Can gig work fix India's healthcare workforce problem?

No. Not the problem people think it fixes.

India's health workforce crisis is a distribution and financing problem before it is a scheduling problem. Gig platforms are very good at solving scheduling. They are structurally poor at solving distribution, because platform economics send supply toward the highest willingness to pay, which is exactly where clinical supply is already densest.

So the honest forecast is this. Gig work will meaningfully expand urban and Tier 2 convenience care, home nursing, diagnostics collection and teleconsultation. It will not, on its own, put a qualified nurse in a village in Koraput.

Key findings
  • Gig platforms solve scheduling. India's workforce crisis is distribution and financing, so the models are aimed at different problems.
  • India has 26.5 registered health workers per 10,000, 16.7 active, and 11.0 active and adequately qualified, against a WHO benchmark of 44.5.
  • 66% of Indians are rural; about 33% of the health workforce serves them. Platform economics push supply the other way.
  • India publishes no reliable healthcare-specific gig workforce data: no headcount, assignment volume, rural share, or safety outcomes.
  • The unresolved issue is accountability, not technology. Fixes exist: episode payment, a named accountable clinician, clinical risk tiering, and a publicly funded rural premium.

One caveat before any of the numbers. India does not publish a reliable estimate of the healthcare gig workforce: not headcount, not assignment volume, not rural share, not safety or continuity outcomes. The gig figures below describe the whole gig economy. So this edition assesses the model's potential and its governance requirements. It does not estimate the current market size, because nobody credibly can.

That does not make gig work irrelevant. It makes it a capacity instrument, not an equity instrument. Treating it as the latter is where policy and investor money will get wasted over the next five years.

And underneath the workforce question sits a governance one, which is the part almost nobody is pricing. When a home nurse takes instructions from a platform, a hospital and a family at the same time, nobody has written down who is accountable. That is not a scheduling detail. It is the thing that will decide whether this sector scales or stalls.

Figure 1
Gig healthcare against the Quintuple Aim
Scored against the five aims health systems are judged on, the Quadruple Aim plus health equity. These are the author's directional scores based on the evidence in this edition, not measured outcomes. Equity is included deliberately, because it is the axis the whole argument turns on.
View the data behind this chart
Gig healthcare scored against the Quintuple Aim (author's directional assessment, 0-10)
AimPromise under ideal governanceLikely delivery, current structure
Health outcomes8.04.5
Patient experience8.57.0
Clinician experience6.54.0
Lower cost8.04.0
Health equity9.02.0

Author's directional assessment based on the evidence discussed in this edition. Not measured outcomes.

Read this way: the shape is lopsided by design. Platforms optimise the transaction, and two of the four aims are properties of the episode, not the transaction.

02 — The numbers

Five numbers that frame the debate

11–26.5
Health workers per 10,000 depending on definition: 26.5 registered, 16.7 active, 11.0 active and adequately qualified. WHO benchmark 44.5
66/33
Two-thirds of Indians are rural. About a third of the workforce serves them
23.5mn
NITI Aayog's projection for the whole gig economy by 2029-30, from 7.7mn. Healthcare's share is not separately published
~5lakh
Gig workers registered on e-Shram, against roughly 1 crore estimated
47cr
Cumulative eSanjeevani teleconsultations as of June 2026. Proves the modality scales, not the marketplace model

The fourth number is the one to sit with. The legal recognition arrived. The registration did not follow. Any thesis that assumes gig healthcare workers will be automatically covered by social security is running ahead of the evidence.

Figure 2
Workforce density against the WHO floor
India's health worker density per 10,000 population versus the WHO benchmark.
View the data behind this chart
Health worker density per 10,000 population, India
MeasureDoctors, nurses and midwives per 10,000
WHO benchmark44.5
Registered stock26.5
Active workforce16.7
Active and adequately qualified11.0

Karan et al., Human Resources for Health (2021), using NHWA 2018 and PLFS 2017-18 data.

