Dementia prevalence estimation among the main ethnic groups in New Zealand: a population-based descriptive study of routinely collected health data

Abstract

Objective

Estimates of dementia prevalence in New Zealand (NZ) have previously been extrapolated from limited Australasian studies, which may be neither accurate nor reflect NZ’s unique population and diverse ethnic groups. This study used routinely collected health data to estimate the 1-year period prevalence for diagnosed dementia for each of the 4 years between July 2016 and June 2020 in the age 60+ and age 80+ populations and for the four main ethnic groups.

Design

A population-based descriptive study.

Setting

Seven national health data sets within the NZ Integrated Data Infrastructure (IDI) were linked. Diagnosed dementia prevalence for each year was calculated using the IDI age 60+ and age 80+ populations as the denominator and also age–sex standardised to allow comparison across ethnic groups.

Participants

Diagnosed dementia individuals in the health datasets were identified by diagnostic or medication codes used in each of the data sets with deduplication of those who appeared in more than one data set.

Results

The crude diagnosed dementia prevalence was 3.8%–4.0% in the age 60+ population and 13.7%–14.4% in the age 80+ population across the four study years. Dementia prevalence age–sex standardised to the IDI population in the last study period of 2019–2020 was 5.4% for Māori, 6.3% for Pacific Islander, 3.7% for European and 3.4% for Asian in the age 60+ population, and 17.5% for Māori, 22.2% for Pacific Islander, 13.6% for European and 13.5% for Asian in the age 80+ population.

Conclusions

This study provides the best estimate to date for dementia prevalence in NZ but is limited to those people who were identified as having dementia based on data from the seven included data sets. The findings suggest that diagnosed dementia prevalence is higher in Māori and Pacific Islanders. A nationwide NZ community-based dementia prevalence study is much needed to confirm the findings of this study.

Correspondence to Dr Gary Cheung; [email protected]

Strengths and limitations of this study

• Routinely collected administrative health data are standardised and provide good national coverage but have limitations.

• Dementia is under-recognised and underdiagnosed, so not all cases will be captured by administrative health data sets.

• New Zealand does not routinely collect primary care dementia data, so this study likely underestimates the prevalence of dementia.

• A community-based dementia prevalence study is needed to determine the true prevalence of dementia.

• Community-based dementia prevalence studies are expensive. This study provides New Zealand-specific dementia estimates.

Introduction

Dementia is a common late-life neurodegenerative condition. In 2015, there were an estimated 46.8 million people living with dementia globally and this number was expected to double every 20 years.¹ The Dementia Economic Impact Report estimated there were 70 000 people living with dementia in New Zealand in 2020.² This estimate was extrapolated from findings of one small New Zealand and three Australian studies, three of which were published over 25 years ago^3–6 and none of them considered New Zealand’s diverse ethnic population, which includes 70.2% European, 16.5% Māori, 15.1% Asian (mainly Chinese and Indian) and 8.1% Pacific Islanders.⁷ A recent New Zealand study reported that the risk factor prevalence and weighted population attributable fraction for dementia are higher in Māori (51.4%) and Pacific Islanders (50.8%) but lower in Asian peoples (40.8%), compared with New Zealand as a whole (47.6%); therefore, dementia prevalence is also likely to vary across ethnic groups.²

Addressing the public health and cost impact of dementia requires an accurate estimation of its prevalence, but there has never been a community-based dementia prevalence study in New Zealand that represents all of the major ethnic groups. Prevalence studies are costly to carry out, so linked electronic health records are increasingly being used to inform the epidemiology of various health conditions, including dementia. Integrated Data Infrastructure (IDI) is a state-of-the-art New Zealand research database holding microdata about people and households.⁸ Data from a number of sources including government agencies, Statistics New Zealand surveys and non-government organisations are linked together to form the IDI. All data in the IDI are deidentified and information that could potentially be used to identify people, such as National Health Index (NHI) identifiers, are encrypted.

A previous New Zealand study linked health data sets (Mortality, National Minimal Data set—Publicly and Privately Funded Hospital Discharges and Pharmaceutical Collection) in the IDI and estimated the prevalence of dementia in the age 60+ population from 2012 to 2015.⁹ The estimated dementia prevalence of 2% was much lower than the 6.9% estimated by the World Alzheimer Report 2015.¹ However, direct comparison of these dementia prevalence estimates cannot be made due to different methodology used in these studies. The IDI has since been expanded to include additional health data sets such as interRAI, which may increase the potential for identification of dementia cases. interRAI is a standardised geriatric assessment mandated by the New Zealand Ministry of Health since 2012 for all people assessed for publicly funded home support services and aged residential care. Approximately, 10% and 40% of all New Zealanders aged 65 years and 85 years or older, respectively, have had an interRAI assessment.¹⁰ Therefore, we would expect its inclusion in a suite of linked data sets to identify more dementia cases and provide a better estimate of dementia prevalence in New Zealand.

This descriptive study aimed to use linked data from seven health databases within the IDI in order to: (1) estimate the crude 1-year period prevalence of dementia in the age 60+ and age 80+ populations for each of the 4 years between 1 July 2016 and 30 June 2020 and (2) calculate age–sex standardised rates to allow more accurate comparison of dementia prevalence across the four main New Zealand ethnic groups (Māori, Pacific Islander, Asian and European).

Materials and methods

This was a population-based descriptive study. We sought permission from Statistics New Zealand to access IDI and to conduct this project between November 2020 and February 2022 (Reference: MAA2020-12).

Identification of dementia cases and study period

Table 1 provides details of each data set and summarises the methods used to identify dementia cases in the seven health data sets (interRAI, Mortality, National Needs Assessment and Service Coordination Information System (SOCRATES), Pharmaceutical Collection, Privately Funded Hospital Discharges, Programme for the Integration of Mental Health Data (PRIMHD) and Publicly Funded Hospital Discharges).

