Cost and Return on Healthcare AI: What the Research Actually Shows
Published cost-effectiveness and economic evaluations of healthcare AI. A short list, which is itself the finding.
Here is the uncomfortable result, and it is the reason this subject exists as its own page.
Healthcare AI has tens of thousands of published papers. Formal economic evaluations of it number in the hundreds. So when a vendor tells you the return is proven, the honest response is that the return is mostly modeled, in a handful of use cases, under assumptions somebody chose.
That is not an argument against buying. It is an argument for writing your own business case instead of accepting theirs, and for treating the published economic work as a source of method rather than a source of numbers you can borrow.
The use cases with real economic modeling behind them cluster tightly: screening programs, colonoscopy, retinopathy, and a few imaging workflows. If your use case is not in that cluster, nobody has done the sums yet and the first honest slide in your business case should say so.
What Kind of Evidence This Is
Not every study carries the same weight. This is the mix behind the list below.
What to Measure in Your Own Setting
A published result is somebody else's hospital. These are the numbers worth tracking in yours.
- Your own baseline first. Most failed business cases are failures of knowing what the process costs today, not failures of the software.
- Cost per avoided event, not license cost. A cheap tool that changes nothing is the expensive option.
- The downstream load a tool creates. More detection means more follow-up, and that capacity is rarely in the original budget.
- Who captures the saving. Time returned to a clinician is only money if something is done with it, and that decision usually sits outside the project.
The Studies
Every entry links to the source record. Summaries are the study's own abstract, shortened but not reworded.
-
Automated Diabetic Retinopathy Image Assessment Software: Diagnostic Accuracy and Cost-Effectiveness Compared with Human Graders.
ObjectiveWith the increasing prevalence of diabetes, annual screening for diabetic retinopathy (DR) by expert human grading of retinal images is challenging. Automated DR image assessment systems (ARIAS) may provide clinically effective and cost-effective detection of retinopathy.
PMID 28024825 ... doi:10.1016/j.ophtha.2016.11.014
-
Cost-effectiveness of artificial intelligence for screening colonoscopy: a modelling study.
BackgroundArtificial intelligence (AI) tools increase detection of precancerous polyps during colonoscopy and might contribute to long-term colorectal cancer prevention. The aim of the study was to investigate the incremental effect of the implementation of AI detection tools in screening colonoscopy on colorectal cancer incidence and mortality, and the cost-effectiveness of such tools.MethodsWe conducted Markov mode...
PMID 35430151 ... doi:10.1016/s2589-7500(22)00042-5
-
Cost savings in colonoscopy with artificial intelligence-aided polyp diagnosis: an add-on analysis of a clinical trial (with video).
Background and aimsArtificial intelligence (AI) is being implemented in colonoscopy practice, but no study has investigated whether AI is cost saving. We aimed to quantify the cost reduction using AI as an aid in the optical diagnosis of colorectal polyps.MethodsThis study is an add-on analysis of a clinical trial that investigated the performance of AI for differentiating colorectal polyps (ie, neoplastic versus non...
PMID 32240683 ... doi:10.1016/j.gie.2020.03.3759
-
An observational study to assess if automated diabetic retinopathy image assessment software can replace one or more steps of manual imaging grading and to determine their cost-effectiveness.
BackgroundDiabetic retinopathy screening in England involves labour-intensive manual grading of retinal images. Automated retinal image analysis systems (ARIASs) may offer an alternative to manual grading.ObjectivesTo determine the screening performance and cost-effectiveness of ARIASs to replace level 1 human graders or pre-screen with ARIASs in the NHS diabetic eye screening programme (DESP).
PMID 27981917 ... doi:10.3310/hta20920
-
Cost-effectiveness of Artificial Intelligence for Proximal Caries Detection.
Artificial intelligence (AI) can assist dentists in image assessment, for example, caries detection. The wider health and cost impact of employing AI for dental diagnostics has not yet been evaluated. We compared the cost-effectiveness of proximal caries detection on bitewing radiographs with versus without AI.
PMID 33198554 ... doi:10.1177/0022034520972335
-
Cost-effectiveness of Artificial Intelligence as a Decision-Support System Applied to the Detection and Grading of Melanoma, Dental Caries, and Diabetic Retinopathy.
ObjectiveTo assess the cost-effectiveness of artificial intelligence (AI) for supporting clinicians in detecting and grading diseases in dermatology, dentistry, and ophthalmology.ImportanceAI has been referred to as a facilitator for more precise, personalized, and safer health care, and AI algorithms have been reported to have diagnostic accuracies at or above the average physician in dermatology, dentistry, and oph...
