Desperate for evidence

When Dave (our prime mister) ran his recent health ‘summit’ – without inviting those who disagree with the Health and Social Care Bill – he announced that research showed cometition is best for health. The research he was referring to is a paper that has, in fact, been ‘announced’ many times since it was first published. It was authored by Zack Cooper of the LSE, and reports that competition leads to ‘moderate but statistically significant’ reductions in patients’ length of stay.

What we know from studying health and care systems is that the focus on costs leads to shorter episodes of care – hence shorter lengths of stay – but also more of them – people are ‘churned’ – and every time people are readmitted they are usually treated as though they are ‘new’ events; creating more revenue for the provider – through commissioning arrangements – but also ensuring there is no continuity of relationship. Furthermore, the consequences, for many cases, are greater costs and greater dependency when people return home or, in the case of many adults, are shunted off to care homes.

Aside from the evidence of what is actually happening on the ground, two academics have published criticisms of Cooper’s research methods. Allyson Pollock wrote about its flaws last June and Ian Greener criticised it last month. See these here:

Competition and saving lives – a reprise

And for his frequent use of the word ‘evidence’, as though to convince, the Secretary of State for health was taken to task by Ben Goldacre, ‘Bad Science’ writer for the Guardian:

Le Grand Drivel

From the same stable (the LSE), economist and ministerial favourite, Julian Le Grand, bangs on about his pet theory that people delivering public services may be either ‘knaves’ of ‘knights’, each with their own forms of motivation. And thus, by somehow assessing which each is, ‘the trick is to… to construct… a ‘robust’ incentive structure’ to fit each, according to Le Grand.

We should remember that the ‘science’ of economics is not a science at all, merely an attempt to explain. Perhaps this is why politicians listen to people like Le Grand; they provide explanations that fit the political narrative. Le Grand ought to read the robust research literature on incentives, which shows that all contingent incentives (do this to get that) lead to less of what you want, not more.

You can read his drivel here:

Motivated to meet targets

Meanwhile, back to the NHS: Polly Toynbee, writing in the Guardian, described an example of motivation at work; a waiting-list clerk in the NHS was ordered to cancel operations for anyone who was already waiting over 18 weeks; and instead, to fill that theatre time with people closest to breaching the 18-week limit. ‘I was told to call people who had already gone over the 18 weeks and pretend there was no longer theatre time for their operation, and not give them a new date.’ She was told not to book anyone already in breach until April, the start of the next financial year. Instead she was told to fill theatre slots with as many short, minor operations as possible.

Read it here: http://www.guardian.co.uk/commentisfree/2012/feb/20/cameron-nhs-cheats-waiting-lists

Would Le Grand count her as a knave? Clearly she was doing as she was told, so were her bosses knaves? No, it’s far simpler than that, they are merely products of the system.

Vanguard’s team working in health blogged on this issue last November. In essence our argument is that better measures, specifically measures of demand and capability (achievement of purpose in customer terms), would lead to better understanding and therefore better performance:


A cop knows the answer

A police officer working in mental health learned the same thing. In his blog he explains that if we were to actually help people who present with mental health problems, the costs would fall. Why don’t we? Because we are focused on the wrong things. See the blog here:


Solid evidence

I was asked to speak on the You and Yours programme on Radio 4 a couple of weeks ago, to give some background to a package they had made of our work with health and social care providers in the South West. The package illustrated how spending time understanding peoples’ real needs, the costs of service provision fall.

I mentioned a similar study we have been helping with in the North West. What we know from that work is that between two thirds and three quarters of NHS demand is what we are calling ‘help me’ demand (largely health and social care issues), as opposed to ‘fix me’ demand (i.e. something gone wrong). What’s more, we now know that if we can configure the services to meet the real ‘help me’ needs, costs fall by about 25%. As I said to an audience in Whitehall recently, ten years ago naively I would have expected a call from the Secretary of State for Health to ask how this has been achieved. But today I know that Whitehall doesn’t do evidence. You can listen to the Radio 4 piece here:

And if you think I am being cynical about Whitehall, please remember that I first gave evidence to a care minister about the problems and opportunities for improvement in 2005 and nothing has changed.

Commissioning drives up costs

At the heart of Dave and Co’s competition strategy is commissioning. It is central to what economists call the ‘efficient market hypothesis’; something for which there is no proof. The economists believe that if you price standardised services, then ‘any willing provider’ might come along and offer to provide them at lower costs than their competitors.

