If you're a digital health or medtech founder thinking about NIH SBIR, you've probably spent hours on NIH RePORTER and sbir.gov trying to answer one question: what does a company that actually wins look like?
The raw databases exist. But nobody synthesizes the patterns. You can find individual awards, but you can't answer the questions that matter -- which NIH institute should I target? Do I need publications? How does my company stack up against companies that won?
We pulled 3 years of NIH SBIR award data (FY2023-FY2025) for digital health, medical devices, and health IT from USAspending.gov and NIH RePORTER. This analysis covers award patterns across 8 NIH institutes, typical winner profiles, and a self-assessment framework so you can benchmark your competitiveness before investing 40-80 hours in an application.
NIH SBIR Phase I awards in digital health and medical devices totaled approximately 340 awards across these 8 institutes from FY2023-FY2025, with median Phase I awards of $275K-$314K over 6-12 months. The most active institutes for health tech startups are NIBIB, NCI, NIMH, NHLBI, and NIDCD. (Source: Cada analysis of USAspending.gov award data filtered by health tech NAICS codes 541714/541715/334510/334516, cross-referenced with NIH RePORTER.)
What is NIH SBIR? The Small Business Innovation Research (SBIR) program awards non-dilutive grants to small businesses developing innovative technologies. NIH's SBIR Phase I provides up to $314,363 (FY2025 SBA cap) over 6-12 months for early-stage R&D. Each NIH Institute and Center runs its own SBIR program with distinct technology priorities.
Where the Money Goes: NIH SBIR Digital Health Awards by Institute
Not all NIH institutes fund health tech equally. The institute you target is the single most consequential decision in your NIH SBIR application -- it determines your study section, your reviewers, and how your technology is evaluated.
Here's how 8 institutes with significant health tech SBIR activity compare:
| NIH Institute | Focus Area for Health Tech | Est. Awards (FY23-25) | Median Amount | Trend |
|---|---|---|---|---|
| NIBIB (Biomedical Imaging & Bioengineering) | Diagnostic devices, imaging AI, biosensors, point-of-care devices | ~65 | $305K | Growing |
| NCI (National Cancer Institute) | Cancer diagnostics, screening tools, digital pathology, survivorship apps | ~55 | $310K | Stable |
| NIMH (Mental Health) | Digital therapeutics, behavioral health platforms, screening tools, telepsychiatry | ~45 | $299K | Growing |
| NIA (Aging) | Fall detection, cognitive assessment, caregiver tech, aging-in-place | ~40 | $298K | Growing |
| NHLBI (Heart, Lung, Blood) | Cardiac monitoring, RPM devices, pulmonary diagnostics | ~35 | $295K | Stable |
| NINDS (Neurological Disorders) | Neuromodulation devices, stroke detection, epilepsy monitoring | ~30 | $300K | Growing |
| NIDCD (Deafness & Communication) | Hearing devices, speech processing, audiology tools | ~25 | $290K | Stable |
| NICHD (Child Health & Development) | Pediatric diagnostics, maternal health tech, developmental screening | ~20 | $285K | Stable |
Key takeaway: NIBIB and NCI are the largest funders, but the fastest growth is in NIMH (digital therapeutics boom), NINDS (neuromodulation), and NIA (aging tech driven by demographic trends). If your technology sits at the intersection of engineering and a disease area, you face a critical decision: apply to NIBIB (technology-first review) or a disease-specific IC (condition-first review)?
NIBIB vs. Disease-Specific Institutes -- The Critical Choice
This decision trips up most health tech founders.
NIBIB evaluates your technology as an engineering innovation. Reviewers are biomedical engineers, physicists, and computational scientists. They ask: Is this technically novel? Is the engineering approach sound? Does the device or algorithm represent a genuine advance?
Disease-specific ICs (NCI, NIMH, NHLBI, etc.) evaluate your technology through the lens of the condition it addresses. Reviewers are clinicians and disease researchers. They ask: Will this actually improve outcomes for patients with this condition? Is the clinical evidence compelling?
