Practice Impact Statement

COVID-19 responses, including lockdowns, have been hypothesised to lead to social isolation for older adults. This study of aged residential care populations found a lower rate of loneliness in Māori, a higher rate of depression in New Zealand Europeans and lower rates of hospitalisation in Māori and New Zealand Europeans during the first wave of COVID-19. Further research is needed to explain our findings.

1 INTRODUCTION

Older adults living in aged residential care (ARC) are the most vulnerable population for being infected by SARS-CoV-2 and developing coronavirus disease 2019 (COVID-19) with high COVID-19–related morbidity and mortality.1 New Zealand implemented strict social gathering restrictions and border control measures soon after its first confirmed COVID-19 case on 28 February 2020.2 It has been postulated that COVID-19 restrictions and the fear generated by the pandemic could have a negative impact on the health and psychosocial well-being of older adults.3, 4

The first New Zealand COVID-19 case was confirmed on 28 February 2020. New Zealand has a 4-level COVID-19 alert system. Table 1 shows the timeline of the events and the public health responses during the first wave of COVID-19, which included a nationwide lockdown. The New Zealand Ministry of Health published COVID-19–specific guidelines for ARC, which include the use of personal protective equipment (PPE), management of residents or staff contracting infection, and prevention and management of COVID-19 outbreaks.5 At all alert levels, if an individual was COVID-19–positive or was symptomatic, admission to the ARC was delayed until they had a negative test. They were to remain in isolation for 14 days and be monitored daily, and appropriate PPE precautions should be taken. At Alert Level 3 and Alert Level 4, which lasted for nearly 2 months, ARC residents were not permitted to leave the premises or have visitors, except for those receiving palliative care.5 Resident outings and visiting, with precautions, were gradually resumed at Alert Level 2. By 22 July 2020, there were 153 infected cases and 16 COVID-19–related deaths linked to five ARC clusters across three regions of New Zealand, accounting for 10.2% of total infected cases and 72.7% of all COVID-19–related deaths in the country.5

TABLE 1. Timeline of the events and responses during the first wave of COVID-19 in New Zealand in 2020 Date Alert levels Comments 28 February 2020 First confirmed COVID-19 case 21 March 2020 Alert Level 2: Reduce The disease is contained, but the risk of community transmission remains 23 March 2020 Alert Level 3: Restrict High risk, the disease is not contained 25 March 2020 Alert Level 4: Lockdown Likely, the disease is not contained 27 April 2020 Alert Level 3: Restrict High risk, the disease is not contained 13 May 2020 Alert Level 2: Reduce The disease is contained, but the risk of community transmission remains 8 June 2020 Alert Level 1: Prepare The disease is contained in New Zealand

There have been reported ethnic disparities of COVID-19 infections in the United States with Pacific populations, among other ethnic minorities, having increased rates of COVID-19 infections compared with the rest of the US population.6 Within New Zealand, there are also significant inequities across ethnic groups, with a higher estimated infection fatality rate from COVID-19 among Māori and Pacific populations compared with New Zealand Europeans.7 Māori have higher rates than non-Māori for disability and chronic diseases, including mental health, cancer, diabetes, cardiovascular disease and asthma,8 which can increase their vulnerability to COVID-19. Older Pacific People in New Zealand are also a particularly vulnerable population with poorer reported physical and mental health, and higher rates of health conditions.9

