Virtual Roundtable: Financing of Primary Health Care and the Central Role and Important Limitations of Capitation

FRIEDEN: Thanks so much. And thanks for joining us. I'm really looking forward to this discussion. Dr. Kara Hanson has led important work assessing financing in health care. And you might say: Why is financing in health care important? Well, to begin with, health care is a huge part of our economies. In addition, it's a huge part of what determines productivity. So, health and economics are tightly related. For decades, we've been trying to strengthen primary health care. And other than a few countries, we haven't made much progress. What Dr. Hanson and her group have looked at is the essential components of financing and governance. This is really what determines whether primary health care both gets the money it needs and gets the structure it needs to spend that money well. So, spending more and spending better in primary health care. And the emerging consensus is that capitation is of fundamental importance to improving access, quality, financing efficiency. Capitation is the simple approach of saying that a health care system will be paid per patient or family per period of time, whether it's a month or quarter or a year. So, I look forward to learning more.

And without further ado–I will say, I didn't do an adequate introduction to Dr. Hanson. But she's very distinguished and has written a series of terrific articles and done very important original work as well as analysis, summary of health care financing. She's really one of the world's top experts in health care financing. So, Kara, over to you.

HANSON: Thank you very much, Tom, for that very kind introduction. So, as Tom alluded, I was involved with a group that put together the Lancet Global Health Commission on financing primary health care that was actually published about eighteen months ago. And for–during the last eighteen months, I have been involved in a number of discussions at the global level, at the country level, invited to contribute to national financing strategies. And so, this is a really nice opportunity to share that work and how it's developed and emerged since the commission with this very distinguished group of participants.

So, first of all, I just wanted to say a bit about what we mean by primary health care, because even that is contested. So those of you who remember Alma-Ata in the 1970s and Alma-Ata–the follow-up declaration on primary health care in 2018, the primary health care approach is this broad approach to organizing services at the first contact level. In its broadest sense, it is this combination of primary care, as we might know it from North America, the first contact with a health care provider, but it also includes multisectoral action and supportive health. So, the water sanitation sector, the education sector, and it also includes community approaches: the things that we might recognize as community health workers. What I'll say is that our report really did focus on the primary care elements of that. And that's really because that's where the evidence lies. There's just–there's lots of evidence about multisectoral action, about community-based delivery modes, but very little out there in terms of the financing arrangements to facilitate those. That's beginning to develop now, but at the time we were looking at that work there was not much there. So, we really focused on what are the different elements of good ways to finance primary care.

Then the next thing we set out to do was understand what's really the challenge. And the challenge–I think we'd recognize from all of our health systems, which is that in general, priority to health is limited. In many countries, governments are spending, you know, two or three percent of GDP on health. It's really not nearly enough to assure universal access to a basic set of services. But within that health budget, hospitals are prioritized over the primary care level. They're much more appealing from a political perspective. They're shiny, they're big, they're equipment–full of shiny equipment, and they're also much more supported by professionals. Right, because your powerful professionals tend to be the highly qualified clinicians who tend to advocate for resources to go to their patch, and there's specialty. So, lots of those political reasons, the primary care level is often really neglected. And I would say that I think we're likely to see that get worse as populations grow richer, grow more exposed to understanding and knowledge about what the possibilities of health care are. There's a real risk that people's expectations drive a further focus on hospital care. So that's something that we really trying to tackle through appropriate financing arrangements for primary care.

So, the consequence of those pressures that we have not enough money going to primary health care, that the resources that are allocated to primary health care often don't make it there, they get kind of siphoned off, whether that's through your legitimate or illegitimate means. They often get reallocated when there's shortages at higher levels of the system. And that means that they just don't reach those frontline providers. And then that PHC funding that reaches those frontline providers is–it's fragmented. And I'll come back to that, because I think it's a really important thing in the developing world, this fragmentation of funding flows. But it's also–it's governed by very rigid budget rules that leave providers very limited autonomy to make decisions about where to focus their efforts, and where to use their resources, and how to solve the problems that they're facing. They're on the spot, which all leads to inefficiencies. And so, the main story, I guess, of our commission is that there's, you know, as Tom kindly flagged for you, that countries need to spend more and spend better on primary health care. So that means allocate more money to primary health care, make sure that those allocations are equitable, and that they reach the providers intact, and that they come with well-aligned incentives for providers to be able to manage priorities locally. And those we would say, are all kind of elements of an approach to financing primary health care that's both centered on patients in terms of service delivery and centered on patients in terms of the arrangements for financing.

So, what we did in this report was we first conducted a landscaping. We used WHO data to confirm that there are very, very low levels of primary health care spending in the developing world. So, we estimated that governments are spending about three dollars per capita in low-income countries, and sixteen dollars per capita on primary health care in lower-middle-income countries. And we didn't want to do our own price tag exercise, there's lots of those around, I can just tell you that those are orders of magnitude too small to be providing a basic primary care package of services, which at best estimates is probably sixty or seventy dollars per capita. So, we're very, very far away from that.

