The next generation of clinician platforms will not only answer questions or document consultations. They will generate strategic insight from clinician behaviour — what clinicians ask, what they struggle with, what they trust, what they repeat, and where the knowledge infrastructure fails them at the point of clinical decision-making.
Doximity's DocInsight is an early and visible signal of this shift. Daniel Brox, who recently joined Doximity, described DocInsight as a new insights, analytics, and strategy offering built with clients, using Doximity's physician engagement and behavioural data to support decisions for life-sciences innovators. He framed it as being built on top of Doximity's existing physician platform, data science capability, and AI-first infrastructure — a natural extension of a platform that already sits inside the daily workflow of 85% of US physicians.
The strategic implication: clinical platforms are no longer just tools that clinicians use. They are becoming market-intelligence infrastructure — where the aggregate patterns of clinician behaviour create commercial, educational, and clinical value that extends far beyond the individual user interaction.
The Scale Behind DocInsight
Doximity's scale creates a behavioural dataset unlike anything traditional HCP research can produce. With 3 million+ registered members, 800,000+ active prescribers using workflow tools in Q4 FY2026, nearly half of those using clinical AI tools, and prompts per user nearly doubling from January to April — the platform captures what physicians actually search for, document, prescribe, reference, and engage with during their working day. This is not survey data collected once a quarter. It is continuous behavioural data generated through real clinical workflow, every day, at population scale.
The 2026 State of AI in Medicine report drew on 3,151 physicians across 15 specialties and found that 94% are using AI or interested, with adoption rising from 47% to 63% between the two study cohorts. Literature search is the leading use case (35%), followed by voice-based documentation (29%). These are survey-level insights. DocInsight operates at a different level: using the continuous behavioural data from daily platform engagement to inform life-sciences strategy, medical education planning, evidence generation, market-access decisions, and product positioning.
The commercial model is a closed-loop flywheel: workflow tools create clinician engagement → engagement generates behavioural data → data feeds insight products (DocInsight) → insight products attract life-sciences partnerships → partnership revenue funds product development → better products attract more clinicians → more engagement → more data → more insight. Each layer reinforces the others. The flywheel accelerates with scale.
Why the UK Version Must Be Different
The US model is built on life-sciences engagement intelligence and physician-platform monetisation at scale. The UK model must account for fundamentally different clinical, regulatory, procurement, and trust dynamics.
Guideline architecture. UK clinical practice is organised around NICE, CKS, SmPC/eMC, MHRA drug safety updates, and SIGN — authoritative sources whose recommendations create specific clinical obligations. Understanding how UK clinicians engage with these sources — where they find guidance helpful, where they find it confusing, where they repeatedly need clarification at specific decision points — is a different analytical task from tracking US physician engagement patterns. The friction points are UK-specific because the guidelines are UK-specific.
Regulatory landscape. UK market entry for clinical AI requires MHRA medical device classification assessment, DTAC compliance, DCB 0129/0160 clinical safety standards, UK GDPR data protection, and NHS procurement readiness. None of these exist in the same form in the US. US life-sciences intelligence — however sophisticated — does not tell a product team whether their clinical AI tool will pass MHRA assessment, meet DTAC criteria, or satisfy NHS governance requirements. UK-specific regulatory intelligence is a separate and essential requirement.
Procurement and adoption. NHS adoption involves institutional procurement, clinical governance approval, information governance assessment, and ICB/Trust-level decision-making — processes that can take months or years and require specific evidence, governance documentation, and clinical champion engagement. The US model of clinician-direct distribution or enterprise SaaS sales does not apply in the same way. Understanding NHS procurement dynamics is not optional for UK market success.
Professional trust dynamics. UK clinician trust is shaped by GMC/GPhC professional accountability, NHS clinical governance, revalidation and appraisal requirements, and a relationship between clinicians and commercial health technology that is structurally different from the US market. UK clinicians are accountable for every clinical decision regardless of what AI tool informed it. Trust must be earned through provenance, transparency, regulatory compliance, and alignment with UK professional standards — not through scale or brand recognition alone.
Where iatroX Insights Fits
iatroX Insights is the advisory, research, and partnership division of iatroX — helping digital-health companies, NHS partners, research groups, medical publishers, and life-sciences organisations design, validate, and scale trustworthy healthcare technology in the UK.
The division is not a UK clone of DocInsight. It is a UK-focused model built around different data (clinical queries grounded in UK guidelines rather than US physician engagement), different outputs (regulatory readiness, guideline-friction intelligence, clinician validation rather than life-sciences engagement analytics), and different trust requirements (NHS governance, MHRA compliance, GMC/GPhC professional accountability).
The four pillars:
UK clinical-intent insights. Aggregated, anonymised analysis of what clinicians ask, check, repeat, and struggle to apply — drawn from the patterns of clinical question behaviour across the iatroX platform. These patterns reveal where the UK knowledge infrastructure serves clinicians well and where it fails them.
Guideline-friction intelligence. Mapping where UK guidance is hard to operationalise in real clinical work — which NICE recommendations generate repeated queries, which prescribing decisions create uncertainty, which referral thresholds cause confusion, which monitoring intervals are ambiguous. A novel intelligence product with no direct UK equivalent.
Regulatory and adoption intelligence. MHRA classification advisory, DTAC readiness assessment, DCB 0129/0160 clinical safety case development, CSO services, claims review, and NHS market-entry preparation — the regulatory infrastructure that UK clinical AI products need and that US intelligence products do not provide.
Custom partnerships and projects. White papers, opt-in clinician validation studies, bespoke clinical AI workflows, educational partnerships, market-entry projects, co-branded research, and clinical AI pilot design — built with partners rather than sold as off-the-shelf reports.
The Trust Boundary
iatroX Insights operates within explicit ethical boundaries: aggregate and anonymised data only, opt-in research where individual clinician participation is involved, no hidden individual profiling, no patient-identifiable data, no manipulation of clinical answers based on commercial relationships, and clear labelling of partnership content. The clinical answer layer (Ask iatroX) is independent of the insight layer — commercial partnerships do not influence clinical answers. This independence is non-negotiable.
The Proposition
Doximity's DocInsight reflects a US platform model built on physician engagement at scale. iatroX Insights is a UK-focused model built around guideline-grounded clinical questions, regulatory readiness, clinician validation, and ethical aggregate insight. Different market, different data, different intelligence — same underlying strategic logic: clinical platforms are becoming intelligence engines.