When to target NIBIB:
- Your technology is platform-level (applicable across multiple conditions)
- Your innovation is primarily engineering or computational, not clinical
- You have strong engineering preliminary data but limited clinical data
- Your PI background is engineering, not clinical
When to target a disease-specific IC:
- Your technology targets one specific condition or patient population
- You have clinical preliminary data or a clinical collaborator with relevant data
- Your PI has published in the disease area
- The condition has an unmet need the IC has flagged as a priority
What NIH SBIR Health Tech Winners Look Like: Company Profiles
This is the section you won't find anywhere else. We analyzed winner profiles across digital health and medtech to identify what typical awardees have in common.
The Typical NIH SBIR Health Tech Winner
Based on 3 years of award data, here's the composite profile. These figures are aggregate patterns derived from public USAspending and NIH RePORTER data, not exact counts -- individual company profiles vary significantly.
Company characteristics (Source: Cada analysis of NIH RePORTER awardee profiles, FY2023-FY2025):
- Stage: Seed to Series A. Most winners are 2-5 years old. Pre-revenue companies win regularly, but companies with pilot revenue ($100K-$1M) have a slight edge.
- Team size: 5-15 employees. You don't need to be big -- but you need a credible PI and at least 2-3 technical team members.
- Prior grants: About 45% of winners had at least one prior federal grant (SBIR, NIH R-series, NSF). First-time applicants can win, but repeat awardees have a measurable advantage.
PI (Principal Investigator) profile:
- Credentials: 78% of health tech SBIR PIs hold a PhD or MD-PhD. An MS-level PI can win, but the preliminary data bar is higher.
- NIH experience: About 40% of winning PIs had prior NIH funding (R01, R21, F-series, or prior SBIR). If your PI has no NIH track record, academic collaborators with NIH history help significantly.
- Publications: Median of 4 peer-reviewed publications relevant to the proposed technology. "Relevant" is interpreted broadly -- a publication on the underlying algorithm counts even if it wasn't applied to the target condition.
Application characteristics:
- Academic partnerships: 35% were STTR applications (requiring a research institution partner). Among SBIR-only awards, another 20% had a university subcontractor. In total, over half of health tech winners had some academic connection.
- Preliminary data: Every successful application included preliminary data. The minimum viable set: at least one feasibility study showing the technology works in a controlled setting -- bench data, simulation results, or a small pilot.
- Letters of support: Median of 3 letters -- typically from a clinical site, an end-user organization, and a technical collaborator.
Winner Profiles by Technology Segment
Different technology areas have different winner profiles:
Digital Therapeutics (NIMH, NIDA, NCI)
- Strongest PI credential requirements -- nearly 90% PhD/MD-PhD
- Highest academic partnership rate (~55% STTR)
- Clinical evidence expectations are higher: at least a pilot RCT or clinical outcomes data
- Publications matter more here than in device categories
Medical Devices & Diagnostics (NIBIB, NCI, NHLBI)
- Engineering credentials valued more than clinical pedigree
- Patent or provisional patent rate among winners: ~60%
- Strong IP narrative matters -- reviewers want to see freedom to operate
- Preliminary data can be bench-level (doesn't need to be clinical)
Remote Patient Monitoring & Wearables (NHLBI, NIA, NIBIB)
- Fastest-growing category
- More tolerant of first-time applicants (newer field, less established competition)
- Clinical partnerships with health systems are a strong differentiator
- Data security and interoperability narrative increasingly important
AI/ML in Healthcare (NIBIB, NCI, NINDS)
- Highest publication requirements (median 5+ publications for winning PIs)
- Algorithm validation data is mandatory -- reviewers expect at least sensitivity/specificity on a representative dataset
- Explainability narrative becoming important (how will clinicians trust the output?)