A second finding is that out-of-pocket payments are still really important for primary care. And medicines are a big share of out-of-pocket payments. So those are, those are two problems, right? So, one of them is that we know that out-of-pocket payments–payments that people make at the time they seek care–are a real burden to households, they're not covered by insurance, they have to pay that first dollar. And they have to pay it at a time when they're most vulnerable, when they're actually ill, or someone in their household is ill. And we talk about the impact of these payments as being catastrophic. So, we think about: What is a catastrophic level of payments for a family? It might be if they spend, you know, ten or twenty or thirty percent of their entire household resources on health care in a period of time. And that happens quite frequently. We often associate those kinds of episodes with tertiary care, major surgical interventions, or cancer, really serious conditions. But actually, the evidence is quite clear that the accumulation of small levels of use, like you might find if you had a chronic condition, like hypertension, or diabetes, or a range of other chronic conditions can very quickly get to that catastrophic level. And so, there's lots of good reasons for thinking about where we target our insurance arrangements or our risk sharing arrangements.

The other kind of element of this picture was the fact that the level of spending actually matters. So, if you just correlate the level of government health spending on primary health care against coverage of key services, there's a strong positive association. There's no question that spending more buys you better outcomes. What I would say is that the distribution around that scatterplot is quite wide. So, for any given level of spending, there's quite a variety of levels of achievement. And that to me, as an economist, suggests that some countries are making better use of their money than others. And we should be looking at how the financing arrangements can help countries to make better use of the money that they're spending: hence the “spend better” element of our mantra.

So, our deliberations as a commission–our commission is made up of policymakers from national governments, it was made from global experts, it was made from representatives of some of the leading global health institutions. And we deliberated over the course of a year, all during COVID. It was a long and drawn-out process made more complex by deliberating at a distance. But what we did over that period of time was work through the three main financing functions. So, we talked about how money is raised for health, how money is allocated for primary health care, and then how it's paid to providers. So, when Tom talked about capitation, which will come to in a moment, but that's one element of the health financing system. So that's one way of moving money from a pool, a bundle, a bucket of money to healthcare providers that carries incentives with it.

So, we looked across these three main functions–I won't talk very much about revenue mobilization, I think that there's lots of good reasons why raising money through compulsory insurance or through taxation is the most efficient way to raise money for health. But that's a kind of a different line of argument. What we really focused on was: What are the range of tools that governments have at their disposal for allocating money to primary health care? And we spent an inordinate amount of time thinking about budgets. That might sound a little bit parochial, but actually budgets are really important, budgets are the embodiment of national priorities. And they set out the plumbing and the way that money moves through the system to reach health care providers. So, we talked a lot about the sorts of policy levers that exist both with budget formulation and budget execution, to shape an increase in the share that can go to primary health care.

But we also talked quite a lot about service delivery arrangements as a way of shaping financial flows as well. So, one way of thinking about this is, if you have a new way of delivering primary health care, suppose you've got–if you're Turkey, and you develop, for the first time, family doctors, you create a new corridor. That becomes something you've got to budget for, it pulls money just by virtue of being new and shiny, it pulls money to the primary care level. Likewise, if you're a community health worker network in Ethiopia, the creation of that corridor creates an infrastructure that will–is capable of pulling resources out to the primary health care level. And there's this really intimate relationship between financing arrangements and delivery arrangements, because financing can stop delivery from being successful, and it can also enable and facilitate new innovative delivery models. So, if we think about one of the newest innovations that's being discussed in sub-Saharan Africa, is the idea of primary care networks where facilities get together with some shared functions and then some decentralized functions. And that, to the extent that financing arrangements can support that, they can be really important in the success of new models. So, we've talked a lot about budget resource allocation.

The next thing we deliberated about was how providers are paid. And in most countries, governments allocate money to a provider through input-based budget. So just a budget that says how much you have to spend on salaries, on drugs, on electricity, on different line items. And that's a–it's quite an easy way of moving money to providers, but it's also very inflexible and allows providers very little autonomy in making decisions about how to configure their services. You could also think about it as funding the supply side rather than funding services. So, just basically funds infrastructure, it doesn't fund services. So, we talked a lot about how you could move from that kind of low-powered incentive structure to one that really captures what we want from primary care. And that's why capitation was such a strong recommendation from the commission.

So, capitation is where providers receive a fixed amount per individual that's registered with them per period of time: per quarter or per year. It has four merits, I think. One is that it starts from a principle of equality. So, your starting position is that every person is allocated an equal amount of money. The second is that you can then adjust that for needs. You could say, “Well, in my population that are registered with this provider, there are some that have higher needs than others.” So, if you're–so you can adjust that by risk factors like age or sex. It also is the only payment mechanism that really rewards providers for looking after their populations, for preventing and promoting health.

