how to “do” PBMA – survivor’s lessons

Making PBMA work in population based accountable systems of care – how to “do” PBMA


Greg Fell / Toni Williams


Background and preface

  • In 2011/12 NHS Bradford and Airedale, with clinical support from across the spectrum, extensively piloted using PBMA methodology to achieve efficiency savings in four clinical areas.
  • This was a strand of QIPP for the PCT. There had been concern that the QIPP programme as was would not deliver cashable savings.
  • We explored the use of PBMA as an alternative to the “standard” approach to QIPP.
  • The detail of how this process worked is described elsewhere (article in preparation)


What we did

  • We tested PBMA as a methodology for making disinvestments (the reverse application of normal PBMA …. “there is a little more cash…what are the most valuable investments”)
  • We tested in 4 areas – respiratory, CVD, vision, diabetes
  • Each of these systems was set a challenge to find efficiencies. The scale of the challenge we gave the clinicians was proportionate to a “cut” in the PB level spend in those areas that was the fair share of the total financial challenge to the PCT view of funds as a total. The 4 year challenge to PCT was £50m and thus the challenge to each of the 4 clinical groups was finding efficiency totalling – £520k (vision), £1.1m (CVD), £550k (respiratory) and £750k in diabetes. Each year, for 4 years.
  • This was a clinically focused process – involving a selected group of senior clinicians . It was “PBMA lite” in reality. Described as follows:-
  • give each of the system focused groups of clinicians a good overview of current spend as was in their area….using PB data, actual activity data (as PB view and contracted activity view never mesh properly!!), prescribing and other date.
  • Give each system a view of outcomes acheived for that spend, both in absolute terms and relative to others,
  • give each system the scope and scale of the financial challenge that system had been set.
  • Develop, though discussion and debate, a clear and unequivocal list had been generated for lower value interventions that could be squeezed out of the system for little or no loss of health
  • We described it as “lite”….as it was expert led process and didn’t involve “proper” economists to crunch loads of data (the clinicians may have disengaged had it been a very data heavy process)
  • that list was then checked out with a wider group of clinciians and some consensus reached.






The failure to realise

  • We think that our effort was a failure, the proposed methods of achieving the efficiency savings were largely not realised.

  • it failed, gloriously…., we think for a number of reasons
  • arguably we didnt have the charisma to pull it off
  • we certainly didnt have the institutional traction or buy in to pull it off
  • the PCT ( and to be fair the system) wasn’t prepared to invest (time not necessarily resources) in IMPLEMENTING…. and actually seeing it through (it was seen as “one of Greg’s crazy ideas”)
  • the clinicians remained keen, but provider trusts weren’t
  • the PCT started to collapse around us…..eye was taken off the ball of implementing
  • the institutions were (still are) fickle……loose interest quickly, move onto something else…. and massively under value the notion of IMPLEMENTATION (and capacity / capability for this)
  • many to most of us (not me) are of the view that there is a single big idea….. we just don’t think it’s true…. there are lots of little ideas…. and thus this takes us back to implementation
  • there was no real mechanism to MAKE it happen…… fragmented commissioning and fragmented system………..
  • is there IS a single big idea…. it is around system design…. ACO and the like…. putting providers together, making the incentives for financial balance be in the same place as clinical decisions. whilst we are of no doubt that ACO type model is NOT a panacea (and the evidence from CBO on the Medicare pilots in the states tells us this clearly…..) it must, in our view, be on the table as one of the “big ideas”

What is this paper.

  • We learned a great deal from this failure
  • Here we assess the reasons WHY we failed, and that factors would need to be in place to hold funding within a population based system.
  • There was always a danger that “all of this right care and PBMA stuff, whilst being very good, is just a bit too fluffy and gives us a bit of breathing time to provaricate, talk nice words, but not actually take any action”


Why this is important

    • Once funding has been allocated by parliament, 70 – 80% of the resource allocation decisions in the NHS are made by the doctors pen.
    • It may be argued that the concept that we call “commissioning” does not have a huge influence on this, or has not to date
    • In the current climate there are weak to no mechanisms to hold a system to within a fixed budget, this will remain so whilst the system is such that “someone else pays” – a form of moral hazard.
    • There is a great deal of interest in the notion of what mechanisms are there to hold people to fixed budgets.
    • We offer our thoughts on the prior thinking, circumstances and preconditions of successfully implementing programme budget approach.
    • This should be read alongside the traditional literature on programme budgeting.
    • It should also be read alongside the literature on alternative and emerging contractual models
    • Lastly we have underplayed the notion of culture in this. Arguably culture is the most important fundamental.