Table 1

Health datasets, their availability periods and dementia case identification in the New Zealand Integrated Data Infrastructure (IDI)

Health datasets	Description	Data availability period	Identification of dementia cases
1. interRAI	Mandated standardised comprehensive geriatric assessment for all publicly funded home support services and aged residential care.	July 2014–June 2021	Diagnosis of Alzheimer’s disease or Dementia other than Alzheimer’s disease
2. Mortality	Data classifying the underlying cause of death for all deaths registered in New Zealand, including all registered foetal deaths (stillbirths), using the WHO Rules and Guidelines for Mortality Coding.	July 1907–December 2018	ICD-9 and ICD-10-AM codes*
3. National Needs Assessment and Service Coordination Information System (SOCRATES)	Used by Ministry-funded Needs Assessment and Service Coordination agencies to record information about clients who are eligible for Disability Support Services.	September 1939–September 2020	Diagnostic codes*
4. Pharmaceutical Collection	Contains information about subsidised dispensed medications processed by the General Transaction Processing System, including demographic information about healthcare users to whom these prescriptions were dispensed.	January 2005–June 2020	Donepezil tablets Rivastigmine patch
5. Privately Funded Hospital Discharges	Subset of fields from the National Minimum Dataset. Includes discharge and event data about privately funded hospital events and demographic data reported for the population cohort.	March 1914–December 2018	ICD-9 and ICD-10-AM codes*
6. Programme for the Integration of Mental Health Data (PRIMHD)†	Contains data about the referral, what services (activities) were provided, and demographic information. Excludes outcomes, diagnosis and legal status data.	November 1974–June 2020	ICD-9 and ICD-10-AM codes*
7. Publicly Funded Hospital Discharges	Subset of fields from the National Minimum Dataset. Includes discharge and event data about publicly funded hospital events (including admissions occurring at private hospitals but are publicly funded) and demographic data reported for the population cohort.	May 1914–December 2020	ICD-9 and ICD-10-AM codes*

*Refer to online supplemental appendix.

†PRIMHD contains information of PRIMHD data and Mental Health Information National Collection (MHINC) data, which was the mental health data collection prior to PRIMHD.

The diagnostic codes for dementia in each database were defined as follows:

• ICD-9 and ICD-10-AM codes for dementia (online supplemental appendix) in Mortality, PRIMHD, Privately Funded Hospital Discharges and Publicly Funded Hospital Discharges. We included ICD-9 codes because they were still being used in some of the health data sets such as Privately Funded Hospital Discharges and Publicly Funded Hospital Discharges.

• Diagnosis of ‘Alzheimer’s disease’ or ‘Dementia other than Alzheimer’s disease’ in interRAI. An interRAI assessment routinely records these two diagnoses. These diagnoses are determined by interRAI assessors who undergo competency assessment to confirm they can accurately record assessment information. interRAI assessors use multiple sources of information to determine diagnoses, for example, referral documentation, person interview, observation and discussion with family, carers or health professionals.

• Funded antidementia medications (donepezil tablets and rivastigmine patch) are used as proxies for a diagnosis of dementia in the Pharmaceutical Collection (see online supplemental appendix for medication codes) as these medications are not prescribed for any condition other than dementia in New Zealand. Donepezil is fully subsidised in New Zealand, while rivastigmine patch is available on a special authority application for funding, meaning certain criteria need to be met for subsidy to be obtained. Only data on subsidised medicines are contained in the Pharmaceutical collection, so we were unable to collect prescribing data regarding galantamine and memantine.

• Diagnostic codes for dementia or dementia subtypes (online supplemental appendix) in SOCRATES.

Data linkage

Statistics New Zealand routinely cleans the IDI population data to avoid an individual having two Unique Person Identifiers. We also used the Structured Query Language (SQL) function ‘COUNT DISTINCT Unique Person Identifier [snz_uid]’ to ensure there were no duplicated individuals. After we identified all dementia cases in the IDI in our study period, we used the SQL ‘JOIN’ function, including ‘LEFT JOIN’, ‘RIGHT JOIN’ and ‘INNER JOIN’, to link the seven health datasets and sociodemographic details. The sociodemographic details include sex, date of birth, deceased date and ethnicity (in the prioritised order of Māori, Pacific Islander, Asian, Middle East Latin American and African (MELAA), Other and European).

Calculation of dementia prevalence

Using the deceased date, we were able to determine the number of dementia cases in the IDI (alive and deceased) in the four study years: 1 July 2016 to 30 June 2017, 1 July 2017 to 30 June 2018, 1 July 2018 to 30 June 2019 and 1 July 2019 to 30 June 2020. We used the same definition of ‘total population at risk’ as was used in the previous IDI dementia prevalence study⁹ where all people who were alive or had died during each year of the study period were included. Each individual could only be counted once in each of the four study years.

TotalIDIpopulationatrisk=TotalnumberofactiveandaliveIDIindividuals+TotalnumberofdeathsinIDI

Statistics New Zealand defines an active IDI case by the presence of at least one activity in one of the following data sets within 12 months: (1) Inland Revenue; (2) Education; (3) Health; (4) Accident Compensation Corporation claims and (5) Aged under 5 and with a New Zealand birth registration or visa.

