RxAnte

The prediction is the original product and remains the thing plans buy: modelling which members will lapse before they lapse, from pharmacy and claims history, is not something a rules engine does well.

The business has since acquired a second half that is not a model at all. The pharmacy service dispenses, packages, delivers and staffs pharmacist reviews, which is logistics and labour. That is the same split this index recorded on Deciphex, where separately saleable software sits beside a staffed service, and the record is scoped accordingly: the analytics are graded here and the pharmacy is described.

Low autonomy in the usual sense and appropriately so. The models rank members; pharmacists conduct the reviews, prescribers change the regimens, and members decide whether to enrol and whether to take anything.

The consequential automated act is selection rather than decision. Who appears on the outreach list determines who is offered intensive support, and that list is generated by a model. Nothing published describes whether a plan can override the ranking, whether clinicians can add members the score did not surface, or what happens to members who decline once and are then deprioritised.

Unusually candid about what the model optimises, which is the disclosure this index values most and which most vendors avoid.

The score is named, the software behind it is named, and the objective is stated in plain terms: identify the members with the greatest cost savings potential so a plan can engineer the return on its programme. The measure specifications the models are scored against are also identified, which lets an outside reader work out exactly what counts as success.

No model architecture, feature description or accuracy figure is published, so a buyer knows what the system is aiming at and not how well it hits. The company describes its analytics as patented, which is a pointer to a public document a diligent buyer could actually read.

Nothing identifies any party in the chain: no model or model family, no hosting arrangement and no sub processor list was located in two passes, and no retention schedule, encryption detail or secondary use position was found. What that unnamed chain holds is a plan's complete pharmacy and claims history across millions of members, which is among the more revealing datasets in healthcare for a reason worth stating precisely: what a person is prescribed discloses conditions they may never have discussed with anyone outside a consulting room, including conditions they have not disclosed to family or employers.

The pharmacy arm adds home delivery addresses and household context on top, so the combined holding places a medication history against a residence. Neither is governed by anything published. The commercial structure raises the pooling question that recurs across payer side products in this index: the value of a targeting model grows with the breadth of populations it has learned from, so a vendor has a structural incentive to pool across plans and a plan has a structural interest in knowing whether that happens. Nothing states whether one plan's member data informs models serving another. Ask that, ask for retention across the analytics and pharmacy arms separately, and ask for a sub processor list.

Better designed than most in this index and short of independent.

The published analysis compares members enrolled in the pharmacy service against matched members who filled prescriptions elsewhere, across total cost of care, unplanned care events and adherence at six months, with a hypothesis stated in advance that savings would concentrate in the high scoring subgroup. A matched control group and a pre stated subgroup hypothesis are real methodological choices and put this above the unattributed percentage claims common here.

Three things hold it at B. The analysis is self published rather than peer reviewed. It was conducted within a health plan belonging to the same organisation as the company's investor, so the study site is not arm's length. And matched controls are not randomisation: members who accept an intensive pharmacy service differ from those who decline in ways matching cannot fully capture.

Graded on an honest basis. No retention schedule, encryption detail or secondary use position was located in this pass.

The data is a plan's complete pharmacy and claims history across millions of members, which is among the more revealing datasets in healthcare: what someone is prescribed discloses conditions they may never have discussed with anyone else. The pharmacy arm adds home delivery addresses and household context on top. Nothing published describes how either is governed.

Graded on an honest basis. No compliance statement or agreement posture was located in this pass.

The structure is two sided in a way a buyer should map. Acting as an analytics vendor to a plan is one relationship; operating a pharmacy that dispenses to that plan's members is another, with its own licensure and its own direct relationship to the patient. Which instrument covers which flow is not described publicly.

Recorded honestly and provisionally: the dedicated trust and security search this index requires was not run in this pass, and no attestation was encountered incidentally.

Operating both an analytics platform across millions of member records and a licensed dispensing pharmacy implies two distinct control regimes, and neither was retrieved.

No device pathway applies and none is claimed.