And finally, unlike other payment mechanisms, it gives them a very predictable revenue source over a period of time. So, they're able to plan the resource use and to plan how they're going to configure and deliver their services because they know how much income they're going to have in each quarter.

Now, capitation on its own has some problems as well. It can lead to undertreatment, providers can skimp on services. And so, in our pathway to a more strategic way of paying providers, we wanted to have capitation at the core but to supplement or complement that with other provider payment mechanisms where there was a special need. So, if you really had a service that was high priority and not being delivered, you could incentivize it through performance targets and pay for performance. If you were dealing with very remote rural providers, you might provide a basic rural allowance just to cover the fixed costs of being located in a rural area. So, the model we set out was one where you've got capitation at the core of a provider payment system. But all around it, you make use of the best tools you have to deal with some of the downsides of capitation, where we sketched out the class system requirements for a more strategic way of providing–of paying providers.

I just want to, before I go on to my last set of points, I just want to comment on fragmentation, because I don't know if there are people in this audience who are involved in some of the big donor agency or global health initiatives that often fund primary care programs in low- and middle-income countries. But one of the consequences of lots of vertical programs and off-budget programs is that providers can often find themselves receiving lots of payments from different sources. And those different sources will often specify a service that they're supposed to pay for, a population they're supposed to pay for, a set of inputs that are only for their programs. And I think this can be very, very difficult for primary care providers to manage. This fragmentation is confusing, doesn't send them strong signals about what's a priority, where should they put their effort. And worse than that, some of the things that we do as a global community to try to address the needs of certain groups is to introduce more of these schemes. So, we introduce a new health insurance scheme for this target group, or we introduce a new vertical program for this problem. We really contribute to this proliferation, and I think that it has really negative consequences for the efficiency with which funds are used at the provider level amongst another whole set of wider consequences.

So, two last points. So one is that we also, in our commission, talked a lot about political economy. And we looked at some of the countries that had made some–had some transformations in their health financing systems or their PHC financing systems. And these transformations often arose at a time of a political opportunity, often together with some economic opportunity. But sometimes they arose through some crisis as well. So being savvy to the political landscape and being able to do two things. One is to see politics not just as a constraint, but an opportunity. But also, have your technical strategies ready so that when the political opening happens, you're ready to go. If you've got a political leader who wants a solution to put in their campaign, be ready to go with a well-designed solution, seemed to be a really good way to have a–we called a “politically informed technical strategy”.

So where did we land in the end? We ended up trying to describe the attributes of a well-functioning primary health care financing system, and we describe it as having four elements. One is the real importance of having public resources at the core. And so that public resources and pooled schemes, so insurance schemes, social health insurance schemes, or big tax schemes should be covering primary health care first, and that should be the starting point for kind of a gradual, progressive universalism. This idea that pooled funds should cover PHC first is really a recognition of the enormous gains to protecting households from that accumulation of primary health care payments, out-of-pocket payments, especially for drugs that can be impoverishing at the household level.

We argued that resources should always be allocated equitably, and that they really need to be protected until they get to providers. That's the beginning of a set of budgetary and public financial management and other guardrails to stop those resources from being diverted from their final destination. And then that these providers should be paid through a system that places capitation at the core, but that also incorporates you know–as systems become more sophisticated, information becomes more widely available –that capitation can then be supplemented with other provider payment mechanisms.

We had a few other recommendations that came from the report, they were slightly more prosaic. I think an important one was the idea that primary health care financing is an all-of-government initiative. It means ministers of health negotiating carefully with ministers of finance to show why they need to spend more on primary health care and how they're going to spend it well. But that also within countries–whether you have large, kind of devolved authorities–that it’s important also to be negotiating with those devolved authorities, because that's where those resource allocation decisions are often made. And they're often very difficult to influence from the central level. But I'll stop there. And I'm happy to hear about which bits of that you'd like to carry on discussing further, Tom.

FRIEDEN: Great. Well, let me start with some definitional issues. Sometimes I think of your approach as “capitation++”: capitation at the core, but plus for basic infrastructure apparent–especially in rural or urban challenged areas–and plus for well-designed incentives, which we'll get to later. We also sometimes talk of total cost-of-care models, like the ACO models in the U.S. where, in a way, you are less worried about the indicators, because it's harder to cheat, basically. Either someone's hospitalized or not, and if you can incorporate the hospitals in the format, or at least capture savings from prevented hospitalizations, to fund primary care more, you have a win-win, where primary care is preventing more disease, preventing hospitalizations, getting more resources. But capturing that is very challenging and not part of just capitation alone. So, can you talk a little bit about capitation versus total cost-of-care?

HANSON: So, I think what you mean by total cost-of-care is that you've got a much greater integration across the primary care and the hospital level. So, in some sense, there's a common budget, shared budget, shared responsibility for spending well, and then a shared gain for managing that well. So, I think that the real essence of that is being is ultimately–the general point is: How do you make sure that your systems at different levels of care aren't fighting against each other, and at best, are all pulling in the same direction?