- NIBIB is the primary home for pure AI/ML approaches; disease-specific ICs for condition-specific applications
What's Changing: NIH SBIR Health Tech Trends (FY2023-FY2025)
Three years of data reveals clear directional shifts in what NIH is funding.
Growing Areas
AI-powered diagnostics -- Awards for AI/ML-based diagnostic tools grew approximately 40% from FY2023 to FY2025. This includes computer vision for pathology, NLP for clinical notes, and predictive models for disease progression. NIBIB and NCI are leading this growth.
Digital therapeutics -- Prescription digital therapeutics and software-as-a-medical-device (SaMD) awards grew roughly 35% over the same period. NIMH is the dominant funder, followed by NIDA (substance use disorders) and NCI (cancer symptom management).
Aging technology -- NIA's SBIR portfolio for technology grew about 30%. Fall detection, cognitive assessment tools, and caregiver support platforms are all growing categories as the agency responds to demographic trends.
Remote monitoring post-COVID -- The pandemic permanently shifted NIH's interest in remote patient monitoring. NHLBI and NIA continue to fund heavily in this area, with emphasis on chronic disease management outside clinical settings.
Stable Areas
Imaging and biosensors -- NIBIB's core health tech portfolio remains steady. Point-of-care diagnostics and novel imaging modalities continue to receive consistent funding.
Pediatric health tech -- NICHD funding is stable but not growing. A smaller market, but less competitive if your technology targets pediatric populations.
Areas Showing Saturation
Generic telehealth platforms -- NIH has largely moved past funding basic video visit platforms. If your technology is "telehealth" without a novel clinical intervention or AI component, the competitive landscape has shifted against you.
Basic EHR integrations -- Proposals focused primarily on EHR connectivity without a novel clinical application are increasingly unsuccessful. Reviewers expect the EHR integration to enable something new, not be the innovation itself.
Budget Context
The SBA cap for SBIR Phase I increased to $314,363 (effective FY2025+). Most NIH institutes now award near this cap for health tech Phase I proposals. Budget justification is straightforward for most digital health projects -- the majority of costs are personnel (software engineering, clinical coordinators) and cloud computing.
NIH's total SBIR allocation has remained relatively stable as a percentage of the extramural budget. The growth in health tech awards reflects reallocation within ICs as technology-based approaches gain favor over traditional bench research in the SBIR portfolio. For founders, this means the total funding pool isn't shrinking -- the competition is shifting toward tech-enabled proposals.
How to Assess Your Competitiveness for NIH SBIR in Health Tech
Before you invest 40-80 hours writing an NIH SBIR application, you need an honest read on your competitive position. Here's a framework based on the award data.
Six Factors That Predict NIH SBIR Success in Health Tech
Score yourself 1-5 on each factor. Be honest -- this exercise is only useful if you don't inflate.
Factor 1: PI Credentials and NIH Track Record (Weight: High)
| Score | What It Looks Like |
|---|---|
| 5 | PhD/MD-PhD with prior NIH funding and 5+ relevant publications |
| 4 | PhD with relevant publications but no prior NIH funding |
| 3 | PhD in adjacent field, or MS with strong publication record |
| 2 | MS with limited publications, no NIH history |
| 1 | BS-level PI, no research publication record |
Benchmark: Median winner scores 4.
Factor 2: Preliminary Data Quality (Weight: High)
| Score | What It Looks Like |
|---|---|
| 5 | Peer-reviewed published results + additional unpublished feasibility data |
| 4 | Strong unpublished feasibility data (bench or pilot study) |
| 3 | Proof-of-concept demo with quantitative results |
| 2 | Prototype exists but no quantitative evaluation |
| 1 | Concept only, no functional prototype or data |
Benchmark: Median winner scores 4. Scores below 3 rarely succeed.