Family relationships are particularly important in determining the health and well-being among Māori and Pacific Peoples, with the family forming the foundation of the Te Whare Tapa Whā and Fonofale holistic models of health in these two cultures, respectively.10, 11 Te Whare Tapa Whā is a Māori model that describes health and well-being as a wharenui/meeting house with four walls: taha wairua (spiritual), taha hinengaro (mental and emotion), taha tinana (physical) and taha whanau (family and social).10 These four dimensions in balance contribute to well-being.10 The Fonofale model of health is a system of well-being that uses the model of the Samoan fale or house to embrace and acknowledge importance of Pacific perspectives. The fale represents overall health, which is comprised of family or aiga (floor or foundation), culture, values and beliefs (roof), which are supported by four pillars or Pou—spiritual, physical, mental and other aspects of well-being. Finally, the fale is surrounded by three elements—environment, time and context, which shapes well-being.11 We hypothesised that strict isolation and restricted family visiting could have a negative effect on the health and psychosocial well-being of Māori and Pacific Peoples living in ARC. interRAI Long-Term Care Facilities (interRAI LTCF) is an internationally developed comprehensive geriatric assessment, which provides information on 250 demographic, clinical and psychosocial factors. In this study, we compared the interRAI LTCF assessments of Māori and Pacific Peoples during the first COVID-19 national response (21/3/2020 to 8/6/2020) with assessments of the same ethnicities during the same period in the previous year (21/3/2019 to 8/6/2019). We also performed a similar analysis with New Zealand Europeans to compare and contrast with the findings for Māori and Pacific Peoples.

2 METHODS

Ethics approval was obtained from the Auckland Health Research Ethics Committee (Reference number: AH3334).

2.1 interRAI data collection and data access

New Zealand has a nationwide mandated interRAI program for all people living in ARC where residents are assessed using interRAI LTCF at least every 6 months. interRAI LTCF assessments are completed by qualified assessors who are in-house health professionals working in ARC facilities. A national competency framework provides quality assurance for interRAI assessment. Most of the ARC facilities in New Zealand continued to use interRAI LTCF throughout the COVID-19 pandemic.

We requested from Technical Advisory Services (TAS) all interRAI LTCF assessments for Māori, Pacific Peoples and New Zealand Europeans, aged 60 years or older, conducted during the first COVID-19 national response (21/3/2020 to 8/6/2020) and the same period in the previous year (21/3/2019 to 8/6/2019) as a comparison. Ethnicity is routinely collected by interRAI LTCF. When a person has multiple ethnicities, TAS uses the prioritised method to report the person's ethnicity.12

2.2 Outcome measures

In addition to demographics (age, sex and marital status), selected interRAI items were chosen to provide key information on physical, cognitive, psychosocial and service utilisation indicators:

2.2.1 Physical health

Self-rated health was reported as ‘excellent’, ‘good’, ‘fair’ or ‘poor’. The Activities of Daily Living (ADL) Hierarchy Scale assesses the loss of ADL ability, with scores ranging from 0 to 6: 0 (independent), 1–2 (mild to moderate dependent) and 3+ (severe dependent).13 Falls in the last 30 days were measured as ‘no fall’ or ‘one or more falls’.

2.2.2 Cognition

The Cognitive Performance Scale (CPS) scores range from 0 (intact) to 6 (very severe impairment).14 CPS was categorised as 0 (intact), 1–2 (borderline or mild cognitive impairment) and 3+ (moderate-to-severe cognitive impairment).

2.2.3 Psychosocial health

Depression Rating Scale has scores ranging from 0 to 14 and was categorised as 0–2 (no-to-minimal depressive symptoms), 3–5 (moderate depressive symptoms) and 6+ (severe depressive symptoms).15 Loneliness was measured as ‘yes’ or ‘no’ to the interRAI item ‘says or indicates that he /she feels lonely’. The Aggressive Behaviour Scale is a measure of aggressive behaviour based on the occurrence of verbal abuse, physical abuse, socially disruptive behaviour and resistance to care. It has scores ranging from 0 to 12 and was categorised as 1–4 (mild aggressive behaviour) and 5+ (moderate-to-severe aggressive behaviour).16

2.2.4 Service utilisation

Hospitalisation in the last 30 days was measured as ‘yes’ or ‘no’.

2.3 Statistical analysis

The Statistical Package for the Social Sciences (SPSS), version 27, was used for statistical analysis. interRAI LTCF assessments of the three ethnicities (Māori, Pacific Peoples and New Zealand Europeans) were analysed separately. Only the last interRAI assessment record of each person during each of the two periods was used for analysis. Chi-squared (χ2) tests were used to compare any differences in discrete variables between the two time periods. Adjusted residuals were calculated to identify specific cells making the greatest contribution (applying the ±2 criteria) to the χ2 test results. The statistical significance level was set at 0.05.

3 RESULTS

There were a total of 12,368 and 13,550 interRAI LTCF assessments from 21 March to 8 June in 2020 and 2019, respectively. A total of 232 and 386 individuals had two interRAI assessments during the 2020 and 2019 time period, respectively, and their first assessment was excluded from the analysis. There were a noticeable lower number of interRAI assessments completed during the first 4 weeks of the first wave of COVID-19 than the same time period in 2019 (Figure 1). The final 2020 sample consisted of 538 Māori, 276 Pacific Peoples and 11,322 New Zealand Europeans, while there were 549 Māori, 248 Pacific Peoples and 12,367 New Zealand Europeans in the 2019 sample. Table 2 shows the demographic details of the three ethnic groups in the 2020 and 2019 samples. There was no statistical difference for age, sex and marital status between the two time periods in any of the three ethnic groups.

Number of interRAI assessments per week from 21 March to 8 June in 2019 and 2020. *This is not a weekly total but is from 6 June to 8 June

TABLE 2. Comparison between the health and psychological well-being of Maori, Pacific Peoples, New Zealand European populations in long-term care during the first wave of COVID-19 in 2020 and the same time period in 2019 Māori Pacific Peoples New Zealand Europeans

2019

n = 549

n (%)

2020

n = 538

n (%)

p-Value

2019

n = 248

n (%)

2020

n = 276

n (%)

p-Value

2019

n = 12,367

n (%)

2020

n = 11,322

n (%)