And so, I think that total cost-of-care model offers the opportunity to do that, to present a set of incentives that are about keeping people to be managed at the most appropriate level of care. And to do that, in this case, by if not an organizational integration, certainly by a financial integration across that boundary between primary care and hospital care. So, I think it sounds like–it sounds very promising. I think it also relies on lots of really good organizational arrangements, right? That whole kind of managing of that referral interface is one that just requires, you know, some system strengths and information systems, and then, you know, audit systems as well to make sure that all the promised alignment of incentives is actually paying off.

FRIEDEN: Yeah. It's very challenging, because we know for some hospitals, there are unnecessary hospitalizations and there are avoidable hospitalizations. And if you could reduce both of those, you could use some of that money to fund primary health care to avoid those avoidable hospitalizations. But that's not necessarily inherent in the capitation approach. And in many parts of the world–although there are unnecessary and avoidable hospitalizations–there are also many people who need hospitalization who don't get it. So, it's not so clear that you'll have just pure cost saving.

HANSON: That’s right, you can't solve it within the system. You need also think about how you use those resources to build improved access to the system. I think also you have to think really carefully about the design of those incentives at the hospital level. We touched very briefly on arrangements that are being piloted in China, where you've got have hospitals being told that they are being directed to support the primary health care level. And that's kind of all fine, except that the fundamental incentives at the hospital level haven't changed. And so, they still are facing, you know, the incentive to draw patients into hospital care and then to over-prescribe medicines and procedures. And so, it's a very–it's a whole system issue, I think, for it to work well. And you can't–these changes in the design system often have unintended consequences.

FRIEDEN: That gets us to the political questions. Let's address that for a minute. In the U.S., we have a very good health care system in Kaiser Permanente. And we've had effort after effort to revise our health care system, and it just hasn't been possible to essentially replicate that model nationally. And that model is a total cost-of-care model. They have long term members, and if they prevent hospitalizations, their financial situation is better. In most of the health care system, if we don't prevent hospitalizations, then the health care system has better financing. And as your commission really pointed out, if we want to strengthen primary health care, we have to fix the financial structure of the system. And if we want to do that, we have to fix the politics and governance. So, what are some ways that all of us who work around the world could think about driving political, financing, economic change, to support more and better money for primary health care?

HANSON: Oh, Tom, it's like, that's like the million-dollar question, isn't it? I wouldn't–I would never dream of commenting on the politics of your own country's system. I can only imagine that it's one that's got a considerable amount of vested interests that makes it very difficult to budget off its equilibrium, right? Because many actors are really quite comfortable in this system as it is. I think we've looked at the role of social movements. If you look at some of the big changes in Brazil, for example–if you go back, not into the most recent past, but into the mid-2000s–when there was a massive commitment to primary health care, a reformulation of their whole delivery system and a financing system to back that up. That came I think, through a combination of social movements, of favorable macroeconomic consequences, which also made it easier to be at the margins, allocating more to lower levels of care. So, it's a little bit hard to feel a bit–not a bit discouraged now as we look at the real financial and fiscal challenges that many countries are facing and thinking: How can we use that to be a jumping–stepping stone onto a system that is more efficient, and directs people to–seeking care at the lowest possible, the most appropriate level?

FRIEDEN: We've seen, you know, Canada apparently strengthened its primary health care system for this reason, they knew as they expanded coverage universally, that they didn't have enough money. And so they made primary health care practitioners the gatekeepers, really, as a way of both improving quality access and controlling costs.

HANSON: That gatekeeping function is really important. And I think in the United Kingdom, where I live and work, you know, those GPs are the gatekeepers to the system, you cannot get to see a higher level of care without going through your GP. And I think with the current kind of crises notwithstanding, I think over time it's been shown to be a really effective way of keeping people being managed at the right level. I think it's kind of all slowing down now as the system is under resourced at the macro level. But that gatekeeping function and then the referral function, so how do you facilitate referrals as well.

Every country I know has a referral system on paper. Most of them are observed in the breach, right? That people will go first to the level of care they wish to seek care at. And the sort of tools that we've tried to apply–things like bypassing fees, you pay more if you don't go through a referral channel–they don't really deter much use because people are very determined to go where they want to go. What it really does mean is that you end up with the most vulnerable groups not able to afford that bypassing and then being left to the most neglected parts of the health care system.

FRIEDEN: I'm going to ask three more questions, then we'll open it up to the audience for any questions they have. The first has to do with incentives within the capitation world. We know that incentives can sometimes result in distortions in the care given or some of them may be fudged. One of the potential approaches for incentives is to have just a start of: Did you see the patients in your panel as a first indicator? Maybe fifty percent in the first year, seventy percent in the second, eighty percent in the third or whatever it is, we're starting from whatever the baseline is. But what are the ways to have reasonable incentives? And related to that, what are the ways to not result in just profusion of paperwork?