  • PBMA is only ONE of a number of facets of Right Care – equally important are other components of the Right Care approach – including variation, value, shared decision making population medicine.



Thanks to Kev Smith for thoughts on CDF

Prior to set up ……….. spend enough time focusing on learning from what others have done


  1. a) What’s the fundamental purpose

  • We spent insufficient time on the question of “what is the fundamental purpose of PBMA”? need to agree in advance whether this is a culture change or management process


  • This can be seen as a vehicle for LT culture change; and a means of achieving high value / high quality services

    • However, can also be seen as a managerial mechanism of stripping money out of a system

  • subtle difference between such a “managerialistic” approach (“get it implemented and deliver the savings”) versus getting culture change embedded in the heads of clinician resource committers (“you can only have new toys if you ditch the old ones”). The need to implement, it is important…….but culture is more important! – both in the way that clinicians behave and in the way that organisations plan. Culture change takes time and needs to be nurtured.


  • Don’t set out to over micro manage implementation,
  • There is thus a balance between using this process to “make savings” and protection of the process to achieve cultural change (“how should we see services in 20 years time”).
  • Protection of individual clinicians, clinical services, the institutions in which they sit will be important to stakeholders.
  • Understand the impact on provider, commissioner, clinician and clinical service. A commissioner perspective will not help providers achieve efficiency savings.
  • A balance re the proportion of time that is spent in identifying and delivering “savings” versus the time that is spent identifying and planning for known challenges in the future.

  1. b) Opening ground rules – Some challenges to address before you even start.

If we save £x, what proportion can be reinvested. Who runs the rules of reinvestment.

Link to social care?

How will you deal with efficiences as a result of windfalls (vs those that require effort)

  • If it is a naturally occurring windfall (eg atorva going off patent) then the commissioner has first dibs on its reallocation.
  • If the efficiency requires some effort or risk (clinical, financial, legal, reputation, other) then the clinical area making the effort should be the area receiving the investment. So if the cardios and GP community made big effort to sort out OMT in angina, to reduce stenting, the the money should be reinvested in say community cardio services (something that is mutual gain for patient, provider, and commissioner, and clinician at coal face)
  • If investment was needed up front…..then the money for this investment comes from the overall budget envelope for that area. It is unfair that patients with other needs get less good care on account of this wonderful xxxxxxxx

Dealing with the questions of intra and inter programme shift before they come up? Process to deal with questions about cross programme reallocations……….spend a little bit less in teeth and a little bit more in eyes etc etc. Intra programme process. .i.e the gastroenterologists can further debate with the liver docs – relative merit of UC / Chrons / Liver disease

  1. c) Agree what’s in scope budget wise

nature and size of the budget is a reflection of 2 issues –

    • how much is enough – economic, epidemiology, HC technology, demographic

  • how much risk are we willing to takeBudget cap MUST be clear – “service line commissioning” The critical importance of absolute budget cap – if people think “we can find some brass” then there is no incentive.What of the save can be reinvested in better services”. financial incentives, whereby clinicians can choose to reinvest a proportion of the savings realised back into services. By what rules does this operate? . if the savings are found out of primary care prescribing should they be reinvested in secondary care medicine?multiple commissioners makes this harder not easier – NHSE commission primary care and specialised. Strength of clinical evidence for different components of the pathway – rehab, prevention, palliation vs therapy and diagnostics (where commercially sponsored studies). Is it about COST or VALUE – How do we measure VALUE. A weakness in our accounting system, in that from a finance perspective we measure COST in the short term, rather than VALUE in the medium to long term. Are we prepared to change this. Would require a fundamental restructure of accounting system, and risk.Perspective and level of focus – agree early……


  • Consider the use of formal tools such as the STAR tool[i] to focus on VALUE
  • get a clear plan of need / want / line in sand that cant be crossed / what is going to be given up
  • timeframe for realising benefit – prevention……different types of prevention.
  • Is it about whole care pathway – prevention through to tertiary / palliation, or specific parts – without complete overview of the whole spend there is the obvious danger that different bits of the service focuses on its own priorities – participants need to be able to debate the relative value of promoting physical activity, versus, medical management (stains etc) versus, angios versus device implementation versus end of life heart failure care versus specialised care (TAVIs/Mitral clips, surgical care) – Do we invest more in CABG, rehab or primary prevention
  • Make clear that THERE WILL BE NO AD HOC INVESTMENT FOUND – even for V HIGH PRIORTY investments,…… significantly intensifies the OC of making poor (and unsustainable) decisions…..magnifies it…..if there is even a suspicion that “money will be found” then there is no incentive for trade offs and no incentive to play……
  • Ensuring clarity re ABSOLUTE budget cap – this is about making it real at the clinical coal face

