key: cord-0268445-1aemhuyk authors: Barker, R. O.; Hanratty, B.; Kingston, A.; Ramsay, S.; Matthews, F. E. title: Changes in health and functioning of care home residents over two decades: what can we learn from population based studies? date: 2020-08-06 journal: nan DOI: 10.1101/2020.08.05.20168740 sha: 3c81b3f2a5c0a2ae2dcd486d568a2e96a17114ac doc_id: 268445 cord_uid: 1aemhuyk Background Care home residents have complex care and support needs, as demonstrated by their vulnerability during the COVID-19 pandemic. There is a perception that the needs of residents have increased, but evidence is limited. We investigated changes in health and functioning of care home residents over two decades in England and Wales. Methods We conducted a repeated cross-sectional analysis over a 24-year period (1992-2016), using data from three longitudinal studies, the Cognitive Function and Ageing Studies (CFAS) I and II and English Longitudinal Study of Ageing (ELSA). To adjust for ageing of respondents over time results are presented for the 75-84 age group. Results Analysis of 2,280 observations from 1,745 care home residents demonstrated increases in severe disability (difficulty in at least two from washing, dressing and toileting). The prevalence of severe disability increased from 63% in 1992 to 87% in 2014 (subsequent fall in 2016 although wide confidence intervals). The prevalence of complex multimorbidity (problems in at least three out of six body systems) increased within studies over time, from 33% to 54% in CFAS I/II between 1992 and 2012, and 26% to 54% in ELSA between 2006 and 2016. Conclusion Over two decades, there has been an increase in disability and the complexity of health problems amongst care home residents in England and Wales. A rise in support needs for residents places increasing demands on care home staff and health professionals. This is an important concern for policymakers when considering the impact of COVID-19 infection in care homes. Care home residents are known to have high needs for health and social care support [1] . There is a perception amongst care home and NHS staff that residents' needs have grown in number and complexity in recent years [1, 2] . However, evidence from epidemiological studies about changes in the resident population is limited. People in care homes take part in some research, but many studies exclude residents, either at the start, or at the point when they move into a care home [3] . Information about UK care home residents is available from a small number of population-based cohort studies, including the Cognitive Function and Ageing Studies (MRC CFAS -here called CFAS I and CFAS II) [4] , and the English Longitudinal Study of Ageing (ELSA) [5] . Analysis of these and other data point to a possible rise in care needs. A study based on ELSA (2002 to 2015) described an increase in the number of health conditions and functional deficits amongst older people who were about to move into a care home [6] . The proportion of older people living with dementia in care settings increased from 56% in CFAS I (1991) (1992) (1993) (1994) to 70% in CFAS II (2004) [7] . Despite nearly all residents in long-term care having functional impairment at both time points, more were chair or bed-bound in CFAS II (34%) compared to CFAS I (22%) [8] . An in-depth understanding of trends in the health and functioning of care home residents is needed as care providers and policy makers strive to meet the needs of this complex population. The COVID-19 pandemic has highlighted the consequences of this gap in our understanding of the health of care home residents and the level of support required. In the absence of a minimum dataset on UK care home residents, this study set out to synthesise data from existing cohort studies of ageing in England in Wales. The aim is to investigate how the health and functioning of care home residents in epidemiological studies in England and Wales has changed over time. This will be done by addressing the question of how the proportion of care home residents experiencing complex multimorbidity, severe disability and poor self-reported health has changed over time. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 6, 2020. . https://doi.org/10.1101/2020.08.05.20168740 doi: medRxiv preprint Data were obtained from MRC-CFAS (CFAS I), CFAS II and ELSA. Full details of the CFAS I [9] and CFAS II study design and methods have been described in detail elsewhere [7] . Briefly, CFAS I and II are both interview population-based cohort studies [8] . The CFAS I interviews were conducted between 1991 and 2003 in five geographical areas in the UK (Cambridgeshire, Gwynedd, Newcastle, Nottingham, Oxford). CFAS II interviews took place between 2008 and 2012 using three of the same geographical areas (Cambridgeshire, Newcastle, Nottingham) and the same study design. ELSA is a panel study of men and women aged ≥ 50 years living in England [5] . Participants are interviewed approximately every 2 years [5] . ELSA does not recruit new participants from care homes, but attempts to collect data on those who transfer into a care home during the follow-up period [6] . The ELSA core datasets for each individual wave were used to identify all participants who had an institutional interview, either in-person or informant, to define the care home population. The ELSA harmonized dataset, which incorporates proxy responses, was then used to investigate the variables of interest. Individuals were included if they were original participants, or refreshment sample participants (partners were excluded) for all interviews from wave 3 to wave 8. In addition, to reflect the ageing