Keeping it Real: Outcomes Based Contracting for Mental And Social Health

ETHOS Issue 26, Nov 2023

An outcomes-based approach to addressing mental health

Be it in high- or low-income countries, mental health is part and parcel of everyday existence. There is no physical health without mental health, and there is no social health without mental health. Social health has to do with one’s ability to perform in the labour market; to do well at school; to raise a healthy baby; to age well without ending up as a significant cost burden to the community.

The Healthy Brains Global Initiative (HBGI) was established to address the global lack of understanding and data around mental health. Such data as exists tends to be from white wealthy communities, outside of which there is a paucity of data. HBGI set out initially to try and fill that gap, raising money to collect the data. It was a struggle because, I think, donors wanted to pay not for data but solutions. HBGI also had no clear operating model for how to gather that data. This is when I started talking to them.

Based on my 23 years of practitioner experience in outcomes- and performance-based contracting, I proposed that if we want to gather this data and learn what needs doing, we must implement a whole new range of mental health services. From these, we can then derive data and learning, that could in turn be shared with others. If we want to deliver these services as effectively as possible, we must attach funding to the outcomes that the programmes deliver for the individuals accessing those services. For each individual, we want to ensure that they achieve the outcomes that they are looking for. The outcomes that we want to contract for are livelihood-linked outcomes: they will be linked to someone's life experience and social wellbeing, while allowing us to track and better understand mental health and its impacts at the same time.

Given the growing global interest in outcomes-based contracting, and since mental health is significantly under-invested globally, this should interest donors. There is a lot of interest among service providers in delivering on outcomes-based contracts, who say: “I deliver big results, pay me for that result and I will do more of it.” There is also a lot of interest among social investors willing to provide the catalytic cash or working capital; but these investors also want returns on their investments. So there needs to be someone at the end of the process who is willing and able to pay for the outcome: each time a child goes back to school; each time a person goes back to work; each time a homeless person is housed. While in some instances this could be government, in many cases, particularly in international development, there is a dearth of people willing to pay for these outcomes. The big institutions are set up to award grants and pay for inputs, not outcomes. At HBGI, we are now looking to establish and run a series of thematic outcomes funds to pool the resources of various philanthropic donors, and then use those funds to contract and pay programmes on the basis of the results that they achieve. We also want to provide technical assistance to governments or others in how to contract for outcomes and how to carry out performance management on such contracts.

An outcomes-based approach is not a cure-all. But it can help show that you can genuinely deliver an impact on individuals’ lives, and ensure that programmes are more likely to achieve results than not. This minimises the resource waste that exists in so many programmes and services, which is particularly important for low-income countries. We know that by attaching the money to the performance, we are going to increase the performance focus within that programme, minimise opportunities for fraud, and more likely increase the outcomes for the individuals accessing these services.

There is no physical health without mental health, and there is no social health without mental health.

We also know that an outcomes-based approach is different from the typical fee-for-service programme which is very much one-size-fits-all: the programme is determined in advance and then people are put through that process. An outcomes-based programme, by design, is an individualised programme. If I'm looking to deliver outcomes for individuals, my service has got to be individualised; I have got to be engaging with that person and understanding their life circumstances to enable them to achieve their outcome. This means that the service itself becomes very localised and contextualised. This also means the programme is not imposing some sort of cultural understanding of mental health. Mental health varies from culture to culture in the way it's perceived, described or experienced. You need to create a service that can reflect and respond to that, rather than imposing some preconceived notion or counselling process that everybody has to squeeze through. This is why mental health is ideally suited to outcomes-based approaches.

Mental health varies from culture to culture in the way it's perceived, described or experienced. You need to create a service that can reflect and respond to that.

Enabling and evaluating success

In outcomes contracts, there is always a question around the metrics: or what it is we are going to pay for. We need the outcomes, as far as possible, to be very clear, tangible and concrete. It should be relevant to the individual in their life, and also something that we can measure and verify, because we want to attach a payment to it.