Dementia prevalence was calculated using the following formula:

${\displaystyle \begin{array}{l}\mathrm{P}\mathrm{r}\mathrm{e}\mathrm{v}\mathrm{a}\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e}=\frac{\genfrac{}{}{0ex}{}{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{o}\mathrm{f}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{d}\mathrm{e}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{c}\mathrm{a}\mathrm{s}\mathrm{e}\mathrm{s}+\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{b}\mathrm{u}\mathrm{t}}{\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{d}\mathrm{e}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{c}\mathrm{a}\mathrm{s}\mathrm{e}\mathrm{s}+\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{d}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{a}\mathrm{s}\mathrm{e}\mathrm{d}\mathrm{d}\mathrm{e}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{c}\mathrm{a}\mathrm{s}\mathrm{e}\mathrm{s}}}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{r}\mathrm{i}\mathrm{s}\mathrm{k}}\end{array}}$

We calculated crude 1-year period dementia prevalence in each of the four study years under the following categories: age 60+, age 80+ and per 5-year age bands from age 60+. Age 60+ was used in this study to allow direct comparison with the age 60+ figures reported by the World Alzheimer Report 2015 and the Walesby et al’s study.^{1 9} We also estimated crude 1-year period dementia prevalence for the four main ethnic groups (Māori, Pacific Islander, Asian and European). Due to the relatively low number of older adults of MELAA and Other Ethnicities living in New Zealand, they were excluded from our interethnic analysis. We estimated dementia prevalence age–sex standardised to the New Zealand IDI population in each of the four study years because of the differences in age and sex profile between different ethnic groups. These were calculated using the following formula to account for changing population numbers each year¹¹:

${\displaystyle \mathrm{D}\mathrm{i}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{l}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{i}\mathrm{z}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}=\frac{\mathrm{\Sigma }\left(\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{m}\mathrm{s}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{s}\times \mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{s}\right)}{\mathrm{\Sigma }\left(\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{s}\right)}}$

${\displaystyle \mathrm{D}\mathrm{i}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{l}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{i}\mathrm{s}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}=\frac{r_1{N}_1+r_2{N}_2+r_3{N}_3+\dots +r_n{N}_n}{N_1+N_2+N_3+\dots +N_n},}$

where, for k=1, 2, …, n,

${\displaystyle r_k=\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{i}\mathrm{n}{k}^{th}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{m}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{u}\mathrm{d}\mathrm{y}\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}$

That is, For each ethnicity, calculate the rate for each gender by the formula:

${\displaystyle r_k=\frac{\mathrm{D}\mathrm{e}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{C}\mathrm{o}\mathrm{u}\mathrm{n}\mathrm{t}\mathrm{f}\mathrm{o}\mathrm{r}{k}^{th}\mathrm{A}\mathrm{g}\mathrm{e}\mathrm{G}\mathrm{r}\mathrm{o}\mathrm{u}\mathrm{p}}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{P}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{f}\mathrm{o}\mathrm{r}{k}^{th}\mathrm{A}\mathrm{g}\mathrm{e}}}$

${\displaystyle N_k=\mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{o}\mathrm{f}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{n}{k}^{th}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{m}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}$

${\displaystyle N_k=\frac{\left(\mathrm{N}{\left(\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}1\right)}_k+\mathrm{N}{\left(\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}2\right)}_k+\mathrm{N}{\left(\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}3\right)}_k+\mathrm{N}{\left(\mathrm{Y}\mathrm{e}\mathrm{a}\mathrm{r}4\right)}_k\right)}{4}}$

$$\begin{gathered} {\displaystyle \begin{array}{l}\left(ie,N_{k}is\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{a}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{g}\mathrm{e}\mathrm{o}\mathrm{f}{k}^{th}\mathrm{a}\mathrm{g}\mathrm{e}\mathrm{s}\mathrm{u}\mathrm{b}\mathrm{g}\mathrm{r}\mathrm{o}\mathrm{u}\mathrm{p}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{h}\mathrm{e} \\ \mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{t}\mathrm{h}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{s}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{s}\mathrm{t}\mathrm{u}\mathrm{d}\mathrm{y}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{i}\mathrm{o}\mathrm{d}\mathrm{s}\right)\end{array}} \end{gathered}$$

${\displaystyle N=\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}\mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{o}\mathrm{f}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{p}\mathrm{o}\mathrm{p}\mathrm{u}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\left(\sum N_k\right)}$

(ie, N is the sum of all the average age subgroup of the total population of the three specific study periods);

${\displaystyle \sum \mathrm{m}\mathrm{e}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{s}\mathrm{u}\mathrm{m}\mathrm{m}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{k}\mathrm{s}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{a}.}$

Results

Identification of dementia cases

Table 2 shows the number of dementia cases identified in each of the seven health data sets. interRAI, Publicly Funded Hospital Discharges, PRIMHD and Pharmaceutical Collection data sets contributed the greatest number of dementia cases. Table 2 also shows the Venn diagrams to illustrate the intersects of these four health data sets.

Table 2

Number of dementia cases identified in the seven health data sets

Data set	1 July 2016 to 30 June 2017 N=41 763, n (%)	1 July 2017 to 30 June 2018 N=43 416, n (%)	1 July 2018 to 30 June 2019 N=44 019, n (%)	1 July 2019 to 30 June 2020 N=44 136, n (%)
1.interRAI	26 415 (63.2)	28 104 (64.7)	28 773 (65.4)	29 127 (66.0)
2.Mortality	3426 (8.2)	5157 (11.9)	2424 (5.5) *	NA
3.SOCRATES	330 (0.8)	342 (0.8)	348 (0.8)	351 (0.8)
4.Pharmaceutical Collection	13 185 (31.6)	14 007 (32.3)	14 598 (33.2)	15 012 (34.0)
5.Privately Funded Hospital Discharges	843 (2.0)	690 (1.6)	471 (1.1)	219 (0.5)
6.PRIMHD	6636 (15.9)	6093 (14.0)	4959 (11.3)	4074 (9.2)
7.Publicly Funded Hospital Discharges	24 747 (59.3)	25 683 (59.2)	25 914 (58.9)	25 317 (57.4)

The counts shown are the total numbers of alive and decreased individuals identified with dementia in each of the study periods. They are not individuals with a new diagnosis of dementia identified in each of the study periods.