HANSON: So, I guess I can say a couple of things. One is there's performance indicators and incentives. I think it's worth kind of slightly distinguishing this. So, the performance indicators, what are we measuring in order to reward good performance? And so there is a good question: What are the right indicators to use to measure performance? I think one thing we also note is that providers are–they're pretty candy, right? They respond–they will do what they're paid for. And so–and they're always a little bit ahead of the payers. And so, whatever reward system is there, it needs to be quite nimble and quite easily adjusted, because you basically will–you'll get what you pay for, but what you want will change over time. And so being able to adjust both the performance indicators and the incentives attached to them.

So, an easy one is, you know, where do you set your performance threshold, right? If you set it at sixty percent the first year, you know, everyone's going to get pretty close to sixty percent. So, you have to be able to raise it in future years as providers get better able to organize themselves around achieving those targets. So, I guess–so I think carefully is the question. How do you design your incentives and your performance indicators? You just design it really carefully to make sure that you–because they will always have unintended consequences? Always. There's no–there is no perfect incentive scheme, they will always lead to behavior which is, you know, at odds in some way with what you're seeking. The judgment is that that's still better than an alternative, which is maybe an incentive to not do very much at all.

FRIEDEN: I guess this, related to that, is the issue of electronic systems, that it's going to be very difficult to have meaningful incentives without reliable digital health care.

HANSON: Absolutely. I know when countries have introduced pay-for-performance systems without those digital systems, it has come at an enormous administrative cost, because you have to both–you have to collect the data and then you have to verify the data. And we know from the generation of field studies–field experiments by the World Bank, that the administrative costs of running a pay-for-performance scheme are considerable. So, I think electronic health records are definitely going to be the way things are improved. And we see that–we see amazing things happening now in terms of the digital–the countries that have introduced massive EHR systems–usually, with an expansion of their financial protection scheme. So, the one I'm thinking of is Indonesia. Indonesia has got 250 million people, it's almost the size of the U.S., eighty percent at least now are covered by their social health insurance system. And they have tens of thousands of providers who are billing electronically. And that is creating this–it is creating an electronic infrastructure that allows you to then work with the billing system to set an appropriate set of performance indicators and incentives to meet them. And so, I'm confident that as these large-scale financial protection systems get going, they expand in coverage, that they will help to create the infrastructure that's needed for fine tuning systems.

FRIEDEN: Great. Let's talk about prioritization. This, I think, is a real challenge, because in many countries, the idea is we're going to do everything for everyone. And in the lower-income countries, that means de facto rationing, that we won't have medications for you, that we will do a little bit of everything for the people who are very adamant about getting care, but for most people, it's going to be very difficult. Are there examples or are there methodologies to prioritize in a way that is a transparent, inclusive planning process and allows you to say: “For this year, or this term, we're going to focus on these conditions, and we recognize our other priorities, but we're going to get to them next.” I've seen that be very difficult for countries because you want to continue the fiction that we're doing everything for everyone.

HANSON: You remind me of what's a very common–so the system for doing that is through cost effectiveness analysis, right? So it is to look at all the interventions that you would wish to provide, to look at the costs, attached to them, the benefits that they generate in terms of DALYs averted or qualities produced and you rank them and you say: “Okay, the first thing we're going to do is take the most cost-effective intervention and deliver it to the entire population that needs it, and we'll see what money's left, and we'll go to the next one.” And so, you sort of go through this ordered set of priorities, where they're ranked by their cost effectiveness. A many, many, many countries have gone through exercises of costing their basic package, basically doing that. So, using cost effectiveness analysis and information on costs and target populations. And it always goes a bit like this: “Okay, so we costed our package, it costs 320 dollars per person, but we've got fifty.” And then, and then the exercise kind of collapses. Where, of course, the answer is to say: “Well do the best job that you can with that fifty.” So, generate the greatest health benefits that are possible with that fifty dollars that you have to spend, so that you are buying–your best buy is you're buying those interventions that will generate you the greatest health gain for every dollar spent.

And that is politically painful, right, to actually–because it's the explicitness of that rationing that I think is so painful. And so, countries tend to kind of fudge it a little bit. But that is, of course, how you could take a gradual developmental approach to expanding benefits as resources become more available. So, you can either–once you–so, my colleague WHO, Joe Kutzin sort of calls this a “unified benefits framework” that says if you can figure out the money that you've got, you figure out which populations it's covering, and if you can systematically use incremental money to expand population coverage, and then to expand coverage beyond the primary care level up to higher levels of care, then you can be sure that you are using the resources that you have at your disposal to generate the greatest health impact. But it's politically really hard. And I think it is that explicitness of saying: “Well, I'm sorry, we can't afford this.” I think people would rather fall on other covert methods of rationing, like through waiting lists, or through having to buy your drugs outside the facility. Rather than say, this is actually what our system can afford.