Read this way: which number you quote changes the argument. Against the registered stock India looks close to halfway; against workers who are active and adequately qualified it is at a quarter of the benchmark. No matching algorithm creates a clinician who does not exist, and gig platforms draw from the active pool, not the registered one.
Figure 3
Where the workforce is, and where the people are
Rural share of population against rural share of the health workforce.
View the data behind this chart
Rural and urban share of population against share of the health workforce
AreaShare of populationShare of health workforce
Rural66%33%
Urban34%67%

Karan et al., Human Resources for Health (2021), 2017-18 data.

Read this way: this is the distribution gap. It is the gap gig work is least equipped to close, because the payment is thinnest exactly where the need is greatest.

03 — Definitions

The word "gig" is doing too much work

Four very different things are being bundled together, and they carry different liability, different economics and different regulatory exposure. A locum radiologist covering weekend shifts is not the same as a phlebotomist dispatched to a house, which is not the same as an agency supplying nurses to a hospital.

ModelExamplePlatform carriesMain risk
Digital task marketplaceCoding, transcription, scribingMatching and qualityData security, accuracy
Remote professional networkTele-radiology, teleconsultationCredentialing and escalationDiagnostic liability
Home-service marketplaceNursing, phlebotomy, physiotherapyFulfilment and field protocolsPatient safety, continuity
Institutional flexible staffingLocum doctors, agency nursesSupply and compliancePrivileging, hospital liability

These four are not one market and they do not deserve one answer. The first is barely healthcare. The last is not really gig, it is staffing with an app. The two in the middle are where the argument actually lives.

The distinction matters commercially. A platform that only matches supply to demand is a discovery business with thin margins and rising regulatory cost. A platform that owns the clinical episode is a care business with defensible economics. Most Indian healthcare gig platforms today are still the first thing.

04 — The case for

Four arguments that hold up

1

There is dormant qualified supply

More than a fifth of qualified health professionals in India are not active in the labour market, and a large share are women who left after caregiving breaks. Flexible, credentialed re-entry is the single most credible workforce unlock gig models offer, and it does not require producing a new doctor.

2

Specialist time is badly utilised

A Tier 3 hospital cannot fund a full-time intensivist. It can fund eight hours a week of one. Tele-radiology and tele-pathology have already proven this works at commercial scale.

3

Home care demand is real and growing

Ageing, chronic disease and hospital cost pressure all push care into the home, and home care is intrinsically shift-based.

4

The verification rail now exists

ABDM gives India a national professional registry, a facility registry and consent-based record exchange. Credential verification, historically the biggest barrier to trustworthy flexible staffing, is becoming a query rather than a project. With one limit worth stating up front: HPR can confirm identity and professional registration. It cannot confirm that a clinician is competent or privileged to perform a given procedure at a given facility. Those are separate steps, and only the first is now cheap.

Figure 4
The dormant supply, and what re-entry could unlock
Registered stock against active workforce for doctors and for nurses and midwives.
View the data behind this chart
Registered, active and adequately qualified health workers per 10,000
CadreRegistered stockActive in labour marketActive and adequately qualified
Doctors8.86.15.0
Nurses and midwives17.710.66.0

Karan et al., Human Resources for Health (2021). The gap between registered and active is the re-entry opportunity.

Read this way: the gap between registered and active is the most addressable number on this page, and flexible work is a genuine lever on it.

05 — The case against

Why the answer is still mostly no

Platforms concentrate where payment concentrates. No commercial matching algorithm will route a physiotherapist to a low-income rural household when the same hour earns three times more in a metro. Gig work redistributes convenience within served markets. It does not extend the market. Fixing that needs public contracting or a rural rate premium, which is a financing decision, not a technology decision.

Continuity is the clinical cost. Outcomes depend on the same clinician seeing the same patient across an episode. Shift-level matching optimises the transaction and quietly degrades the episode. The clinical literature on continuity is consistent for conditions like diabetes control, post-surgical recovery and mental health counselling, though no Indian platform-level outcome data exists to quantify the effect for gig staffing specifically. The plausible damage shows up as repeat consultations and unplanned readmissions, which never appear on a platform's dashboard.

