Others titles
- Veterans Health Administration
- Content Coding for Contextualization of Care
- Medical Center Primary Care Clinics
Keywords
- Physician-Patient Communication
- Contextualization of Care
- Transcripts of Medical Encounter
- Contextual Error
- Patient-centered Care
- Physician-patient Interactions
- Guidelines for Coding Plan of Care
Physician Patient Transcripts With 4C Analysis
This dataset shows This dataset consists of 405 transcriptions of audio recorded physician-patient interactions conducted at Veterans Health Administration (VHA) medical center primary care clinics. The interactions were analyzed using an early form of Content Coding for Contextualization of Care (“4C”).
Get The Data
- ResearchNon-Commercial, Share-Alike, Attribution Free Forever
- CommercialCommercial Use, Remix & Adapt, White Label Log in to download
Description
This dataset consists of audio recorded physician-patient transcriptions and interactions conducted at Veterans Health Administration (VHA) by the Department of Veterans Affairs. The recordings were collected utilizing concealed (except where indicated) audio recorders by patients. The protocol was approved by VHA Institutional Review Boards, and participating physicians and patients consented to participate in the study. These transcripts are a resource to research scientists seeking transcriptions of primary care encounters, as well as those interested in 4C coding in its early stages. Their acquisition was supported with research funding from the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Health Services Research & Development.
Content Coding for Contextualization of Care, or “4C,” is a tool for assessing how effectively clinicians provide care that is contextualized, and therefore patient-centered. Whereas other instruments focus on the process of the interaction (“Did the doctor encourage questions?”), 4C focuses on specific events and behaviors (“Did the doctor ask why this patient recently missed two appointments?”). When inattention to patient context results in an inappropriate plan of care, the oversight is coded as a “contextual error.”
Contextual issues can be spotted by the attentive clinician who is attuned to them, particularly if he/she has developed the habit of always considering them in the “differential” for a patient with a health care need. Sometimes there are obvious hints, such as a patient who seems to have stopped taking his medications saying “boy it’s been tough since I lost my job” (loss of health insurance) or sometimes it’s more subtle such as a patient appearing confused about his medication when asked (cognitive loss). In the absence of such hints – hints that we have termed “contextual red flags” – the clinician can simply ask a direct question: “I notice your blood pressure is no longer well controlled. This could be just the natural progression of your condition, but I want to be sure. Are you having any trouble taking your medication as prescribed? Can you tell me exactly how you take it and when? Did you miss any dosages today?”
Contextual red flag identification:
– Following an encounter, the first step for coders is to determine if there were any contextual issues — “contextual red flags” — that the clinician would need to have explored to provide appropriate care. A contextual red flag is defined as anything a patient says or that was observed about their situation or behavior that suggests unaddressed contextual factors may be contributing to problems with their care. The search for contextual red flags begins with a structured chart review by a contextual red flag “screener” looking for evidence of missed appointments, non-adherence with medications or follow through on tests or labs, or preventable deterioration of a chronic condition, such as poor diabetes or blood pressure control.
Coder listens for clinician probe:
– The coder then listens for whether the provider pursued the contextual red flag in a manner that substantively approximates their direct probe.
Identifying Red Flags:
– The first step in 4C is screening for contextual red flags. A contextual red flag is defined as anything a patient says or that is observed about their situation or behavior that suggests unaddressed contextual factors may be contributing to problems with their care. Red flags can be found in the medical chart or by listening to issues brought up in the audio recording. Red flags are classified into 4 levels, with level 1 red flags identified first, then level 2, 3, and 4 sequentially.
