[N.b. I used Claude to help format this post; all arguments are mine]
A response to Rethink Priorities HSS report & F/U, with some inspiration from @NickLaing, @Berke and others.
--
There is a category error sitting at the centre of the EA Forum conversation on community health workers (CHWs); workers who bring care straight to their neighbor’s doorsteps. Rethink Priorities’ HSS report models “CHW programs” as an intervention, scores them at roughly 1,000x in Coefficient Giving terms, and notes they do not meet the 2,000x bar. The conclusion that follows is that CHWs are an interesting but second-tier bet.
But CHWs are not an intervention. CHW programs are the delivery platform through which most of the interventions EA already funds—bed nets, vitamin A, deworming—reach billions.
Modelling the platform against its own throughput suggests the model for the throughput wasn’t built to deliver.
A platform, not an intervention
A community health worker is not an intervention any more than a nurse is. Whether a CHW program is cost-effective depends, like any other delivery mechanism, on a) what it delivers and b) how it is designed.
What the platform delivers
A modern, professional CHW program is a primary care delivery point, not a single-disease vertical. The WHO competency-based curriculum—the closest thing the field has to a current job description–sets out the portfolio: childhood illness (diarrhoea, pneumonia, malaria), maternal and newborn health, family planning, nutrition screening, immunisation support, TB and HIV case-finding, NCD screening (hypertension, diabetes), mental health first response, neglected tropical disease case management, and health promotion.
The same worker, in the same household visit, delivers whatever subset the program is designed for.
Any cost-effectiveness analysis that attributes the full fixed cost of that worker to one of those service lines is going to produce an incomplete number.[1]
Facility-based care alone cannot reach everyone. Hospitals struggle to serve remote communities, and in most high-burden countries the realistic counterfactual to a CHW is not a clinic visit — it is no contact with the health system at all.
Where researchers have compared CHW-delivered care head-to-head with facility-based care, the CHW route has been consistently cheaper. Most CHW programs evaluated for HIV, TB, malaria and for reproductive, maternal, newborn and child health were cost-effective against the facility comparator.
The median annual cost to deliver primary care via a CHW platform—across 380 scenarios in 130 studies—is US$0.59 per capita. That is not cost per DALY, and not cost per visit. It is the full annual unit cost, per person per year.
The median cost per beneficiary is $10.03, but of course the cost varies by service package. Integrated horizontal platforms—many service lines through one worker—sit at a median of just $6.02 per capita per year to reach those not close to a facility. [2]
HIV, TB, malaria and RMNCH range more widely; in both areas, CHW programs were cost-effective in over 80% of scenarios assessed.
Bundling is where the savings compound. When Liberia folded NTD case management into an existing CHW platform, cost per diagnosis fell up to tenfold versus running NTDs as a standalone vertical.
The mechanism is platform economics. A CHW’s fixed costs—salary, supervision, supplies, transport, data—exist whether she delivers one service line or ten. Vertical analyses load the whole stack onto a single disease; horizontal analyses spread it across everything the worker delivers in one visit.
How the platform is designed
Pool salaried, supervised, well-supplied, skilled CHWs together with volunteer, unpaid, under-trained ones, and the average tells you little. The “critical components” literature—including the work by yours truly that Rethink cites—has been clear for a decade: compensation, training, supervision, supply-chain integration, and career progression are not optional. Averaging two different things: poorly supported volunteer initiatives and high-performing professional platforms isn’t consistent with the documented 4x variation in ROI.
Contrary to what Nick Laing’s piece from April ’25 implies, this isn’t a workforce that is “often not cost-effective.” Like nurse-led clinics—which Nick rightly champions!—it is a platform that is cost-effective when designed properly and not cost-effective when it isn’t.
The relevant question is not whether to fund “CHWs” but which design pattern to fund, and at what level of the system.