FRIEDEN: It's discouraging in part because if systems were upfront about what they can't afford to do, they might be able to get more money from the public and the government. I have one last question–we'll open it up for questions. If there aren't questions, I have five other questions to ask, but raise your hand. If so, we'll come on with instructions in a moment, but just give us a quick sense of what remains to be done, both in terms of what we need to learn and what we need to do to improve financing of primary health care.

HANSON: So, I think that the first agenda is that one that you alluded to, when your comments about that total cost of care model. It’s this–the importance of aligning financing arrangements across levels of care. And I think that is, and we looked, while acknowledging the need to align across levels of care, we really focused just on that primary care level. And that's not enough. You can create really distorting incentives by focusing only on one care level. I think there's a really interesting, both a knowledge and implementation agenda around understanding how do you best achieve that smooth transition across this financing–the financing arrangements across those institutional organizational boundaries.

Another thing I think we didn't crack and our commission, which I think is really important, is different ways of covering medicines in primary health care. So, in a whole set of countries across the world, at least fifty percent of out-of-pocket payments is on medicines. And it's interesting that the essential drugs world has really not thought about different ways of providing publicly-financed medicines, whether that's through pharmacies with prescriptions and billing centrally, or whether it's through dispensing health facilities, or I think there must be a number of different ways in which this can be done. But there's not a big evidence base on alternative ways of covering medicines through primary health care packages, other than the dispensing health facility, which is the other one that we're probably all quite familiar with.

I think the other one that I'm just really stymied by is: What do we do about primary health care providers, particularly about private primary health care providers? So, I'm thinking about places where, in large cities in low- and middle-income countries, where there is a proliferation of partly-qualified, semi-qualified private health care providers that are providing services for payment. Some of those services are dangerous. Often they are really inefficient for patients because patients end up following this trajectory of care, seeking care from lots of different providers, repeating tests, repeating diagnostics, spending a lot more out-of-pocket than they need to. And I think some way of introducing some greater regulation, but also just some kind of rationalization of that healthcare market so that users can navigate it more effectively is really important. And I have not seen good models for managing primary providers at scale. There's lots of experiments, kind of donor-funded projects that show that you can contract private providers to do this or that, but I've not really seen anything that I think can operate at scale across a wide range of health conditions.

FRIEDEN: And the answer may be what rich countries have done over the last hundred years: you professionalize, you regulate, and you incorporate into the delivery system.

HANSON: Precisely. And what we do know, if you look across the OECD, what's really important in determining whether a country has universal health coverage or not, is the money, right? It's whether–it's public funds. It's not the public or private mix of the provider market. So, you have places like France, where you've got a predominance of private providers, providing primary care and hospital care, and that works just fine. There's lots of different combinations on the provider side that are perfectly well suited in a well-regulated environment with well-managed contracting facilities. But it's really important that the money is public money, because without that, you end up with all kinds of problems of risk selection.

FRIEDEN: Great. Let me turn it to CFR just to repeat the instructions for how to raise your hand. And then I will keep asking questions if there aren't any more from the audience.

OPERATOR: As a reminder, to ask a question, please click on the raise hand icon on your Zoom window. When you're called on, please accept the unmute now prompt, then proceed with your name and affiliation, followed by your question. If you'd like to view the roster of registered participants for this meeting, please click the link in the Zoom chat.

FRIEDEN: So, let's go a little bit further on the issue of PHC supplies. What we see in so many countries is they may be able to hire staff, they may even be able to pay staff, but they often don't have the basics: medications, equipment, maintenance contracts. And that leads to a total lack of confidence and then the leapfrogging to the hospitals. Do you see models that would help address that?

HANSON: I think that getting more money into the system, I think that's where the sort of the prioritizing primary health care becomes so important. Often, there's not that much give in the health care worker budget. But there is a need to be able to finance the basics in terms of medical supplies and drugs and equipment. I guess we're talking about lower-middle and middle-income countries, there often is enough money to get their supplies there. In the lowest income countries, I think that's where countries are more likely to be relying on some of the global funds for that. And what's really important is that those resources are moved in ways that don't distort the system. And there's a lot of kind of thinking about how do you–if you do have suitable donor funding for those sorts of supplies: How do you arrange your transition arrangements once countries do get richer to make sure that the benefits of those external funds continue to be that the coverage achieved through those external funds continues to be achieved and sustained without recreating waste, actually?

FRIEDEN: Next, you discussed the issue of primary care networks, and we've discussed gatekeeping functions, and the need for referral interface. This is pretty broken in a lot of countries, rich and poor. What are some best practices and ways forward, what's the latest in primary care network development and what works?

HANSON: So good questions, not an area that I am totally up to date with. I know there are some interesting experiments going on with new models of care in Ghana and in Kenya, where they are developing these primary care network models and developing the financing arrangements to support them. I think it's probably a little bit too early to say, and I think they are very promising because they offer that–those kind of economies of scale and scope, right? I think if you have those together with some of the kind of newer technologies and opportunities of telemedicine and things that can reduce the burdens of travel on patients, I think there's some good prospects for those models to be very effective in providing better quality primary care.

