Precision clinical analytics

Reduce avoidable variation, mortality and harm. Deliver higher quality care at lower cost.

C2-Ai has developed, successfully tested and proved demonstrable benefits of our unique clinical Ai systems in 11 countries including the UK and USA – helping hospitals reduce avoidable harm and mortality, generate signficant savings on operating expenditure and reduce clinical negligence claims by up to 10%
Hospitals and health insurance companies face serious issues:
  • Variability in patient care (increasing treatment costs)
  • Unclear how to recognise best practice
  • Difficult to reduce treatment costs
  • Need to improve quality of care
  • Desire to reduce clinical negligence claims

There are no real alternatives.  Business information systems in hospitals are at best smoke alarms and we’ve seen up to 95% false positives on issues we cover.   You cannot measure performance by comparing care and safety metrics to an average across hospitals. 

A hospital performing higher risk surgery will see higher complication rates than those that do not.  Typical BI systems will flag the first hospital because they do not consider the actual patients being treated, their physiology, risks of the procedure etc. 

We risk adjust accurately for each patient and can tell which hospitals, specialties, consultants etc. are doing well, where the hospital has issues and what the causes are. 

Spotlight on UK (12 months)

In the UK alone we estimate our systems have avoided harm and mortality (shown right), as well as money saved on the direct costs of treatment created from avoidable harms.

These figures ignore reductions in other costs, complaint handling and negligence/malpractice related litigation and awards.

Lives saved
Harms avoided

Why is our validated and Ai-backed approach different?

Our systems can provide real-time, risk-adjusted assessments for individual patients going forward

That’s like driving a car and having all the information you need to help you make turns, choose the right route etc.




Business Information systems are reviewing average performance across all patients with no/limited risk-adjustment and up to 12 months after the events

That’s like driving a car looking in the rear view mirror with binoculars at a journey you were on a year ago!


Why our Ai-backed, risk adjusted results go beyond current approaches

Where quality is reported on the basis of mortality rates and readmissions, the true picture may be very different.

We understand and show the difference between unadjusted mortality and the true picture when case-mix and complexity of patients treated is taken into account.

In the diagram on the right, notice that in the period when overall mortality is at its highest (A), the case-mix adjusted quality is nevertheless good.

Points plotted below the yellow line represent better than predicted outcomes, benchmarked against international as well as UK practice.

Our unique systems deliver:

  • risk adjustment individualised for every patient retrospectively and in real-time
  • lowered avoidable harm and mortality
  • significantly reduced complaints and clinical negligence related costs
  • fast (4 weeks) feedback on observed to expected hospital performance for surgery and the care of deteriorating patients
  • significant savings relating to the reductions in avoidable harm and mortality
  • prospective assessment of the risk of complications and mortality, individualised to patients pre-operatively – identifying patients that should be optimised and help the hospital manage the individualised risk of complications 
  • faster resolution of issues that cost money, lives and harm in hospitals
  • lower readmissions
  • lower complaints and clinical negligence related costs
  • direct cost savings per hospital
  • lower costs per treatment and variability 
  • lower conversion of day surgery resulting in in-patient admissions



million patient records
countries providing patient data
Countries we are operating in
years of research



surgical complications
avoidable harm triggers
% of surgical interventions
% of a patient's diagnoses



average potential saving/hospital
complaints/negligence claims
% reduction in AKI
return on investment in systems

Our effective analytical algorithms have been developed, validated and refined over 14 years. They are backed up by detailed clinical research and built on our unique, proprietary referential dataset. 

Our approach is proven to work across different geographies and is not only in use in hospitals globally but has also been used by regulatory authorities and in governmental studies of harm and mortality in hospitals.

Typical BI systems do not adjust for a patient’s complications and the risk and severity of any treatments.  So a hospital doing a great job of higher risk surgeries often flag up with these systems because the associated complication rates are also naturally higher.
We understand and our accurate risk adjustment means we can identify where the real issues are without false positives.
A commonly used information system in the US does not look at every patient and relies on surgeons to voluntarily report outcomes.  This skews results.  For example, we have seen the rate for anastamotic leakage reported by this system at only 50% of the rate we see in our global datasets.  
That system is not providing a representative or accurate view of performance.  We do.


Some systems’ mortality scores are affected by the order clinical codes are entered into them as they don’t consider all codes for the patient.  Entering codes for the severest codes first means some systems will reduce the ‘weighting’ of the related deaths in any assessment.  In some cases we’ve even seen these systems ‘gamed’ to mask real performance issues rather than resolve them.
We consider every clinical code for all patients and our systems cannot be skewed or gamed.

What people say about our solutions

C2-Ai have the most robust software approach to comparing safety and quality across hospitals, systems and physicians that I have ever seen. The algorithms are backed up by years of published international research. I believe their approach could be most useful as a solution for providers across any network.

James C Bonnette, MD
Executive Vice President, the Advisory Board (USA).

“It took 2 years and a very costly investigation to deal with a competence issue in our organisation some time ago. We set CRAB® the blind challenge of seeing if they could have found the problem in our historical data. They did so in 20 minutes. Needless to say, we have invested in the system

Dr. Michael Roberts
Chief Medical Officer, Northland District Health Board, New Zealand

CRAB® can identify outcomes that are better than expected, as well as those that are worse, and thus can be used as an improvement tool as well as to assure clinicians and others of the standard of care being provided, and to measure productivity

Lord Ara Darzi

“CRAB® predictions have proved accurate in my primary external research validation of the system. I currently don’t know of any other electronic system in use that can deliver this kind of overall and detailed qualitative feedback to the department and the individual surgeon. It has been a great benefit for our clinic and helped develop our work on patient safety.”

