India Diabetes Care Report 2026

1 · Hybrid care model

A doctor-led, protocol-driven, structured program journey

Lillia’s hybrid care model extends the clinic doctor with AI-enabled digital tools and human coaches (health coaches, nutritionists, exercise coaches and diabetes educators), empowering the doctor to practise evidence-based medicine.

Lillia’s own proprietary care pathway is designed around the latest American Diabetes Association (ADA) principles and finalised together with our clinic partners, so the right intervention reaches the right patient at the right moment.

The model is built around four core elements:

In-depth metabolic analyticsIn depth analysis of metabolic profiles based on high density data eg interstitial glucose data (every 3 minutes), SMBG readings, lab reports, food logs and other health parameters.

Personalized lifestyle interventionsTargeted interventions focused on sustainable micro-habit changes eg meal sequencing, balancing of macronutrients, portion control, daily step count targets, post meal walks etc.

Proprietary journey management and coachingA structured model focused on habit formation, reinforcement and modulation to drive sustained behavioural change.

Integrated care delivery modelA coordinated, doctor-led approach combining clinic doctors, human-in-the-loop coaching and an agentic AI model operating within defined clinical guardrails, with appropriate human oversight and escalation to ensure safe, consistent and personalized care delivery.

The care pathway is built to serve every patient persona the clinic sees: those newly diagnosed and those living with diabetes for over a decade, users who are severely uncontrolled and those already at target, users who are on medicines and those who never have been, and patients seen in Tier 1, Tier 2 and Tier 3 clinics or online.

Start

Doctor consultation sets the clinical plan

Day 1

Baseline HbA1c, goal setting and CGM begin, and health tracking on Lillia’s mobile application

Day 6 and 15

Human coaches turn CGM reports, food logs and activity logs into a personalised action plan

Day 7, 17 and 30

Early check-ins build the habits that decide completion

Day 60 and 90

Check-ins and joint consults sustain progress and adjust the plan

Day 120

HbA1c and doctor consultation confirm outcomes and set the next step

2 · Participant demographics

Four in five users enrol with uncontrolled diabetes, and half already have a complication

Uncontrolled diabetes remains the majority presentation: 80% of users enrolled above 7%, against 76% in the 2025 report.

80% of users enrolled with HbA1c above 7%
39% enrolled with HbA1c above 9%
20% enrolled with HbA1c below 7%

HbA1c at enrolment. 2026: 3,099 users. 2025 figures are from last year’s report cohort of 1,918 users, shown for comparison.

Switch the year to compare

Report year

Tier 2 and Tier 3 cities remain a substantial part of the base, accounting for 47% of users.

47% of users are from Tier 2 and Tier 3 cities, as Tier 1 grew faster
A new online-only pathway now serves almost one in four users

Users by city tier. 2026: 2,397 clinic users with a known city; 702 users (23%) on the online pathway.

Switch the year to compare

Report year

Working-age adults remain the core cohort; the share under 40 rose to 19% from 17%.

54% are aged 40 to 59, unchanged from 2025
19% are under 40, up from 17% in 2025

Share of users by age group

Half of users report a complication at enrolment; neuropathy is the most common.

Neuropathy is the most common, at 13% of users
Retinopathy 4%, nephropathy 3%
Other or unspecified complications affect 35% of users

Complications recorded for 3,098 of 3,099 users; 49.6% have none.

Diabetes complications at enrolment, share of users

Most users with diabetes also have associated health conditions, most commonly obesity, hypertension and dyslipidemia.

57% of users are overweight or obese; 25% have hypertension
16% have dyslipidemia and 10% a thyroid disorder

Comorbidities at enrolment, share of users

3 · Program impact

In the GLP-1 era, lifestyle change and structured care still drive glycaemic control

Addressing root causes through nutrition, activity and behaviour change continues to drive glycaemic control.

The average HbA1c fell by 1.3 points, from 8.8% to 7.5%, compared with 1.2 points in 2025.

Four in five users ended the program with a lower HbA1c
The share of users with HbA1c below 7% rose from 20% at baseline to 42% at program end

Average HbA1c, baseline to program end

Outcomes by baseline severity

Select a group

Baseline HbA1c

Improved control coincided with reduced medication.

Among users whose medication status was recorded at both enrolment and program end:

40% of users reduced or discontinued their medication, while 9% required an increase
10% stopped all their diabetes medication, and 72% of those reached below 7%
Users who were never on medication also lowered their HbA1c, and 52% of them reached target (below 7%)

Medication status at program end

Greater weight loss, greater HbA1c reduction.

Among users starting at 7 to 9%, a loss of 10% or more of body weight corresponded to a 1.5-point reduction, against 0.5 points with no weight loss.

Users with baseline HbA1c 7 to 9%.

HbA1c reduction in points, by weight lost

4 · Drivers of impact

What makes the difference

Whether a user completes the four-month program is decided in the first 30 days.

68% of users who did all three early check-ins completed the four-month program
42% completed with two check-ins, 20% with one

Completion rate by early check-ins done in the first 30 days.

Completion rate, by early check-ins done

Meaningful improvements, with or without in-person clinic visits.

1.5 points with online doctor consultations
1.3 points in Tier 1 clinics; 1.2 in Tier 2 and Tier 3

702 users had their doctor consultations online through the clinic chain's video OPD, with Lillia between consultations.

HbA1c reduction in points, by where the user saw the doctor

Earlier intervention, greater HbA1c reduction.

1.8 points within a year of diagnosis, 1.1 points after 10 years or more
HbA1c fell in every duration band; the largest reduction was in users who joined within a year of diagnosis

HbA1c reduction in points, by time since diagnosis

AI companion and human coach

Users who messaged the AI companion on 11 or more days had an HbA1c reduction of 1.6 points; users who never used it, 1.1
Users who used both the AI companion and a human coach: 1.6 points. Human coach only: 1.3. No chat: 1.2

927 users subscribed to the AI companion since July 2025; 75% used it (699 of 927). 1,276 users (40%) messaged a human coach.

Switch the view

Show

The Lillia care team

The experts driving impact

Certified nutritionists and diabetes educators

The takeaway

In the GLP-1 era, lifestyle change and structured care still drive glycaemic control.

Four in five of the 3,099 users who completed the four-month program ended it with a lower HbA1c, an average fall of 1.3 points
The largest improvement was in severely uncontrolled users (HbA1c above 9%): 2.5 points
With better control, medication could be reduced or stopped: 40% of users with a recorded medication status did so
Meaningful improvements, with or without in-person clinic visits

Voices of leadership

“By combining proven clinical results with a scalable hybrid model, we're positioning diabetes care for global value-based transformation.”

Dr. Arati Joshi · Chief Medical Officer

“Behind every improved glycemic reading is a doctor we support, across 300+ partner clinics, managing diabetes care pathways that are lowering health risks for patients today and tomorrow.”

Dr. Jacob Mathew · Clinical Director

Results are observational, drawn from the 3,099 users who completed the four-month program; they describe real-world outcomes, not a controlled trial.