p-Value Age 60–69 106 (19.3) 107 (19.9) 0.927 47 (19.0) 45 (16.3) 0.389 622 (5.0) 551 (4.9) 0.613 70–79 212 (38.6) 198 (36.8) 89 (35.9) 118 (42.8) 2371 (19.2) 2206 (19.5) 80–89 188 (34.2) 192 (35.7) 91 (36.7) 88 (31.9) 5462 (44.2) 4922 (43.5) 90+ 43 (7.8) 41 (7.6) 21 (8.5) 25 (9.1) 3912 (31.6) 3643 (32.2) Sexa Female 331 (60.3) 313 (58.7) 0.600 136 (54.8) 155 (56.2) 0.761 8177 (66.1) 7387 (65.7) 0.489 Male 218 (39.7) 220 (41.3) 112 (45.2) 121 (43.8) 4188 (33.9) 3856 (34.3) Marital statusa Married/Civil Union/De facto 91 (16.6) 100 (18.8) 0.346 76 (30.6) 82 (29.7) 0.816 2955 (23.9) 2780 (24.7) 0.138 Other 458 (83.4) 433 (81.2) 172 (69.4) 194 (70.3) 9410 (76.1) 8463 (75.3) Self-rated health Excellent 23 (4.2) 24 (4.5) 0.960 7 (2.8) 3 (1.1) 0.574 295 (2.4) 260 (2.3) 0.285 Good 288 (52.5) 292 (54.3) 118 (47.6) 141 (51.1) 5878 (47.5) 5495 (48.5) Fair 98 (17.9) 90 (16.7) 45 (18.1) 47 (17.0) 3116 (25.2) 2804 (24.8) Poor 20 (3.6) 17 (3.2) 6 (2.4) 9 (3.3) 805 (6.5) 672 (5.9) Could not (would not) respond 120 (21.9) 115 (21.4) 72 (29.0) 76 (27.5) 2273 (18.4) 2091 (18.5) ADL hierarchy scaleb 0 (independent) 152 (27.7) 159 (29.6) 0.792 53 (21.4) 45 (16.3) 0.332 2909 (23.5) 2623 (23.2) 0.659 1–2 (mild to moderate dependent) 194 (35.3) 186 (34.6) 72 (29.0) 85 (30.8) 4300 (34.8) 3914 (34.6) 3+ (severe dependent) 203 (37.0) 193 (35.9) 123 (49.6) 146 (52.9) 5157 (41.7) 4785 (42.3) Falls in last 30 days No falls 461 (84.0) 470 (87.4) 0.111 211 (85.1) 238 (86.2) 0.707 9876 (79.9) 8935 (78.9) 0.074 ≥1 fall 88 (16.0) 68 (12.6) 37 (14.9) 38 (13.8) 2491 (20.1) 2387 (21.1) Cognitive Performance Scale (CPS) 0–1(intact) 69 (12.6) 53 (9.9) 0.365 26 (10.5) 27 (9.8) 0.568 1520 (12.3) 1367 (12.1) 0.842 1–2 (borderline or mild cognitive impairment) 241 (43.9) 244 (45.4) 105 (42.3) 106 (38.4) 5789 (46.8) 5293 (46.7) 3+ (moderate-to-severe cognitive impairment) 239 (43.5) 241 (44.8) 117 (47.2) 143 (51.8) 5058 (40.9) 4662 (41.2) Depression Rating Scale 0–2 (no-to-minimal) 437 (79.6) 428 (79.6) 0.197 205 (82.7) 218 (79.0) 0.322 9527 (77.0) 8564 (75.6) 0.028d 3–5 (moderate) 79 (14.4) 89 (16.5) 37 (14.9) 45 (16.3) 2061(16.7) 1973 (17.4) 6+ (severe) 33 (6.0) 21 (3.9) 6 (2.4) 13 (4.7) 779 (6.3) 785 (6.9) Says or indicates that he / she feels lonelyc No 506 (92.2) 514 (95.5) 0.021 236 (95.2) 261 (94.6) 0.758 11,509 (93.1) 10,555 (93.2) 0.632 Yes 43 (7.8) 24 (4.5) 12 (4.8) 15 (5.4) 855 (6.9) 765 (6.8) Aggressive Behaviour Scale 0 (nil) 289 (52.6) 272 (50.6) 0.787 134 (54.0) 140 (50.7) 0.490 8289 (67.0) 7468 (66.0) 0.218 1–4 (mild aggressive behaviour) 209 (38.1) 213 (39.6) 84 (33.9) 107 (38.8) 3200 (25.9) 3031 (26.8) 5+ (moderate-to-severe aggressive behaviour) 51 (9.3) 53 (9.9) 30 (12.1) 29 (10.5) 878 (7.1) 823 (7.3) Hospitalisation in last 30 days No 489 (89.1) 498 (92.6) 0.046 225 (90.7) 248 (89.9) 0.737 11,204 (90.6) 10,407 (91.9) <0.001 Yes 60 (10.9) 40 (7.4) 23 (9.3) 28 (10.1) 1163 (9.4) 915 (8.1) Note Missing data—a2019: n = 2 (<0.01%), 2020: n = 84 (0.7%); b2019: n = 1 (<0.01%); c2019: n = 3 (<0.01%), 2020: n = 2 (<0.01%); dAdjusted residuals: 0–2 (no-to-minimal) = 2.5; 3–5 (moderate) = 1.6; and 6+ (severe) = 2.0.

The outcome measures between the two time periods in the three ethnic groups are shown in Table 2. Fewer Māori residents reported feeling lonely in 2020 (2020: n = 24, 4.5%; 2019: n = 43, 7.8%, p = 0.021). New Zealand European ARC residents reported more severe depressive symptoms in 2020 (no-to-minimal depressive symptoms—2020: n = 8564, 75.6%; 2019: n = 9527, 77.0%, adjusted residual = 2.5; and severe depressive symptoms—2020: n = 785, 6.9%; 2019: n = 779, 6.3%, adjusted residual = 2.0, p = 0.028). Māori and New Zealand European ARC residents had lower rates of hospitalisation (in last 30 days) during the first wave of COVID-19 than in the comparative period (Māori—2020: n = 40, 7.4%; 2019: n = 60, 10.9%, p = 0.046; and Europeans—2020: n = 915, 8.1%; 2019: n = 1163, 9.4%, p < 0.001). There was no statistical difference for other outcome measures between the first wave of COVID-19 and the comparative period in the three ethnic groups.