Venn diagrams illustrating the intersects of the four health datasets with the greatest number of dementia cases: A (orange)=interRAI; B (green)=Publicly Funded Hospital Discharges; C (blue)=Pharmaceutical Collection; D (purple)=Programme for the Integration of Mental Health Data.

*Mortality data only available until December 2018.

NA, Mortality data not available; PRIMHD, Programme for the Integration of Mental Health Data; SOCRATES, National Needs Assessment and Service Coordination Information System.

IDI populations

Table 3 shows the IDI populations from this study and Statistics New Zealand’s national population estimates which give the best measure between census dates of the population size.¹² We found that the IDI 60+ and 80+ populations are higher than the Statistics New Zealand population estimates; whereas the IDI total (all ages) populations are lower than the Statistics New Zealand population estimates. The Statistics New Zealand population estimates are based on estimated natural increase (births minus deaths) and estimated net migration (migrant arrivals minus migrant departures). The IDI populations are not estimates; they represent the true number of people who have at least one activity in one of five data sets (including health) within 12 months. It is possible that older adults have more contacts with health services and, therefore, are more likely to be captured by the IDI. There is also some concern about the quality of the 2018 census.¹³ There were no missing IDI data for age and sex; ethnicity data were missing in 0.2% and 0.6% of the age 60+ and age 80+ populations, respectively.

Table 3

Integrated Data Infrastructure (IDI) total populations, age 60+ populations and age 80+ populations, compared with Statistics New Zealand population estimates from 30 June 2016 to 30 June 2020

	1 July 2016 to 30 June 2017	1 July 2017 to 30 June 2018	1 July 2018 to 30 June 2019	1 July 2019 to 30 June 2020
IDI total (all ages) population	4 770 897	4 843 773	4 909 617	4 984 836
Statistics New Zealand population estimate*	4 813 600	4 900 600	4 979 200	5 090 200
Difference†	42 703 (0.9%)	56 827 (1.2%)	69 583 (1.4%)	105 364 (2.1%)
IDI age 60+ population	1 001 367	1 031 247	1 061 691	1 095 192
Statistics age 60+ population estimate*	977 600	1 005 900	1 039 600	1 082 500
Difference†	−23767 (−2.4%)	−25347 (−2.5%)	−22091(−2.1%)	−12692 (−1.2%)
IDI age 80+ population	188 580	193 113	197 775	204 720
Statistics age 80+ population estimate*	168 800	172 300	177 400	185 200
Difference†	−19780 (−11.7%)	−20813 (−12.1%)	−20375 (−11.5%)	−19520 (−10.5%)

*Source: https://infoshare.stats.govt.nz/ Population estimates for the last month of the study period.

†

Calculation of dementia prevalence

Table 4 shows the crude 1-year period dementia prevalence for each study year by age and ethnicity in 5-year age bands from age 60. Māori and Pacific Islanders have higher crude prevalence than Europeans in each of the 5-year age bands from age 60 to 95+ across the 4 years; while Asian people have lower crude prevalence than Europeans in each of the 5-year age bands from age 60 to 89 across the 4 years.

Table 4

Crude dementia 1-year period prevalence per 5-year age bands from age 60 in the four main ethnic groups

Dementia cases by age range and ethnicity, n (% of IDI population)		1 July 2016 to 30 June 2017	1 July 2017 to 30 June 2018	1 July 2018 to 30 June 2019	1 July 2019 to 30 June 2020
60–64	All ethnicities	1077 (0.4)	1116 (0.4)	1140 (0.4)	1134 (0.4)
	Māori	195 (0.7)	195 (0.6)	207 (0.6)	204 (0.6)
	Pacific Islander	66 (0.6)	63 (0.6)	66 (0.6)	69 (0.6)
	European	708 (0.4)	729 (0.4)	738 (0.4)	726 (0.4)
	Asian	75 (0.3)	81 (0.3)	87 (0.3)	93 (0.3)
65–69	All ethnicities	1977 (0.8)	2010 (0.9)	1995 (0.8)	2004 (0.8)
	Māori	294 (1.4)	330 (1.5)	339 (1.4)	360 (1.4)
	Pacific Islander	93 (1.1)	90 (1.0)	117 (1.3)	126 (1.3)
	European	1455 (0.8)	1446 (0.8)	1383 (0.8)	1359 (0.8)
	Asian	93 (0.5)	102 (0.5)	102 (0.5)	117 (0.5)
70–74	All ethnicities	3513 (1.9)	3825 (2.0)	3993 (2.0)	4062 (1.9)
	Māori	435 (3.3)	483 (3.3)	519 (3.3)	561 (3.4)
	Pacific Islander	180 (3.3)	210 (3.7)	210 (3.4)	213 (3.2)
	European	2679 (1.8)	2880 (1.8)	3006 (1.8)	2985 (1.8)
	Asian	162 (1.5)	180 (1.5)	180 (1.4)	195 (1.3)
75–79	All ethnicities	6102 (4.4)	6360 (4.5)	6465 (4.4)	6489 (4.3)
	Māori	609 (6.7)	654 (7.0)	726 (7.5)	708 (7.2)
	Pacific Islander	273 (7.4)	291 (7.7)	285 (7.2)	306 (7.5)
	European	4863 (4.2)	5037 (4.3)	5055 (4.2)	5088 (4.1)
	Asian	273 (3.6)	291 (3.7)	300 (3.7)	282 (3.3)
80–84	All ethnicities	8193 (9.0)	8484 (9.0)	8715 (8.9)	8922 (8.7)
	Māori	594 (12.0)	684 (13.1)	720 (13.2)	771 (13.0)
	Pacific Islander	273 (13.1)	300 (13.8)	330 (14.3)	357 (14.5)
	European	6846 (8.8)	7005 (8.8)	7134 (8.7)	7206 (8.3)
	Asian	357 (7.8)	378 (7.4)	420 (7.6)	465 (7.9)
85–89	All ethnicities	9648 (15.9)	9786 (16.0)	9741 (15.8)	9501 (15.3)
	Māori	393 (17.6)	456 (18.5)	492 (19.2)	480 (18.3)
	Pacific Islander	240 (23.7)	261 (24.9)	255 (22.9)	252 (21.5)
	European	8601 (15.9)	8610 (16.0)	8475 (15.7)	8223 (15.2)
	Asian	306 (15.4)	321 (14.2)	354 (14.3)	387 (13.7)
90–94	All ethnicities	6402 (22.8)	6807 (23.6)	6885 (23.4)	6732 (22.4)
	Māori	171 (25.6)	186 (25.5)	222 (26.5)	231 (26.2)
	Pacific Islander	108 (31.3)	123 (32.5)	144 (35.8)	156 (35.6)
	European	5922 (22.9)	6237 (23.6)	6243 (23.3)	6024 (22.2)
	Asian	132 (20.9)	174 (24.2)	204 (24.7)	231 (25.0)
95+	All ethnicities	2532 (30.0)	2718 (30.4)	2763 (30.1)	2952 (30.5)
	Māori	42 (31.1)	60 (35.1)	63 (36.2)	63 (33.9)
	Pacific Islander	33 (50.0)	39 (59.1)	48 (55.2)	54 (66.7)
	European	2355 (29.9)	2505 (30.3)	2502 (29.5)	2682 (30.0)
	Asian	45 (27.3)	57 (28.4)	69 (31.5)	81 (32.9)

