ClinicalTrials.gg
CompletedNCT07647159Updated Jun 15, 2026

Voice Assistant for Outpatient Neurology Visits

An interventional study of AI-based voice assistant Service for automated pre-filling of outpatient neurology visit protocol in EMIAS in Burnout, Professional, Documentation and Medical Records Systems, Computerized, sponsored by Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department. Completed at 1 site in Russia. Open to participants aged 18 Years and older. Per ClinicalTrials.gov, last updated 2026-06-15.

Sponsored by Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department · Not applicable, Interventional, and Health services research

How we know

From the registry’s dates

  • Registered 7 months after the study started (first participant enrolled Oct 2025, registered May 2026).

Help build ClinicalTrials.gg — short surveys, remote interviews or testing, in your language. Take part →

Phase
Not applicable
Study type
Interventional
Enrollment
320
Allocation
Not applicable
Ages
18 Years and older
Sex
All
01

Study summary

Documentation duties account for a substantial portion of an outpatient physician's working time and reduce time available for direct patient interaction. Voice assistants based on automatic speech recognition and large language models are being developed to automate medical documentation across clinical specialties. However, such ambient AI-based services have not been systematically validated in Russian-language outpatient neurology practice integrated with a regional electronic health record platform.

This pilot, multicenter, prospective, before-after study evaluated the feasibility and preliminary effectiveness of a voice assistant Service designed to automatically pre-fill the structured outpatient neurology visit protocol in the Moscow regional medical information system (EMIAS). The Service implements a pipeline of streaming speech-to-text transcription, two-speaker diarization, and large language model-based mapping of the dialogue between the physician and the patient onto the fields of the standardized neurology examination protocol.

Five neurologists at five outpatient clinics in Moscow participated. The study comprised three stages: (1) baseline timing of consultations without the Service; (2) timing of consultations with the Service after a two-week adaptation period, with parallel evaluation of transcription and pre-fill quality and of physician and patient satisfaction; and (3) statistical analysis. Three hundred twenty consultations were timed (160 per stage). A stratified random sample of 30 audio-recording / generated-protocol pairs was used to evaluate Service quality; free-text fields were rated on a 5-domain Likert questionnaire and on a 10-point visual analogue scale, and binary fields were rated dichotomously to derive sensitivity, specificity, accuracy, Jaccard index, and false-positive rate. Patient satisfaction was assessed by the modified Patient Satisfaction Questionnaire 8 (PSQ-8); physician feedback was assessed by a custom questionnaire (including the Net Promoter Score) and by semi-structured in-depth interviews with thematic analysis using grounded theory.

The primary outcomes were the change between stages in (a) the time of focused physician attention to the patient and (b) the time spent filling and editing the protocol. Secondary outcomes addressed total consultation time, transcription quality (Word Error Rate), expert-rated quality of pre-filled fields, patient satisfaction, and physician satisfaction.

The study was conducted under the framework of the Moscow Healthcare Department experiment on the use of digital innovation technologies in health care (Order No. 153 of 21 February 2025), and was approved by the local independent ethics committee.

Read the detailed description

Rationale and evidence gap. Documentation burden is widely recognized as a leading driver of professional burnout among physicians and reduces the time available for direct patient interaction. International evidence on ambient AI scribes - systems that combine automatic speech recognition (ASR) and large language models (LLM) - shows consistent reductions in documentation time and stable or improved patient satisfaction. However, evidence specific to Russian-language outpatient neurology practice integrated with a regional electronic health record is lacking. Neurologists are among the medical specialties with the highest reported burnout rates internationally; the specifics of Russian-language medical terminology, the structured neurological examination, and integration with a regional medical information system preclude direct extrapolation of international findings.

Technical implementation of the Service. The Service implements a distributed pipeline: streaming audio capture by an EMIAS client module; REST-API transmission of audio chunks to the speech recognition service with two-speaker diarization (physician/patient) and contextual re-clarification of streaming transcripts; asynchronous routing of transcription results through Apache Kafka topics to a pre-fill subsystem; large-language-model-based mapping of the full transcript onto a JSON schema of the structured outpatient neurology examination protocol (document class code 23951) using clinical reference dictionaries (including ICD-10 via the MKB10_ACTIVE_SHORT_INFO_V2 reference table); and return of the structured pre-filled protocol to the physician's user interface for review, editing, and signing. Audio is recorded via an active HD-capsule microphone placed at the physician's workstation (OGG Vorbis, 16 kHz, ≥ 64 kbit/s, mono). A priori quality targets fixed in the technical documentation before the pilot were: Word Error Rate of streaming transcription ≤ 10 %; sensitivity, specificity, accuracy, and Jaccard index for binary protocol fields ≥ 0.9; mean expert rating on a Visual Analogue Scale ≥ 8 out of 10; share of consultations with protocol generation time ≤ 35 seconds ≥ 90 %; physician Net Promoter Score ≥ 0 %.