What the Somerset, and our other work in health, has taught us is that commissioning drives costs up. The reason is simple: when you commission services they are specified and the specifications don’t meet peoples’ actual needs. It is a classic failure-to-absorb-variety problem. If you listened to the ‘You and Yours’ piece, you’ll have heard examples of the phenomenon.

The eminence grise

So who are the thinkers behind Dave and Co’s plans for the NHS? This time it’s the Mail on Sunday that has the story. McKinsey, the pre-eminent scale consultancy is the thought-leader for health service reform:


More bad news from SW1

The failure of scale designs to absorb variety is at the heart of the failure of SW1 (the joint venture between IBM and two South West councils). Just like the NHS, SW1 has pursued a strategy of industrialising services (call centres, back-offices, standardised work etc). Sadly, the council’s leader responsible for getting out of the mess thinks out-sourcing services will be the better answer. See the latest report and his views of the next steps here:


Out-sourcing: a disaster in the making

The councillor struggling with SW 1should take a drive north where one of our largest metropolitan councils has been on the out-sourcing path for some years. An enlightened mole has been corresponding with me about what’s going on there and I summarise our conversation:
The idea of out-sourcing arose in 2006 when the costs of the call centre were growing. All regular readers will know the reason call centre costs were growing was because it was full of failure demand, caused by moving all telephone work out of services and into the call centre to comply with the 2005 target. I asked if they knew about the extent of failure demand. He went off to ask and he was told that the change-programme people estimated that 35% of all unanswered calls (‘abandoned’ in call centre parlance) were return calls. Yes, you may want to read that again. No knowledge of the nature or extent of failure demand and a reliance on the abandon rate as an indication of what might be preventable. It’s because the numpties worry about service levels – picking up the phone – and that’s all the data they have. Of no use whatsoever.

He tells me the change planning people are spending their time manipulating spreadsheets to come up with a ‘palatable’ benefits case. To get to break-even over the coming years the change planners have assumed they will get rid of failure demand (yes, something they have no knowledge of). My intrepid reporter pushed on the matter of what volume of demand is failure demand and, eventually, he was told it would be the 35%; when he pushed again, to ask about how they knew this was the right number, as by their own admission they knew nothing about true volumes, he was told ‘that it felt about right’. He went on to ask how the failure demand was going to be removed and the answer was: that will be the responsibility of the managers running the services.

I know – you couldn’t make it up. But it gets worse…

Not giving up, my mole got a manager from a care service involved in the conversation. He knew that 63% of all calls to the service went unanswered (the reported abandon rate). The plan is to move all these calls to the call centre, as this will ensure they will be picked up (as the argument goes). The manager asked my mole to critique her slides which had been put together for the change-programme board. He explained the basics of failure demand and pointed out that at no point did her plan describe what the problem(s) are and picking up the failure demand is not the same as getting rid of it.

The mole continues: ‘Recognising I was right she dragged over a manager from the change programme [and he said] that it doesn’t matter, the money has already been budgeted for and the slides are only really to confirm the business case. When I pushed him on the lack of a defined set of problems and asked how they intended to spend the money if they don’t actually know what needs fixing, he said that the general view at the leadership level was that the causes of the poor performance were due to capacity issues in the workforce. Demand is not a concept even discussed.’

The mole gave up. As you would. It illustrates the nonsense of following an ideological wrong-headed view, something Dave and Co ought to worry about. The management factory is all about delivering the plan, there is little appetite for understanding what works.

The facts of the matter are these: it is common to find failure demand running as high as 80% of all demand into care services. Instead of sending all these calls to a call centre, understanding demand leads to a service design that works and demand volumes drop (failure demand drops to almost zero). If only they knew.

How do we change thinking?

The change-programme people and the managers of this metropolitan council believe a number of things that are, simply, false. These beliefs are all derived from what I call the core paradigm which, in the case of call centres, is: how many calls come in, how many people do I have and how long do they take to handle the calls?

While the most obvious counterintuitive truth is that if large volumes of demand are failure demand you’ve missed the point, a less obvious one is that 95% of the peoples’ performance is governed by the system. True for care services and all other services.