IDI, Integrated Data Infrastructure.

Table 5 shows that the crude 1-year period prevalence was 3.8%–4.0% in the age 60+ population and 13.7%–14.4% in the age 80+ population across the four study periods. Table 5 also shows the crude and age–sex standardised 1-year period prevalence in the four main ethnic groups for the age 60+ and age 80+ populations. After age–sex standardisation to the total IDI study population to account for the interethnic differences in age and sex profiles, dementia prevalence in Māori is 34%–46% higher in the age 60+population and 16%–29% higher in the age 80+ population compared with Europeans; while dementia prevalence in Pacific Islanders is 58%–70% higher in the age 60+ population and 49%–63% in the 80+ population compared with Europeans.

Table 5

Crude and age–sex standardised dementia 1-year period prevalence in the four main ethnic groups: age 60+ and 80+ populations

	Age 60+				Age 80+
	1 July 2016 to 30 June 2017	1 July 2017 to 30 June 2018	1 July 2018 to 30 June 2019	1 July 2019 to 30 June 2020	1 July 2016 to 30 June 2017	1 July 2017 to 30 June 2018	1 July 2018 to 30 June 2019	1 July 2019 to 30 June 2020
Dementia cases by ethnicity, n (% of IDI population)
All ethnicities	39 444 (3.9)	41 088 (4.0)	41 682 (3.9)	41 793 (3.8)	26 772 (14.2)	27 795 (14.4)	28 104 (14.2)	28 107 (13.7)
Māori	2733 (3.4)	3048 (3.6)	3288 (3.6)	3378 (3.5)	1200 (15.0)	1386 (16.2)	1497 (16.6)	1545 (16.1)
Pacific Islander	1266 (4.0)	1377 (4.2)	1455 (4.2)	1533 (4.2)	654 (18.7)	723 (19.7)	777 (19.9)	819 (19.7)
European	33 429 (4.2)	34 449 (4.2)	34 536 (4.2)	34 293 (4.0)	23 724 (14.3)	24 357 (14.4)	24 354 (14.2)	24 135 (13.7)
Asian	1443 (2.2)	1584 (2.2)	1716 (2.2)	1851 (2.2)	840 (11.4)	930 (11.2)	1047 (11.6)	1164 (11.8)
Dementia cases by sex, n (% of IDI population)
Female	23 697 (4.5)	24 450 (4.5)	24 717 (4.4)	24 591 (4.3)	17 268 (15.6)	17 757 (15.7)	17 826 (15.5)	17 676 (14.9)
Male	15 747 (3.3)	16 656 (3.4)	16 980 (3.4)	17 205 (3.3)	9507 (12.3)	10 038 (12.5)	10 278 (12.5)	10 431 (12.1)
Age-sex standardised dementia prevalence by ethnicity (%)
Māori	5.1	5.4	5.5	5.4	16.4	17.5	18.0	17.5
Pacific Islander	6.0	6.3	6.2	6.3	21.0	22.3	22.0	22.2
European	3.8	3.9	3.8	3.7	14.1	14.2	14.0	13.6
Asian	3.4	3.5	3.5	3.4	13.1	13.1	13.4	13.5

IDI, Integrated Data Infrastructure.

Discussion

This is the first study to date that estimates the prevalence of dementia in New Zealand using linked administrative data from seven health datasets. Our study found an estimated crude 1-year period dementia prevalence of 3.8%–4.0% in the age 60+ population and 13.7%–14.4% in the age 80+ population, respectively. The age 60+ figures are nearly double that of the estimated prevalence of 2.0% in a previous IDI study that used four of the seven health data sets included in our study.⁹ We only ascertained cases of identified and coded dementia but a substantial number of people with dementia living in the community will remain unidentified and/or uncoded for dementia.¹⁴ A previous meta-analysis found the pooled rate of undetected dementia in the community, including residential care settings, of high-income countries was 61.7% (95% CI 55.0% to 68.0%).¹⁵ In addition, a diagnosis of dementia may not get entered into clinical records by clinicians and, therefore, is undercoded in the relevant health datasets. Therefore, the true dementia prevalence in New Zealand is likely to be higher than that found in this study.

