Abstract：
OBJECTIVE: Assisted living (AL) is a significant and growing congregate care option for vulnerable older adults designed to reduce the use of nursing homes (NHs). However, work on excess mortality in congregate care during the COVID-19 pandemic has primarily focused on NHs with only a few US studies examining AL. The objective of this study was to assess excess mortality among AL and NH residents with and without dementia or significant cognitive impairment in Alberta, Canada, during the first 2 years of the COVID-19 pandemic, relative to the 3 years before.
METHODS: Population-based, retrospective cohort study.
METHODS: Residents who lived in an AL or NH facility operated or contracted by the Provincial health care system to provide publicly funded care in Alberta between January 1, 2017, and December 31, 2021.
METHODS: We used administrative health care data, including Resident Assessment Instrument - Home Care (RAI-HC, AL) and Minimum Data Set 2.0 (RAI-MDS 2.0, NHs) records, linked with data on residents\' vital statistics, COVID-19 testing, emergency room registrations, and hospital stays. The outcome was excess deaths during COVID-19 (ie, the number of deaths beyond that expected based on pre-pandemic data), estimated, using overdispersed Poisson generalized linear models.
RESULTS: Overall, the risk of excess mortality [adjusted incidence rate ratio (95% confidence interval)] was higher in ALs than in NHs [1.20 (1.14-1.26) vs 1.10 (1.07-1.13)]. Weekly peaks in excess deaths coincided with COVID-19 pandemic waves and were higher among those with diagnosed dementia or significant cognitive impairment in both, AL and NHs.
CONCLUSIONS: Finding excess mortality within both AL and NH facilities should lead to greater focus on infection prevention and control measures across all forms of congregate housing for vulnerable older adults. The specific needs of residents with dementia in particular will have to be addressed.
METHODS: Population-based, retrospective cohort study.
METHODS: Residents who lived in an AL or NH facility operated or contracted by the Provincial health care system to provide publicly funded care in Alberta between January 1, 2017, and December 31, 2021.
METHODS: We used administrative health care data, including Resident Assessment Instrument - Home Care (RAI-HC, AL) and Minimum Data Set 2.0 (RAI-MDS 2.0, NHs) records, linked with data on residents\' vital statistics, COVID-19 testing, emergency room registrations, and hospital stays. The outcome was excess deaths during COVID-19 (ie, the number of deaths beyond that expected based on pre-pandemic data), estimated, using overdispersed Poisson generalized linear models.
RESULTS: Overall, the risk of excess mortality [adjusted incidence rate ratio (95% confidence interval)] was higher in ALs than in NHs [1.20 (1.14-1.26) vs 1.10 (1.07-1.13)]. Weekly peaks in excess deaths coincided with COVID-19 pandemic waves and were higher among those with diagnosed dementia or significant cognitive impairment in both, AL and NHs.
CONCLUSIONS: Finding excess mortality within both AL and NH facilities should lead to greater focus on infection prevention and control measures across all forms of congregate housing for vulnerable older adults. The specific needs of residents with dementia in particular will have to be addressed.
摘要：
目标：辅助生活（AL）是一种重要且不断增长的针对弱势老年人的聚集护理选择，旨在减少养老院（NHs）的使用。然而,关于COVID-19大流行期间集合护理超额死亡率的研究主要集中在NHs，只有少数美国研究检查AL。这项研究的目的是评估艾伯塔省有或没有痴呆或明显认知障碍的AL和NH居民的超额死亡率。加拿大,在COVID-19大流行的头两年，相对于3年前。
方法：以人口为基础，回顾性队列研究。
方法：居住在AL或NH设施中的居民，由省医疗保健系统运营或签约，在2017年1月1日至2021年12月31日之间在艾伯塔省提供公共资助的护理。
方法：我们使用了行政医疗保健数据，包括居民评估工具-家庭护理(RAI-HC，AL)和最小数据集2.0(RAI-MDS2.0，NHs)记录，与居民生命统计数据相关，COVID-19检测，急诊室登记,和住院。结果是COVID-19期间的过量死亡(即，根据大流行前的数据，超出预期的死亡人数)，估计,使用过分散泊松广义线性模型。
结果：总体而言，ALs的超额死亡率风险(校正后的发生率比率[95%置信区间])高于NHs(1.20[1.14-1.26]vs1.10[1.07-1.13]).每周超额死亡高峰与COVID-19大流行波同时发生，在诊断为痴呆或显著认知障碍的人群中，死亡人数更高，AL和NHs。
结论：在AL和NH设施中发现超额死亡率应导致在所有形式的弱势老年人聚集住房中更多地关注感染预防和控制措施。必须特别解决痴呆症居民的具体需求。
方法：以人口为基础，回顾性队列研究。
方法：居住在AL或NH设施中的居民，由省医疗保健系统运营或签约，在2017年1月1日至2021年12月31日之间在艾伯塔省提供公共资助的护理。
方法：我们使用了行政医疗保健数据，包括居民评估工具-家庭护理(RAI-HC，AL)和最小数据集2.0(RAI-MDS2.0，NHs)记录，与居民生命统计数据相关，COVID-19检测，急诊室登记,和住院。结果是COVID-19期间的过量死亡(即，根据大流行前的数据，超出预期的死亡人数)，估计,使用过分散泊松广义线性模型。
结果：总体而言，ALs的超额死亡率风险(校正后的发生率比率[95%置信区间])高于NHs(1.20[1.14-1.26]vs1.10[1.07-1.13]).每周超额死亡高峰与COVID-19大流行波同时发生，在诊断为痴呆或显著认知障碍的人群中，死亡人数更高，AL和NHs。
结论：在AL和NH设施中发现超额死亡率应导致在所有形式的弱势老年人聚集住房中更多地关注感染预防和控制措施。必须特别解决痴呆症居民的具体需求。