Rethink’s way forward
For all the buzz about CHWs, the design pattern with the strongest evidence is the most neglected. A professional CHW platform is salaried (not volunteer), skilled (competency-based training to a published curriculum), supervised (relevant supervisor-to-CHW ratio), supplied (the commodities the job requires), and built into the public primary care system.
So the smart bet is the one Rethink itself names in its own report: “Help fund and establish a public health program until it is transitioned to national government ownership.”
For the whole vertical portfolio EA has spent fifteen years building—bed nets, SMC, vitamin A, deworming, child immunisation—that pathway is the scale opportunity.
How countries can get there
How does a country move from volunteer CHWs to a paid, professional, government-owned workforce? Three conditions hold the problem in place, and an EA funder can act on each one.
Guidelines. Until recently, a health minister had no authoritative international consensus describing a well-designed CHW program That has changed. The WHO 2018 guideline, the aforementioned curriculum, and proCHW guidance co-authored by Africa CDC and other norm-setters now give a minister both the cover to make a major workforce decision and the spec to make it well.
Start up capital . Even a willing government faces a chicken-and-egg problem: standing up a salaried workforce needs startup capital before the recurrent domestic budget line can absorb it. The Global Fund and others move money at that scale, often catalyzed by private actors.
Domestic constituency. Professional CHW associations are what turn a presidential commitment into a recurrent budget line that outlasts the president who made it.
Funding a delivery NGO to ship commodities saves lives now, and it is essential. Funding the workforce, the policy uptake and the government adoption that let a country deliver those commodities itself is the complementary lever that outlives any single donor.
This is a bet not dissimilar to hits-based giving: a large, durable return (i.e. a permanent, domestically-financed platform that scales all the verticals and outlives any single donor) with outcomes hard to attribute to any one funder.
It also has the one thing a hit almost never does: a decade of cost-effectiveness evidence already behind the intervention. The only open risk is whether a government finishes the handover, but if it does it pays for the whole portfolio.[3]
A closing thought
Berke framed the underlying choice well on the Forum:
“Does success for EA look like hundreds of cost-effective NGOs distributing health commodities across South Asia and Sub-Saharan Africa until 2100 — or governments able to keep their people from dying of preventable conditions without depending on funders far away?”
For any funder serious about scale, it’s the second. And professional CHW platforms are the pathway.
- ^
Rethink’s CHW model, for instance, uses integrated child-survival RRs at full credit, but loads the fixed worker cost against that one outcome category. It’s likely that that CHW is also delivering services affecting, e.g., adult morbidity, non-fatal iCCM DALYs, contraception, NCD case-finding, mental health. Platforms should likely use multi-outcome accounting to reflect these real-world conditions.
- ^
To my point in this article: the $0.59,$10.03, $6.02 figures sit on quite a bit of heterogeneity! They are medians across 130 studies, with wide variation in setting, package, and study quality. I think the central tendency is robust enough to act on. I don’t think any single number should be quoted without that caveat, and I’d welcome a re-analysis that weights by study quality.
- ^
Most EA models are static; handover, adoption, and durability are not modelled. In the HSS piece, 997x is the marginal-DALY price under a permanent(?)-donor assumption. The catalytic case is the temporary-donor price: donor cost goes to zero after handover, government carries the platform for decades, and the accounting compounds across every category it produces (see footnote 1)
Thanks @Madeleine Ballard. As you know I’m a big fan of the Pro-CHW movement and all the work you do. I agree facility care can’t reach everyone. I agree we need pro CHWs across Africa and like you say CHWs are often the main people doing the work facilitating EA funded vertical programs like mosquito nets and deworming. Sometimes this platform isn’t recognised, and the cost borne by governments and donors who support these CHWs isn’t appreciated properly in calculations—which I think I’ve pointed out to GiveWell before.
I also agree that EA doesn’t pay nearly enough attention to gradual government adoption of the most cost-effective interventions, and I hope we’re going to see more discussion about this in the near future.
When looking at RP’s report, this was what I said in context.
”I agree that CHWs are an essential, scalable part of many LMIC health systems. CHWs however are touted to be very cost-effective when they often aren’t. This statement made my eyes pop a little. “Many studies have found CHW programs to be cost-effective by various metrics. In Mozambique, the annual cost per beneficiary was just $47.12.
That number $47.12 I quoted from the RP report is quite different from $10 you’ve quoted as the median cost per capita here and doesn’t scream cost-effectiveness. The report which your $10 number came from wasn’t out at the time I wrote that response. $10 per beneficiary still seems high to me but it’s hard to compare to other interventions (see below)
One thing I’m confused about, is that I don’t think a cost as low as $0.59 per capita is possible for a salaried CHW. How did the researchers get to this number? Trying to be as conservative as possible—here’s my sense check. The ceiling for the no. of people a CHW can cover on a door-to-door model is perhaps 800. A lower end CHW salary might be $40 a month. Then we add commodity and supervision costs (excluding digital/connectivity costs). I’ve estimated this at around 20 people tested for malaria, 10 given treatment for malaria and 5 each treated for pneumonia and diarrhoea. Often CHWs would treat more people than this, which would mean higher costs.
- Salary (12x40) = $480
- Malaria tests 20 a month (20 x $0.5 x 12 months ) = $120
- Malaria treatment 10 a month (10 x $0.3 x 12) = $36
- Pneumonia treatment 5 a month (5 x $0.5 x 12) = $30
- Diarrhoea ORS + zinc 5 a month (5 x $0.5 x 12) = $30
- Supervision ? $60 a year
= $756 / 800 people = $0.95 per capita per year under pretty conservative assumptions. I might be missing something here though!
Also “Per capita cost” and “Cost per beneficiary” are useful metrics for governments to decide if CHWs are affordable/cheap enough, but not useful metrics for cost-effectiveness comparisons. What can these metrics be compared to? We can’t compare this metric to any other intervention. I’ve considered we could use a metric such as “cost of universal primary care per-capita” which would combine the cost of CHWs + other services, which could then be compared to facility only based models + extra transport costs. This would be tricky though…
For CHWs a more straightforward “Cost per DALYs averted” would be more useful to compare with other EA funded interventions, or a more thorough GiveWell style CEA.
Also on this point “Any cost-effectiveness analysis that attributes the full fixed cost of that worker to one of those service lines is going to produce a misleading number.” I don’t think Givewell attributes the full fixed cost of CHWs in CEAs. They might apportion some of the cost, or perhaps none. I often have the opposite criticism that they don’t always appreciate the underlying cost of government health workers used in mass-distribution campaigns. I could be wrong here though I haven’t looked closely recently.
Again I’m a huge fan of CHWs and your work and I believe in your mission. I’m just challenging whether your numbers make sense, and I think we need more than per-capita/per-beneficiary numbers to compare CHWs to other cost-effective interventions under EA frameworks.
Thanks so much, Nick! We published a massive multi-paper cost-effective review only weeks after you shared your original post! You can dig into all the papers/numbers/methodology here: https://joinchic.org/resources/cost-effectiveness/
TL;DR: The numbers are medians across 380 scenarios in 130 studies. As I said in the footnote, data is heterogeneous (mostly due to platform design differences, including salaries). Ranges are big and depend on context, but would say a) your $0.95 BOTE is a fair floor for a professional program at modest catchment, but b) that your earlier take (CHWs are “often not cost-effective,” they treat “few patients per month,” they can only treat “a handful of conditions in young children”) is not reflected in these data.
Re: DALYs: you’ll see in the papers that vertical evidence is already in EA terms (e.g. there are DALY numbers for iCCM) but none of the 42 horizontal integrated scenarios in the BMJ GH piece report cost per DALY. The research I’d most want EA to fund is a prospective platform-economics evaluation: does bundling 6–8 service lines on one salaried CHW compound DALYs and lower cost-per-DALY the way fixed-cost-sharing predicts? (or to your point, maybe not?)
My bigger point though is that maybe this is already a good enough bet based on the vertical DALY evidence we both already accept (i.e. why not apply hits-based giving on the handover: catalytic financing to professionalize a salaried, skilled, supervised, supplied workforce via national, government). Esp, as this has what a hit almost never has, which is a decade of cost-effectiveness evidence already behind it. If the handover lands, the donor pays once and the government runs the whole portfolio forever (and the vertical $/DALY figures are the floor).
Have added 2 footnotes re:the above.
I think It wouldn’t cost much at all to make forward a pretty robust cost-effectiveness model for a CHW which rolls out a wide range of interventions. (I think Living Goods +- others might well have decent models already here?). I think you could even build this yourself? Some of the package would be easy to do because data is there (malaria, diarrhoea, pneumonia treatment, family planning antenatal care), while screenings and referrals are much harder to quantify and might have to be left out of the analysis pending better data.
I agree the bet argument is pretty good if the goal is government adoption. Regardless of the nuance of cost-effectiveness, CHWs will always be more cost-effective than most health things govt. could do and it would likely displace less cost-effective things. Unfortunately I don’t think EA funders have seriously considered scale through govt. something worth chasing as a bet, but I really like the idea.
You might want to explain what ‘CHW’ stands for?
D’oh! Thank you, I’ve added to the first sentence :)
Thanks for posting this Madeleine, it is great to see people from outside the traditional EA global health space engaging here!
This isn’t an area I’m super familiar with, but I’ll try to throw in some questions/thoughts to perhaps draw out the argument a bit, because I think this is valuable to think about!
Let’s assume for the sake of discussion here that the creation of a CHW program in an area where core CHW-delivered care (like vaccination, malaria bed nets, SMC, vitamin A, deworming, etc.) is completely unavailable is cost-effective at a typical EA bar.
I think it is interesting that the recommended thing to get funded notably slightly different/more indirect, which is to fund policy change to get governments to pay for the creation of more professional CHW programs. I think I’d be really interested in hearing more about the evidence base behind this recommendation (e.g., the systematic review you linked pertains more to academic/NGO interventions designed to improve CHW performance, rather than efforts to improve government rollout of CHWs). Questions I’d be really interested in hearing some more about:
What evidence do we have of externally funded health systems strengthening campaigns successfully changing government implementation of CHWs, in a way that has both (a) created professional, paid CHW programs and (b) which has then actually changed health outcomes in an identified way afterwards?
I imagine this has worked somewhere—I’m only vaguely familiar with CHIC, but from that vague knowledge I know CHWs have been gaining steam. That said, it certainly hasn’t worked everywhere. My impression for instance is that lots of funders have spent quite a lot of money trying to get a Nigerian CHW program off the ground only to end up with some policy documents that look nice on paper but CHWs not actually getting paid or doing anything meaningful most of the time.
What has defined the difference between contexts where externally funded programs to change government behavior have worked and where they have not? How can we predict in advance which areas are worth spending money on and which are not?
I think it is then interesting to revisit our assumption at the top here. The counterfactual we’re talking about here is probably not zero treatment to CHW treatment. It probably looks more like a reasonably competent government rolls out CHWs in an area that has some existing primary healthcare services—in this situation, how many more people get treatment? At what marginal additional cost? Is that marginal benefit worth that cost?
Super possibly! But just flagging that it isn’t as straightforward as the base case assumption we make might be.
I think intuitively we expect this to be really cost-effective in places that are underserved/hard to reach—but those places probably don’t have super competent government in the first place, and so is our notional health systems strengthening campaign going to help? Maybe! Maybe not! Would love to hear more thoughts.
Anyway in closing—you guys should put together an EA-style CEA of this! I think that’d be the best way to make this case.