  • micro – service line, meso – PB level – PBMA / PPBS / specialty, macro – QIPP across whole h economy, generic px, staff ratios, prim care quality etc
  • Across multiple organisations? Multiple providers and 1 (or more) commissioners
  • Addressing the interface and overlaps with other stuff – easy in dermatology / harder in diabetes? many different (and overlapping / abutting / interfacing) pathways – stroke, vascular surgery, AF, palpitations, prevention worstreams might look radically different across different agencies.


  1. d) Six fundamental principles for getting the programme budget and system overview sorted. Get them agreed across your system.

    • explicit criteria based on health need

  • consider needs and cost together


  • consider alternatives that are feasible and consider cost, resources and strategies TOGETHER
  • independent analytic staff at v senior levels to advise policy and strategy
  • multiyear service and financial plan
  • conduct open and transparent analysis – available to any interested party
  • The data on spend and outcomes isn’t perfect. And not great match for how sec care organizes their world view – specialties, acute med, medical v surgical. How can we asses the actual costs per each part of our programmes? PB data not accurate enough (will improve as switches to being expression of commissioners and not providers) How what do you think best to use as ‘currency’ for comparison?
  • Don’t spending too long poring over data. We didn’t. Its not perfect. Deal with it. Combining data on (needs adjusted) spend and outcomes is important. the difference between “good enough” analysis vs perfect. information will always be imperfect……………………
  • There is a significant difference between PCT view of spend and what providers may actually allocate to depts. PBMA categories are not an easy fit with specialties and the underpinning service infrastructure (eg critical care / pain medicine / acute surgery / geriatrics etc – none of which fit neatly into PBMA categories)

2              Implementing – process and decision making


  1. a) The decision making process

    • don’t over complicate the process
    • Plan for, but don’t assume there will be, rational decision making – The difference between Rational Actor, Organisational Paradigm and political paradigm
    • Don’t assume that those who commit resource actually understand how money moves around the system, or the fundamental differences between cost / value / different types of efficiency. (scope for developing a ‘How money in the NHS work: a guide for senior clinicians’ – these exist. A ‘How the army runs: a guide for senior commanders’ was a key recommendation from US military.)

  • Ensuring a system influences the day job – critical. The principle reason is that “it’s nice, but all a bit fluffy and doesn’t really translate into ACTION on our real world coal face.   How do we contract for it”. This is related to the difficulty of turning “culture change stuff” into something that actually changes contracts and clinical behaviour.


  • Decision making – keep to the basic issues / don’t over complicate / Plan for, but don’t assume there will be, rational decision making – The difference between Rational Actor, Organisational Paradigm and political paradigm (Essence of Decision)
  • Making it stick in the context of the world of contract management

  1. b) Day to day challenges

    • Turning proposals into contracts, pathways & real reduction in absolute spend of commissioner is v. challenging. But don’t try to precisely project how much this process will save. focus on sensible clinical engagement in the consideration of “where we can save”

  • “new stuff” will easily blow any savings out of the water, thus maybe we need to plan to “save” even more.


  • Danger of reduction in spend in one area leading to inflationary pressure elsewhere.       “Whack a mole”.
  • Whilst there might be clinician engagement, providers as organisations need to be willing to participate, in implementation
  • Small changes in individual’s clinical behaviour is the focus– aggregated to big effect, rather than ‘big cuts’.
  • Focus on ‘doability’ – can we actually achieve change?
  • Clinician engagement in this planning is difficult – for them it requires time away from patient care, this requires resource and as a min incurs opportunity cost for the provider or employer of that clinician. Some stakeholders may not want to / be able to / be allowed to / have time to properly participate
  • Avoid over focus in 1 particular area: primary v secondary care, prescribing v interventions, prevention v treatment.
  • It is an iterative process……not get it right first time round…..but the outcome and the process around which the outcome hangs….


  1. c) the hit list, wish list, shift list

    • CRITICAL that you go into the workshop with a hit list…….dont expect the participants to do so……
    • what incentive has expert members to come to table with hit list (more likely to come to table with a wish list……to expand sphere of influence)….not going to naturally advocate for disinvestments in their own budget……

  • hit list (shift list) needs to be fully populated and discussed prior to starting on wish lists…..


