SYNTHETIC RECORD, NOT A REAL MEDICAL RECORD

Review checklist: Gwen Ifill (HR-ML032)
All records · Back to the record · Dossier (sources.md) · Terminology · Report a problem · Comments
Everything a reviewer needs to judge, gathered from the record. Release status: draft.
- 9judgment calls in the dossier
- 2protocol deviations
- 5orders and referrals without results in the record
- 1evidence items to check (retrospective or sensitive claims, unconfirmed diagnoses)
Review status
| Review | Status | Reviewer | Date | Notes |
|---|
| schema validation | not_started | | | Run scripts/build.ps1 -Conformance. |
| clinical review | not_started | | | Awaiting clinician review; check the combined biopsy and ultrasound pathway, the FIGO 2023 stage IIIC2ii with the MMRd molecular class, the choice of dostarlimab with carboplatin and paclitaxel, MLH1 methylation instead of germline testing, the immune-related hypothyroidism managed without stopping dostarlimab, and the survivorship schedule. |
| provenance review | not_started | | | Recent-death rule: only what she made public, what was reported during her life, and obituaries and official statements: The Washington Post, CNN, PBS NewsHour, NPR, CBS News and the Associated Press. The stage, histology and treatment of her cancer were never made public; everything clinical is synthetic. |
| safe messaging review | not_started | | | No mental health or substance claim is made; mood screens are synthetic and below the threshold for a safety assessment. Her death is recorded once, without detail. No relative is named in the modern episode, and no health fact, decision or genetic result is attributed to any real person. |
Judgment calls (dossier editorial resolutions)
- Recent-death rule. Ifill died in 2016, so the rule applies. The case uses only what was reported publicly during her life (the Associated Press report of her April leave) and in obituaries and official statements at her death (The Washington Post, CNN, PBS NewsHour, NPR, CBS News). She kept her illness private; the cause of death was given by her family, and the timing of the diagnosis, the reason for the April leave and her work through chemotherapy come from friends quoted in the obituaries (CLM-0006, graded probable). No sick visit is added. Her death is recorded once (CLM-0007), in plain words. No claim concerns mental health or substance use; the modern mood screens are synthetic and below the threshold for a safety assessment.
- What is not public. The stage, histology, molecular result and treatment of her cancer were never made public, beyond friends’ references to chemotherapy. Every clinical detail in the modern episode (the bleeding, the biopsy and its histology, the mismatch repair result, the nodal stage, the surgery, dostarlimab, the hypothyroidism, all values) is synthetic and is not a claim about her real illness.
- Real people. No relative is named; the family member who gave the cause of death is referred to only as her family. No decision, statement or health fact is attributed to any relative. Her co-anchor and the president she interviewed appear only in history claims about her career; friends quoted in the obituaries are cited as the obituaries’ sources and have no role in the modern episode. The modern family history is left empty so that no health fact is attributed to a real relative, and the Lynch syndrome work-up never assigns a genetic result to anyone: the tumor’s MLH1 loss is explained by promoter methylation, so germline and cascade testing are not needed.
- Conflicts. (1) Wikipedia states that she died of breast and endometrial cancer, citing a Politico report that could not be fetched; the Washington Post obituary, CNN, NPR and PBS mention only endometrial cancer or cancer, so breast cancer is not used. (2) Her return from the spring leave is dated to May by a friend quoted by CNN, to around Memorial Day by the network’s plan in May, and by the start of June, when she interviewed President Obama; CLM-0005 says she was back by the start of June. (3) The New York Times obituary and the Britannica entry could not be fetched and are not cited.
- Index age and dates. The index visit on January 5, 2026 maps to January 5, 2016, when she was 60: shortly before the April leave that was the first public sign of her illness, and within the period a friend later gave for the diagnosis. The synthetic date of birth is September 29, 1965 (shift 10 years). The chronic-care visits on January 20, February 24, April 28 and June 30, 2026, the surgery on February 3, and the preventive visit on August 11, 2026 (age 60, so a periodic adult preventive visit) map to the same days of 2016, before her death. The claim about the diagnosis and treatment is linked to the November 2016 reports, so the intake documents do not show it. There is no past medical or family history from the public record before the index date.
- Synthetic clinical picture and teaching points. The presenting problem is postmenopausal bleeding (N95.0), resolved at the consultation when the biopsy gives the diagnosis. The diagnosis is endometrioid, low-grade (grade 2) and mismatch repair deficient, staged FIGO 2023 IIIC2ii at surgery with the molecular class MMRd (POLE wild-type, p53 wild-type), chosen so that the treatment teaches the current standard for mismatch repair deficient advanced disease (dostarlimab with carboplatin and paclitaxel, the RUBY regimen, which covers stage IIIC2 regardless of measurable disease; pembrolizumab with the same chemotherapy is the alternative). The teaching points on equity are explicit in the plan and the teaching objective: Black women in the United States have roughly twice the endometrial cancer mortality of White women, with more non-endometrioid and p53-abnormal tumors, later stage, and delays in evaluating bleeding, and the disparity is not fully explained by subtype and stage (Clarke 2022; Doll 2020). The modern episode models the opposite of delay: any bleeding after menopause is examined and sampled at the first visit, together with transvaginal ultrasound, as the 2026 ACOG update recommends instead of relying on a thin endometrium. Qualitative results (pathology, immunohistochemistry, methylation, POLE, hepatitis and HIV serology, imaging, mammogram) appear only in procedure and note text. Height (165 cm) and weights are synthetic; BMI 26 to 27 is not coded as a problem.