An important secondary finding of our study is the marked difference in dementia prevalence across New Zealand’s four main ethnic groups. Our age–sex standardised dementia prevalence in 2019–2020 was 5.4% for Māori, 6.3% for Pacific Islander, 3.7% for European and 3.4% for Asian in the age 60+ population, and 17.5% for Māori, 22.2% for Pacific Islander, 13.6% for European and 13.5% for Asian in the age 80+ population. This suggests dementia prevalence in Māori is over 34% higher in the age 60+ population and over 16% higher in the age 80+ population compared with Europeans; and for Pacific Islanders, it is over 58% higher in the age 60+ population and over 49% in the 80+ population compared with Europeans.

Ethnic differences in dementia prevalence have been observed in previous international studies. For example, a systematic review of 114 US studies found age 65+ dementia prevalence ranged from 6.3% in Japanese Americans, 12.9% in Caribbean Hispanic Americans, 12.2% in Guamanian Chamorro and 7.2%–20.9% in African Americans.¹⁶ Another meta-analysis of US studies also found African Americans had a higher dementia prevalence than Caucasians,¹⁷ which is consistent with a UK systematic review where older African-Caribbean people had a higher prevalence of dementia than the White British population.¹⁸ We found Māori, the indigenous people of New Zealand, have a higher dementia prevalence than the New Zealand European population, and this finding is consistent with a previous systematic review, which found indigenous populations from Canada, Australia, USA, Guam and Brazil had higher dementia prevalence than non-indigenous populations.¹⁹ International literature suggests higher rates of dementia in indigenous populations are associated with lower education level and poorer health conditions.²⁰

Our finding that Māori and Pacific Islanders have higher rates of dementia than Europeans and Asians aligns very well with a recently published study, which found Māori and Pacific Islanders have higher burden of dementia risk factors (such as lower education, hypertension, obesity and smoking) compared with European and Asian populations living in New Zealand.²¹ The estimated population attributable fraction (ie, the potential reduction in dementia prevalence if a particular risk factor was eliminated) was highest in Māori (51.4%) and Pacific Islanders (50.8%), compared with Europeans (47.6%) and Asians (40.8%). The lower dementia risk and prevalence in Asian ethnic groups warrants further investigation and possible separation into Indian and Chinese subpopulations as they are likely to have different risk factor profiles.²²

Implications for future research

Our case identification methods mean we are likely to be detecting only those dementia cases that were assessed in secondary/tertiary inpatient settings (and were also coded in clinical records), received an interRAI assessment and/or received an antidementia medication. In New Zealand, dementia is not coded in secondary/tertiary outpatient settings, only in inpatient settings. Individuals early on in their dementia/cognitive impairment care pathway and known only to primary care will not be identified from our case identification methods, due to a lack of access to primary care data in New Zealand. Likewise, many people living with dementia may not be identified by our case identification methods if they have never been diagnosed and are cared for by family at home, so have not had an interRAI assessment to access publicly funded home support services or aged residential care. In order to make an accurate estimation of all people with dementia living in the community, there needs to be a New Zealand community-based dementia prevalence study. This will be critical to (1) ascertain the true prevalence of dementia, (2) compare the characteristics of individuals with dementia accessing health services with those who are not and (3) test the diagnostic accuracy of using New Zealand administrative data to estimate dementia prevalence and incidence in the future.

Implications for policy

Our results (and international studies of ethnic differences in dementia prevalence) challenge the traditional methods of estimating national dementia prevalence for policy, and reinforces the notion that a ‘one size fits all’ approach is not appropriate.²³

Dementia prevalence data are used in the estimation of economic impacts, including the costs of formal and informal care for people living with dementia, and to inform service planning. Māori, Pacific Islander and Asian families are generally inclusive, have a strong obligation to care for their elders at home and are reluctant to admit their loved ones to aged residential care.^24–28 Māori and Pacific Islanders present with dementia at a younger age than New Zealand Europeans²⁹ and may live at home cared for by their families for many years.^{30 31} Given our findings of higher rates of dementia in Māori and Pacific Islanders, dedicated and culturally appropriate resources allocated to meet the formal and informal care needs of Māori and Pacific Islanders and families living with dementia are essential.

Strengths and limitations

This is the first study to report New Zealand dementia prevalence figures that approximate to what we would expect based on the previous estimations^{1 2} and the population attributable fraction estimations in a recent New Zealand study.²¹ This study is not a replacement for a fully powered community-based dementia prevalence study but a good proxy measure that we might be able to use if it is shown to be valid by future research.

There are several limitations relating to the use of routinely collected health data that need to be acknowledged. First, we have already mentioned the issues of underdiagnosis, undercoding of dementia and the lack of capture of dementia cases in primary care, which are likely to have contributed to an underestimate of dementia prevalence. However, the estimated dementia prevalence across the four study years is relatively stable, suggesting these seven health data sets are reliable for dementia case identification once a dementia diagnosis is recorded on them. In addition, the limitations of the data in this study applied equally across all ethnic groups; therefore, the differential prevalence between ethnic groups is likely to be a true difference, even if the total number of dementia cases was underestimated. Since a fully powered community-based dementia prevalence study is very resource intensive, there is a role for using routinely collected health data to monitor the prevalence and incidence of dementia over time. However, the methods would have to be validated against real-world epidemiological data to prove its accuracy. Second, we did not report dementia subtypes in this study because subtyping of dementia is likely to be inaccurate as most people have mixed pathologies³² and we are more concerned with the overall prevalence of all-cause dementia. Third, stigma around dementia is often an issue for non-European populations in New Zealand, which could lead to inequitable access of services and, therefore, proportionally greater underdiagnosis of dementia in some ethnic groups.^{25 28 33} There is also evidence that Māori, Pacific Islanders and Asians have lower rates of accessing dementia community care and aged residential care.² Therefore, they are less likely to be registered in the administrative health data sets included in this study. However, both Māori and Pacific Islanders have higher rates of diabetes and hypertension than European²¹; dementia diagnosis might be more likely to be considered alongside their vascular risk factors in these ethnic groups. It has also been shown that the optimal cut-off of a cognitive screening tool was lower for Māori than non-Māori,³⁴ which potentially could result in cultural bias in assessment and misdiagnosis of dementia. Fourth, since Māori and Pacific Islanders present with dementia at a younger age than New Zealand Europeans,²⁹ a more in-depth examination of the interethnic differences in dementia prevalence in the under 60 population is likely needed.

