Procedure, Device, and Radiotherapy Exposure Definitions

Definition and scope

Construction of non-drug treatment exposures from completed procedures, implanted-device records, or delivered radiotherapy, with explicit distinction between planned care, orders, claims, and actual delivery. The target population, use, and analysis remain protocol decisions; this card explains the choices and their consequences rather than approving a study.

Operational specification

Define the intervention and comparator at a clinically meaningful level: procedure technique and indication, device model/version and implantation, or radiotherapy course/site/fraction delivery. Record eligibility, treatment decision and delivery dates, code/version lists, laterality/site, completion, revision/removal, operator or center, and the start of risk. An order or authorization is not a completed intervention.

Analysis and diagnostics

Reconcile facility and professional claims into episodes, validate the identification algorithm, align time zero and baseline covariates, and address treatment selection, provider effects, learning curves, calendar changes, and informative discontinuation. If comparing completed courses, avoid requiring future survival/completion to assign baseline treatment.

When NOT to use and common errors

Do not infer implantation from a consultation code, combine device generations without assessment, treat fraction records as independent patients, or classify a treatment strategy using future completion while starting follow-up earlier.

Worked example and implementation boundary

A synthetic radiotherapy extract has 12 delivered-fraction records for four distinct courses after reconciling patients and course identifiers. Calculate mean fractions per course; this does not establish adequate dose or course completion. The descriptive mean is three recorded fractions per course. Delivered dose and clinical completeness need their own evidence. The code performs that explicitly stated calculation with synthetic inputs. It is not a complete fitted analysis and does not demonstrate the identification assumptions, estimate sampling uncertainty, or certify application in another population.

Data-source and reporting requirements

Preserve the provenance, dates, units, and observation rule for each input. Claims capture services and transactions under observable benefits; absence of a record is not proof that an event or resource use did not occur. EHR and registry measurements depend on the collection process and may miss care outside the system. Linked sources add reconciliation and selection questions. Primary collection can improve a missing domain but still requires valid measurement and follow-up. Report which of these limitations affects the stated use, which evidence supports the chosen inputs, and what remains uncertain.

Worked example

Scenario

A synthetic radiotherapy extract has 12 delivered-fraction records for four distinct courses after reconciling patients and course identifiers. Calculate mean fractions per course; this does not establish adequate dose or course completion.

Dataset

Synthetic inputs with units defined in the scenario; no patient data.

Steps

Result

The descriptive mean is three recorded fractions per course. Delivered dose and clinical completeness need their own evidence.

Trade-offs

Code example

delivered_fraction_records = 12
courses = 4
result = delivered_fraction_records/courses
assert abs(result - 3.0) < 1e-9
print(result)

delivered_fraction_records <- 12
courses <- 4
result <- delivered_fraction_records/courses
stopifnot(abs(result - 3.0) < 1e-9)
print(result)

data worked_example;
  delivered_fraction_records = 12;
  courses = 4;
  result = delivered_fraction_records/courses;
  put result=;
run;

Citations

1. [1]McCulloch P, et al. No surgical innovation without evaluation: the IDEAL recommendations. The Lancet. 2009.
2. [2]Hernán MA, Robins JM. Using big data to emulate a target trial when a randomized trial is not available. American Journal of Epidemiology. 2016;183(8):758-764.