PMID 35289862 ... doi:10.1001/jamanetworkopen.2022.0269
-
Cost-effectiveness analysis of a risk-adapted algorithm of plerixafor use for autologous peripheral blood stem cell mobilization.
Historically, up to 30% of patients were unable to collect adequate numbers of peripheral blood stem cells (PBSCs) for autologous stem cell transplantation (ASCT). Plerixafor in combination with granulocyte colony-stimulating factor (G-CSF) has shown superior results in mobilizing peripheral blood (PB) CD34+ cells in comparison to G-CSF alone, but its high cost limits general use.
PMID 22922211 ... doi:10.1016/j.bbmt.2012.08.010
-
Artificial Intelligence and Diabetic Retinopathy: AI Framework, Prospective Studies, Head-to-head Validation, and Cost-effectiveness.
Current guidelines recommend that individuals with diabetes receive yearly eye exams for detection of referable diabetic retinopathy (DR), one of the leading causes of new-onset blindness. For addressing the immense screening burden, artificial intelligence (AI) algorithms have been developed to autonomously screen for DR from fundus photography without human input.
PMID 37729502 ... doi:10.2337/dci23-0032
-
Systematic review of cost effectiveness and budget impact of artificial intelligence in healthcare.
This systematic review examines the cost-effectiveness, utility, and budget impact of clinical artificial intelligence (AI) interventions across diverse healthcare settings. Nineteen studies spanning oncology, cardiology, ophthalmology, and infectious diseases demonstrate that AI improves diagnostic accuracy, enhances quality-adjusted life years, and reduces costs-largely by minimizing unnecessary procedures and opti...
PMID 40858882 ... doi:10.1038/s41746-025-01722-y
-
Cost-effectiveness of automated external defibrillators on airlines.
ContextInstallation of automated external defibrillators (AEDs) on passenger aircraft has been shown to improve survival of cardiac arrest in that setting, but the cost-effectiveness of such measures has not been proven.ObjectiveTo examine the costs and effectiveness of several different options for AED deployment in the US commercial air transportation system.Design, setting, and subjectsDecision and cost-effectiven...
PMID 11572741 ... doi:10.1001/jama.286.12.1482
-
Cost-effectiveness of implementing automated grading within the national screening programme for diabetic retinopathy in Scotland.
AimsNational screening programmes for diabetic retinopathy using digital photography and multi-level manual grading systems are currently being implemented in the UK. Here, we assess the cost-effectiveness of replacing first level manual grading in the National Screening Programme in Scotland with an automated system developed to assess image quality and detect the presence of any retinopathy.MethodsA decision tree m...
PMID 17585001 ... doi:10.1136/bjo.2007.120972
-
Cost-effectiveness of automated telephone self-management support with nurse care management among patients with diabetes.
PurposeThis study evaluated the cost-effectiveness of an automated telephone self-management support with nurse care management (ATSM) intervention for patients with type 2 diabetes, which was tested among patients receiving primary care in publicly funded (safety net) clinics, focusing on non-English speakers.MethodsWe performed cost analyses in the context of a randomized trial among primary care patients comparing...
PMID 19001303 ... doi:10.1370/afm.889
-
MLH1 promoter hypermethylation in the analytical algorithm of Lynch syndrome: a cost-effectiveness study.
The analytical algorithm of Lynch syndrome (LS) is increasingly complex. BRAF V600E mutation and MLH1 promoter hypermethylation have been proposed as a screening tool for the identification of LS. The aim of this study was to assess the clinical usefulness and cost-effectiveness of both somatic alterations to improve the yield of the diagnostic algorithm of LS.
PMID 22274583 ... doi:10.1038/ejhg.2011.277
-
Effectiveness and cost-effectiveness of letters, automated telephone messages, or both for underimmunized children in a health maintenance organization.
BackgroundImmunization rates have improved in the United States, but are still far from the national 90% goal for the year 2000. There is scant evidence about the effectiveness and costs of automated telephone messages to improve immunization rates among privately insured children.ObjectiveTo evaluate the effectiveness and cost-effectiveness of sending letters, automated telephone messages, or both to families of und...
PMID 9521970 ... doi:10.1542/peds.101.4.e3
-
Cost-effectiveness of artificial intelligence screening for diabetic retinopathy in rural China.
BackgroundDiabetic retinopathy (DR) has become a leading cause of global blindness as a microvascular complication of diabetes. Regular screening of diabetic retinopathy is strongly recommended for people with diabetes so that timely treatment can be provided to reduce the incidence of visual impairment.
PMID 35216586 ... doi:10.1186/s12913-022-07655-6
-
Cost-effectiveness of automated external defibrillator deployment in selected public locations.