4 DISCUSSION

We did not find any immediate negative impact of the first wave of COVID-19, which included a nationwide lockdown, on the health and psychosocial well-being among older Māori and Pacific Peoples living in ARC. On the contrary, we found lower rates of loneliness and hospitalisation among Māori residents during the first wave of COVID-19. New Zealand European ARC residents also had a lower rate of hospitalisation during the first wave of COVID-19, but they reported more severe depressive symptoms.

International studies suggest ARC residents were often emotionally distressed, and felt disconnected and socially isolated during COVID-19 lockdown.17, 18 For example, residents living in an Israeli continuing care retirement community reported their community was transformed from a place of pleasure and activity to what felt like a prison where they were cut off from the outside world during the lockdown period.18 Similarly, ARC residents in Malaysia described feelings of entrapment and disconnection from the outside world during COVID-19.17 We found a higher rate of severe depressive symptoms but no significant change to the level of loneliness among New Zealand European residents during the first wave of COVID-19 in New Zealand. A New Zealand survey conducted with adults aged 18–90 years during Alert Level 4 lockdown in April 2020 found higher levels of moderate-to-severe psychological distress than prepandemic national health survey in all age groups.19 Another New Zealand study found people aged between 65 and 74 years had a significantly higher level of depressive symptoms than prepandemic normative data; however, the level of depressive symptoms in the 75+ age group was not statistically significantly different from prepandemic normative data.20

The psychosocial health (depression, loneliness and behaviour) of Māori and Pacific Peoples was largely unaffected in our study. Our findings suggest the COVID-19 pandemic and the associated national response may affect different ethnic groups differently. A recently published New Zealand study exploring staff experiences of living-in with people with dementia in an ARC facility during COVID-19 lockdown found an increase in the quantity and quality of interactions between the staff and residents.21 They also observed a reduction in falls and some positive changes in residents' behaviour including improved mood. A previous review concluded that residents have a better mood and delayed functional dependence when staff treat and interact with them empathically and humanely.22 During the lockdown, there were likely less interruptions and more consistent routines in care provision, which has been shown to be beneficial for people with dementia.23 A study during COVID-19 lockdown in New Zealand has also found ‘silver linings’ with improved social and community cohesiveness, with reference to the strong government public health messaging as a potential positive contributor.24 However, the longer-term impact of the COVID-19 response on the health and psychosocial well-being of ARC residents will need to be investigated. Monitoring both the immediate and delayed psychological effects of lockdowns and personal quarantine has been identified as an important area to help provide evidence for mitigation strategies and to improve the mental health of the public health in terms of pandemic preparedness.25