Accountability dissolves across three parties. When a home nurse follows instructions from a platform, a hospital and a family, and something goes wrong, nobody has written down who was responsible. Indian jurisprudence on platform liability in clinical settings is thin. My expectation, not a fact: the first serious adverse-event case to reach judgment will set the economics for the entire sector.

Cost savings are frequently transferred, not created. Hospitals see a lower per-shift rate. What they often do not price: orientation time, documentation errors, higher cancellation rates, permanent-staff resentment and rework. The right unit of measurement is cost per completed clinical episode, not cost per shift.

And ratings are not competence. Patient ratings track punctuality and politeness. They do not track whether the nurse recognised early sepsis.

Figure 5
Where the gap between promise and delivery sits
The distance between what platform models claim and what the evidence currently supports.
View the data behind this chart
Gap between promise and likely delivery, by aim (author's assessment)
AimPromise minus likely delivery
Health equity7.0
Lower cost4.0
Health outcomes3.5
Clinician experience2.5
Patient experience1.5

Derived from the directional scores above. Illustrative, not measured.

Read this way: the claims that survive scrutiny are about capacity and utilisation. The ones that do not are about equity and cost.

06 — The framework

Gig suitability is a function of clinical risk

This is the practical part, and it is the part a board should be asking for. Not "are we using gig staff" but "which tier, and who signs off."

GREEN

Medical coding, transcription, scribing, scheduling, translation, de-identified data labelling, outreach.Governance needed: data security and quality audit

AMBER

Teleconsultation, physiotherapy, home sample collection, routine home nursing, counselling follow-up, medication adherence, remote monitoring, locum OPD.Governance needed: credential verification, protocol adherence, a named supervising clinician, indemnity

RED

ICU staffing, emergency medicine, labour and delivery, surgery and anaesthesia, chemotherapy, neonatal critical care, unsupervised invasive procedures.Governance needed: hospital-controlled internal float pool only. No open marketplace.

Anyone marketing a red-tier service through an open marketplace is selling regulatory risk with a clinical wrapper.

Figure 6
Clinical risk against gig suitability
Each role plotted by clinical risk and how well it tolerates shift-level matching.
Read this way: the diagonal is the governance boundary. Work below it can be matched. Work above it needs an employment relationship and a named accountable clinician.

07 — Policy

What the law changed, and what it did not

Changed. The four labour codes came into force on 21 November 2025. The Code on Social Security, 2020 gives gig and platform workers statutory definition for the first time, with an aggregator contribution of 1 to 2 percent of annual turnover, capped at 5 percent of payments made to gig and platform workers, feeding a social security fund. The rate and commencement date still have to be notified. Union Budget 2025-26 committed e-Shram registration, identity cards and AB-PMJAY coverage for platform workers. Karnataka legislated a platform gig worker welfare framework in 2025. The DPDP Rules were notified in November 2025, which puts real obligations on any platform touching patient data.

Did not change. Rules for several state laws remain unframed. Registration coverage is a fraction of the eligible population. Benefit portability across platforms is still a design document rather than a working system. A parliamentary committee has had to recommend making e-Shram registration mandatory for aggregators, which tells you voluntary uptake failed.

The gap to watch: an aggregator levy funds a pool. It does not deliver a benefit to a named worker. Between those two points sits the entire implementation risk.

Figure 7
Registration against the eligible population
Gig workers registered on e-Shram versus the government's own estimate of the gig workforce.
View the data behind this chart
Gig workers registered on e-Shram against the estimated gig workforce
MeasureLakh workers
Estimated gig workforce, 2024-25~100 (about 1 crore)
Registered on e-Shram, Dec 2025~5

Ministry of Labour and Employment replies via PIB. Whole gig economy, not healthcare-specific.

Read this way: this is the distance between a right on paper and a benefit in a bank account.
Figure 8
The projected growth the policy has to keep up with
NITI Aayog's gig workforce projection to 2029-30.
View the data behind this chart
Projected growth of India's gig and platform workforce
YearWorkers (millions)
2020-217.7
2024-25~10
2029-30 (projected)23.5

NITI Aayog, India's Booming Gig and Platform Economy (2022). Whole gig economy; healthcare's share is not separately published.