- Preventive visit. A periodic adult preventive visit at 60 (CPT 99396 with 99497 in plain text), six weeks after the end of combination treatment and on the day of the second maintenance dose. Judgment calls: cervical screening not indicated after total hysterectomy for cancer, with no vaginal cytology; average-risk colorectal screening because the mismatch repair loss is acquired; PCV20 under the current recommendation for all adults from 50; recombinant zoster and Tdap given; bone density deferred to 65 with no added risk factor; influenza and COVID-19 vaccines in the autumn. Advance care planning records only her general preferences and an unnamed agent.
- Focused visits. The four chronic-care follow-ups (two weeks with the diagnosis and consultation, three weeks after surgery with cycle 1, cycle 4, and the end of combination treatment with the first maintenance dose) are documented as focused visits on the NCCN uterine neoplasms guideline, the CMID endometrial cancer monograph, FIGO 2023, the RUBY trial and dostarlimab label, the ASCO immune-related adverse event, antiemetic, neuropathy and hepatitis B guidance, and the 2026 SGO surveillance statement. One deviation is recorded instead of an invented result: the morning cortisol that the CMID protocol calls for at baseline and before each checkpoint inhibitor cycle was deferred at cycle 1 (she arrived after the morning window) and then missed before cycles 2 and 3 because the request did not reach the infusion center’s standing orders; no harm resulted (normal at cycle 4 and at the end of treatment, no symptoms), and the standing orders were corrected. Synthetic findings introduced: grade 1 paclitaxel neuropathy at cycle 4, a TSH just above range at cycle 4, and grade 2 immune-related hypothyroidism at the end of combination treatment, treated with levothyroxine while dostarlimab continued.
- Consultation. The gynecologic oncology referral, written at intake when only the bleeding was known, and the consultation by Miriam Delacroix, MD (a fictional consultant), on the day of the two-week visit, are S. The consultant answers the five referral questions in line with what the record shows afterwards: minimally invasive hysterectomy with sentinel node mapping and removal of the enlarged pelvic and para-aortic nodes; MLH1 methylation, POLE and FIGO 2023 staging with the molecular class; dostarlimab with carboplatin and paclitaxel if nodal disease was confirmed, which it was; genetic counseling, with germline testing only if methylation did not explain the MLH1 loss; the immune-related monitoring that found her hypothyroidism; and the survivorship schedule.
Protocol deviations
Checks the monitoring protocol called for that the schedule did not perform. Each is recorded with its outcome and corrective action, never filled with an invented result.
| Visit | Measure | Interval | What happened |
|---|
2026-02-24 Progress note 03 (C-CDA) | Morning cortisol (immune-related adrenal and pituitary disease) | Baseline and before each checkpoint inhibitor dose | Not drawn: she arrived after the morning window, when a cortisol cannot be read. To be drawn before 9 a.m. before cycle 2 in the infusion center. No symptom of adrenal insufficiency |
2026-04-28 Progress note 04 (C-CDA) | Morning cortisol (immune-related adrenal and pituitary disease) | Baseline and before each checkpoint inhibitor dose | 12.6 ug/dL today, normal. A deviation: the baseline cortisol deferred at cycle 1 was not drawn before cycles 2 and 3 either, because the request was not carried into the infusion center's standing orders. No harm resulted: today's value is normal and she has had no symptom of adrenal insufficiency. Corrective action: morning cortisol added to the standing orders before every remaining dose; no earlier values are recorded |
Orders and referrals without results in the record
| Visit | Type | Item | Status | Detail |
|---|
2026-01-20 Progress note 02 (C-CDA) | Laboratory | MLH1 promoter methylation | Ordered | Sent on the biopsy today |
2026-01-20 Progress note 02 (C-CDA) | Laboratory | p53 immunohistochemistry and POLE sequencing (molecular classification) | Ordered | p53 wild-type pattern on the biopsy; POLE to be sequenced on the surgical specimen |
2026-01-20 Progress note 02 (C-CDA) | Laboratory | Hepatitis B, hepatitis C and HIV screening | Ordered | Hepatitis B surface antigen, core antibody and surface antibody, hepatitis C antibody, and HIV drawn today |
2026-08-11 Progress note 06 (C-CDA) | Prevention | Breast cancer screening | Ordered | Mammogram every two years from 40 to 74; none in the past two years on file; booked in two weeks |
2026-08-11 Progress note 06 (C-CDA) | Prevention | Colorectal cancer screening | Ordered | No screening on file; the tumor's MLH1 loss is explained by promoter methylation, so average-risk screening applies; stool immunochemical test chosen, with colonoscopy if positive |
Evidence to check
Retrospective claims (R)
None.
Sensitive claims
| Claim | Grade | Statement |
|---|
| CLM-0007 | D | She took a second leave shortly before her death and died at a hospice in Washington, D.C., on November 14, 2016, at 61. Her family gave the cause as endometrial cancer. |
Diagnoses not confirmed
None.
Notes to read
Every note in the record, with its type. All clinical content is synthetic.
Abnormal results
| Date | Result | Value | Flag |
|---|
| 2026-01-05 | Hemoglobin | 11.8 g/dL (low) | Low |
| 2026-02-24 | Hemoglobin | 11.6 g/dL (low) | Low |
| 2026-04-28 | Hemoglobin | 10.6 g/dL (low) | Low |
| 2026-04-28 | TSH | 4.8 mIU/L (high) | High |
| 2026-06-30 | Hemoglobin | 11.4 g/dL (low) | Low |
| 2026-06-30 | TSH | 11.8 mIU/L (high) | High |
| 2026-06-30 | Free T4 | 0.7 ng/dL (low) | Low |
| 2026-08-11 | Total cholesterol | 204 mg/dL (high) | High |