There are a number of established and internationally accepted measures of mental health and mental wellbeing, but these are generally self-reported. And they may not get to the heart of what mental health means to an individual and their ability to function and engage with their social life. We have seen during the COVID pandemic how social isolation can lead to a breakdown in the social network, which in turn leads to a deterioration of mental health. We have seen on the back of COVID, a range of social impacts that are clearly related to mental health, such as among young people an increase in self-harm, an increase in suicide, an increase in failure at school, and so on.

So we need to recognise that socialisation—enabling individuals to return to their social environments—is in itself going to be part of the desired outcome of such programmes. We can measure this in some ways: such as whether or not they return to school, work or social activities. On many of our programmes, we look to define these real-life outcomes, and then measure mental health alongside it, and over time learn a little bit more about their interplay. This will help in other ways: preventing further deterioration or further harm, for instance, lends itself clearly to outcomes-based models.

One way to approach evaluation is to look at measuring outcomes across a population. We can start with a baseline, say of current levels of school attendance, self-harm, suicide ideation and how children are behaving and performing in schools. And then against these baselines, we can determine what outcomes we might be looking to achieve—what difference we might be looking to make—over a 12-, 24- or 36-month period—and then attach payments to these goals. In the particular context of a school population, you have a captive group and could introduce some regular measurement processes to assess social wellbeing within that environment over time, which we can attach payment to.

We talk about ‘incentivising’, but such programmes do oblige the service provider who is aiming to achieve the outcomes to go in and listen to the young people, engage with them, and try to understand exactly what the core issues are. We might then find them starting to innovate around youth mentors, for instance, to enable engagement and real outcomes for the young people, such as coming up with appropriate warning signs to identify when an individual is in distress.

Letting Beneficiaries Choose Their Desired Contracted Outcomes

An impact bond in Greater Manchester, in the UK, sought to prevent homelessness among high-risk young people. The associated contract had two fixed outcomes, with a unit price attached for each, as well as three more variable outcomes—A, B and C—each also with a price attached. The young person themselves decide what those outcomes are. From a menu of options, they choose the three that are important to them, and how to prioritise them.

The actual individual beneficiaries on the programme determine what outcomes they want from the service delivered to them. This empowers them in the process, returning to young people a sense of agency over their circumstances.

But you cannot design a programme in Geneva and impose that in Ghana to address such issues. Solutions have to be developed on the ground, around the individuals one is trying to help and their social circumstances. The challenge is how to do this and at the same time ensure that that intervention is actually going to deliver an outcome for them.

We should not fall into the trap of coming up with one model, such as for warning signs of distress, that is then imposed across different schools or other contexts. Instead, we must create the conditions for communities, such as schools, to identify within their population what it would mean to increase their wellbeing over the next 12, 24 or 36 months. It could look like increased social cohesion and social strength, better communication and so on: then we can have other sub-measures, such as warning signs, that could then be implemented according to what has been identified as important to them.

Bottom-up approaches and personalised, context sensitive solutions require a structure, to ensure that the performance is real, accountable and delivers what we want.

That said, bottom-up approaches and personalised, context-sensitive solutions require a structure, to ensure that the performance is real, accountable, and delivers what we want; that the funds are spent in the right way, with clear metrics tied to what it is we are looking to pay for. These will be a mix of outputs and outcomes. There will also need to be strong performance management systems in place. We need to ensure that we have visibility over what is being delivered, and have a way of tracking, reporting on and reviewing what is going on, on an ongoing basis.

Getting good at outcomes contracting and performance management

Performance management is not contract management. It is not a bureaucratic process of counting pieces of paper, but an ongoing engagement with the programme, and with the experiences of the community one is trying to help, to ensure that as many of them as possible achieve the outcomes we want.