Read this way: the workforce triples while the enrolment machinery is still being built. The mismatch is the story.

08 — The fix

What actually fixes it, mechanism by mechanism

Naming the problem is the easy half. Each failure above has a specific, already-available remedy. None of them requires a new technology.

The six parts have to hold together. Remove one and the chain fails.
1Flexiblecapacity 2Verifiedprofessionals 3One accountableclinician 4Interoperablerecords 5Portablebenefits 6Fairalgorithms
Strip out any one of these and you have a staffing arbitrage wearing a health-tech logo.
01Supply goes where money is, not where need is
Price the rural assignment
A public-contracted rural premium, routed through PM-JAY or a state scheme.
Why it worksPlatforms follow margin. If the underserved hour pays more than the metro hour, the algorithm routes there on its own. You are not fighting the incentive, you are reversing it.
02Continuity breaks across shifts
Pay per episode, not per visit
A 30 or 90-day episode payment instead of a per-visit fee.
Why it worksUnder per-visit pricing, a handover costs the platform nothing. Under episode pricing, the platform absorbs the cost of fragmentation, so continuity becomes its problem instead of the patient's.
03Nobody owns the adverse event
One named accountable clinician per episode
Named in writing, recorded in the record, carrying indemnity, before the first assignment.
Why it worksAccountability does not dissolve because three parties are involved. It dissolves because nobody wrote a name down. Writing one down costs nothing and changes who answers the phone at 2am.
04Credentials are unverifiable at speed
Verify against HPR, then privilege separately
Registry check for identity and qualification; facility privileging for scope.
Why it worksThese prove different things. HPR says this person is who they claim and is registered. Privileging says this person may do this procedure here. Collapsing them is how an unqualified hand reaches a patient.
05High-risk work reaches open marketplaces
Tier every role before contracting
Green matched freely, amber with supervision and indemnity, red inside a hospital float pool only.
Why it worksRisk is not uniform, so governance should not be either. Tiering lets you move fast on the 60% that is low risk without pretending an ICU shift is a delivery job.
06Benefits are funded but not delivered
Aggregator-obligated registration, portable benefits
Registration becomes the platform's duty, and entitlements follow the worker between platforms.
Why it worksA levy funds a pool. A pool is not a benefit. Voluntary registration already failed, which is why a parliamentary committee had to recommend making it mandatory.
07Ratings substitute for competence
Publish clinical metrics, not volume
Adverse events per 1,000 assignments, escalation response time, 30-day unplanned admissions, same-professional episode rate.
Why it worksRatings measure politeness and punctuality. They cannot detect a missed sepsis. Publishing the clinical numbers is also the cheapest defence a platform has when the first case lands.
08Algorithms decide livelihoods invisibly
Explainable allocation and a route of appeal
A stated reason for deactivation, reduced shifts or a lower rate, and a way to contest it.
Why it worksFor a clinician a suspension is a livelihood question and a licence question at once. Healthcare will face this before food delivery does, because the stakes on both sides are higher.
Figure 9
The scorecard nobody is publishing
The metrics that would actually settle the argument, against what platforms currently report.
Read this way: a platform announcing 100,000 completed visits has told you nothing. Rural share is the test of the whole thesis. If it is not rising, it is a convenience business.

Gig work in Indian healthcare is going to grow whether or not it is governed well. The choice is between a credentialed, risk-tiered, episode-owning model and an unrestricted marketplace that discovers its limits through an adverse event.

The formula is not complicated: flexible capacity + verified professionals + one accountable clinician + interoperable records + portable benefits + fair algorithms. Strip out any one of those six and you have a staffing arbitrage wearing a health-tech logo.

The question I would put to a board: if gig models mainly serve markets that can already pay, should India be regulating them as a workforce solution at all, or as a consumer convenience sector that happens to touch clinical care?

← All perspectives

EVERY FRIDAY · FREE

Enjoyed this issue of Healthcare Pulse?