There are a few limitations to be acknowledged. First, ARC residents included in the two time periods were two separate cross-sectional samples. Although the median length of stay in ARC is estimated to be 10 months (New Zealand's Health Quality & Safety Commission, 2021, personal communication), 5266 residents (Māori: 281; Pacific Peoples: 109; and Europeans: 4939) appeared in both time periods. We decided not to use a within-subject study design on these 5266 residents because 47.4% of them had a diagnosis of dementia recorded on their interRAI assessment. Since dementia is a neurodegenerative disorder, we would expect a decline in some of study outcomes used to measure activities of daily living, cognition, behaviour and falls between the two time periods. Second, the sample sizes of Māori and Pacific Peoples were much smaller than New Zealand Europeans; we might not have the statistical power to detect any real underlying differences between the pre- and post–COVID-19 time periods in these two ethnic groups. Third, this study was specifically designed to examine the impacts of COVID-19 on Māori and Pacific ARC residents. We did not include Asians, the third largest ethnic group that had interRAI assessments in New Zealand in 2019–20: Europeans, 87.8%; Māori, 5.5%; Asians, 3.7%; and Pacific Peoples, 2.6%. Future New Zealand research on the impact of COVID-19 pandemic on the ARC populations will need to include Asians in their analysis. We are currently planning to extend this study by examining the 2021 data and studying the longer-term effects of COVID-19 pandemic. Fourth, there were a total of 12,586 Māori, Pacific Peoples and New Zealand Europeans included in our sample during the first wave of COVID-19. The New Zealand Aged Care Association estimated there were 34,646 residents living in New Zealand ARC facilities on 31 March 2020.26 It is important to highlight that ARC residents are assessed using interRAI LTCF at least every 6 months. Therefore, not every New Zealand ARC resident would have had an assessment during our two study time periods, and they were excluded from this study. Fifth, the total number of interRAI assessments completed during the first wave of COVID-19 was reduced by 1264 from the same time period in 2019, particularly in the first 4 weeks. Higher workload and/or work-related stress adjusting to the pandemic and the COVID-19-specific guidelines for ARC could have impacted on the number of interRAI LTCF assessments completed, as well as the process and accuracy of the assessments. For example, rapport building is likely to be more difficult with interRAI assessors wearing PPE. It is beyond the scope of this study to examine these potential impacts from COVID-19. We are aware some local qualitative research is currently underway to study the impacts of COVID-19 on ARC staff, residents and families, as well as the role of digital technology in promoting social connectedness.

Our study is one of the first studies using a globally validated assessment tool and nationwide data to examine the impact of the first wave of COVID-19 on ARC residents. New Zealand experienced one of the lowest cumulative COVID-19 case counts, incidence and mortality in its first wave of COVID-19 due to the timely implementation and rapid escalation of national COVID-19 suppression strategies.27 These strategies resulted in an unprecedented reduction of influenza and other respiratory viral infections in 2020,28 which could have led to better health status and explained our finding that there was a reduction in hospitalisation in Māori and New Zealand Europeans during the first wave of COVID-19.

5 CONCLUSIONS

COVID-19 lockdowns have been hypothesised to lead to social isolation for older adults. We found a lower rate of loneliness in Māori but a higher rate of severe depressive symptoms in New Zealand European ARC populations during the first wave of COVID-19. Further research, including qualitative studies with ARC staff, residents and families, is needed to explain our findings and the ethnic group differences.

ACKNOWLEDGEMENTS

The authors would like to thank Nikolai Minko, Health Quality Intelligence Principal Data Scientist, New Zealand's Health Quality & Safety Commission, for analysing some of the aged residential care data presented in this manuscript.

CONFLICTS OF INTEREST

Professor Debra Waters is the Editor-in-Chief of the Australasian Journal on Ageing. The other authors declare no conflicts of interest.

The data that support the findings of this study are available on request from interRAI New Zealand, Technical Advisory Services Limited: https://www.interrai.co.nz/data-research-and-reporting/requesting-interrai-data/

REFERENCES

1Thompson D, Barbu M, Beiu C, et al. The impact of COVID-19 pandemic on long-term care facilities worldwide: an overview on international issues. Biomed Res Int. 2020; 2020: 1- 7: doi:10.1155/2020/8870249 2 New Zealand Doctor. Timeline – Coronavirus – COVID-19. https://www.nzdoctor.co.nz/timeline-coronavirus. Accessed August 5, 2021. 3 World Health Organization. Mental health and psychosocial considerations during the COVID-19 outbreak. https://www.who.int/docs/default-source/coronaviruse/mental-health-considerations.pdf. Accessed August 5, 2021. 4Miller EA. Protecting and improving the lives of older adults in the COVID-19 era. J Aging Soc Policy. 2020; 32(4–5): 297- 309. doi:10.1080/08959420.2020.1780104 5Ma'u E, Robinson J, Cheung G, Miller N, Cullum S. COVID-19 and long-term care in New Zealand: impact, measures and lessons learnt. https://ltccovid.org/wp-content/uploads/2020/08/The