sample, individuals who were below the age of 65 at each interview wave were excluded. In CFAS I and II, participants living in council residential or nursing homes, private nursing homes or long stay hospitals defined the care home population. CFAS I and II used informant interviews to supplement respondent information in the case of cognitive or physical frailty impairing the interview. Respondent and interview information was merged. We explored changes in self-reported health, levels of comorbidity and disability. Self-reported health was reported as excellent, good, fair or poor in ELSA and CFAS studies. For the multimorbidity and disability domains, we selected core variables that were common to CFAS and ELSA, to allow comparison across datasets. These core variables were combined into new variables, which were used to make inferences about multimorbidity and disability. The variable groupings are shown in tables 1 and 2 in the supplementary material. Disability was measured as difficulty in undertaking activities of daily living. Activities relevant to care home residents were included, such as help with going to the toilet. Variables less relevant to care home residents, higher on the hierarchy of disability [10] , for example the ability to do shopping and prepare a meal, were excluded. Participants were classified as having severe disability if they had difficulty, or needed assistance, with two out of the three domains -washing, dressing and going to the toilet. In the multimorbidity domain, we selected variables that have been shown to be risk factors for functional decline in older adults [11] . The medical conditions in the core variables were grouped according to body system: cardiovascular, cerebrovascular, musculoskeletal, endocrine, respiratory or cognition. The resident was classified as having comorbidity in a domain if they had a medical condition related to that particular body is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 6, 2020. . https://doi.org/10.1101/2020.08.05.20168740 doi: medRxiv preprint system. Complex multimorbidity was defined as having a comorbidity in at least three out of six body systems [12] (table 2, supplementary material). STATA15 was used to conduct a repeated cross-sectional analysis study. At the mean time point for each round of data collection, the prevalence of core and derived variables was calculated. In CFAS I and II, the interview question about medical conditions was sometimes phrased according to responses at a previous wave of data collection ('since we last saw, has your Doctor told you…'). In these instances, responses from previous rounds of data collection were considered in order to derive the prevalence. Baseline (cross sectional) weights were released into both studies. To adjust for longitudinal attrition inverse probability weighting was calculated for each wave taking account of age, sex, health status, disability, self-reported health, and care home status at previous wave. To account for item non-response within an interview, multiple imputation was undertaken. Each study had its own multiple imputation model for all variables that are used to calculate self-reported health, multimorbidity and severe disability, together with care home status, age and sex. Multiple imputation by chained equations using 50 imputation samples were used. Logistic regression was used to model the relationship between the presence of each variable with age and wave of interview. The main analysis presents predicted probabilities of each variable within the age group 75-84 years, which are used in the analysis to adjust for the longitudinal nature of the data and as small numbers make age-standardisation unstable. A sensitivity analysis to account for including the same individuals at more than one interview presents only the crosssectional results from the individuals the first time they were interviewed in a care home (see supplementary appendix). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 6, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 6, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted August 6, 2020. . https://doi.org/10.1101/2020.08.05.20168740 doi: medRxiv preprint Health status of UK care home residents: a cohort study Care home medicine in the UK-in from the cold Out of sight, out of mind? A review of data available on the health of care home residents in longitudinal and nationally representative cross-sectional studies in the UK and Ireland What are the Cognitive Function and Ageing Studies? Cohort profile: the English longitudinal study of ageing Changes over time in the health and functioning of older people moving into care homes: analysis of data from the English Longitudinal Study of Ageing. Age and ageing A two-decade comparison of prevalence of dementia in individuals aged 65 years and older from three geographical areas of England: results of the Cognitive Function and Ageing Study I and II Who Lives Where and Does It Matter? Changes in the Health Profiles of Older People Living in Long Term Care and the Community over Two Decades in a High Income Country Cohort profile: the Medical Research Council Cognitive Function and Ageing Study (CFAS) Losing the Ability in Activities of Daily Living in the Oldest Old: A Hierarchic Disability Scale from the Newcastle 85+ Study Risk factors for functional status decline in community-living elderly people: a systematic literature review The Nursing Home Minimum Data Set Assessment Instrument: Manifest Functions and Unintended Consequences-Past, Present, and Future Office for National Statistics. Changes in the Older Resident Care Home Population between