Methodological design choices. A non-randomized single-group before-after design was selected as appropriate for a first-line feasibility evaluation in real clinical practice: the same physicians and clinics provide their own baseline reference in Stage 1, eliminating between-physician confounding while preserving ecological validity. A two-week post-deployment adaptation period was prespecified to mitigate purely learning-curve effects on Stage 2 metrics. Timing of consultation stages was performed by trained experts of the Scientific and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Healthcare Department on the basis of three synchronized data sources for each consultation: (i) automatic EMIAS event logs from protocol creation to signing; (ii) screen recording of the physician's workstation (VocoScreen NG, version 4.01.1); and (iii) webcam recording of the physician's face and gaze direction. The tri-source synchronization reduces systematic bias common to single-source or self-reported timing studies.

Quality evaluation framework. A stratified random sample of 30 paired audio recordings and Service-generated protocols (six per physician) was used to evaluate transcription and pre-fill quality. Reference (gold-standard) transcripts were prepared manually by an expert with more than five years of experience. Free-text fields of the protocol were rated by four independent neurology experts on a five-domain Likert questionnaire (Relevance, Accuracy, Completeness, Conciseness, Linguistic correctness) and on a 10-point Visual Analogue Scale; binary fields were rated dichotomously to construct a confusion matrix and derive sensitivity, specificity, accuracy, the Jaccard index, and the false-positive (hallucination) rate. Inter-rater agreement among the four experts was quantified by Gwet's AC2, preferred over Cohen's κ for categorical classifications with asymmetric prevalence. Word Error Rate was computed in Python with the jiwer library against the reference transcripts.

Analytical approach. Quantitative between-stage comparisons used Student's t-test or the Mann-Whitney U test depending on distribution, with Pearson's χ² for categorical variables and Cohen's d as the effect-size measure (α = 0.05, two-sided). Qualitative analysis of semi-structured in-depth interviews with each participating neurologist followed the grounded theory approach of Strauss and Corbin, with open, axial, and selective coding.

02

Conditions studied

  • Burnout, Professional
  • Documentation
  • Medical Records Systems, Computerized
  • Speech Recognition Software
  • Health Services Research
  • Neurology

Keywords

  • voice assistant for clinicians
  • speech recognition
  • large language model
  • ambient AI
  • electronic health record
  • neurology
  • outpatient appointment
  • patient satisfaction
  • documentation burden
  • pilot study
03

In context

Burnout, Professional

218 studies on the registry are indexed under Burnout, Professional; 35 are open to participants now.

This study's enrollment of 320 is above the median of 80 across 162 interventional studies indexed under Burnout, Professional.

Browse Burnout, Professional studies →

Lead sponsor

Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department is the lead sponsor of 11 studies on the registry; 5 are open to participants now.

Counted across the registry records on this site, refreshed daily.

04

Who can participate

Ages eligible
18 Years and older
Sexes eligible
All
Accepts healthy volunteers
No

Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.

Inclusion criteria

  1. First-time outpatient neurology visit conducted by a participating physician according to the approved consultation script.
  2. Age 18 years or older.
  3. Signed to participate in the study.
  4. Signed consent to processing of personal data.
  5. Clinical condition that allowed the patient to complete the modified Patient Satisfaction Questionnaire 8 (PSQ-8).

Exclusion criteria

Exclusion Criteria:

  1. Consultation longer than 45 minutes.
  2. Consultation shorter than 5 minutes, not requiring a full examination (e.g., prescription renewal, clarification of follow-up questions only).
  3. Technical failures preventing valid interpretation of the consultation or its timing (microphone failure, malfunction of the recording software, malfunction of the Service).
  4. Deviation from the approved consultation script.
Terms in these criteria, in plain language(1)

Underlined terms in the criteria open these too. Explained by clinicaltrials.gg, not part of the registry record. The study team decides who can take part.

Informed consent

Before you join, the study team explains what taking part involves — its purpose, what happens, the risks and possible benefits — and answers your questions, so you can decide freely. Consent is usually recorded on a form (the informed consent form, ICF) signed by you or, where the law allows, by someone legally authorized to decide for you.

You can keep asking questions during the study, you are told of new information that matters, and you can choose to leave at any time.

05

Study design

Phase
Not applicable
Primary purpose
Health services research
Allocation
Not applicable
Intervention model
Single group
Masking
None (open label)
Enrollment
320 participants (actual)

Study arms

  • Experimental
    Outpatient Neurology Visits with AI Voice Assistant

    All participating neurologists conducted outpatient consultations using the Service after a two-week adaptation period. Each consultation followed the approved script; the Service captured the audio, transcribed it in real time, generated a pre-filled visit protocol via a large language model, and returned it for physician review and signing.