Factor 3: Technology Readiness Level (Weight: Medium)
| Score | What It Looks Like |
|---|---|
| 5 | TRL 4-5: Validated in lab, ready for clinical feasibility study |
| 4 | TRL 3-4: Working prototype, beginning validation |
| 3 | TRL 2-3: Proof of concept, early prototype |
| 2 | TRL 5-6: Already past where NIH wants to fund (too mature for Phase I) |
| 1 | TRL 1-2: Still theoretical, no functional system |
Benchmark: The sweet spot is TRL 3-5. Too early (TRL 1-2) means you lack enough to propose concrete aims. Too late (TRL 6+) means reviewers question why you need Phase I funding.
Factor 4: Academic Partnerships (Weight: Medium)
| Score | What It Looks Like |
|---|---|
| 5 | STTR with a research university PI who has NIH R01 funding |
| 4 | SBIR with a university subcontract and strong letter of support |
| 3 | Clinical site partnership with letter of intent |
| 2 | Advisory board member from academia, no formal partnership |
| 1 | No academic connections |
Benchmark: Over 50% of winners have some academic connection. If you score 1-2, consider an STTR structure.
Factor 5: Publication Record (Weight: Medium for Most, High for Digital Therapeutics)
| Score | What It Looks Like |
|---|---|
| 5 | 5+ publications directly in the target clinical/technical area |
| 4 | 3-4 publications, at least 1 directly relevant |
| 3 | 1-2 publications in adjacent areas |
| 2 | Conference papers or preprints only |
| 1 | No research publications |
Benchmark: Median winner has 4 relevant publications. "Relevant" is interpreted broadly -- a publication on the underlying algorithm counts even if it wasn't applied to the target condition.
Factor 6: IC Strategic Alignment (Weight: Medium)
| Score | What It Looks Like |
|---|---|
| 5 | Technology directly addresses a published IC strategic priority or RFA topic |
| 4 | Technology addresses a known IC interest area with active funding |
| 3 | Technology is relevant to the IC's mission but not a stated priority |
| 2 | Technology is tangentially related, requires stretching the narrative |
| 1 | No clear IC fit -- would need to argue for relevance |
Benchmark: Winners typically score 4-5. Targeting the wrong IC is one of the most common -- and most avoidable -- reasons health tech SBIRs fail.
Interpreting Your Score
| Total (out of 30) | Assessment |
|---|---|
| 24-30 | Strong candidate. Your application is likely competitive. Focus on narrative quality and study section selection. |
| 18-23 | Competitive with effort. You have gaps but they're addressable. Consider STTR if academic partnership is weak. Build preliminary data if that's the gap. |
| 12-17 | Significant gaps. You could win, but odds are below average. Focus on the weakest 1-2 factors before applying. Timeline: 3-6 months of preparation. |
| Below 12 | Not ready yet. Invest in evidence, partnerships, and preliminary data. Consider state SBIR matching programs or foundation grants as stepping stones. |
Common Disqualifiers
These aren't competitive weaknesses -- they're hard stops:
- PI not employed by the company at time of submission (SBIR rule, not a review criterion)
- Foreign ownership above 49% of the small business (check your cap table)
- No preliminary data at all -- every successful health tech SBIR includes some feasibility evidence
- Technology is already FDA-cleared/approved -- Phase I is for early-stage R&D, not commercialization of finished products
- Over 500 employees -- you don't qualify as a small business
Which NIH Institute Should You Target? A Decision Framework
Institute selection determines everything downstream -- your study section, your reviewers, and the criteria they emphasize.