Conclusion

This study provides valuable insights into dementia prevalence in New Zealand. It provides the strongest evidence so far that dementia prevalence is higher in Māori and Pacific Islanders, which is likely to be a result of the higher prevalence of dementia risk factors in these populations. As the study relied on administrative data, a carefully designed nationwide New Zealand dementia prevalence study is needed to validate the findings of this study (and, thus, provide evidence that using routinely collected health data is an effective method for future surveillance of dementia prevalence) but to also provide further evidence regarding the extent and impact of dementia on families and society.

Data availability statement

Data may be obtained from a third party and are not publicly available. Access to the data used in this study can be applied through Statistics New Zealand. https://www.stats.govt.nz/integrated-data/apply-to-use-microdata-for-research/.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Ethics approval was obtained from the Auckland Health Research Ethics Committee (Reference: AH2810).

Footnote

Contributors GC and CR-R conceptualised the research. GC, CR-R, ET, EM, AC, BR and SC contributed to the study methodology. ET performed data collection, data linkage and statistical analysis. GC and CR-R supervised ET. GC drafted the manuscript. CR-R, ET, EM, AC, BR and SC contributed to the writing and critiqued the manuscript. GC accepts full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish.

Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Competing interests None declared.

Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

References

¹ Prince M, Wimo A, Guerchet M. World Alzheimer report 2015: the global impact of dementia, 2015. Available: https://www.alzint.org/u/WorldAlzheimerReport2015.pdf [Accessed 09 Feb 2022 ].

² Ma’u E, Cullum S, Yates S. Dementia economic impact report 2020, 2021. Available: https://cdn.alzheimers.org.nz/wp-content/uploads/2021/09/Dementia-Economic-Impact-Report-2020.pdf [Accessed 09 Feb 2022 ].

³ Campbell AJ, McCosh LM, Reinken J, et al. Dementia in old age and the need for services. Age Ageing 1983; 12: 11–16. doi:10.1093/ageing/12.1.11 http://www.ncbi.nlm.nih.gov/pubmed/6846087

⁴ Henderson AS, Jorm AF, Mackinnon A, et al. A survey of dementia in the Canberra population: experience with ICD-10 and DSM-III-R criteria. Psychol Med 1994; 24: 473–82. doi:10.1017/S0033291700027446 http://www.ncbi.nlm.nih.gov/pubmed/8084942

⁵ Kay DW, Henderson AS, Scott R, et al. Dementia and depression among the elderly living in the Hobart community: the effect of the diagnostic criteria on the prevalence rates. Psychol Med 1985; 15: 771–88. doi:10.1017/S0033291700005006 http://www.ncbi.nlm.nih.gov/pubmed/4080881

⁶ Smith K, Flicker L, Lautenschlager NT, et al. High prevalence of dementia and cognitive impairment in Indigenous Australians. Neurology 2008; 71: 1470–3. doi:10.1212/01.wnl.0000320508.11013.4f http://www.ncbi.nlm.nih.gov/pubmed/18799785

⁷ Statistics New Zealand. 2018 census, 2020. Available: https://www.stats.govt.nz/information-releases/2018-census-nz-stat-tables [Accessed 09 Feb 2022 ].

⁸ Statistics New Zealand. Integrated data infrastructure, 2020. Available: https://www.stats.govt.nz/integrated-data/integrated-data-infrastructure [Accessed 09 Feb 2022 ].

⁹ Walesby KE, Exeter DJ, Gibb S, et al. Prevalence and geographical variation of dementia in New Zealand from 2012 to 2015: brief report utilising routinely collected data within the integrated data infrastructure. Australas J Ageing 2020; 39: 297–304. doi:10.1111/ajag.12790 http://www.ncbi.nlm.nih.gov/pubmed/32394527

¹⁰ interRAI New Zealand. Annual report 2018/19, 2019. Available: https://www.interrai.co.nz/assets/Documents/9169ec695a/201819_interRAI-Annual-Report-WEB.pdf [Accessed 09 Feb 2022 ].

¹¹ Schoenbach V. Standardization of rates and ratios, 1999. Available: http://www.epidemiolog.net/evolving/Standardization.pdf [Accessed 09 Feb 2022 ].

¹² Statistics New Zealand. Population, 2020. Available: https://www.stats.govt.nz/topics/population [Accessed 09 Feb 2022 ].

¹³ Statistics New Zealand. Overview of data quality ratings, interim coverage and response rates, and data sources for 2018 census. Available: https://www.stats.govt.nz/reports/overview-of-data-quality-ratings-interim-coverage-and-response-rates-and-data-sources-for-2018-census [Accessed 20 May 2022 ].