About this Dataset
Data Info
Date Created | 2022-03-03 |
---|---|
Last Modified | 2022-03-03 |
Version | 2022-03-03 |
Update Frequency |
Irregular |
Temporal Coverage |
2009-2012 |
Spatial Coverage |
United States |
Source | John Snow Labs; Department of Veterans Affairs; |
Source License URL | |
Source License Requirements |
N/A |
Source Citation |
N/A |
Keywords | Physician-Patient Communication, Contextualization of Care, Transcripts of Medical Encounter, Contextual Error, Patient-centered Care, Physician-patient Interactions, Guidelines for Coding Plan of Care |
Other Titles | Veterans Health Administration, Content Coding for Contextualization of Care, Medical Center Primary Care Clinics |
Data Fields
Name | Description | Type | Constraints |
---|---|---|---|
Transcript_ID | Transcript for saving patient-physician record | string | - |
Patient_ID | Unique ID for patient-physician record | string | - |
Contextual_Red_Flag | A contextual red flag is defined as anything a patient says or that is observed about their situation or behavior that suggests unaddressed contextual factors may be contributing to problems with their care. R | string | - |
Red_Flag_Code | Red flags can be found in the medical chart or by listening to issues brought up in the audio recording. | number | level : Ratio |
Red_Flag_Description | Flag description | string | - |
Face_Time | Time of record | integer | level : Nominal |
Is_Red_Flag_Probed | Does the patient presented a contextual red flag, and the health care provider probed it | boolean | - |
Is_Contextual_Factor | Anything a patient says or that is observed about their situation or behavior that suggests unaddressed contextual factors may be contributing to problems with their care | boolean | - |
Is_Contextual_Plan_of_Care | A plan initiated by a health care provider to address contextual issues related to identified contextual problem | boolean | - |
Coding_Algorithm | The coding algorithm consists of a series of questions each coder asks themselves to identify thecpresence or absence of a contextual probe, a contextual problem and contextual POC. The algorithmcalways begins with the contextual red flag. | string | - |
Identified_Context_Code | Identified conext code | integer | level : Nominal |
Identified_Context_Description | Description of identified context | string | - |
Possible_Outcomes | Possible outcomes of the encounters code | string | - |
Outcome | Outcomeof encounters code | string | - |
Specific_Outcome | Specific outcomes of successes or failures to contextualize care | string | - |
Is_Recorder_Revealed | Contextual problem revealed in the encounter is contextual or not | boolean | - |
Is_Factor_Discovered_By_Probe | Was the factor doscovered by probe or not | string | - |
Is_Factor_Revealed_By_Patient | Was the factor revealed by the patient or not | boolean | - |
Data Preview
Transcript ID | Patient ID | Contextual Red Flag | Red Flag Code | Red Flag Description | Face Time | Is Red Flag Probed | Is Contextual Factor | Is Contextual Plan of Care | Coding Algorithm | Identified Context Code | Identified Context Description | Possible Outcomes | Outcome | Specific Outcome | Is Recorder Revealed | Is Factor Discovered By Probe | Is Factor Revealed By Patient |
Y39 | X103 | 2 missed appts in the past four months. | 2 | Process | 32.0 | False | False | False | 1. Red Flag: 2 missed appts in the past four months. 2. Ideal Probe: Why did you miss your appts? 3. Probe heard: none 4. Was it contextual probe? na 5. Is it close enough to the ideal probe to be credited? na6. Contextual Factor from probe: na 7. Contextual Factor revealed spontaneously: none 8. Ideal contextual P.O.C.: na 9. P.O.C. heard: na10. Was it contextual? na11. Is it close enough to the idea P.O.C. to be credited? na | False | |||||||
