Two numbers sit next to each other in the government’s newest health accounts, and they do not tell the same story. In 2021-22, out-of-pocket spending accounted for 39.4% of India’s total health expenditure, the lowest share ever recorded. A year later, in 2022-23, it was back up at 43.4%.
Neither number is a mistake. Both come from the same source: the National Health Accounts (NHA) Estimates for India 2022-23, released by the Union Health Ministry on 27 May 2026. It is the tenth such report, prepared by the National Health Accounts Technical Secretariat at the National Health Systems Resource Centre, using the internationally recognised System of Health Accounts 2011 framework.
The gap between the two years is not a policy reversal and it is not a data error. It is what happens when emergency government spending rises sharply for one year, and then does not repeat. The 2021-22 low was produced by the pandemic response. The 2022-23 rebound is what the system looks like once that response ended.
This is not, by itself, a failure story. Nor is it a success story dressed up as bad news. It is a measurement with a real answer: emergency spending buys a temporary reduction in what households pay directly for healthcare, and when that spending stops, the household share of the bill goes back up.
The decade of real progress
Start with what is unambiguously true. Between 2013-14 and 2022-23, out-of-pocket expenditure (OOPE) as a share of India’s total health expenditure fell from 64.2% to 43.4%, according to the NHA 2022-23 estimates. That is a decline of nearly 21 percentage points over nine years, and it is the central achievement the new report documents.
That fall did not happen by accident. It tracks a sustained rise in government health spending across the same period, on every measure the NHA framework tracks.
- Government Health Expenditure (GHE) rose threefold in absolute terms, from Rs 1.30 lakh crore in 2013-14 to Rs 3.85 lakh crore in 2022-23, per the Health Ministry’s figures.
- GHE as a share of GDP rose from 1.15% to 1.43%. On the newer GDP series with base year 2022-23, the Ministry puts the 2022-23 figure at 1.48%.
- GHE as a share of Total Health Expenditure (THE) rose from 28.6% to 43.7%, an increase of almost 15 percentage points, meaning government now covers close to half of everything spent on health in the country.
- GHE as a share of General Government Expenditure rose from 3.78% to 4.89%, showing health claiming a larger slice of the state’s overall budget, not just growing in rupee terms.
- GHE per capita rose nearly 2.7 times, from Rs 1,042 to Rs 2,786, according to the NHA estimates.
Put together, these numbers describe a state that spent more on health, spent it more consistently as a share of a growing economy, and directed enough of it toward primary and preventive care to move the national out-of-pocket ratio by 21 percentage points. That is a real structural shift, not a rounding effect, and it deserves to be stated plainly before the more complicated part of the story begins.
The decade at a glance
| Indicator | 2013-14 | 2022-23 | Change |
|---|---|---|---|
| OOPE share of Total Health Expenditure | 64.2% | 43.4% | Down nearly 21 percentage points |
| GHE share of Total Health Expenditure | 28.6% | 43.7% | Up almost 15 percentage points |
| GHE share of GDP | 1.15% | 1.43% (1.48% on new GDP series) | Up roughly 0.3 percentage points |
| GHE per capita | Rs 1,042 | Rs 2,786 | Up nearly 2.7 times |
| Social Security Expenditure share of THE | 6% | 9.9% | Up 3.9 percentage points |
| Private Health Insurance share of THE | 3.4% | 9.2% | Up 5.8 percentage points, nearly threefold |
Every figure in this table comes from the NHA 2022-23 estimates as released by the Health Ministry. They are estimates prepared on a defined accounting framework, not a census of every rupee spent on health in the country, a distinction worth holding onto through the rest of this article.
Why a tenth consecutive report matters
The NHA 2022-23 release is the tenth in a continuous series, covering estimates from 2013-14 onward. That continuity is easy to overlook and it is what makes the rest of this analysis possible at all.
A single year’s health expenditure figure, taken alone, says very little. It cannot distinguish a structural improvement from a one-off surge, because there is nothing to compare it against. A decade of figures prepared on the same framework, by the same technical secretariat, can.
The framework itself is the System of Health Accounts 2011, an international standard for classifying health spending by source, by scheme, and by function. Using a common standard is what allows a figure like the out-of-pocket share to mean the same thing in 2013-14 as it does in 2022-23.
The expert group behind the estimates draws on the Ministry of Statistics and Programme Implementation, NITI Aayog, the Ministry of Jal Shakti, the National Health Authority, and research institutions including the National Institute of Public Finance and Policy, the National Council of Applied Economic Research, and the Institute of Economic Growth.
That breadth matters for a specific reason. It is the consistency of method across ten reports, rather than any single year’s headline, that lets a reader say with confidence that 39.4% was an anomaly and 43.4% is a baseline. Without the series, both numbers would just be numbers.
What “out-of-pocket” actually means at the point of care
Out-of-pocket expenditure, in the NHA framework, is money a household pays directly at the point of receiving care: not reimbursed by an employer, not paid by insurance, not covered by a government scheme. It is the amount a family hands over, or borrows, when someone is sick.
This distinction matters because it is the single clearest available measure of whether a health system exposes households to financial risk. A country can spend a great deal on healthcare in total and still leave families exposed, if most of that spending arrives as a bill the patient pays themselves, in the moment, with whatever cash or credit is on hand.
That is the mechanism behind medical impoverishment. It does not require a rare or catastrophic diagnosis. A single hospitalisation, or a lengthy course of treatment for a chronic condition, can consume a year of savings for a household that has no insurance and no cushion beyond what it has already put aside.
Assets that exist on paper do not always help in that moment. A family land holding tied up in a boundary dispute or a stalled title transfer cannot be sold or pledged for a loan when a hospital bill arrives, whatever its notional value. India’s land records and property dispute problem means an asset can be locked out of reach at the exact moment a household needs to convert it into cash.
This is why the OOPE share of total health expenditure is treated as a headline indicator by health economists and by the government itself. It is not a proxy for how much a country spends on health. It is a proxy for who is exposed when spending happens.
What tends to fall on the household, and what does not
| Type of spending | Typically out-of-pocket? | Why it matters |
|---|---|---|
| Outpatient consultation | Usually paid directly; rarely covered by insurance in India | Frequent, small payments that add up steadily across a year and are easy for a household to defer if there is no cash on hand |
| Medicines | Often paid directly, especially outside government facilities | Recurring costs for chronic conditions accumulate month after month, with no single moment where insurance or a scheme intervenes |
| Diagnostic tests | Frequently paid directly, particularly in the private sector | A round of tests ordered before treatment can begin can itself be a sizeable upfront cost, before any actual care has started |
| Hospitalisation | More likely than the other categories to be covered by insurance or a government scheme | The category most likely to produce a single large bill, and the one public schemes such as AB PM-JAY are specifically built to reduce |
| Follow-up care | Often paid directly, especially once an insurance claim or treatment episode has formally closed | Rehabilitation and repeat consultations after the main treatment can extend a household’s financial exposure past the point where coverage stops |
The pattern in this table explains why a national OOPE share can fall even while individual households still face real exposure. Government and insurance coverage tends to concentrate on hospitalisation, the most visible and most catastrophic category, while outpatient care, medicines and diagnostics remain largely a direct household cost.
The COVID anomaly, in detail
The lowest out-of-pocket share India has ever recorded, 39.4% in 2021-22, was not produced by a durable change in how the health system is financed. It was produced by a spike in emergency government spending during the pandemic, and the government’s own release says so.
According to the NHA 2022-23 estimates, government health spending in 2021-22 rose to 1.84% of GDP, the highest level in the ten-year series, as the state managed the pandemic response. That figure covers the emergency response packages of the period and the mass vaccination programme, both large, both concentrated in a single financial year.
The Health Ministry’s own release describes that additional spending explicitly as a one-time measure. That is a significant admission, made by the government about its own data, not an inference imposed on the numbers by an outside reader.
Because government spending rose so sharply that year, the household’s share of the total fell correspondingly: OOPE as a share of THE dropped to 39.4% in 2021-22. It was the direct arithmetic result of the denominator changing, not of a permanent redesign of who pays for what.
The following points summarise the sequence:
- 2021-22: government health spending rises to 1.84% of GDP to fund pandemic response and mass vaccination, both described by the Ministry as one-time.
- 2021-22: with government spending unusually high, OOPE’s share of total health expenditure falls to 39.4%, the lowest point in the series.
- 2022-23: the one-time pandemic spending does not repeat at the same scale, and OOPE’s share rises back to 43.4%.
None of this means the 2022-23 figure represents backsliding. It means the 39.4% figure was never a stable floor to begin with. Reading it as the achieved level, and treating the return to 43.4% as regression, misreads what produced the 2021-22 number in the first place.
The more defensible reading treats 43.4% as the current, non-emergency baseline, still 21 percentage points below where the decade started, and treats 39.4% as a one-year artifact of crisis-scale spending that this year’s data confirms did not carry forward.
What is carrying the load instead
If emergency government spending was a one-off, what has been doing the durable work of reducing household exposure over the decade. Two channels stand out in the NHA data, and both come with a caveat about who they actually protect.
Social Security Expenditure on healthcare, as a share of total health expenditure, rose from 6% in 2013-14 to 9.9% in 2022-23. This category, per the NHA framework, includes government-funded health insurance schemes such as AB PM-JAY, medical reimbursements paid to government employees, and other social health insurance programmes.
Private Health Insurance’s share of total health expenditure rose from 3.4% to 9.2% over the same period, nearly tripling. That is a genuine expansion of financial protection for the people who hold such policies.
It is also a different kind of protection from public provision, and the difference matters for who actually benefits. A government scheme extends coverage by eligibility rules set by the state. A private policy extends coverage only to the household that can afford the premium, meet the underwriting requirements, and get through the paperwork to make a claim.
That gap between a scheme existing on paper and a family actually being able to use it is not unique to health insurance. It is the same gap documented in India’s right-to-service laws, twenty years on: an entitlement written into policy does not automatically translate into something a household can walk in and use when it needs it.
None of this is an argument against private insurance’s growth. It is a reminder that a rising private-insurance share, on its own, tells you about the depth of coverage among people already able to buy into the market. It says less about the households furthest from that market, who remain the most exposed regardless of what the national aggregate shows.
Primary care: the spending most likely to hold
One category of government spending stands out as the more durable, structural lever, distinct from both the pandemic surge and the insurance expansion: primary health care.
Government expenditure on primary health care more than doubled over the decade, from Rs 0.5 lakh crore to Rs 1.4 lakh crore, according to the NHA 2022-23 estimates. That is not a one-time spike tied to an emergency. It is sustained investment in the part of the health system that treats people before illness becomes serious enough to require hospitalisation.
The logic is straightforward. Hospitalisation is expensive, unpredictable, and the category most likely to produce a catastrophic bill for an uninsured household. Primary care, delivered close to where people live, is where conditions get caught and managed before they escalate to that point.
A few reasons primary care spending behaves differently from emergency or insurance spending:
- It is recurring by design, budgeted year over year rather than triggered by a single crisis, which makes it more likely to persist once scaled up.
- It reduces the number of expensive hospitalisation episodes a population needs in the first place, rather than only softening the cost of episodes that already occurred.
- It reaches households before a health event becomes a financial event, which is the point in the sequence where out-of-pocket exposure is smallest and easiest to prevent.
If the goal is a lower OOPE share that holds without another pandemic-scale spending surge, primary care investment is the component of the decade’s progress most likely to deliver it, precisely because it does not depend on a repeatable emergency to keep working.
What 43.4% still means
Even after a decade of genuine progress, 43.4% is not a small number. It means that for every ten rupees spent on health in India in 2022-23, more than four rupees were paid directly by households, out of pocket, at the point of care, according to the NHA 2022-23 estimates.
That is the current baseline the country is working from, after the pandemic-era low has washed out of the data. It is lower than 64.2% by a wide margin, and it is still a level at which a large share of the health system’s cost sits directly on the people using it, rather than pooled through government or insurance.
For a household without savings set aside for exactly this purpose, that exposure is not abstract. It shows up as a choice between paying for care and paying for something else, a choice that the financial fragility documented in why a PhD applies for a peon’s job makes clear is closer to the norm for many Indian households than a one-off exception.
What the number does not tell you
A national OOPE share is an average across a very large population, and averages hide as much as they reveal. Being rigorous about this indicator means being honest about its limits, not just citing the headline figure.
- The national share says nothing about distribution. A rural household with no nearby government facility and a wealthy urban household choosing a private hospital both contribute to the same 43.4%, despite facing entirely different exposure.
- It says nothing about who within the population bears the cost. The NHA framework measures aggregate expenditure, not how that burden is distributed across income groups, states, castes, or genders.
- It cannot capture care that was never sought. If a household decided not to visit a doctor, not to fill a prescription, or not to pursue a recommended test because it could not afford to, that decision produces no expenditure at all, and so it does not appear anywhere in health expenditure data.
That last point deserves emphasis. A health accounting framework can only measure money that moved. Care that a family needed but did not seek, because the cost was a barrier before it ever became a transaction, leaves no trace in the NHA’s numbers. The absence of that data point is itself a limitation of the measure, not evidence that the problem it would describe does not exist.
This is also why the NHA estimates should be read as exactly that: estimates, built on a defined accounting methodology, not a transaction-by-transaction census of every rupee spent on health in the country. They are the best available systematic measure, and they are still a model of the system, not the system itself.
What would make the fall durable
The decade-long fall in India’s out-of-pocket health share, from 64.2% to 43.4%, is real, and the government spending increases behind it, tripled GHE, near-tripled per-capita spending, a doubled primary care budget, are documented in the same NHA report that shows the 2022-23 rebound.
The 2021-22 figure of 39.4% is the lesson inside that same data. It shows what is achievable when government spending rises sharply, and it shows, just as clearly, that a level reached through one-time emergency spending does not hold once the emergency ends. The Health Ministry’s own framing of that spending as one-time makes this the government’s own conclusion, not an outside critique of it.
The distinction that matters going forward is between spending that responds to a crisis and spending that is built into the base. Primary care investment, sustained social security coverage, and steady growth in the government’s per-capita and GDP-share numbers, the trends already visible across 2013-14 to 2022-23, are the categories that do not require another emergency to keep moving the number down.
Whether the next NHA report, covering the years after 2022-23, shows the OOPE share continuing to fall from 43.4%, or holding roughly steady, will be a reasonable test of whether the decade’s progress was built on those durable foundations, or whether households are waiting for the next crisis-scale spending surge to bring their share down again, temporarily, until it rises once more.