¹⁴ Zhu Y, Chen Y, Crimmins EM, et al. Sex, race, and age differences in prevalence of dementia in Medicare claims and survey data. J Gerontol B Psychol Sci Soc Sci 2021; 76: 596–606. doi:10.1093/geronb/gbaa083 http://www.ncbi.nlm.nih.gov/pubmed/32588052

¹⁵ Lang L, Clifford A, Wei L, et al. Prevalence and determinants of undetected dementia in the community: a systematic literature review and a meta-analysis. BMJ Open 2017; 7: e011146. doi:10.1136/bmjopen-2016-011146 http://www.ncbi.nlm.nih.gov/pubmed/28159845

¹⁶ Mehta KM, Yeo GW. Systematic review of dementia prevalence and incidence in United States race/ethnic populations. Alzheimers Dement 2017; 13: 72–83. doi:10.1016/j.jalz.2016.06.2360 http://www.ncbi.nlm.nih.gov/pubmed/27599209

¹⁷ Steenland K, Goldstein FC, Levey A, et al. A meta-analysis of Alzheimer's disease incidence and prevalence comparing African-Americans and Caucasians. J Alzheimers Dis 2016; 50: 71–6. doi:10.3233/JAD-150778 http://www.ncbi.nlm.nih.gov/pubmed/26639973

¹⁸ Adelman S, Blanchard M, Livingston G. A systematic review of the prevalence and covariates of dementia or relative cognitive impairment in the older African-Caribbean population in Britain. Int J Geriatr Psychiatry 2009; 24: 657–65. doi:10.1002/gps.2186 http://www.ncbi.nlm.nih.gov/pubmed/19235788

¹⁹ Warren LA, Shi Q, Young K, et al. Prevalence and incidence of dementia among Indigenous populations: a systematic review. Int Psychogeriatr 2015; 27: 1959–70. doi:10.1017/S1041610215000861 http://www.ncbi.nlm.nih.gov/pubmed/26088474

²⁰ de Souza-Talarico JN, de Carvalho AP, Brucki SMD, et al. Dementia and cognitive impairment prevalence and associated factors in Indigenous populations: a systematic review. Alzheimer Dis Assoc Disord 2016; 30: 281–7. doi:10.1097/WAD.0000000000000140 http://www.ncbi.nlm.nih.gov/pubmed/26840546

²¹ Ma'u E, Cullum S, Cheung G, et al. Differences in the potential for dementia prevention between major ethnic groups within one country: a cross sectional analysis of population attributable fraction of potentially modifiable risk factors in New Zealand. Lancet Reg Health West Pac 2021; 13: 100191. doi:10.1016/j.lanwpc.2021.100191 http://www.ncbi.nlm.nih.gov/pubmed/34527984

²² Mukadam N, Sommerlad A, Huntley J, et al. Population attributable fractions for risk factors for dementia in low-income and middle-income countries: an analysis using cross-sectional survey data. Lancet Glob Health 2019; 7: e596–603. doi:10.1016/S2214-109X(19)30074-9 http://www.ncbi.nlm.nih.gov/pubmed/31000129

²³ Brayne C, Miller B. Dementia and aging populations-A global priority for contextualized research and health policy. PLoS Med 2017; 14: e1002275. doi:10.1371/journal.pmed.1002275 http://www.ncbi.nlm.nih.gov/pubmed/28350794

²⁴ Cheung G, Appleton K, Boyd M, et al. Perspectives of dementia from Asian communities living in New Zealand: a focus group of Asian health care professionals. Int J Geriatr Psychiatry 2019; 34: 1758–64. doi:10.1002/gps.5189 http://www.ncbi.nlm.nih.gov/pubmed/31389087

²⁵ Cheung G, Su AY, Wu K, et al. The understanding and experiences of living with dementia in Chinese new Zealanders. Int J Environ Res Public Health 2022; 19: 1280. doi:10.3390/ijerph19031280 http://www.ncbi.nlm.nih.gov/pubmed/35162305

²⁶ Dudley M, Menzies O, Elder H. Mate wareware: understanding ‘dementia’ from a Māori perspective. NZ Med J 2019; 132: 66–74.

²⁷ Fakahau T, Faeamani G, Maka M. Pacific people and dementia. Auckland: Tongan Advisory Council, 2019.

²⁸ Krishnamurthi RV, Dahiya ES, Bala R, et al. Lived experience of dementia in the New Zealand Indian community: a qualitative study with family care givers and people living with dementia. Int J Environ Res Public Health 2022; 19: 1432. doi:10.3390/ijerph19031432 http://www.ncbi.nlm.nih.gov/pubmed/35162453

²⁹ Cullum S, Mullin K, Zeng I, et al. Do community-dwelling Māori and Pacific peoples present with dementia at a younger age and at a later stage compared with NZ Europeans? Int J Geriatr Psychiatry 2018; 33: 1098–104. doi:10.1002/gps.4898 http://www.ncbi.nlm.nih.gov/pubmed/29766582

³⁰ Cullum S, Varghese C, Coomarasamy C, et al. Predictors of mortality in Māori, Pacific Island, and European patients diagnosed with dementia at a new Zealand memory service. Int J Geriatr Psychiatry 2020; 35: 516–24. doi:10.1002/gps.5266 http://www.ncbi.nlm.nih.gov/pubmed/31957058

³¹ Cullum S, Varghese C, Yates S, et al. Predictors of aged residential care placement in patients newly diagnosed with dementia at a new Zealand memory service. Journal of Long Term Care 2021; 39: 24–32. doi:10.31389/jltc.46

³² Savva GM, Wharton SB, Ince PG, et al. Age, neuropathology, and dementia. N Engl J Med 2009; 360: 2302–9. doi:10.1056/NEJMoa0806142 http://www.ncbi.nlm.nih.gov/pubmed/19474427

³³ Martinez-Ruiz A, Huang Y, Gee S, et al. Individual risk factors for possible undetected dementia amongst community-dwelling older people in New Zealand. Dementia 2020; 19: 750–65. doi:10.1177/1471301218786277 http://www.ncbi.nlm.nih.gov/pubmed/29989431

³⁴ Zawaly K, Moyes SA, Wood PC, et al. Diagnostic accuracy of a global cognitive screen for Māori and non-Māori octogenarians. Alzheimers Dement 2019; 5: 542–52. doi:10.1016/j.trci.2019.08.006 http://www.ncbi.nlm.nih.gov/pubmed/31650011