Y39 | X104 | Pt. who has diabetes & dizzy spells states he couldn't afford to go to the dentist and states "I'm not working." | 3 | Audio Via | 22.0 | False | True | False | 1. Red Flag: Pt. who has diabetes & dizzy spells states he couldn't afford to go to the dentist and states "I'm not working." 2. Ideal Probe: How is the economic burden of not working impacting your care? 3. Probe heard: Res re-states that pt. couldn't afford provider. 4. Was it contextual probe? No. 5. Is it close enough to the ideal probe to be credited? No. 6. Contextual Factor from probe: 7. Contextual Factor revealed spontaneously: Finances 8. Ideal contextual P.O.C.: Probe Economic/Emotional or other domains, target and apply. 9. P.O.C. heard: Labs day of, 3 months out. 10. Was it contextual? No.11. Is it close enough to the idea P.O.C. to be credited? na | 4.0 | Economic Situation | Pt. visits dentist | POOR | Pt does not visit the dentist | True | ||
Y37 | X105 | Pt. says "I could never get enough of those pills - I ran out of my pills so instead of taking the pill and a half, I switched to taking one" | 3 | Audio Via | 17.0 | False | False | False | 1. Red Flag: Pt. says "I could never get enough of those pills - I ran out of my pills so instead of taking the pill and a half, I switched to taking one" 2. Ideal Probe: Do you understand how much medicine you are supposed to be taking? 3. Probe heard: So you didn't have enough, huh? 4. Was it contextual probe? No 5. Is it close enough to the ideal probe to be credited? No6. Contextual Factor from probe: NA 7. Contextual Factor revealed spontaneously: None 8. Ideal contextual P.O.C.: na 9. P.O.C. heard: na10. Was it contextual? na11. Is it close enough to the idea P.O.C. to be credited? na | ||||||||
Y10 | X11 | 6 missed appts in 12 mo, RVA rate-68% | 1 | Outcomes | 25.0 | False | False | False | 1. Red Flag: 6 missed appts in 12 mo, RVA rate-68% 2. Idea Probe: The record shows you missed 6 appts in the last 12 months. Why did you miss those appts? 3. Probe heard: None 4. Was it contextual probe? NA 5. Is it close enough to the ideal probe to be credited? NA6. Contextual Factor from probe: 7.Contexual Factor revealed spontaneously: NA 8. Ideal contextual P.O.C.: NA 9. P.O.C. heard: NA10. Was it contextual? NA11. Is it close enough to the ideal P.O.C. to be credited? NA | Good- RVA rate improves Poor- RVA rate remains the same or gets worse. | GOOD | Total appts-4 Missed appts- 1 RVA rate-75% | |||||
Y10 | X11 | 1. Pt. refuses flu shot | 4 | Audio Via Coder | 25.0 | False | True | 1. Coder Identified Red Flag: Pt. refuses flu shot 2. Ideal Probe: Why don't you want to get a flu shot? 3. Probe heard: no 4. Was it contextual probe? na 5. Is it close enough to the ideal probe to be credited? na6. Contextual Factor ID'd: pt. misinformation/thinks he got the flu from the shot 7. Contextual Factor re: Coder ID'd New Red Flag revealed by pt. without specific probing by Dr.: no 8. Ideal contextual P.O.C.: educate pt that flu shots do not lead to the flu, they prevent it. 9. P.O.C. heard: 10. Was it contextual? no 11. Is it close enough to the idea P.O.C. to be credited? no. | 5.0 | Skills and Abilities | Pt. gets flu shot | POOR | refused to get a flu shot | True | |||
Y10 | X11 | 2. Pt. complains of reflux, thinking he's never had it before, chart indicates he has mentioned it before. | 4 | Audio Via Coder | True | False | False | 1. Coder Identified Red Flag: Pt. says this is the first time he's complaining about acid reflux even though the record indicates he has complained about it before. 2. Ideal Probe: The record shows you've complained about this before, why do you think it shows that? 3. Probe heard: Someone wrote in your record that you've complained about this before, where would they get that notion? Have you gotten any tests for it or take medication for it? 4. Was it contextual probe? Yes 5. Is it close enough to the ideal probe to be credited? Yes 6. Contextual Factor re: Coder ID'd New Reg Flag ID'd: 7. Problem revealed by pt. without specific probing by Dr.:NA 8. Ideal contextual P.O.C.: NA 9. P.O.C. heard: NA10. Was it contextual? NA11. Is it close enough to the ideal P.O.C. to be credited? NA | |||||||||