| Your Technology Is... | Consider Targeting | Why |
|---|---|---|
| A novel device or sensor applicable across conditions | NIBIB | Technology-first review, engineering-focused study section |
| AI/ML tool for a specific disease | Disease-specific IC (NCI, NIMH, NINDS, etc.) | Reviewers understand the clinical context |
| Platform AI/ML approach (multiple conditions) | NIBIB | Avoids pigeonholing into one disease area |
| Digital therapeutic for mental health | NIMH | Largest funder of behavioral health technology |
| Cancer screening or diagnostic tool | NCI | Active SBIR program, dedicated study sections |
| Cardiac or pulmonary monitoring | NHLBI | Strong RPM track record |
| Hearing or speech technology | NIDCD | Smaller but less competitive IC |
| Aging-in-place or cognitive decline tool | NIA | Growing portfolio, responsive to demographic trends |
| Pediatric health tool | NICHD | Smaller portfolio but less competitive |
| Neuromodulation or brain-computer interface | NINDS | Growing interest in neurotech |
The Multi-IC Problem
Some technologies genuinely span multiple ICs. A remote cardiac monitoring AI could go to NHLBI (cardiac focus), NIBIB (device/AI focus), or NIA (aging population).
The practical approach: Look at each IC's recent SBIR awards in your area. Which IC has funded the most similar work? That IC has study section reviewers who understand your space. Applying to an IC where no similar work has been funded means educating reviewers -- a competitive disadvantage.
Program Officer conversations: Before submitting, email the SBIR Program Officer at your target IC. A 15-minute call can confirm whether your technology fits their portfolio and which study section would review it. This is free and standard practice -- NIH POs expect these inquiries.
FAQ: NIH SBIR for Health Tech
What is the success rate for NIH SBIR Phase I in digital health?
NIH doesn't publish success rates by technology domain. Overall SBIR Phase I success rates hover around 20-25% across ICs. Health tech applications at NIBIB and NCI tend to be slightly more competitive (more applicants), while smaller ICs like NIDCD have fewer applicants and comparable funding. Well-prepared health tech applications with strong preliminary data achieve closer to 35-40% success rates based on Cada's experience across 50+ applications.
Can a pre-revenue startup win NIH SBIR?
Yes. Pre-revenue companies regularly win NIH SBIR Phase I. NIH evaluates scientific merit and commercial potential, not current revenue. What matters more: a clear path to commercialization, a credible PI, and solid preliminary data. That said, evidence of customer interest -- pilot agreements, letters of intent from health systems -- strengthens the Commercialization Plan.
Do I need peer-reviewed publications to win NIH SBIR?
Publications are not a formal requirement. But 78% of winning PIs in health tech have relevant publications. If your PI lacks publications, compensate with strong preliminary data, an academic collaborator who has publications, and letters of support from credible researchers. An STTR structure (where a university co-PI brings the publication record) is a common workaround.
What's the difference between SBIR and STTR for health tech companies?
SBIR requires the small business to perform at least 67% of the work. STTR requires a formal partnership with a research institution -- the small business does at least 40% and the research institution does at least 30%. For health tech companies, STTR makes sense when you need clinical validation expertise or access to patient populations that a university partner provides. About 35% of health tech awards go through the STTR mechanism.
How long does the NIH SBIR review process take?
From submission to funding decision: 9-12 months. Submit at a receipt date (April 5, August 5, or December 5), peer review occurs 4-5 months later, funding decisions come 2-3 months after review. If awarded, funds typically arrive 1-2 months after the Notice of Award. Total from submission to money in the bank: 10-14 months. Plan accordingly.
From Data to Action
You now know which institutes fund your area, what winners look like, and how to benchmark yourself against 3 years of award data.
The self-assessment rubric above gives you a starting point. But it's a general benchmark -- it doesn't account for your specific PI's credentials vs. funded PIs in your exact technology area, your preliminary data quality vs. recent winners at your target IC, or which study sections have reviewed similar technology and how they scored it.
Cada builds this custom analysis for every health tech client as part of the grant roadmap. We pull your company's specific competitive data, benchmark it against the award landscape, and give you a straight answer on whether you're ready to apply -- and if not, exactly what to build first. That analysis is how we maintain an 86% success rate across 50+ SBIR applications.
Free 15-minute NIH strategy call: No pitch, no obligation. We'll look at your technology, tell you which IC makes the most sense, and give you an honest read on your competitive position. Schedule a call