Wilhelmina Ekström, MD, PhD
Senior Consultant, Karolinska University Hospital, Sweden

“To have reports of this quality dropping on to my iPad is a real joy”

Marcus Bankes
Consultant Orthopaedic Surgeon

“This exercise is not about making data/surgeons/departments look good, but about being accurate so that performance can truly be assessed”

Steve Corbett
Consultant Orthopaedic Surgeon and Clinical Lead

“CRAB has allowed real time review of data, which has raised awareness and led to change in both clinical practice and hospital culture. I think it will become an essential part of the appraisal and governance structures of secondary care.”

David Williams
Consultant Surgeon, Northern Devon Hospital Trust

“We use CRAB to provide us with a detailed monthly audit report of each surgeons complications, adjusted against CRABs risk methodology, in order to monitor outcomes performance. Our Clinical Leads meet monthly and review any complications that arise in order to inform and learn from the evidence. We find CRAB provides outcomes intelligence quickly, enabling us to be ‘on top’ of outcomes immediately.
One of the significant gains from using CRAB is a massive improvement in our coding from an engaged clinical workforce”

Brian Wells
Former Director of Orthopaedics, GSTT

“CRAB just makes sense. It presents data in a way that is easy to understand and interpret. It has been immensely useful for me both personally in my appraisal and in my role as a Clinical Director. It helps me to pick up early warning of problems with intelligence that can be believed and acted upon”

Jeremy Cundall
Consultant Colorectal and General Surgeon - Executive Director, CDDFT

“CRAB® is 100% better than any solution available to us at the moment. It has turned out to be a very useful tool in analysing and understanding our case-mix and where our complications are occurring.”

Per Svedmark MD, PhD
Senior Consultant, Stockholm Spine Centre, Sweden

“We are thrilled to receive this award, which reflects a great deal of hard work and support from within and […] the support of CRAB® Clinical Informatics (C2-Ai) who enabled us to benchmark our Trust’s AKI rates against national levels – and then measure the significant impact of our AKI Programme, which coincided with a significant and sustained fall in AKI rates our Trust, especially across surgical wards”

Dr Jonathan Murray
Renal Consultant at South Tees Hospitals NHS Foundation Trust, UK

The problem with [HSMR systems] is that they tell you there might be a problem, but not where or why. CRAB® tells you exactly what and where the problem is, and even which patients are involved. Then you can do something about it.”

Dr. Aresh Anwar
Medical Director Royal Perth Hospital, Australia

“CRAB has provided data for the consultants to really understand their outcomes for patients in a user friendly way.”

Alison Diamond

“CRAB is generating trusted data which we can use to flag up areas of concern. From there we are able to take action in a much more sophisticated way than we have in the past.”

Timothy Ho
Medical Director, Frimley Health NHS Foundation Trust

“I have worked with C2-Ai for the past seven years, and their insights have shone a light on the quality and safety of patient care that other less sophisticated analyses have not been able to provide”

Consultant in Emergency Medicine, John Radcliffe Hospital, Oxford
Accurate benchmarking of outcomes was a real challenge to us as an independent hospital with limited access to big data sets, however our work with CRAB analytics has provided invaluable quality assurance. The risk adjusted reporting has provided confidence that our outcomes are better than comparable organisations and the level of detail enables us to focus on improvements in specific areas.  It was particularly useful during our regulatory inspection and follow up meetings with the CQC to show how this strengthens our clinical governance and contributes to our ‘effective’ and ‘well led’ domains.  
Dr Jenny Davidson
Director of Governance, King Edward VII's Hospital.

Our Ai-backed systems

Click on a heading to learn more

CRAB helps hospitals identify and resolve issues relating to avoidable harm and mortality. We provide:

  • Timely information rapidly and accrue related savings in opex etc.
  • Accurate identification of issues (rather than false positives) as results are adjusted for individual patient’s physiology and pre-existing conditions as well as any procedural risk.
  • Fast indication of problems (as quickly as four weeks to allow for considering readmissions within 30 days).
  • Help for hospitals to recognise there is a real issue and then drill-down to the root cause.

A clinician uses Compass to assess the risks of complications and the real risk of mortality based on patient’s physiology, existing triggers and the prospective operation. This can be done in around one minute and then can be used to:

  • Avoid treating unsuitable patients and acting on predicted complication risks to reduce harms and mortality
  • Stops issues before they arrive (pre-operative and specific areas such as AKI and sepsis)

Hospitals and finance directors under pressure to make savings the clear choice is to cut services and staff, but this can place them in direct conflict with the founding principles of healthcare and the professionals who deliver it.

Our ground-breaking model built around our core algorithms and analytical engine is called the Harris Unit.

It creates a sophisticated but accessible ‘profit and loss’ to show on a single page the economic impact (positive and negative) of the quality of care across a hospital.

We have been validated around the World

We receive patient data from 46 countries and operate in 11.

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from Youtube
Consent to display content from Vimeo
Google Maps
Consent to display content from Google
Consent to display content from Spotify
Sound Cloud
Consent to display content from Sound