    Device: AI-based voice assistant Service for automated pre-filling of outpatient neurology visit protocol in EMIAS

Interventions

  • DeviceAI-based voice assistant Service for automated pre-filling of outpatient neurology visit protocol in EMIAS

    The Service is software using artificial intelligence technologies that automatically pre-fills the structured outpatient neurology visit protocol in the Moscow regional medical information system (EMIAS) based on the audio-recorded dialogue between physician and patient. The pipeline comprises: (1) streaming audio capture by the EMIAS audio-recording client module, (2) chunk-wise transmission to the speech recognition service via REST API, (3) routing of transcription results through Apache Kafka topics to the pre-fill subsystem, (4) large language model-based mapping of the full transcript onto the JSON schema of the target document using clinical reference dictionaries, and (5) return of the structured pre-filled protocol to the physician's user interface. Audio is recorded via an active HD-capsule microphone placed at the physician's workstation. The physician reviews each pre-filled field, edits when necessary and signs the protocol.

06

What researchers measure

Primary outcomes

  1. Change in physician focused-attention time on the patient

    Time (in minutes) of focused physician attention on the patient, including visual contact and active interaction during examination, measured from three synchronized sources (EMIAS logs, screen video, webcam video)

    Time frame: Up to 14 weeks: pre-intervention measurements collected per consultation during study Weeks 1-2 (Stage 1); post-intervention measurements collected per consultation during study Weeks 5-14 (Stage 2, after a 2-week adaptation period).

  2. Change in protocol-filling time

    Time (in minutes) spent filling and editing the visit protocol, including keyboard input, mouse clicks, copying and pasting, measured from the three synchronized sources.

    Time frame: Up to 14 weeks: pre-intervention measurements collected per consultation during study Weeks 1-2 (Stage 1); post-intervention measurements collected per consultation during study Weeks 5-14 (Stage 2, after a 2-week adaptation period).

Secondary outcomes

  1. Change in total consultation time

    Total time (in minutes) from creation to signing of the visit protocol, derived from EMIAS event logs

    Time frame: Up to 14 weeks: pre-intervention measurements collected per consultation during study Weeks 1-2 (Stage 1); post-intervention measurements collected per consultation during study Weeks 5-14 (Stage 2, after a 2-week adaptation period).

  2. Transcription quality (Word Error Rate, WER)

    WER (%) of automatically generated transcripts versus expert reference transcripts on a stratified random sample of 30 consultations (6 per physician), computed with the jiwer Python library

    Time frame: Up to 14 weeks: assessed once per consultation on a stratified random sample of 30 audio recordings collected during study Weeks 5-14 (post-intervention period).

  3. Expert-rated quality of free-text protocol fields

    Mean rating of free-text protocol fields by four independent neurology experts using a modified expert evaluation questionnaire ("E-5"), scored on a 5-point Likert scale ranging from 1 ("completely disagree") to 5 ("completely agree"). Each of the "Complaints", "History of present illness", and "Recommendations" sections of 30 Service-generated protocols is rated across five domains: Relevance, Accuracy, Completeness, Conciseness (scored on an inverted scale, where higher values denote less redundant content), and Linguistic correctness. Each domain score ranges from 1 to 5; for all five domains, higher scores indicate better field quality (higher scores reflect a better outcome).

    Time frame: Up to 14 weeks: assessed once per protocol on a stratified random sample of 30 Service-generated protocols from consultations conducted during study Weeks 5-14 (post-intervention period).

  4. Quality of pre-filling for binary protocol fields

    Sensitivity, specificity, accuracy, Jaccard index, and false-positive (hallucination) rate for binary fields of 30 auto-generated protocols rated by four independent experts

    Time frame: Up to 14 weeks: assessed once per protocol on a stratified random sample of 30 Service-generated protocols from consultations conducted during study Weeks 5-14 (post-intervention period).

  5. Overall protocol quality (Visual Analogue Scale)

    Mean expert rating of the overall quality of each Service-generated visit protocol on a Visual Analogue Scale (VAS) ranging from 1 to 10. The scale anchors are 1 = "lowest possible quality" and 10 = "highest possible quality"; higher scores indicate better protocol quality (higher scores reflect a better outcome). Each of the 30 sampled Service-generated protocols is rated independently by four neurology experts; the mean of all expert ratings across protocols is reported.

    Time frame: Up to 14 weeks: assessed once per protocol on a stratified random sample of 30 Service-generated protocols from consultations conducted during study Weeks 5-14 (post-intervention period).

