Background The RCOPP was a national programme aimed at enabling older people to stay independent and well at home without the need for care and support. Funding was made available for health boards and local authorities to work together to design and implement interventions aimed at shifting the balance from institutional care to community based care. This study aims to describe emergency admissions to hospital and average length of stay in hospital in Glasgow City during the RCOP.

Methods Monthly crude and standardised rates of emergency admission and bed days between April 2011 and March 2015 were calculated, for residents of Glasgow City aged 65 years+ attending Glasgow City hospitals. Multilevel Poisson models for emergency admissions nested by datazone were created, adjusting for sex, 5-year agegroup, area-level deprivation (SIMD quintile), season, month and month squared. Similarly, models were run for outcome bed days, with emergency admissions as the offset term.

Results Emergency admissions first rose then fell between April 2011 and March 2015. When modelled, both month (RR for month 12 relative to month 1 and 95% CI: = 1.02 (0.999, 1.04)) and month squared (RR for month 12 relative to month 1 = 0.998 (0.997, 0.998)) were significant, indicating a rise in admissions from April 2011 until March 2012 and a fall thereafter, with risk of admission falling below the baseline figures by March 2013 and continuing to reduce thereafter. Bed days reduced across the entire study period. Standardised rates per population and per admission fell over time, however the reduction in rates per admission (ie length of stay) levelled out towards the end of the period. When data were modelled for outcome bed days with admissions as an offset term, both month (RR for month 12 relative to month 1 and 95% CI: = 0.94 (0.94, 0.95)) and month squared (RR for month 12 relative to month 1 = 1.001 (1.001, 1.001)) were significant, indicating a fall in average length of stay from April 2011, plateauing around January 2014 and rising slightly thereafter.

Conclusion Overall, number of emergency admissions and bed days for those aged 65 years+ reduced during the RCOP period. Average length of stay however increased latterly. One interpretation may be that the RCOP resulted in admissions being avoided and that these are likely to have been shorter stay admissions. Alternatively, a reduction in admissions may have allowed patients to stay longer in response to a reduced demand for beds.