FRIEDEN: That relates to the issue of gatekeeping. How does gatekeeping get to be politically palatable? I think, you know, this idea that we provide everything for everyone and you can go anywhere results in the migration to hospitals. It's not an irrational choice on the part of patients, because hospitals are more likely to have diagnostics, more likely to have skilled staff, more likely to have treatment. And if you need hospitalization, you'll be there. So how could a country that wants to strengthen the gatekeeping function effectively do it?

HANSON: Well, there’s two issues. I think one is, unless you increase confidence in your primary care level, then no amount of process is going to stop people from going where they want to go. So, I think it really is an issue of improving the quality and the allocation of resources to the primary care level so that people use–that is the place of choice for most uncomplicated, kind of first line illness. I think the other tool that governments have at their disposal, there are some service design decisions that can be taken in different ways, right? You could–you can choose to locate your HIV testing facilities at the hospital level or at the primary care level. Maybe if you locate them at the lower level, you will still–you’ll be able to both attract resources and equipment and staff to that level, but also keep patients at that level. You know, the alternative, which is to keep investing all of the diagnostic capacity at the hospital level is certain to keep patients flowing up. So, I think that there's got to be some of that kind of decentralization of technical capacity so that people have the confidence to use primary care level first.

FRIEDEN: I have one more question, and then we'll turn to Greenberg. To what extent do you think part of the challenge is the disconnect between felt needs and maximum health impact? We see this very starkly in our hypertension treatment programs. Hypertension is asymptomatic. And, a sore shoulder, a bad back, a headache, a toothache, these are all symptomatic conditions. People will go, they will pay for it, whether public or private. But you have to kind of generate demand for immunizations, for example, or deworming, or other things.

HANSON: Well, I think yeah, I think this kind of hidden burden of hypertension is a great example, right? So, unless people are using primary care and are routinely being screened, we have–we're going to continue to have this large undiagnosed population of people with a health condition that–without treatment–is going to become very expensive and hard to manage effectively. So, I think there's something about kind of a return to that notion of primary care as a–almost as a family practice, isn't it? As a place where all members of households are welcome and where there is–where there are routine screening programs. And I guess where it is something that's high priority like that, for which there isn't much demand, that's a great candidate for a pay-for-performance system, isn't it? Because that way you can really, you can capture and incentivize the providers to go out and screen people. You capture their efforts is what you want to do. So, you use your pay-for-performance where you think the constraint can be addressed by greater provider effort.

FRIEDEN: Great, let's turn to Greenberg. I don't know if–what the salutation is, but over to you. Please introduce yourself and a brief question.

Q: Good morning, I really enjoyed this session. Given the disruption that COVID introduced to the entire medical system and created a lot of virtual medicine, what would be the role of a virtual primary care system whereby if a patient needed a specific contact for a physical exam or taking blood pressure, this could be allocated to someone in their neighborhood? How would you see that as playing a role and taking advantage of the disruption of COVID which can be a very powerful stimulus in many ways, good and bad?

HANSON: That's a great question. I think that is the big question about COVID, isn't it? Sort of: What do we want to keep and retain in terms of our learning? What were the things that actually, they were good enough at the time, but we need to revert? Here in the UK, there's been some–it's been very interesting discussions about remote consultations for primary care. And a lot of GPs have really, really maxed out on them because they find them a really–a really efficient way of seeing patients and seeing patients quickly. And there is a bit of a backlash for patients who haven't been able to articulate their condition as clearly as others. So as with all of these things, I think they have that sort of virtual consultation backed up by a real consultation is very promising as long as it's well targeted, and I don't know quite–I'm not quite confident that we know enough to be able to target it at those who are best protected by–or to protect people from some of the downsides of it.

But it seems–I think it's amazing to be able to see a GP or to talk to GPs. You never used to be able to do that. I can–most my problems I can resolve over the phone. And that ability to do that at my convenience without having to travel into town to go to the practice is fantastic. So, there are clearly lots and lots of possibilities that are opened up by distanced, you know, whether that's, you know, phone or video or otherwise.

I think those benefits are even greater if we think about places that have really strong, really big access problems. So, remote areas of the U.S., remote areas of Canada, remote areas of poor countries, to be able to harness those technologies, to bring health care closer to those who need it, I think it's fantastic opportunity. It's just a matter of picking the things that we really want to retain. I would say also and–keeping the thinking also how we keep the adaptations to the financing arrangements that supported them. So, I remember at the very beginning in Canada, having to change their billing system to make virtual consultations eligible for payment. That wasn't–they had to change that so that doctors could be paid for remote consultations. And then that's easy enough to retain once you've got it working.

FRIEDEN: And you now have some primary care physicians quite concerned that they may be spending hours on answering patients email questions, and how does that get paid? Another argument for capitation, really. Because you end up having to spend so much effort collecting information rather than providing care. Anthony Yuen?