  • Make sure you get details on HOW to IMPLEMENT! And challenges in practical implementation
  • expert members might find it hard to see beyond sectional and factional interests – specialisation, sub specialisation vs general practice and public health? Specialisation and sub specialisation might mean that one opth or optom may not be able to speak for the whole programme budget area
  • NEED to consider the OC, and knock on impact of proposals on WHOLE pathway.,….and other pathways
  • most of the ideas that are claimed to be cost neutral will probably work out to be not so….and need net investment to make work

  • disinvestments must be close to home……………relatively narrowly defined PB ……. (?is whole of resp too big ?? or whole of CV……… the PB is defined in terms of relatively similar input requirements then the clinicians can better understand the need for their own restraint…… The budget stricture as close as possible to the clinicians – ie problems of vision is complex and multi faceted.

  • Interface between money, quality and prevention.
  • consider the various options for achieving savings in particular areas. Prescribing costs; Care pathway redesign; Services to stop or thresholds to impose; Invest to save (subject to specific caveats); Pathways and service changes that may results in ‘a little bit less health, for substantially less investment.’
  • “gold standard” vs “pewter standard” pathways – what we can actually afford.


3              Moving it forward, if we were doing it again


  1. a) Dealing with the practical real world issues

  • deal with power, factions, entrenched views, weak levers to effect change – there will be entrenched views and factions in this debate. Don’t expect CVD clinicians to be always enthusiastic about lifestyle and prevention, or palliation. Eg CV recommendations has not picked up on the cardiac rehab element / stroke / prevention.
  • What is the contractual mechanism to make it stick – prime contractor, alliance contracting, other…..what place and role does different payment for activity have.
  • Whichever model is taken, organisational support – strategic and operational – is needed.
  • who carries the risk and how this is done – what if the implementation model overspends on its planned budget?


  1. b) Consider are the attributes of a successful system for PBMA

  • Low number of (clinical… principal committers of NHS resources) decision makers Simple system….. resources can ONLY be spent on a definitive list of specified interventions
  • Limited number of interventions
  • Political support
  • Institutional support
  • Definitive administrative means of way of fixing the budget and constraining it
  • Consider difference between macro planning concepts (PB, HNA etc) that are much easier context of a closed system with fixed budget overall and a true single payer. This is set against a world of multiple CCGs and the split with LA and NHSE, commercial provision.
  • Learning from the CDF – arguably an excellent example of PBMA. Look at the behaviour change which CDF has brought

  • cancer docs now have to convince other cancer docs of the value of their pet scheme
  • spend is linked to outcome- so if no one gets a second dose we know that it’s not worth funding that one dose
  • with tight control over use is prevented- and there is an incentive on clinicians to stop as soon as no effect is seen


  1. c) Addressing decision maker moral hazard

  • inherent incentive to spend someone else’s money… sec care commenced scrips vs primary care Px
  • Responsibility for service, vs responsibility for system – eg Stroke Prevention in AF has to include stroke care – acute and ?rehab / DM to include DM MI and Strokes, DMO, DR, feet etc etc.
  • What about prevention – esp how to deal with shared risk factors….where do you draw the lines of the circle…



  1. d) Organisational context

organisational and strategic leadership key

Governance – especially over multiple commissioners, providers and clinical areas.


  1. e) six improvements to process & output

  • ensuring the data is fit for purpose. Recognising that the data will never be perfect, and that data is only one driver of decision.
  • consider two, three or more years worth of budget planning critical – esp for investments without immediate payoff.
  • Dealing with investments that might require an out of programme trade off (if we spend more on army, we spend less on navy)… (if we spend more on hearts we may have to spend less on teeth)
  • Tighter management and sequencing of the steps in a process
  • Addressing the disconnects between ‘strategic vision’ and ‘day to day real world’ – in NHS speak there is also the disconnect between the application of PPBS type system and the management of contract; especially in a non integrated, market driven system.
  • Ensuring that PPBS is NOT seen as a substitute for the collective wisdom of the clinicians and system leaders – experience counts! Or a replacement for the technical knowledge of scientific advisors, or the real world knowledge of clinciians.
  • Senior clinicians should ve an enhanced role in shaping the look and feel of the whole spend and programme





[ii] Planning-Programming Budgeting System US DOD. The fundamental ideal behind PPBS was decision making based on explicit criteria of the national interest in defence programs, as opposed to decision making by compromise among various institutional, parochial, or other vested interests in the Defence Department.

Enthoven AC; Wayne Smith K. How Much is Enough? Shaping the Defense Program 1961-1969. USA: RAND Coporation, 2005: 33



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s