In Afghanistan after 2001, there were no health services in the country. All the basic essential health services were contracted out to international NGOs, who were on budget reimbursement contracts with a very tight specification of what they had to deliver per clinic. When I started working there in 2017, we looked to shift all these services to performance-based contracts. To begin with, I engaged with 50 different stakeholders across government and funders and asked them what success looked like—and received 50 different answers. There was no cohesive definition of success. The crucial intervention was to pull them back and get everyone to agree on what successful performance would look like—such as better health services delivered to more people— which we could then use to drive the design of the performance contracts and how these programmes would be performance managed.

This is what we need to get public servants better at: developing a good definition of what success looks like. To ensure that the definition is relevant, we need to move from bureaucratic definitions to ones that are real for the individuals concerned. Indeed, let’s engage those individuals in developing those definitions. We need to determine what success looks like, and engage the people to whom it matters, from a range of different disciplines, in defining that success. Then we use that to cohere efforts around and structure the response that comes after: if that is what success looks like, then what are the steps we need to take to achieve that? And if those are the steps, what are the skillsets that we currently have, mapped against what we need? And then how do we upskill towards that? We build capacity in this way.

If you can shift the conversation to one around a clear definition of success, you can shift a culture within a system to one of performance, because everyone then begins talking about the achievement of that success. Within a properly implemented outcomes contracting environment, you also generate a lot of rich data, because you are constantly tracking, recording and reviewing what's going on. So rather than doing impact evaluation, which is disconnected from delivery, you embed a regular ongoing activity to review what's going on. You focus on how you are on track to achieve this given level of success, based on what has been collectively defined, and you change the language that people use.

Success Determined by Those Who Live It

HBGI has a very active Lived Experience Council—a group of people with intimate lived experience of poor mental health—whom we seek input from each time we reach a relevant touchpoint.

Likewise, we are about to start a project with YMCA Global, where we will spend 18 months training a group of young people to speak to other young people, to ask: What is concerning you right now and what would be a meaningful outcome? These findings can then be used to inform future definitions of success as young people see it.

Some well-meaning service providers may have been delivering a service for decades in a particular way. They push through a fixed process with a paternalistic view of the people they are supposed to be serving or working with, considering themselves the experts who know what’s best. Many of these organisations do not even know how many people they are working with at any one time, because they have not been tracking the data, nor their progress towards the outcomes that the individuals need to achieve. But once they are led to think about these questions, they begin to change their mindsets away from one-off performance measures. Most organisations feel empowered by this and change the way they operate within their contract. They begin to step back from their paternalistic view and to recognise the people they are working with as assets.

Within a properly implemented outcomes contracting environment, you will also generate a lot of rich data, because you are constantly tracking, recording and reviewing what's going on.

Several factors underlie success in using an outcomes-based approach. First, do not do outcomes contracting just for the sake of it. Instead, you need to start off with an understanding of what you are looking to achieve, and then find the best tool to get you there. Second, it is important not to overcomplicate this. Often, when everybody weighs in, you can end up with a very complex beast of a solution. Make sure to keep it real for the individuals you are trying to achieve success for. Keep that definition of success as clear and simple as possible. Keep bringing everybody back to that simple definition and strip out the complications, so you can have lean solutions.

If you have people talking and thinking about performance and success, and about what's important for the person on the frontline, then you are taking a step in the right direction.

ABOUT THE AUTHOR

Richard Johnson is Chief Executive of the Healthy Brains Global Initiative (HBGI), which looks to address the global gap in mental health services and understanding by pooling philanthropy to contract and pay for outcomes, and to provide technical assistance to governments and other organisations. He has chaired eleven Social Impact Bonds, including a Development Impact Bond in Palestine, and has a new youth employment programme starting soon in South Africa. He has run for-profit organisations delivering large scale outcomes contracts in the UK, and has served as a World Bank Senior Consultant, and a Senior Advisor at the Global Fund.

Back to ETHOS Issue page