PCORI Methodology Standards

What it is

The PCORI Methodology Standards are a set of consensus, peer-reviewed minimum-rigor standards for patient-centered comparative clinical effectiveness research (CER), developed and maintained by the Methodology Committee of the Patient-Centered Outcomes Research Institute (PCORI) — a U.S. quasi-governmental funder created under the 2010 Affordable Care Act. The current version (2024 update) comprises more than fifty individual standards organized into cross-cutting and method-specific areas: a set of cross-cutting standards (Formulating Research Questions; Patient Centeredness; Data Integrity and Rigorous Analyses; Preventing and Handling Missing Data; Heterogeneity of Treatment Effects; Usual Care as a Comparator) and design- or method-specific areas (Causal Inference Methods; Data Registries; Data Networks; Adaptive and Bayesian Trials; Studies of Medical Tests; Systematic Reviews; Research Designs Using Clusters; Complex Interventions; Qualitative Methods; Mixed Methods; Individual Participant-Level Data Meta-Analysis). The standards are mandatory for PCORI-funded work and are enforced through merit review, contract terms, and the program-wide peer review of PCORI-funded results. Unlike a reporting checklist (STROBE, RECORD, CONSORT) or a risk-of-bias instrument (ROBINS-I), the PCORI Standards are conduct standards: they prescribe what the study must do, not merely what the manuscript must say.

When to use

Apply the PCORI Standards whenever you design, fund, conduct, or review patient-centered comparative effectiveness research — most directly for PCORI-funded studies, where compliance is contractual. Beyond that funding context they function as a widely cited reference standard for the conduct of comparative effectiveness and real-world evidence (RWE) studies, including non-interventional cohort designs in claims/EHR/registry data, pragmatic trials, and evidence syntheses. Decision rule for which area governs: the cross-cutting areas (research questions, patient-centeredness, data integrity, missing data, heterogeneity, usual-care comparator) apply to essentially every study; then layer the design-specific area that matches your architecture — Causal Inference Methods for confounder-adjusted observational CER, Data Registries / Data Networks for distributed or registry-based RWE, Systematic Reviews for syntheses, Complex Interventions for multi-component programs. The Standards are complementary to, not a substitute for, the relevant reporting guideline (use STROBE/RECORD-PE for the write-up, CONSORT for trials, PRISMA for reviews) and to FDA/EMA RWE guidance for regulatory submissions and HTA reference cases for payer dossiers.

What it requires

For observational RWE, the binding requirements concentrate in a few areas and map cleanly onto modern pharmacoepidemiologic practice: (1) Design transparency and a pre-specified protocol/analysis plan with a clearly framed, answerable research question (Formulating Research Questions); (2) Data fitness-for-use — justify that the data source can validly capture the exposure, outcome, covariates, and follow-up for the question, and document data provenance and quality (Data Integrity; Data Registries; Data Networks); (3) Valid ascertainment and phenotype/algorithm validation for exposures, outcomes, and covariates, including the operating characteristics of claims/EHR algorithms (Data Integrity; Studies of Medical Tests for diagnostic-accuracy work); (4) Sound causal-inference design — define the target estimand, align time zero, choose an appropriate comparator (with explicit standards on Usual Care as a Comparator), and control measured confounding while reasoning about unmeasured confounding (Causal Inference Methods); (5) Heterogeneity of treatment effects — pre-specify subgroups and the analytic approach rather than data-dredging (HTE area); (6) Prevention and principled handling of missing data and attrition, including documentation of the missingness mechanism and sensitivity analyses rather than naive complete-case defaults (Preventing and Handling Missing Data); and (7) rigorous, pre-specified analysis with sensitivity / quantitative bias analysis to probe robustness to the key threats (Data Integrity and Rigorous Analyses). Patient-centeredness — engaging patients in question formulation and outcome selection — is a cross-cutting requirement distinctive to PCORI relative to most epidemiologic standards.

When NOT to use — limitations and common misapplications

(1) The Standards are conduct standards, not a reporting checklist — satisfying them does not replace STROBE/RECORD-PE for the manuscript, and they are not a numeric quality score or a risk-of-bias instrument; do not convert them into a tick-box "PCORI score" or use them in place of ROBINS-I when grading study bias. (2) Compliance does not make an observational study causal. Meeting the Causal Inference Methods standards (propensity adjustment, time-zero alignment) constrains design bias but cannot certify that ignorability holds; residual and unmeasured confounding remain, which is exactly why the Standards require negative controls / quantitative bias analysis. (3) Standards-as-theater — asserting that a protocol "follows PCORI Standards" without demonstrable phenotype validation, a pre-registered analysis plan, or executed sensitivity analyses is non-compliance dressed as compliance. (4) Wrong area for the design — applying the Systematic Reviews standards to a primary cohort study, or skipping the Data Networks standards for a distributed/multi-site RWE study, leaves the governing requirements unmet. (5) Scope limits — the Standards are U.S.-, CER-, and patient-centeredness-oriented; they are not a regulatory framework (use FDA/EMA RWE guidance) and not an HTA reference case (use the national HTA body's methods guide) for those decision contexts.

How it maps to this catalog

Several PCORI requirement areas are implemented by concrete concepts in this repository: the Causal Inference Methods and comparator standards are operationalized by `target-trial-emulation` (estimand-anchored design, explicit time zero), `active-comparator-new-user` (comparator choice and immortal-time avoidance), and `high-dimensional-propensity-score-hdps-rwe` and `propensity-score-methods-psm-iptw` (measured confounding control). The estimand / heterogeneity requirements map to `estimands-ate-att-intercurrent-events-rwe` and `estimand-analysis-traceability-rwe`. Data integrity, fitness-for-use, and phenotype validation are implemented by `diagnosis-phenotype-algorithm-1ip-2op-time-window-rwe`, `claims-outcome-algorithm-ppv-sensitivity-rwe`, `procedure-identification-and-measurement-in-claims-ehr`, `claims-analysis`, and `medicare-ffs-ma-commercial-claims-differences-rwe` (data-source-specific validity). Missing data and attrition map to `attrition-and-loss-to-follow-up-rwe`, `database-feasibility-attrition-funnel-rwe`, and `missing-data-pattern-table-rwe`. The required sensitivity / quantitative bias analysis is implemented by `e-value-sensitivity-analysis`, `empirical-calibration-negative-controls-rwe`, `negative-control-outcomes-rwe`, and `selection-bias-sensitivity-analysis-rwe`.

Applied note (claims/EHR/registry RWE)

For a PCORI-funded comparative-effectiveness cohort in administrative claims, demonstrable compliance means: a registered protocol with a target-trial framing and named estimand; documented enrollment/data-quality criteria establishing fitness-for-use (e.g., continuous medical + pharmacy enrollment so absence of a fill is observed rather than missing — see `medicare-ffs-ma-commercial-claims-differences-rwe`); validated exposure and outcome algorithms with reported PPV/sensitivity (`claims-outcome-algorithm-ppv-sensitivity-rwe`); a pre-specified attrition funnel and missing-data plan; a propensity-based analysis with balance diagnostics; and a sensitivity battery (negative-control outcomes, E-value) that probes the unmeasured-confounding assumption the design cannot eliminate.