Across five conversational chart audits spanning cardiology, gastroenterology, dermatology, hematology, and oncology, the same structural gap shows up every time: a standard chart audit records the prescribing decision, but not the clinical reasoning, access friction, or competitive dynamic that actually produced it. What differs by therapy area is what that hidden reasoning turns out to be.
That consistency is the point of this post. It's easy to read one chart-audit case study and wonder if the finding is specific to that indication. It isn't the methodology transfers, even though what it uncovers is different every time. Here's what showed up in each of five therapy areas, and the pattern underneath all five.
What a checkbox would show: a patient had a bleeding event on anticoagulation, and the physician continued the same drug.
What the conversation revealed: in two of ten cardiologist interviews on anticoagulation maintenance, the physician had identified a co-precipitant - an NSAID or aspirin, not the anticoagulant itself - stopped it, and made an explicit, active risk-benefit calculation to keep the patient on therapy. Separately, nine of ten patients in the sample were on the same drug (Apixaban), but the reasons for staying on it broke into four entirely different logics: one physician following a guideline algorithm, one citing a specific trial by name, one reasoning from accumulated personal experience, and one simply because the patient was stable and asymptomatic.
Commercial implication: those four physicians are indistinguishable in prescribing data but represent four different audiences for messaging, a physician on guideline autopilot needs a different message than one anchored to trial data or one operating on personal conviction.
What a checkbox would show: "Tremfya preferred over Skyrizi, switch reason: efficacy / disease control."
What the conversation revealed: in the cases where Skyrizi was passed over, the actual driver was payer friction and induction architecture, not efficacy - subcutaneous induction clears insurance approval more easily than an IV option, and that friction, not clinical preference, was the deciding factor for at least one physician. In the cases where Skyrizi won, it was the only defensible class option for a patient with a malignancy history where JAK inhibitors and anti-TNFs carried documented risk.
Commercial implication: the two Skyrizi stories point to two different commercial responses, payer support and patient-services investment for the access battle, and targeted safety messaging for the malignancy-history patient profile, neither of which is visible if the audit only records "switched for efficacy."
What a checkbox would show: "Humira to Bimzelx switch, reason: inadequate efficacy."
What the conversation revealed: personal clinical experience, not trial data, drove the switching decision in most cases, and the urgency behind switching was tied specifically to the risk of irreversible scarring from ongoing disease activity, not a data-driven efficacy comparison. One dermatologist described switching a patient after two years of only partial control despite an "inadequate efficacy" label that didn't capture where on the body lesions were still active or how long partial control had been tolerated before the physician acted.
Commercial implication: peer-to-peer case sharing carries more commercial weight than HCP promotion in this indication, physicians who've used a therapy are already convinced; the barrier is exposure, not evidence.
What a checkbox would show: nothing yet - the regimen (Darzalex + Teclistamab) wasn't approved at the time of these interviews, so no traditional chart audit could even see it. See Pre-launch Intelligence: Unlocking Rx Decisions in Multiple Myeloma for the full findings.
What the conversation revealed: physicians were already pre-sorting Darzalex+Teclistamab as a narrow option before approval. Three of five patients discussed were already DARA-exposed in first-line therapy, making them ineligible for the regimen under its trial criteria, an exclusion invisible in any prescribing record. Physicians also framed the drug relative to CAR-T eligibility and manufacturing timelines, and to patient goals-of-care (a treatment-free interval versus a lower-commitment chronic regimen), not as a standalone choice.
Commercial implication: the addressable second-line population is narrower than the label suggests, and academic-versus-community adoption patterns were already diverging before a single prescription existed, commercial intelligence a post-launch chart audit would surface months too late.
What a checkbox would show: a choice between Rybrevant+Lazcluze, Tagrisso monotherapy, or Tagrisso plus chemotherapy. See The Decision Behind the Prescription in 1L EGFR+ NSCLC for the full findings.
What the conversation revealed: physicians stratify by disease burden, CNS involvement, and resistance risk, not a simple drug-versus-drug comparison and at least one prescription was lost purely because a subcutaneous formulation wasn't available and the physician had tolerability concerns about the IV alternative. That's a structural formulation and access barrier, not a clinical or efficacy judgment, and it doesn't show up anywhere in prescribing data.
Commercial implication: a competitive threat here isn't necessarily about clinical differentiation, it can be as specific as a formulation gap that a chart audit surfaces and a claims database cannot.
Checkbox-format data records the what. Conversational chart audits recover the why. That holds regardless of therapy area, what changes is the specific shape the "why" takes: an active risk calculation in cardiology, payer and formulary friction in gastroenterology, personal conviction and urgency in dermatology, pre-launch positioning dynamics in hematology, and a formulation access barrier in oncology.
The through-line that matters for a brand team evaluating this methodology for their own indication: the specific finding is never the same twice, but the structural gap a checkbox format leaves behind is remarkably consistent. If your therapy area has any complexity in how physicians actually decide - competing evidence, access friction, multiple valid options, a pre-launch dynamic, that's exactly the kind of decision a conversational chart audit is built to surface.
Three of the five reads above - ulcerative colitis, hidradenitis suppurativa, and multiple myeloma are drawn from ZoomRx's whitepaper on where conversational chart audits catch what other research methods miss. For the full case studies, download it below.
If you're weighing whether a conversational chart audit would surface something new in your own therapy area, ZoomRx's Patient Chart Audits page has more detail on the methodology.