Q: Hi, my name is Anthony Yuen, I'm with the Clinton Health Access Initiative. Really appreciate the talk today. You mentioned the importance of public domestic funding, but I was wondering on the donor side, in your view, what models or examples of bilateral and multilateral development assistance for health would you point to as being most effective in supporting recipient countries’ primary health systems? Thank you.

HANSON: That's a great question, Anthony. I was at the Montreux meeting a couple of weeks ago, where I was discussing, yet again, the possibilities of, the equivalent of what we used to call in the old day days, “basket funding”. Basically of, you know, for those governments that are able to plan and then be accountable for their budgets, that surely it makes the most sense to give the money to the government and get them to do your work according to a single plan that's been agreed and to be accountable to a commonly agreed set of indicators. And that works well, where you have a fairly trusting relationship between national governments, their populations–so that you know that the governments are working in the interests of their populations–and where they've developed trusting relationships with their donors. I think you're in a different world where either one of those trusted relationships aren't working. So, if you have a government–and they're often related, right? It's government is not working in the interests of its population, or where the donors are worried about misuse or fungibility in the use of donor funds. So, I suppose there you do have to introduce tighter controls, but as an ideal to work towards I think that's a really important one. And the more that donors can move away from having their own program with their own funding stream with their own indicators with their own monitoring visits, I think the easier it is for governments to focus on delivering the services that they're supposed to be delivering.

FRIEDEN: I guess on that issue, I've seen mixed reviews of some of the basket funding initiatives, because sometimes it's very difficult to see the outcomes and the money kind of disappears. And you're not really sure if you've accomplished much. I wonder whether–

HANSON: That feels like bad design of your outcome though, doesn't it? So, I think with a well-designed program with indicators that you agree on, and that you can measure, that you can measure, well, then you shouldn't in a sense, you shouldn't care that much how they're produced, right? It shouldn't be–you are effectively contracting on outcomes rather than contracting on inputs. We resort to contracting on input, on how money is actually spent, where we don't understand the production process, or we don't trust the outcome measurement.

FRIEDEN: Great. Well, we're coming to the end of our time. Would you like to make a couple of brief comments before I make a couple of comments summarizing some of the key lessons from what I've heard from your wisdom?

HANSON: No, I don't think I’ll say very much. Or except, other than I guess maybe just to know one more piece–one piece of work that's following on from the work of the commission is worth–that we're just about to start which, which it's actually it stems from a measurement issue. So, we used in our calculations for the landscaping, we use the Global Health Expenditure Database estimates of primary health care expenditure. And those are all predicated on a common definition of what PHC is across the world. And that common definition of functions doesn't correspond all that well to the actual PHC service delivery platforms in different settings. So–and if you don't have an expenditure figure that's attached to a country-specific model of primary health care, it makes it very difficult for a government to make a commitment: “We're going to increase primary health care expenditure by this much”, and then to be held accountable for that commitment. Because it's not–you need to measure something that is relevant at the local level. So we're just about to embark on a–I think quite an interesting exercise at looking at how those primary health care platforms operate in different types of government settings, whether they're centralized or decentralized, and look at the different ways that are available to capture PHC budgets and expenditure to make that a more useful way of enabling governments to make commitments and then to honor those commitments to increasing their spending on primary health care.

FRIEDEN: Great. Well, I've learned and here are seven things that I've learned from this conversation. One, countries are spending way too little. Three dollars or so per capita in low-income, sixteen dollars per capita in low-middle-income versus an estimate of at least sixty or seventy dollars per capita per year that would be needed. Two, that rigid rules are problematic, they make it harder to use even that limited money as effectively as possible. Three, that catastrophic spending–ten to thirty percent of household income–is far too common and a major result of inadequate primary care services. Four, that budgets and other tools are very important. Your budget is your plan. And so, if you put things into the budget, like a new category of health care worker, that gives you some ways forward.

Five, that aligning incentives across different levels of care is extremely important, but unfinished business. Six, that electronic systems are going to be very important to allow us to reduce paperwork and look at some of the most important indicators. And seven, that public funding, not public provision, is most predictive of an effective system. And we need to think of regulations, upgrading professionalism and other models to think about upgrading the resources that are going into private health care, because patients are paying often quite a bit for substandard care in many countries. And is there a way to try to bring that into the ambit of a publicly funded system? Kara, any last word for you before we close?

HANSON: That was just such a beautiful summary. Thank you very much.

FRIEDEN: Well, thank you for the work that you do. Thanks to the audience for your participation. And we look forward to continuing to see how primary health care can get better as the most important–but unfortunately most neglected–aspect of our health care system. Thank you all very much. And we hope that it will not be neglected so much longer, because coming out of COVID, one of the main lessons is: we must have primary health care at the center of our system, or we’ll have less resilient populations and less ability to find and stop outbreaks. So, thank you all very much.