For a short time I worked for a company in its dying years. I saw it go bust. I got a brown envelope stuffed with money and, far more valuable, a lesson in management stupidity.
It was a small company here in the UK, but owned by a big foreign parent company. The parent company wanted things done in a certain way. All of us got a bonus every year based, in part, on customer satisfaction. So we worked hard to ensure that we had satisfied customers.
Every year the customers who were unhappy would go over to the competition. The happy ones who were left would fill in our survey. Customer satisfaction levels rose. We got bigger bonuses.
By the time the company went bust, our pitifully few remaining customers were extremely satisfied indeed. Didn’t we do well!
Over at Lake Cocytus, The Shrink asks questions about quality measurements in a mental health service. Ah, how it takes me back…
The two fundamental and closely related issues with quality measurement are what you want to use the measure for, and what you plan to measure against.
It’s possible to use measurements for various things. I’ll consider just a couple here: service improvement and management smugness.
If you want to use the measure to improve the service, then you have to know in advance what you mean by ‘improve’. It’s no use defining ‘improve’ as getting better scores on the measure. That’s just circular. A circular definition of ‘improve’ will mean that you go round in circles becoming more and more distant from reality, just like that company I worked for.
Equally, it’s no use measuring against your own system of measurement. Again that’s just circular. If you really want to know know whether you are improving or not, you have to measure against something external.
For example, in a mental health service you could define ‘improve’ to mean more of the things that make patients feel they are being well cared for, like respect, compassion, consistency.
Take respect. How do you know how much respect patients are getting, other than by using your measure? You don’t. There’s no external measure of respect that you can compare with. Measuring respect and things like it is bound to be circular. You can’t create respect by measuring it.
You might want to use the measure to increase management smugness. This is not far-fetched.
Suppose you have senior managers approaching retirement or looking for promotion, and suppose their bonuses, pensions and prospects depend on the quality measure. Would it be reasonable for them to put in place a measure designed to increase management smugness in the short term, even if as a result the service collapses in the longer term?
Or suppose that ultimately management equates to politics, and smugness equates to votes. Does a measure that increases management smugness seem reasonable?
To increase management smugness you need a measure that is easily manipulated. It has to rise every year. (Oh, the nostalgia…) It’s important, therefore, not to have any external benchmark. The measurement system must measure against itself, and then the results can be, essentially, anything you want them to be.
The two intentions that you might have, service improvement and management smugness, therefore lead to opposing measurement strategies. How can it be that a better service makes for unhappy managers, and smug managers makes for a worse service. That paradox is created when you define quality in terms of a measurement system in the first place. If you want to improve the service and have happy managers at the same time, then you cannot manage by measurement. (How to manage a service other than by measurement is a story for another time.)
When it comes to the practicalities of measuring quality, the best tactics also depend on your intentions. Realistically, not all patients are equally interesting.
If you want to improve the service, patients who already think the service is doing its job are not very interesting. The really interesting patients are the ones who have gone to the competition.
Where’s the competition? It’s me. It’s the drinks section at the supermarket, the off-licence, the pub. It’s the local drug dealers and divorce lawyers. It’s the prison. It’s Samaritans. It’s the cemetery. That’s where the patients are who can tell you what you need to know to improve the service.
And if you want to increase management smugness, avoid asking those patients. They’ll ruin your results. Concentrate instead on the ones who regularly attend their appointments. Make them feel special, part of a club. Send them glossy magazines about the Trust, and invitations to special members-only events. Keep your happy customers on board and they’ll give you great survey results year after year. (I’m not making this up — I’m on a local NHS Trust’s ‘keep on board’ mailing list, and I get all their fancy PR drool.)
The question The Shrink poses is: “What matters?” What matters is the market. Not the phoney market being created within a playpen of government regulation — that’s a decoy — but the real-world market.
The market is where the people are who could be your patients but aren’t. They choose. And in the act of their choosing a market comes into being. That’s what matters.
Incidentally, the decoy confuses many, including the otherwise formidable Dr Grumble, who rejects the Department of Health line on the NHS only to swallow the Treasury line on banking together with its hook and sinker. (Antidote here: What Really Happened?)
The market is this, now, here. It always has been. You want to know what to measure? Measure your competitors. You want quality mental health care in the NHS? Put the prison service out of business. Put Samaritans out of business. Put me out of business. I dare you.