For clinicians
Shape the vocabulary so one clinic's account means the same as another's. Your corrections are the point, not an afterthought.
The ‘aha moment’ has no code.
The medical field has standard codes for the drug, the dose, and the diagnosis. It has no widely accepted way to record what a patient actually experiences during an interventional session, so it rarely reaches the data.
Standards bodies and large institutions write these vocabularies today, and the industry must work with whatever they decide. Arcametric’s mission is to give practitioners a unified voice in setting the standards, starting with a shared structure that adapts to your workflow. Users can give feedback and request additions, so the vocabulary can grow out of what clinicians actually document. The more practitioners who record the same way, the better the chance the field adopts it.
Adoption
A standard takes hold when enough practitioners record the same way.
Today clinicians describe the same session in their own words, so two records of one treatment rarely carry the same meaning. That makes comparing results across sites far harder than it should be. A shared structure fixes that, and it spreads as practitioners adopt it.
The alternative
Someone will define these terms, and it may be an institution that does not treat patients.
Medical vocabularies usually come from a standards body or a large research institution. That body publishes a definition, and the practices that report data then work with it, whether or not it matches how they practice. A practitioner who stayed out of the process still works with the result.
Where we start
Records stay comparable across sites, even when workflows differ.
Structure is the starting point of the platform, not a report we assemble afterward. Preparation, dosing, integration, and follow-up go into the same shared fields, so the records line up before anyone tries to compare them. We publish that layout openly as the Coding Landscape, and it maps to the code sets research already uses: RxNorm for medications, LOINC and SNOMED CT for observations and measures. It does that without forcing a breakthrough into a box labeled "symptom."
The evidence
We mapped which concepts already have codes and which do not.
Each point below is a concept a practitioner may need to record during a session. A point sits higher when the concept matters more to the treatment, and further left when the code sets cover it less. That puts the worst problems in the upper left corner.
In the software today
Practitioners can record these experiences as structured data.
During a session, one tap marks a meaningful moment in the session record, timed to the second. When the experience becomes clear, the practitioner selects it from a set list: unity, an entity encounter, geometric imagery, and more.
The record keeps what was described and who described it. It never turns an experience into a diagnosis or an adverse event.
Privacy
The records carry no patient identity in the first place.
Arcametric provides no field for a patient's name, so a record is de-identified from the moment a clinician creates it. Any shared result also sits behind a small-cell suppression floor, which lowers the chance that one person stands out inside an aggregate.
Who builds it
The field should define its own vocabulary.
Researchers, biostatisticians, and medical-vocabulary specialists all have a claim on this, and so do the clinicians who were in the room. They watched the session and know what happened, so their corrections carry evidence a committee usually lacks. Contributors help set what each term means.
For researchers
Review the method and lend your name. We will walk you through it in a technical briefing, with no sales process.
FAQ
Common questions.
What exactly do you pool across sites, and what stays with the clinic?
Arcametric stores structured, de-identified clinical information in a shared platform. When a clinic has agreed to participate in cross-site analysis, Arcametric may combine selected fields to produce aggregate statistics. These fields may include treatment type, medication or device, route, dose, assessment scores, coded safety events, session timing, and follow-up status.
The clinic retains the identity layer. This includes patient names, dates of birth, addresses, phone numbers, email addresses, medical record and insurance identifiers, billing records, local communications, narrative clinical notes, and the local link between a person and their Arcametric PT code. Cross-site pooling does not mean sharing the clinic’s complete patient chart.
How do you keep data from many clinics from identifying a person?
The first protection is structural. Arcametric does not provide a field for entering a patient’s name, and the clinic keeps the identity-to-code link in its own system. Access to clinical records is also limited by site and user role.
Removing names is not enough by itself to guarantee that someone cannot be identified from other details. Before Arcametric releases broader pooled information, it must also verify rules for minimum group sizes, rare combinations, dates, geographic detail, site identifiers, and other information that could narrow a group to one person. No public row-level network dataset is currently offered, and no one should call pooled information release-ready until Arcametric has documented and tested those disclosure controls and the approval process. Formal HIPAA de-identification requires either Safe Harbor or Expert Determination when the HIPAA Privacy Rule applies.
Do I need patient consent or ethics-board approval to contribute or use this?
It depends on what you are contributing and how the information will be used.
Using the Coding Landscape as a reference, or submitting a coding correction that contains no patient information, generally does not involve patient consent or ethics-board review. Using Arcametric for routine clinical documentation is also not automatically a research activity.
Contributing patient-derived information to benchmarking, formal research, publication, or an external data release is different. Depending on the purpose, applicable law, agreements, and whether investigators can identify individuals, the activity may require patient consent, HIPAA authorization, an institutional review board or ethics-board review, a waiver, or a documented determination that the activity is not human-subjects research. The clinic, the research institution, or its authorized review body makes that determination. Arcametric does not replace that review.
How is the Coding Landscape different from registries and outcome frameworks that already exist?
A registry collects records under a defined program. An outcome framework identifies what should be measured and when. The Coding Landscape addresses a different problem: how to represent each part of an interventional mental health record so that information from different systems and clinics means the same thing.
It connects existing standards to their proper roles, such as RxNorm for medications, LOINC and SNOMED CT for measurements and observations, ICD for diagnoses, and MedDRA for safety reporting. It also identifies concepts, especially subjective treatment experiences, that do not yet have a broadly accepted code.
The Coding Landscape does not replace registries, validated instruments, or established code systems. It provides an organizing and crosswalk layer that can make the information used by those systems more consistent and interoperable.
Is the Coding Landscape published and citable, and how are changes decided?
Anyone may cite the Coding Landscape as a web-based working reference. It is not a peer-reviewed publication, an official standard, or a replacement for the source systems it references. It does not currently present a DOI or formal publication record. A citation should include the page title, Arcametric as the publisher, the page address, and the date you accessed it. Check each external code against its current authoritative source, because code systems change over time.
At this stage, Arcametric maintains the working proposal and makes the final editorial decisions about what appears in it. Corrections need evidence, provenance, and a clear mapping rationale. Arcametric records accepted changes in versioned releases and a public change log. The organizations that own RxNorm, LOINC, SNOMED CT, MedDRA, ICD, and other standards retain authority over their own code systems. Arcametric may submit proposals or change requests, but it cannot declare an external standard on its own.
Who builds this, and who decides?
Arcametric coordinates the platform and maintains the current Coding Landscape proposal. The work involves clinicians, clinical researchers, biostatisticians, medical-vocabulary specialists, data architects, privacy and ethics reviewers, and the organizations responsible for established code systems.
Product and working-draft decisions currently remain with Arcametric. Clinical and coding claims need evidence, documentation, and an open path to correction. External standards bodies decide what becomes part of their official systems. As the review community grows, Arcametric should publish reviewer roles, conflicts of interest, decision records, version history, and the evidence used for each material change.