Hospitalization Friction

Of ₹470.2 crore that hospitals billed against Loop's broker book in the latest twelve months, the insurers paid ₹320.6 crore. The remaining ₹98.1 crore — 20.9% of the billed amount — was deducted before payment, and a further ₹51.5 crore fell away in repudiations and partial settlements, for a total claimed-to-incurred gap of ₹149.6 crore. Some of that gap is the system working as designed: tariff negotiation, package rates, genuine exclusions. Much of it is not visible to the employee until the hospital cashier hands over a bill that the policy was supposed to cover and didn't. The year-end claims slide reports the incurred number — ₹320.6 Cr "settled". It does not report the ₹98.1 Cr the household absorbed to get there.

One rupee in five is deducted before the insurer pays#

The arithmetic is a three-step waterfall. Hospitals claim ₹470.2 Cr. Insurers deduct ₹98.1 Cr as non-payable line items. They incur ₹320.6 Cr. The deduction is not a rounding error — it is the second-largest number in the policy, behind only the amount actually paid.

The deduction is also broadly distributed, not concentrated in a few disputed files. 64.4% of all claims carry at least one deduction. Only 34.2% of claims pass through with the full billed amount paid and nothing repudiated. The median deducted claim loses 15.7% of what the hospital billed; the average deducted claim loses ₹20,722. For most households, "fully insured" and "the policy pays the whole bill" are not the same statement.

The bigger the hospitalization, the deeper the cut#

The instinct is that small claims get nickel-and-dimed and big claims get waved through. The data shows the opposite at the top end. Deduction rate falls from 15% on the smallest claims to a 14% trough on ₹25-50K admissions, then climbs steadily — and on claims above ₹5 lakh it reaches 32.4%.

The rupee figure is the part that matters to a household. A family whose member is admitted for a ₹6-lakh cardiac or oncology event does not lose 1.2% of it the way a viral-fever claim does. They lose, on average, ₹2.68 lakh — a third of the bill — at the exact moment they have the least capacity to argue line items. The deduction rate is regressive in the way that matters: it bites hardest where the bill is largest and the family is most exposed.

Reimbursement is the expensive door#

How a claim is settled changes how much of it survives. Cashless — where the insurer pays the network hospital directly at discharge — loses 18.0% to deduction. Reimbursement — where the family pays the hospital, then files for the money back — loses 24.9%. The gap widens further when the claim is still being processed.

Two mechanisms drive the gap. At a network hospital, the rates are pre-negotiated and the pre-authorisation has already aligned the bill to the benefit grid, so fewer line items get rejected at the back end. In reimbursement, the family pays whatever the hospital charges — often a non-network hospital with no negotiated tariff — and the deduction happens after the money has already left their account. The cashless patient never sees the deducted amount as cash out of pocket; the reimbursement patient pays it first and is told later it won't come back.

The effect compounds on large claims. On admissions above ₹5 lakh, reimbursement loses 42% to deduction against 28.5% for cashless. The single most expensive thing a family can do is pay for a major hospitalization out of pocket at a non-network hospital and file for reimbursement — which is also, for parents admitted in an emergency far from a network facility, the most common thing they end up doing. The reimbursement route's friction compounds downstream, too: the Claims Experience page shows it is also the route that draws five times the query rate.

The deductions are mostly small print, not denial#

The ₹98.1 Cr of deductions is overwhelmingly benefit-design mechanics, not the insurer rejecting valid medical claims. Outright policy exclusions account for under 2% of the deducted rupees. The bulk is consumables, copay, and the room-rent-linked proportionate cut.

Three line items — consumables, copay and room-rent-linked cuts — account for roughly six in every ten deducted rupees. None of them is a clinical dispute. Non-medical consumables are gloves, syringes, surgical disposables, administrative charges — the items a benefit grid simply does not recognise as payable, regardless of whether the surgery needed them. Copay is the share the policy contractually assigns to the patient. And the room-rent cut is the most punishing of the three, because it doesn't stay confined to the room bill.

The consumables line deserves its own flag, because it is the most defensible-sounding deduction that is hardest for a patient to control. A surgical admission generates consumables by clinical necessity, not by choice. The patient cannot decline the gloves. When the policy excludes them, the cost simply transfers — from the insurer's incurred figure to the household's out-of-pocket — without any decision the family could have made differently.

Every layer of the policy loses a fifth, and the spouse layer loses most#

Deduction is near-uniform across who the claim is for — every relationship layer loses roughly a fifth — but the spouse layer is cut hardest, at 23.1%.

The spouse layer's higher rate is a maternity artefact. Maternity admissions carry heavy consumables and frequently breach room-rent ceilings, and the package-rate logic on delivery is among the tightest in the grid — the sub-limit mechanics the Maternity & Delivery page details in full. The parent layer, by contrast, carries the highest absolute deduction per claim — ₹16,925 — because parent admissions are the largest claims, and large claims are cut deepest, as the size table already showed. The two readings stack: parents file the most expensive claims, and the most expensive claims lose the most. The parent layer is both the biggest line in the bill and the most exposed to the deduction structure on it.

What the policy pays and what the household pays are different numbers#

The report's claims pages keep returning to one structural fact: the group policy is consumed by a household, not an employee. This page adds the cost dimension. The household doesn't just bear the clinical load of the dependent layer — it bears the deduction load on every layer.

Three facts the 2026 data establishes: