On the effectiveness of care co-ordination services aimed at preventing hospital admissions and emergency attendances

Abstract

Finding alternatives to hospital admissions for older people and helping them to live for as long as possible in their own homes is a key objective of UK health and social policy. However, there is a lack of proof as to whether such alternatives actually work and are cost-effective. The research reported in this paper is based on an evaluation of a care co-ordination service with the aim helping people to remain at home and preventing unnecessary hospital admissions and A&E attendances. An initial evaluation found that the effectiveness of the service was at best marginal. The new method of evaluation described in this paper tracks patient attendance at A&E departments and hospital stays 12 months before they are accepted into the care co-ordination service and evaluates the resultant savings in health care activity. It finds that the service results in between 14 and 29 saved hospital bed days per client per year and between three and eight A&E attendances. Whilst the service does not arrest functional decline in individuals, the incidence of falls is significantly reduced, and that the effect on quality of life is neutral to broadly favourable. It finds that mortality levels are higher than in the general population of similar age but this is probably due to selection effects because clients are unhealthier from the outset, and that in the 90+ age group there is no significant difference. Without the benefit of a control group it is not possible to confirm the results with certainty, but corroborating independent evidence is provided that supports the conclusions reached.

Fig. 1

Average admissions to hospital per client per month before and after ICCS intervention showing standard error bars

Fig. 2

Average admissions to hospital per client per month before and after ICCS intervention after splitting the data into four series

Fig. 3

Monthly average admission activity per ICCS client in the 24 month pre-referral period and the 9 month post-referral period

Fig. 4

The pattern of admissions pre- and post-ICCS intervention with fitted trend curves applied (key: A= actual admissions in previous 12 months; B= actual admissions 12 months post-ICCS intervention; C=prevented admissions based on constant growth assumptions resulting from ICCS intervention; C+D = prevented admissions based on accelerated trend resulting from ICCS intervention)

Fig. 5

Bed-day utilisation in the 12 months pre- and post-referral to ICCS after splitting the data into four series

Fig. 6

Average bed-days consumed per client per month pre- and post-ICCS referral including prevention effects

Fig. 7

Trend in average length of stay pf clients admitted to hospital pre- referral and post-referral

Fig. 8

Pattern of average A&E visits per client per month based on samples with different start dates and post-referral history

Fig. 9

Average A&E attendances per client per month pre- and post-ICCS referral including prevention effects (one attendance = £200)