Behavioral and Care-Delivery Intervention Exposures

Definition and scope

Specification of intervention components, delivery, receipt, intensity, and context for behavioral, service, or organizational interventions, distinguishing the intervention from the strategy used to implement it. The target population, use, and analysis remain protocol decisions; this card explains the choices and their consequences rather than approving a study.

Operational specification

Describe the intervention theory, active components, delivery mode, provider, target recipients, dose/intensity, timing, fidelity, adaptations, comparator care, and setting. Separate offer/assignment from actual receipt and sustained engagement. Define whether exposure is at patient, clinician, clinic, or system level and whether spillover is expected.

Analysis and diagnostics

Align the causal contrast with assignment, receipt, or adherence and choose a design that addresses selection and time-varying engagement. Track implementation and context without conditioning a primary assignment contrast on post-assignment fidelity by default. Analyze clustered delivery and interference where relevant, and document how routine records distinguish nonreceipt from missing documentation.

When NOT to use and common errors

Do not define exposure by completing a future number of visits while counting earlier survival, equate enrollment with engagement, assume a program name means identical components at every site, or attribute secular service changes to the intervention without a credible comparison.

Worked example and implementation boundary

A synthetic program offers sessions to 100 eligible patients; 60 attend at least one session under a prespecified receipt definition. Calculate receipt among those offered, separate from any clinical effect. Observed receipt is 0.60. An effect of offer and an effect of sustained participation are different targets. 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 program offers sessions to 100 eligible patients; 60 attend at least one session under a prespecified receipt definition. Calculate receipt among those offered, separate from any clinical effect.

Dataset

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

Steps

Result

Observed receipt is 0.60. An effect of offer and an effect of sustained participation are different targets.

Trade-offs

Code example

received = 60
offered = 100
result = received/offered
assert abs(result - 0.6) < 1e-9
print(result)

received <- 60
offered <- 100
result <- received/offered
stopifnot(abs(result - 0.6) < 1e-9)
print(result)

data worked_example;
  received = 60;
  offered = 100;
  result = received/offered;
  put result=;
run;

Citations

1. [1]Skivington K, et al. A new framework for developing and evaluating complex interventions: update of Medical Research Council guidance. BMJ. 2021.
2. [2]Curran GM, et al. Effectiveness-implementation Hybrid Designs. Medical Care. 2012.
3. [3]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.