ‘Finding meaningful and enjoyable activities for the person you support’ (37%, n = 62). Several unmet information needs were reported, including ‘Being aware of clinical trials and research that we could be involved in’ (39%, n = 66) and ‘Understanding which symptoms are caused by dementia and which are not’ (37%, n = 62). A number of carers reported difficulties in accessing the health and aged care system, including ‘Navigating the My Aged Care website’ (34%, n = 58), ‘Accessing a home care package with enough help for the person you support’ (30%, n = 50) and ‘Accessing dementia-specific care or services’ (29%, n = 49).

Discussion

Prevalence of unmet needs

This study found that carers of people with dementia experience a high level of unmet needs across the domains examined. Over 85% of carers reported at least one unmet need for help on the UNI-C, with a median of 16 needs reported. The top unmet need was endorsed by over 50% of carers. These results are comparable with studies from other countries in which 76%–97% of carers reported at least one unmet need across other unmet needs measures (Black et al., 2013; Zwingmann et al., 2019).

Types of unmet needs reported

Australian clinical practice guidelines recommend that carers of people with dementia be provided with information about accessing appropriate services, such as those available through Dementia Australia and community aged care services, and that they have access to programs to build their skills and provide coping strategies to support the carer’s own wellbeing (Guideline Adaptation Committee, 2016). Our study found that carers have substantial unmet needs across each of these areas, suggesting significant gaps in the support provided to these individuals.

Consistent with recent reviews, carers most frequently reported unmet needs related to their wellbeing, and in particular their feelings and emotions (Holt Clemmensen et al., 2021; Queluz et al., 2020). These findings also align with an Australian study, in which carers rated a lack of time for self, social isolation, exhaustion, and the emotional toll of caregiving among the most significant impacts of caregiving (Saunders, 2012). In the current study, just over half of carers reported needing help with finding the emotional energy needed to provide support, while 31%–45% reported needing help with feeling angry or frustrated, stressed, sad, and lonely. At least one-third reported needing help with getting enough sleep, maintaining social activities or hobbies, and explaining to others what they are going through. These findings may reflect that the demands of the caring role leave limited time and energy for carers to devote to self-care (Oliveira et al., 2019). These challenges may be especially present for carers of people with dementia (Teahan et al., 2021). Carers have identified that informal supports, such as those provided by family, friends and social support groups may assist in relieving some of the emotional toll by providing positive encouragement, giving an opportunity to release stress and providing a sense of security (McCabe et al., 2016). Conversely, receiving a lack of support from these sources can lead to feelings of frustration, anger, and resentment (Landmark et al., 2013). Carers should be supported to initiate and maintain contact with family, friends and other informal supports (e.g. carer support groups) to reduce the emotional burden.

Large proportions (37%–42%) of carers reported unmet needs related to keeping the person they support motivated and engaged. Apathy is one of the most prevalent and challenging behaviours observed among people with dementia (Craig et al., 2005; Lyketsos et al., 2002), and is associated with increased carer distress (Van Reekum et al., 2005). A systematic review of non-pharmacological interventions found that therapeutic activities such as multisensory stimulation and creative activities may be effective in reducing apathy, however the quality of evidence was poor (Brodaty & Burns, 2012). Our findings highlight the need to develop and rigorously test the effectiveness of interventions involving carers to increase motivation and engagement among people with dementia, and to provide support to carers in implementing evidence-based interventions.

In line with other studies (Holt Clemmensen et al., 2021; Mollica et al., 2020), carers frequently identified that they required help with accessing or utilising formal care services, such as accessing dementia specific care, and adequate aged care services. Just over one-third of carers indicated they needed help with using the My Aged Care website, which was introduced in 2013 as the main gateway to accessing aged care services in Australia. Slightly fewer carers identified difficulties with accessing adequate home care services from community aged care providers. These findings highlight concerning gaps in the provision of health and aged care services to people with dementia in the community, adding to the strain experienced by informal carers.

Carers expressed a need for more information about the symptoms and treatment of dementia. A scoping review found that information needs of carers of people living with dementia vary depending on the severity of dementia (Soong et al., 2020). This suggests that information needs evolve over time and should be regularly reassessed by health care professionals.

Limitations

Compared to national data, our sample included more carers who were providing support to a partner or spouse as compared to a parent and who were older (Australian Institute of Health and Welfare, 2012). Furthermore, three-quarters of carers in this sample were providing support to a person who had been diagnosed with dementia more than 2 years ago, and most rated the dementia symptoms as moderate. These discrepancies may limit the generalisability of findings to carers who are supporting a parent, who are younger and who are supporting someone with a more recent diagnosis. However, it should be noted that we used carer ratings of dementia symptom severity rather than a validated measure which limits the ability to assess generalisability in this regard. The cross-sectional design of the study limits the ability to assess how carer needs change over time to identify the time points at which intervention would be most appropriate. Future longitudinal studies would allow for an examination of how carers’ needs change as symptoms progress and what characteristics of carers or the individuals they support are predictive of certain needs.

It is difficult to compare these results to other studies given the lack of standardised and psychometrically robust measures specific to carers of people living with dementia. Although the measure used in this study was developed utilising a rigorous and accepted measure development process with extensive input from end users to ensure relevance and comprehensiveness of items, its psychometric properties have not been examined and is an area for future research. While we did not examine the extent to which needs were unmet by comparing proportions identifying that they needed ‘a little more’ vs ‘a lot more’ help, it may be possible to utilise this more sensitive approach in clinical settings to assist in identification of the most pressing needs for individual carers.

Conclusions

With rising numbers of people with dementia living in the community, there is an increasing reliance on informal carers to provide support to these individuals. Consequently, there is a need to understand the areas in which carers need support to maximise their ability to provide care while also maintaining their own wellbeing. Our study found that unmet needs are highly prevalent among carers of people with dementia. Carers’ most commonly reported unmet needs span a broad range of areas, including emotional impacts, dealing with apathy, accessing appropriate dementia support services, and information needs. Developing and rigorously testing interventions to support carers across each of these areas may assist in reducing needs and subsequently alleviating burden and stress.

Ethics approval statement

The study was approved by the Human Research Ethics Committees of Hunter New England Health (17/05/17/4.07) and the University of Newcastle (H-2018-0308).

Acknowledgements

The authors would like to thank the carers who gave their time to contribute to this study; William Browne and Elaine Todd for contributing to development of the unmet needs measure; Sandra Dowley for data management; and Matthew Clapham for statistical support.

Disclosure statement

The authors have no conflicts of interest to declare.

Data availability statement

The data that support the findings of this study are available from the corresponding author [EM], upon reasonable request.

Additional information

Funding

This work was supported by a National Health and Medical Research Council Dementia Research Team grant (APP1095078) and infrastructure funding from the Hunter Medical Research Institute. Allison Boyes is supported by a National Health and Medical Research Council Early Career Fellowship (APP1073317). Mariko Carey is supported by a National Health and Medical Research Council Boosting Dementia Leadership Fellowship (APP1136168).