There is a very good question to ask at the beginning of every project: “What does success look like?” If it’s not possible to define the required outcome, we should not be doing the project.
The outcome will be necessarily at the end of the project, so the next question is: “What do I need before I can work on the final outcome?” and so on to the beginning of the project, defining deliverables as we go.
I’ve planned clinical trials using a whiteboard and lots of sticky notes bearing the names of deliverables in just this manner, linking them all into a logical network by starting at the end and working forwards.
Traditionally however, contracts awarded to contract research organisations (CROs) are structured on a unitised basis, rather than as a network of deliverables.
For those not well versed in contracting, this means that projects are built up from libraries of activities, each of which has a unit cost attached, and sometimes a standard duration.
CROs with long histories have good experience of how much each activity costs. Bids can thus be generated quickly, and at the bid defence stage much of the horse trading comprises slotting activities in and out of the model.
The business development specialists in CROs will complain at my simplistic description, but I am reflecting my own experience of consulting for sponsors who have asked me to support them when commissioning major studies.
That experience has told me that unitised contracts can create a number of advantages for the CRO and not many for the sponsor; let’s have a look at some of these.
An expensive business
A global CRO confided to me not very long ago that change orders were more profitable than the baseline contract. It’s not too difficult to see why. Once the study is under way the sponsor is heavily committed, because changing the contractor is going to be highly disruptive and expensive.
So there is an incentive for the CRO to inflate the cost of any change. But also, plugging in extra activities is a cheap thing to do for CROs, because most of the negotiation was done at the bid stage.
To put it simply (again), if for example the project is running late it will require more periodic site monitoring visits. The cost per visit was negotiated and agreed at the outset so the sponsor has no option but to accept the agreed unit cost.
It is often not fully appreciated just how expensive this process can be for a sponsor. In my contract management support role, for multiple sponsors, much of my time was taken up with checking claims by CROs for additional costs.
To return to the site visit example, this meant that someone had to check that the visits being claimed had actually happened. But a site visit wasn’t a single activity; as well as the visit itself, there was writing the report, getting it reviewed, and signing it off.
So before approving the cost, all these items had to be verified by someone, and that someone has to be paid to do it. I have known sponsors to be quite sanguine about the expectation that costs will escalate.
Indeed, at the bid stage both parties expect this, but it’s really an elephant in the room. Yes, the contract wording does cater for project change, but without estimates of how much extra cost and lateness are likely.
I suspect this is because sponsors don’t like to face the possibility that anything can go wrong, but deep down they know it usually does. In my example of the unit cost of a periodic site visit, I missed out an important element – that of expenses.
Years ago sponsors became paranoid at the possibility that CROs were ripping them off by inflating expenses (as some of them probably were). So the practice of separating them as ‘pass-through’ costs became the norm.
You can see how this entails more cost for the sponsor, because all the expenses have to be checked against the pass-through element on the invoice.
The sponsor has to employ someone to do that. But the CRO is not going to handle expenses for free either, so there is usually an activity in the budget called ‘administration’ or more commonly ‘project management’, typically 15% of the whole. We end up with double-checking of expenses at double the cost.
Cost per outcome
All this stems from a focus on work done rather than on results achieved. But a sponsor surely isn’t interested in the effort expended by the CRO; only the results have any business benefit.
Indeed, one of the big current themes in the projects fraternity is benefits management. If you can’t define the benefit of any activity, don’t do it.
We could take that a stage further – if we are OK with the expected benefit, do we really care about the split between effort and pass-through expense? Well I don’t. I am, though, critically interested in the package cost for the item of benefit in the plan. In other words, the benchmark is cost per outcome.
This argument of course doesn’t only apply to drug development; it applies to any industry, and not just to R&D either. To my mind, a sponsor who insists on micro-managing a contractor by tracking every task and every expense is making their life unnecessarily difficult.
Yet sponsors have paid me to set up systems to do just that. They had to do this because by the time I came on board they were well down the road of a unitised contract with no chance of going back.
There can be no doubt that healthcare delivery is also an industry, albeit a largely public sector one in the UK. Social media are currently seething with the widely-held belief that the Conservative party intends to privatise the National Health Service.
I don’t propose to get into a political argument here, but I do want to look at how contracts work in the NHS, to see if there is anything for us to learn in pharma.
In reality, outsourcing in the NHS is not new. It has always been obliged to buy in certain services. But the Health and Social Care Act 2012 places the onus on commissioners to invite tenders from any qualified provider.
As I say, nothing new in this, other than the pressure to do it, and the NHS ‘internal market’ goes back decades. Hence there should be reasonably long experience of negotiating contracts.
