BACKGROUND: evidence from inspection programmes suggest that the quality of care provided by individual care homes for older people is very variable. Aside from periodic inspection, there is limited information that is routinely collected and can be used to monitor quality. OBJECTIVES: to describe a method for using routine hospital data on admissions of older people as means for monitoring quality of care within a care home. To explore how this might be applied and used. METHODS: we linked hospital admissions to care homes using postcode matching and analysed hospital admission data as a time series, using the Cumulative Sum (CUSUM) technique to detect unusually high rates of admission. RESULTS: if we develop the CUSUM so that the number of times it falsely signals a high rate of admissions is limited to a rate of 0.1% per year, the chances of successfully detecting a doubling of the admission rate within 2 years will range from 48% for the smaller homes to 96% for the larger homes. CONCLUSION: monitoring tools using data on admissions to hospital are both possible and feasible, particularly for the larger homes. However, due to data limitations, users need to be careful about how they interpret triggers and thus ensure follow-up is appropriate. Some of the problems caused by using routine national data can be overcome if care homes used their own information for local monitoring.
BACKGROUND: evidence from inspection programmes suggest that the quality of care provided by individual care homes for older people is very variable. Aside from periodic inspection, there is limited information that is routinely collected and can be used to monitor quality. OBJECTIVES: to describe a method for using routine hospital data on admissions of older people as means for monitoring quality of care within a care home. To explore how this might be applied and used. METHODS: we linked hospital admissions to care homes using postcode matching and analysed hospital admission data as a time series, using the Cumulative Sum (CUSUM) technique to detect unusually high rates of admission. RESULTS: if we develop the CUSUM so that the number of times it falsely signals a high rate of admissions is limited to a rate of 0.1% per year, the chances of successfully detecting a doubling of the admission rate within 2 years will range from 48% for the smaller homes to 96% for the larger homes. CONCLUSION: monitoring tools using data on admissions to hospital are both possible and feasible, particularly for the larger homes. However, due to data limitations, users need to be careful about how they interpret triggers and thus ensure follow-up is appropriate. Some of the problems caused by using routine national data can be overcome if care homes used their own information for local monitoring.
Authors: Jennifer K Burton; Charis A Marwick; James Galloway; Christopher Hall; Thomas Nind; Emma L Reynish; Bruce Guthrie Journal: Age Ageing Date: 2019-01-01 Impact factor: 10.668
Authors: Anna Schultze; Chris Bates; Jonathan Cockburn; Brian MacKenna; Emily Nightingale; Helen J Curtis; William J Hulme; Caroline E Morton; Richard Croker; Seb Bacon; Helen I McDonald; Christopher T Rentsch; Krishnan Bhaskaran; Rohini Mathur; Laurie A Tomlinson; Elizabeth J Williamson; Harriet Forbes; John Tazare; Daniel J Grint; Alex J Walker; Peter Inglesby; Nicholas J DeVito; Amir Mehrkar; George Hickman; Simon Davy; Tom Ward; Louis Fisher; David Evans; Kevin Wing; Angel Ys Wong; Robert McManus; John Parry; Frank Hester; Sam Harper; Stephen Jw Evans; Ian J Douglas; Liam Smeeth; Rosalind M Eggo; Ben Goldacre Journal: Wellcome Open Res Date: 2021-04-27