The regulatory machinery that shapes this product is the Medicare Star Ratings system, and it deserves stating because it defines what the models chase. Medication adherence measures carry triple weight and bear on roughly half of a plan's total star performance, and star performance determines bonus payments. So the commercial value of this product is set by a federal quality programme rather than by a clinical outcome.

The pharmacy arm sits under an entirely different regime: state pharmacy licensure, dispensing rules and mail order requirements. A buyer evaluating the combined offering is evaluating a regulated dispensing operation as well as software.

No subgroup performance, monitoring policy or fairness analysis was located, and on this record the objective function itself is the governance question.

The score ranks members by the cost their non adherence would generate. That is a legitimate way to run a programme with finite pharmacists, and it is not the same as ranking by clinical need. A member whose lapse is dangerous but inexpensive ranks below one whose lapse is costly, and the two are not the same people. The company deserves credit for naming the criterion openly rather than describing it as risk; this index would rather grade an honest objective than guess at a one.

The measure set narrows it further. The adherence measures that drive star performance cover a small number of drug classes, so a member non adherent to a medication outside those classes generates no measurable value and is correspondingly less likely to be surfaced. The incentive is precise, and so is what it leaves out.

Nothing published addresses whether outreach, enrolment or outcomes differ by race, income, language or rurality, in a programme aimed explicitly at medically complex and vulnerable members.

This vendor states plainly what its model optimises for, and that is the disclosure this index values most on an allocation product because it is the one almost nobody makes. The score is named, the software behind it is named, and the objective is given in ordinary language: identify the members with the greatest cost savings potential so a plan can engineer the return on its programme.

The measure specifications the models are scored against are also identified, so an outside reader can work out exactly what counts as success. Candour of that kind should be credited, and it should also be read for what it discloses.

An allocation model optimised for savings potential is not optimised for clinical need, and those diverge in a predictable direction: a member with serious medication problems and little scope for cost reduction ranks below one whose adherence gap is expensive to the plan. The vendor is not hiding this, it is selling it, and a plan choosing the product is choosing that objective. What is absent is any measure of how well it hits the target it names.

No model architecture, feature description, calibration or accuracy figure is published, and no warranty, indemnity or remediation commitment was located. The analytics are described as patented, which points to a public document a diligent buyer could read. Ask for calibration by subgroup, and what happens to members who rank low on savings and high on risk.

The record system is not the relevant integration and the record is graded on that basis. The models run on plan side pharmacy and medical claims, and the outputs go to plan quality teams and to the company's own pharmacists.

The interesting connection is at the other end. Optimising a regimen requires the prescriber to agree, and the pharmacy service states that it works with prescribers and caregivers on regimen changes. How that contact happens, and whether anything reaches the prescriber's own system rather than arriving as a call or a fax, is not described.

Not described. No hosting model, region or retention position was located for the analytics.

The pharmacy operation has a physical footprint that the analytics does not, since medications are dispensed, packaged and shipped from a licensed facility, and that changes the operational picture for a plan evaluating national coverage.

Nothing published: no price, no mechanism, no unit of sale for either half of the business.

The structure raises a question that should be asked directly rather than inferred. The company's analytics identify which members would benefit most from intensive pharmacy support, and the company also owns the pharmacy that provides it. That is a coherent product and it is also a vendor scoring the demand for its own service. A plan should establish whether the analytics can be bought without the pharmacy, whether the score behaves the same way for plans that use a different pharmacy, and how each side is priced, since a per member analytics fee and a dispensing margin are entirely different commercial objects.

Concentrated on one buyer and deep within it. The centre of gravity is Medicare Advantage prescription drug plans, where the star measures create the strongest incentive, with coverage extending to HEDIS and patient experience measures for other lines.

Stated reach is substantial: more than 14 million lives under the analytics, and a claim to manage Part D star performance for 70 percent of the lives in five star plans nationally. Clinically the scope follows the star measures, so a small number of chronic medication classes dominate, while the pharmacy service targets medically complex members on multiple daily medications. United States only, which is inherent since the product exists because of how this quality and payment system works.