  6. Patient satisfaction (modified PSQ-8)

    Mean scores on the modified Patient Satisfaction Questionnaire 8 (PSQ-8) - an 8-item self-administered patient satisfaction instrument adapted to outpatient neurology - across seven reported domains: Comfort during the consultation; Perceived physician attention (focus on the patient versus on documentation); Understanding of the physician's explanations; Patient involvement in clinical decision-making; Trust in the prescribed treatment; Understanding of subsequent management tactics; and Knowledge of actions to take in emergency situations. Each item is rated on a 5-point Likert scale ranging from 1 ("absolutely not") to 5 ("yes, completely"). Each domain score ranges from 1 to 5; higher scores indicate greater patient satisfaction (higher scores reflect a better outcome).

    Time frame: Up to 14 weeks: pre-intervention questionnaires collected per consultation during study Weeks 1-2 (Stage 1); post-intervention questionnaires collected per consultation during study Weeks 5-14 (Stage 2).

  7. Physician satisfaction (Net Promoter Score)

    Net Promoter Score (NPS) for physician satisfaction with the Service, calculated from a single 0-to-10 likelihood-to-recommend question administered to each participating neurologist at the end of Stage 2. The item anchors are 0 = "would never recommend" and 10 = "would definitely recommend". Respondents are classified as Detractors (score 0-6), Passives (score 7-8), or Promoters (score 9-10). The Net Promoter Score is calculated as (% Promoters) - (% Detractors). The NPS ranges from -100 (all respondents are Detractors) to +100 (all respondents are Promoters); higher scores indicate greater physician willingness to recommend the Service (higher scores reflect a better outcome).

    Time frame: At the end of study (study Week 14): single measurement collected from each participating physician upon completion of Stage 2.

07

Study locations

1 site
  • Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department
    Moscow, 125051, Russia
08

References and documents

Individual participant data

Plan to share: No — Individual participant data will not be made publicly available due to the requirements of Federal Law No. 152-FZ of the Russian Federation "On Personal Data", the conditions of the signed informed consent forms, and the decision of the local independent ethics committee that approved the study protocol.

No publications or documents are linked to this record.

09

Timeline

Tracking since Sep 25, 2026
ElapsedEstimatedEstimated, not yet reachedSimilar studies: half by the dot; the band is the middle halfHalf of similar studies reached it by the dot; the band is the middle half, from a quarter to three quarters

Ran from Oct 2025 to primary completion in Jan 2026, 3 months.

Of 128 non-industry interventional studies with no phase under Burnout, Professional, half reached primary completion within 11 months of starting (the median), counting those still running; a quarter within 6 and three quarters within 22. This study's 3 months is below that.

Counting every sponsor, half of 132 studies reached it within 11 months; a quarter within 6 and three quarters within 23.

How these figures are worked out, and which studies are left out of them.

Registered
May 27, 2026
First posted
Jun 15, 2026
StartActual
Oct 26, 2025
Primary completionActual
Jan 30, 2026
CompletionActual
Jan 30, 2026
Last update
Jun 15, 2026
What these dates mean
Start.
When the first participant enrolled; while estimated, when the study expects to open to recruitment.
Primary completion.
When the last participant was examined or treated to collect data for the study's main question, its primary outcome. A study that stopped early gives the date its data collection stopped.
Completion.
When the last participant's last visit took place, for every outcome, secondary ones and side effects included: the end of the study. Often later than primary completion, sometimes the same day.
Actual and Est.
Actual means the registry says the date has happened. Est. (estimated) is a date the sponsor plans or expects. It can still move, and the sponsor is expected to change it to actual once it has passed.
No changes since tracking began. The registry record was last updated on Jun 15, 2026, before this site started recording changes on Sep 25, 2026. Its history is on ClinicalTrials.gov ↗
10

Registry details

Key details

Study ID
NCT07647159
Lead sponsor
Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department
Collaborators
Information and Analytical Center of the Moscow Healthcare Department
Responsible party
Sponsor
First posted
Jun 15, 2026
Start date
Oct 26, 2025
Primary completion
Jan 30, 2026
Completion
Jan 30, 2026
Last update
Jun 15, 2026

Study contacts

Yuriy Vasilev, MD, PhD
study director · Research and Practical Clinical Center for Diagnostics and Telemedicine Technologies of the Moscow Health Care Department

Oversight

Data monitoring committee
No
FDA-regulated drug
No
FDA-regulated device
No
View the source record on ClinicalTrials.gov ↗

Not currently enrolling

This study is completed, as verified in Jun 2026. You cannot join it, but the record below documents what was studied.

Follow this study

Get an email when the registry record changes — status, dates, results — or when someone posts here.

Sign in to follow

Discussion

Questions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.

Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.

Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.

Start the discussion