The said Act is quite heavily reliant on outcomes, although rather vague as to how they will be achieved. Evidence-based clinical practice is not mentioned anywhere. We should expect that the outcomes for both the pharmaceutical R&D contract and the healthcare delivery contract are closely aligned.
Both sectors are striving to be patient-centred. For a clinical trial, the desired outcome is to verify a patient benefit from the medicine. For healthcare delivery, the patient outcome may be assessed via quality of life instruments, although of course more specific measures are also used.
Making care integration work
I have identified some progress towards a more modern type of contract in the NHS, as reported by the King’s Fund. Their Commissioning and Contracting for Integrated Care report highlights “holding providers to account for outcomes” as one of three main ambitions for contract managers.
To make care integration work, the author Rachel Addicott also identifies outcomes as comprising one of the three underlying principles of organisational integration, along with service integration and alignment of currencies, to streamline payments between providers.
There is some evidence that this is not just an academic exercise from a health-related think tank. The Oxfordshire Clinical Commissioning Group has issued a work programme on outcomes-based commissioning that apparently has attracted attention from several other CCGs.
It initially aims at care for mental health and for older people, both identified as having especially heavy demands on services. A wide range of advantages is expected to accrue, but I am particularly taken with the prediction that things will be simpler to implement.
A focus on the outcome will – it is said – mitigate against the high degree of fragmentation that prevails at present. With clinical trials becoming immensely complex, this struck a chord. I’m not suggesting that the more modern contract format will make trial protocols simpler, but it could signal a move away from the need to model every protocol item in the contract.
I also like the idea that the Oxfordshire CCG model releases providers to design services “to suit the needs of patients, under a single contract, with fewer process targets”. We have seen cases of suboptimal clinical trials resulting from unimaginative protocols, such as the notorious TG1412 disaster, and much of the time these have been dictated by sponsors.
Why not cut the brief to the CRO right down to the outcomes required, and ask for their ideas? This approach is very close to the project definition process that I taught to trainees for many years. The problem was that most of them had no scope to do that, and just had to conduct protocols that had already been set in stone.
So what type of contract is the norm among CROs these days? Ronald Openshaw, chief executive of Simbec-Orion, doesn’t see any significant trend towards an outcomes basis. He gives the example of a single study in a relatively uncommon condition, where the sponsor company was certain that it could be carried out with five investigator sites.
Simbec-Orion, with their experience in that therapeutic area, knew that 15 sites would be needed. The gap could not be bridged.
I suggested that both parties were not going to agree on the number of sites, but could agree on the outcome, so that should have been the basis of the negotiation. Openshaw feels it would be ‘wonderful’ if a client were to come to him with just a desired outcome, but nobody ever does.
He points out that a purely outcomes-based contract would not work as a business proposition as all CROs need cash flow – nobody can wait until the end before getting paid. I think this is a given, and contracts would ideally be structured with a mix of outcomes and interim deliverables as milestones.
Where there’s a will there’s a way
There is a way though in which a CRO can buy into the results of a trial, by accepting equity in lieu of some fees.
Sponsors seem to be constrained by the fear of delegating too much responsibility, and indeed they are legally responsible under medicines legislation. A profit-sharing deal should allay some of that fear, and of course is not new in the pharmaceutical industry. But I think it only really affects the financial clauses of the contract.
Sponsors and CROs of course have different objectives. TranScrip Partners supports many pharma companies in the management of R&D projects (among other activities), and senior partner Dr Mark Watling alludes to the CROs’ love affair with the item-of-service contract.
Perhaps they rely on sponsors’ carelessness or naiveté when signing off a contract – anything they have missed immediately triggers a change order, at higher cost. Because both parties are so risk-averse, Watling doesn’t see the unitised contract disappearing.
Maybe it can be improved by building in rigorous timelines, but whatever its structure it depends critically on the quality of project management within both client and contractor (a recurring theme of mine!). Not that CROs are always comfortable with the escalating complexity of contracts these days, made much worse by the increasing influential involvement of legal departments.
Nevertheless, I have not detected a clear refusal by CROs to consider outcomes as contract targets, contrary to my expectation. Any reluctance seems to stem more from sponsors, who are fearful of relying on the experience of CROs who have an appropriate specialism.
It does appear that there are some learnings available from what is going on in the NHS, although healthcare delivery is not the same as R&D. Making things simpler is usually good, and outcomes can help towards that.
The hurdles to be jumped seem to be rather more mental than methodological ones.
With R&D costs continuing to sound alarm bells around the world, it behoves pharmaceutical companies to be creative and explore innovation not just in the drugs themselves, but in the way they are developed.
Les Rose is a freelance clinical science consultant and medical writer