Thanks so much, Justin, for taking the time to read and engage w/these great questions! A couple of thoughts:
1) Re: example of ntl level evidence: I used to live/work in Liberia and it’s a great example of such a transition. They adopted proCHWs based on advocacy + pilot evidence from Konobo district post ebola in 2016. Malaria prevalence then fell from28% (2011) to 10% (2022) across nationally-representative surveys. Multi-driver (nets, seasonal malaria chemoprevention, artemisinin therapies, the Liberian program all contributed), the platform is the delivery channel. David Walton of PMI said it was one of the fastest declines they ever documented.
On a more meta scale: 15+ years ago only a handful of countries had proCHW policy (Brazil, Ethiopia). People used to laugh at the suggestion :) Now 55+ countries have such policies following coordinated, multi-country 10 yr campaign targeting international guidelines, funding, and ntl policy. Contribution, not attribution etc etc but after 100 yrs of basically no movement whatsoever on this, it’s something.
2. Predictors: Literal PhDs written on this! (Finnemore & Sikkink on norm diffusion = excellent). We at CHIC have two realist synthesis papers on this in process (one on how CHW associations organize to shift government behavior, e.g. mechanisms = a combined legal/association/protest pressure in a political opening), one on the political economy of domestic CHW financing, e.g. mechanisms = design choices at program inception etc.)
Some common ingredients: a ministerial champion; a unified professional voice/constituency (e.g. see Chenoweth & Stephan on the 3.5% threshold);
3. Counterfactual: Not sure I agree. The 2015 Konobo baseline (from the Liberia example above): a near-total dose-response collapse with distance to facility...odds of even one antenatal visit dropped to 0.04 in the farthest quartile. i.e. the hard-to-reach half of rural sub-Saharan Africa is closer to “no formal contact” than to “competent ministry already delivering care.”
4. A broader observation: The donor’s operative counterfactual is policy stuck vs policy moves: paper policy plus volunteer workforce plus indefinite donor dependence, OR a 5–7 year catalytic grant plus the four predictors gets a country onto a professional workforce on a recurrent domestic budget for 30+ years. Liberia made the transition; the 32% to 10% drop is the outcome.
We are all well served by remembering:
a) Unit cost is a lever, not a constant: antiretroviral therapy went from 10k to 90 bucks per patient per year in a decade—due to funded, coordinated efforts
b) Domestic financing is also a lever: the proCHW dashboard I linked above shows proCHW policy does not track w/gdp per capita (e.g. if Liberia did it...!).
Yes to the EA-style cost-effectiveness analysis! Would love help building it. If anyone is interested in this or wants to offer a view on what it needs , I’d love to connect.
Many thanks again and hopefully talk more!
Interesting. Been thinking about CHWs lately and how they are quite cosy effective for skilled HWF in the diaspora looking to offset social losses due to their migration. Will read this again. Thanks!
Thanks, Nzube, for taking the time to read and share your experience. Would love to hear more of your thinking re: offsetting social losses
This is probably the clearest takedown of the “CHW as intervention” category error I’ve read.
From a county health office perspective here in Kenya, there’s one piece that stands out: the transition to government ownership usually doesn’t hit a technical wall or even a funding wall. It hits a political one.
A local government that runs a CHW program well gets far less visible credit than one that opens a shiny new clinic. The clinic gets the ribbon-cutting, the speeches, the photos. A strong CHW program just quietly keeps people from getting sick, which is exactly the point, but politically it’s invisible.
That mismatch is why so many “successful pilots” never go anywhere. Donors fund the launch, governments signal support, everyone says the right things. Then the pilot ends, CHWs stop getting paid, and attention shifts to the next district.
The model you’re describing only holds if someone inside the Ministry of Health has a real career incentive to keep it alive. That’s the piece missing from most EA cost-effectiveness papers. Not because it’s impossible to measure, but because almost nobody is measuring it at all.
Worth asking: who inside the Ministry actually owns the platform after the handshake? If the answer is no one with a budget line, the evidence doesn’t matter.
Hey @Albert Oyawa thanks for the contribution—on the forum here. These are decent points but this is clearly written by AI, and any AI use needs to be stated up-front here.
Appreciate it, Albert! The point your making about health being political and not just technical is well taken