First posted on April 20th 2012
This might seem to be a silly question, but in the current climate, when most stories are reported with a mix of suspicion and negativity, will we recognise success if it hits us in the face? Those twinges of doubt have started to crowd out some of my own optimism.
Let’s ask some hypothetical questions and play a gedanken experiment.
Let’s start with a typical frail elderly population receiving a mix of services across the health and care spectrum. Now let us suppose that we can allocate each individual to one of three groups.
The first group of stalwarts grew up to be very grateful for whatever services they have received. Their biggest fear is to be seen to be a burden on those around them, and they will soldier on through thick and thin. They are proud of the way they have managed to struggle through adversity and have to be on the verge of death’s door before they will seek out care. As they have aged and become more frail, their resolute spirit still drives them to put off that call for help, even when in their heart of hearts, they know they should.
Our second group are the worriers. Their fear is of being struck down so that they become dependent on others. They worry that each croak and ache might be the first step along a slippery road. In their heyday, they may have been the “worried well”. Seeking reassurance from the experts. Not necessarily hypochondriacs, but still watching out for unexpected signs. As this group becomes more frail, their anxiety levels may be rising. They will be in and out of the GP’s surgery walking out with something more important than a prescription – they will have a sense of ease, at least for a while. As the loneliness of age bites, they may find that the health system becomes a gateway to basic human contact, not just provision of care.
Then we have a third group who are neither the worriers, nor the stoics. Those with good, strong family and neighbour networks will often come in this camp. They have a sense of when to get reassurance from within their own networks and when to seek professional help. When they need to consult a professional, help is often at hand to smooth over the difficulties of navigation. As they become more frail, their frequency of turning to professionals increases, but only proportionate to their frailty.
Now, we experts in the strengths and shortcomings of the care system, observe how our fragmentary care system could be improved through a more unified, seamless approach to services, with a bit of help on navigation, more sharing of information and all the clever stuff we build in to oil the wheels from our perspective as professionals. So we decide to run a new model of joined-up care. But, being the intelligent folk we are, we decide that we need to control for variability where we can. We choose to divide our trial population into our three respective groups, so that we can compare the results between them.
I just know you’ve already made the mental leap, but, please, stay with me!
In our wonderful. integrated care pilot, group 1 makes more use of the services because the new model we have put in place does precisely what it is meant to do – it captures the need at the right time, rather than late. For this group, integrated care should reduce risk, improve both mortality and morbidity. By capturing the need early, it should reduce length of stay, but might very well increase admissions. Rightly! So integration increases demand!
But group number 2 receives reassurance in a more appropriate way, reducing the demand that we experts think is inappropriate. They probably make less use of the service, but with a better level of targeting. By providing more appropriate reassurance, they reach the front doors only when there is a higher risk. Integration reduces overall demand, but leads to a higher conversion rate of admission per attendance!
Quelle surprise? Demand from group 3 is relatively unaffected by our carefully developed improved service, though the better joining up will reduce levels of frustration.
So, which of our three trials succeeds in giving us a definitive picture of the effectiveness of integrated care? Group 1 makes more use, and is likely to be happy. Group 2 makes less use – equally happy. Group 3’s demand unaffected, but still happy.
Now let’s leave this hypothetical construct behind and return to reality. What can we really learn from the evaluation of integrated care pilots which is widely trumpeted as showing that integration doesn’t work? Has it helped us to define success? Nick Goodwin, in commenting for the King’s Fund, has gone a long way to putting the work more carefully into context, avoiding the simplistic solutions which have been too prevalent in other headlines. Now can we move this debate along to a more productive consideration of just what we should be doing to design care around the patient, and therefore what we need to know to evaluate whether we have achieved success or not?
There is one single, unequivocal message. The effects of integration are unpredictable and unmeasurable, UNLESS defined by the effect on the individual. Success is defined by the patient or service user, not by the service provider.