ObjectiveThe American Heart Association (AHA) recommends an automated external defibrillator (AED) be considered for a specific location if there is at least a 20% annual probability the device will be used.
PMID 12950484 ... doi:10.1046/j.1525-1497.2003.21139.x
-
Consolidated Health Economic Evaluation Reporting Standards for Interventions That Use Artificial Intelligence (CHEERS-AI).
ObjectivesEconomic evaluations (EEs) are commonly used by decision makers to understand the value of health interventions. The Consolidated Health Economic Evaluation Reporting Standards (CHEERS 2022) provide reporting guidelines for EEs. Healthcare systems will increasingly see new interventions that use artificial intelligence (AI) to perform their function.
PMID 38795956 ... doi:10.1016/j.jval.2024.05.006
-
Economic evaluation of the one-hour rule-out and rule-in algorithm for acute myocardial infarction using the high-sensitivity cardiac troponin T assay in the emergency department.
BackgroundThe 1-hour (h) algorithm triages patients presenting with suspected acute myocardial infarction (AMI) to the emergency department (ED) towards "rule-out," "rule-in," or "observation," depending on baseline and 1-h levels of high-sensitivity cardiac troponin (hs-cTn).
PMID 29121105 ... doi:10.1371/journal.pone.0187662
-
Automated vs. manual cerebrospinal fluid cell counts: a work and cost analysis comparing the Sysmex XE-5000 and the Fuchs-Rosenthal manual counting chamber.
IntroductionCerebrospinal fluid (CSF) cell counts are traditionally performed by manual microscopy using the Fuchs-Rosenthal counting chamber. This procedure is time-, labour- and cost-intensive and requires experienced laboratory staff.MethodsThe Sysmex XE-5000 haematology analyzer offers a channel to quantify the total cell count of body fluids.
PMID 21668655 ... doi:10.1111/j.1751-553x.2011.01339.x
-
Cost-Effectiveness of AI for Risk-Stratified Breast Cancer Screening.
ImportancePrevious research has shown good discrimination of short-term risk using an artificial intelligence (AI) risk prediction model (Mirai). However, no studies have been undertaken to evaluate whether this might translate into economic gains.ObjectiveTo assess the cost-effectiveness of incorporating risk-stratified screening using a breast cancer AI model into the United Kingdom (UK) National Breast Cancer Scre...
PMID 39235813 ... doi:10.1001/jamanetworkopen.2024.31715
-
From KIDSCREEN-10 to CHU9D: creating a unique mapping algorithm for application in economic evaluation.
BackgroundThe KIDSCREEN-10 index and the Child Health Utility 9D (CHU9D) are two recently developed generic instruments for the measurement of health-related quality of life in children and adolescents. Whilst the CHU9D is a preference based instrument developed specifically for application in cost-utility analyses, the KIDSCREEN-10 is not currently suitable for application in this context.
PMID 25169558 ... doi:10.1186/s12955-014-0134-z
-
[Analysis of cost-effectiveness in the diagnosis of fever of unknown origin and the role of (18)F-FDG PET-CT: a proposal of diagnostic algorithm].
AimTo analyze the costs of Fever of Unknown Origin (FUO) prior to the PET-CT study. To determine the effectiveness of PET-CT in the diagnosis of FUO. A proposal of diagnostic algorithm.Material and methodsA retrospective study was performed that included 20 patients who had been studied between January 2007 and January 2011, with a mean age of 57.75 years and FUO diagnosis.
PMID 23067686 ... doi:10.1016/j.remn.2011.08.007
-
Cost-Effectiveness of Artificial Intelligence Support in Computed Tomography-Based Lung Cancer Screening.
BackgroundLung cancer screening is already implemented in the USA and strongly recommended by European Radiological and Thoracic societies as well. Upon implementation, the total number of thoracic computed tomographies (CT) is likely to rise significantly.
PMID 35406501 ... doi:10.3390/cancers14071729
-
Using a Machine Learning System to Identify and Prevent Medication Prescribing Errors: A Clinical and Cost Analysis Evaluation.
BackgroundClinical decision support (CDS) alerting tools can identify and reduce medication errors. However, they are typically rule-based and can identify only the errors previously programmed into their alerting logic. Machine learning holds promise for improving medication error detection and reducing costs associated with adverse events.
PMID 31786147 ... doi:10.1016/j.jcjq.2019.09.008
Study records come from Europe PMC, which indexes PubMed, MEDLINE, PMC and preprint servers. Titles, journals, years, identifiers, citation counts and abstracts are reproduced from the source record and are not rewritten here. Listing a study is not an endorsement of its conclusion, and citation count measures attention rather than quality. This page is a starting point for your own reading, not clinical guidance.