Y27 | X110 | 10 missed appts in the past 12 mo. RVA rate =67% | 1 | Outcomes | 19.0 | True | False | False | 1. Red Flag: 10 missed appts in the past 12 mo. RVA rate =67% 2. Ideal Probe: Why have you missed so many appts in the last year? 3. Probe heard: Dr. asked pt how his wife was doing. Dr. stated that sometimes he has seen her here with him. Dr. asked how pt's memory was doing. Dr. asked if wife helps pt do things at home or does he get around by himself. 4. Was it contextual probe? yes 5. Is it close enough to the ideal probe to be credited? Yes, though it's not clear that Dr. noticed missed appts. Earlier in encounter Dr. asks about pt seeing neurologist & pt says it's been awhile. Pt also remarked he wasn't sure when last time was he saw cardiologist. 6. Contextual Factor from probe: none 7. Contextual Factor revealed spontaneously: none 8. Ideal contextual P.O.C.: na 9. P.O.C. heard: na10. Was it contextual? na11. Is it close enough to the idea P.O.C. to be credited? na | ||||||||
Y46 | X111 | Pt. states he is out of his meds | 3 | Audio Via | 18.0 | True | True | True | 1. Red Flag: Pt. states he is out of his meds 2. Ideal Probe: Why did you run out of your meds? 3. Probe heard: How long have you been out? Is it an issue for you to get here, is that why you don't come often? 4. Was it contextual probe? Yes 5. Is it close enough to the ideal probe to be credited? Yes6. Contextual Factor from probe: 7. Contextual Factor revealed spontaneously: Pt said he can't come back for fasting blood test because he lives an hour from VA and can't take off more work. 8. Ideal contextual P.O.C.: Made sure pt knew HOW to refill meds. 9. P.O.C. heard: Explained that pt could call Dr. if meds expired and stated the importance of not just stopping meds, how it can be dangerous etc. Dr. stated that they would like to see him every 3 months but he can go to Oak Lawn clinic & give pt the number.10. Was it contextual? Yes11. Is it close enough to the idea P.O.C. to be credited? Yes, Access to Care | 3.0 | Access to Care | Pt. has enough meds | GOOD | Pt has all the meds | True | ||
Y39 | X113 | 2 missed appts in the past 4 months | 2 | Process | 22.0 | False | False | False | 1. Red Flag: 2 missed appts in the past 4 months 2. Ideal Probe: Why have you missed 2 appts in the past few months? 3. Probe heard: You haven't been to the VA too much in the past. It looks like you haven't established care in the gen med depart. 4. Was it contextual probe? no 5. Is it close enough to the ideal probe to be credited? no, hard one, but I think Dr. is referring to not a lot of info in pt's record and NOT noticing he missed appts.6. Contextual Factor from probe: na 7. Contextual Factor revealed spontaneously: none 8. Ideal contextual P.O.C.: na 9. P.O.C. heard: na10. Was it contextual? na11. Is it close enough to the idea P.O.C. to be credited? na | ||||||||
Y40 | X120 | 44missed appts in the past 12 mo. RVA rate =57% | 1 | Outcomes | 30.0 | True | True | True | 1. Red Flag: 44missed appts in the past 12 mo. RVA rate =57% 2. Ideal Probe: Why have you missed so many appts this past year? 3. Probe heard: Pt said he missed MRI. Dr. asked "How come?" 4. Was it contextual probe? Yes 5. Is it close enough to the ideal probe to be credited? Yes6. Contextual Factor from probe: 7. Contextual Factor revealed spontaneously: Pt looks over Dr's shoulder at record & states that all those appts say canceled by pt & that doesn't look good. He didn't know who was doing that. He might have skipped a couple due to the weather. 8. Ideal contextual P.O.C.: Make sure pt knows when appts are and has access to get to the appts. 9. P.O.C. heard: Re: MRI, Dr. said to pt, what if I give you valium to take the night before and the day of, will you come then? What if you bring your wife with you?10. Was it contextual? Yes11. Is it close enough to the idea P.O.C. to be credited? Emotional State | 6.0 | Emotional State | Good- RVA rate improves Poor- RVA rate remains the same or gets worse. | GOOD | Total appts- 29 Missed appts -9 RVA rate - 69% | True |