New care home admission following acute hospitalisation: a mixed methods approach
Item statusRestricted Access
Embargo end date30/11/2019
Burton, Jennifer Kirsty
MetadataShow full item record
Care home admission following acute hospitalisation is a lived reality across Scotland, experienced by over 8,000 people annually. The aim of this thesis was to develop an understanding of new care home admission following acute hospitalisation. Methods and findings from the mixed-methods approach are presented in three parts. Part One: Identifying relevant research – includes a review of quality assessment tools for systematic reviewing; a systematic review and meta-analysis of quantitative data from observational studies of predictors of care home admission from hospital; and a methodological chapter on developing a search filter to improve accessibility of existing research findings supported by the findings of an international survey of care home researchers. The systematic review identified 53 relevant studies from 16 countries comprising a total population of 1,457,881 participants. Quantitative synthesis of the results from 11 of the studies found that increased age (OR 1.02 per year increase; 95%CI 1.00-1.04), female sex (OR 1.41; 95%CI 1.03-1.92), dementia & cognitive impairment (OR 2.14; 95%CI 1.24-3.70) and functional dependency (OR 2.06; 95%CI 1.58-2.69) were all associated with an increased risk of care home admission after hospitalisation. Despite international variation in service provision, only two studies described the model of care provided in the care home setting. The survey identified that there is a lack of shared terminology in the published literature to describe settings for adults who are unable to live independently in their own homes and require care in a long-term institutional setting. A search filter to identify relevant research could help to overcome differences in terminology and improve synthesis of existing research evidence. Part Two: Exploring current clinical practice – reports the findings of a retrospective cohort study of new care home admissions from hospital using case-note review methodology accompanied by findings from inductive thematic analysis of a single dataset from a qualitative case study design exploring the experiences of a patient, their family, and practitioners (n=5). The cohort study (n=100) found a heterogeneous picture with long hospital admissions (range 14-231 days), frequent transfers of care (31% experienced three or more transfers), varied levels of documented assessment and a lack of documented patient involvement in the decision-making processes. The qualitative interviews allowed the patient voice to emerge, alongside the professional and family narrative which dominated case-note documentation. Inductive thematic analysis identified nine major themes exploring how decisions are made to discharge individuals directly into a care home from the acute hospital setting: biography & personality; professional role; family role; limitations in local model of care; ownership of decision; risk; realising preferences; uncertainty of care home admission process; and psychological impact of in-hospital care. Part Three: Harnessing routinely-collected data – includes the challenges of identifying care home residency at admission and discharge from hospital, presenting analysis of the accuracy of Scottish Morbidity Record 1 (SMR01) coding in NHS Fife and the Community Health Index (CHI) Institution Flag in NHS Fife and NHS Tayside. This is followed by a descriptive analysis of the Scottish Care Home Census (2013-16) as a novel social care data source to explore care home admissions from hospital and the methodology for a data linkage study using these data. Identifying care home residents in routine data sources is challenging. In 18,720 admissions to NHS Fife, SMR01 coding had a sensitivity of 86.0% and positive predictive value of 85.8% in identifying care home residents on admission. At discharge the sensitivity was 87.0% and positive predictive value was 84.5%. From a sample of 10,000 records, the CHI Institution Flag had a sensitivity of 58.6% in NHS Fife and 89.3% in NHS Tayside, with positive predictive values of 99.7% and 97.7% respectively. From 2013-16, of 21,368 admissions to care homes in Scotland, 56.7% were admitted from hospital. There was significant regional variation in rates of care home admission from hospital (35.9-64.7%) and proportion of Local Authority funded places provided to admissions from hospital (34.4-73.9%). Those admitted from hospital appeared to be more dependent and sicker than those admitted from home. This thesis has established a series of challenges in how care homes and their residents are identified. It has questioned the adequacy of the evidence to guide practitioners and sought to raise the profile of this vulnerable and complex population and how best to support them in making decisions regarding admission from the acute hospital. It has progressed our understanding of this under-explored area and proposes a programme of future mixed-methods research involving patients, families, practitioners and policy-makers.