I mentioned meeting Hayley Johnson in a previous newsletter, she who runs a sales organisation where such an idea is so counter to convention it is considered completely bonkers, after all sales is all about people right?

Many people asked me to post the ‘Hayley video’ on line, so I did, you can watch it here:

If the managers dumped their change-programme staff and got out into the work they’d learn some things that would help them make a big difference.

The DWP don’t get it

Regular readers will know of my doubts that the Universal Credit will be deliverable on-line, my argument being that an on-line service won’t deal with variety and thus will fail to help those in need and if those people have the energy it will create high volumes of failure demand.

The change-programme managers for the Universal Credit are now back-tracking on their ‘digital by default’ mantra. In the Guardian (again!) we find a man called Steve Dover, director of major programmes at the DWP saying there would be a ‘back office to deal with the more vulnerable in society’, though it is expected to be ‘thinner and there will be a massive web-enabled internet channel for the vast majority of the transactions that will be done for universal credit’.

But Steve doesn’t know how big this ‘back-office’ (in fact a call centre) will be, because he has no understanding of the problem. Speaking at the Government ICT Summit in London he added that other methods of delivery would have delayed the project until February or March 2015, budgets would have doubled and ‘there would have been huge arguments along the way with our partners, our service providers, and within the departments between all the different silos, so policy, business design, IT, implementation, release and operations’.

Complete nonsense. We have offered to design a service that works, with people at the front end (as only they can absorb variety) and it would have meant none of the hundreds of millions being spent on computer systems (an eye-watering £500m announced last month). But the DWP plan is to dispense with people providing benefits locally. Steve’s admission that there will be some means to help those who need it is merely a back-up for his cunning but wrong-headed, cost-obsessed plan.

You can read Steve Dover’s views here:

Down at the coal face

A reader writes:

‘Here in social housing, we are clenching our cheeks (both sets!) for the arrival of the real time computer system for Universal Credit. At a meeting in Manchester in November 2011, the DWP told us that they were using the ‘Agile System’ to manage the implementation of the IT. I was present with a full room of housing professionals from CEO s downwards and once the DWP reps left, the level of dissent was like nothing I’ve ever seen in 15 years in the sector. The focal point was that payments were to be made to families 4 weeks in arrears – akin to how one is paid for work – when in fact, by law, social housing tenants have to pay their rent on the Monday beginning of each week as stated in many tenancy agreements. The thing is John, 4 weeks arrears is when most landlords have actually got into serious action mode (including evictions) for those who don’t pay their rent. And, according to those present who work with the current system, it is more like 8 weeks before many tenants get their housing benefit paid.

No doubt you’ve heard a lot of this stuff already but it’s as if by ideologically classing those on benefits as ‘work-shy’ and needing to know what being in work feels like, they’ve overlooked how the system of paying for your rent actually works.

ALL the landlords present at that meeting in Manchester said that they would not change their rent arrears procedures to accommodate the DWP ‘s Universal Credit payments to claimants 4 weeks in arrears. Nor will my own organisation for that matter.’
There you go Steve, sort that one out.

Shared services

Meanwhile, more evidence of the other ‘economy of scale’ disaster, shared services, only mounts. While Dave and Co think sharing services is the saviour of our fortunes, people notice that it doesn’t work. Herewith two examples from over the pond and I can confidently predict some real shockers in the UK will come to light soon.

Academics notice

‘[Shared Services] was supposed to be streamlining and simplifying our lives, and what it’s done is made it much more complicated, said Benjamin Foster, a professor in the Near Eastern Languages and Civilizations Department. Everything takes about two times as long. We resent the down-skilling of departmental administrative personnel. We don’t see how that can be more efficient or cheaper.’ Read more here:”


Citizens notice

People sceptical about what was delivered in New Jersey:

Mums notice

Although not an example of shared services I couldn’t leave this out. Netmums, a favourite of Dave and Co, noticed the nonsense of risk assessment strangling common sense:

Smaller is cheaper

I mentioned this last year, but it is appropriate to mention it again. Research shows smaller councils are cheaper, economy of scale is a myth:

Click to access TOI-ReporttothePeople-20110110.pdf

An inspector writes

Regular readers will know I am a fan of Inspector Guilfolye, a practicing police officer who manages policing as a system. Hats off to him for writing a thorough academic critique of targets and performance management. I recommend it: