1 1. Purpose

This SOP provides a standard framework for the design, conduct, analysis, and reporting of observational real-world evidence (RWE) studies using routinely collected data.

It is intended for studies using:

The primary objective is to estimate treatment, exposure, or comparative-effectiveness effects as validly as possible while addressing major sources of bias inherent in nonrandomized data.

2 2. Research Question and Causal Objective

The research question should be specified before analysis.

General form:

Among [target population], what is the effect of [treatment/exposure A] compared with [treatment/exposure B] on [outcome] during [follow-up period]?

The study should define:

When causal interpretation is intended, the analysis should emulate the key elements of a hypothetical target trial as closely as possible.

3 3. Target Trial Framework

Whenever applicable, prespecify the following target-trial components:

  1. eligibility criteria;
  2. treatment strategies;
  3. treatment assignment procedure;
  4. time zero;
  5. follow-up;
  6. outcome;
  7. causal contrast / estimand;
  8. analysis method.

The observational emulation should align eligibility, treatment assignment, and start of follow-up at the same time point to minimize time-related biases.

4 4. Data Source and Data Provenance

Describe:

Important limitations of each source should be documented.

Examples include:

5 5. Study Population and Cohort Definition

Define inclusion and exclusion criteria before outcome analysis.

Typical criteria may include:

A cohort flow diagram should document the number of individuals excluded at each step.

6 6. Index Date / Time Zero

The index date is the date on which:

These elements should be aligned as closely as possible.

Incorrect alignment can produce:

7 7. Exposure or Treatment Strategy Definition

Specify:

For comparative-effectiveness studies, a new-user active-comparator design should be considered when feasible because it often improves comparability and temporal clarity.

8 8. Outcome Definition

Specify:

Outcome definitions should be based on clinically meaningful and reproducible algorithms.

Where possible, use validated diagnosis/procedure algorithms.

9 9. Baseline Period and Covariates

The baseline period should be prespecified, for example:

Baseline covariates may include:

Covariates should be defined using information available before or at time zero.

10 10. Confounding Control

Because treatment or exposure is not randomized, confounding must be explicitly addressed.

10.1 10.1 Confounder Selection

Potential confounders should be selected using:

  • clinical knowledge,
  • prior literature,
  • temporal ordering,
  • causal diagrams / DAGs,
  • and known predictors of treatment and outcome.

Variables should not be included solely because they are statistically significant or imbalanced.

10.2 10.2 Confounding by Indication

Special attention should be given to confounding by indication.

For example, patients receiving a more intensive therapy may be systematically sicker than patients receiving the comparator treatment.

Important markers of disease severity should therefore be identified and included when available.

10.3 10.3 Post-Exposure Variables

Post-exposure variables should not automatically be adjusted for.

If the target estimand is the total treatment effect, variables affected by treatment should generally not be included in the baseline confounder set.

If the objective instead concerns a direct or hypothetical effect, the estimand and analytic method should be explicitly defined.

11 11. Descriptive Analysis

Baseline characteristics will be summarized by treatment/exposure group.

Continuous variables:

Categorical variables:

Standardized mean differences (SMDs) should be used to assess baseline imbalance.

An absolute SMD below approximately 0.10 is commonly used as a practical indicator of acceptable measured balance.

Baseline imbalance alone does not determine whether a variable is a confounder.

12 12. Propensity Score Estimation

When propensity-score methods are used, estimate:

\[ PS(X) = P(A=1 \mid X) \]

where \(A\) is treatment/exposure and \(X\) is the set of measured baseline covariates.

Possible approaches include:

Example:

ps_model <- glm(
  treatment ~ age + sex + bmi + hba1c +
    prior_cvd + ckd + smoking,
  family = binomial(),
  data = analysis_data
)

analysis_data$ps <- predict(
  ps_model,
  type = "response"
)

13 13. Propensity Score Adjustment Strategies

Common methods include:

13.1 13.1 Matching

Examples:

  • 1:1 nearest-neighbor matching,
  • variable-ratio matching,
  • caliper matching.

Balance should be assessed after matching.

13.2 13.2 IPTW

For estimation of the average treatment effect:

\[ w_i = \begin{cases} 1/PS_i, & A_i=1 \\ 1/(1-PS_i), & A_i=0 \end{cases} \]

Stabilized weights may be preferred.

Extreme weights should be evaluated and may be truncated or trimmed using prespecified rules.

13.3 13.3 Overlap Weighting

Overlap weighting may be considered when treatment groups have limited overlap or when the clinical target population is the region of common support.

14 14. Positivity and Overlap

Assess whether patients in all covariate strata have a reasonable probability of receiving each treatment.

Evaluate:

Serious positivity violations should be reported and may require:

15 15. Covariate Balance Assessment

After matching or weighting, assess balance using:

Example:

library(cobalt)

bal.tab(
  treatment ~ age + sex + bmi + hba1c +
    prior_cvd + ckd + smoking,
  data = analysis_data,
  weights = analysis_data$weight,
  un = TRUE,
  thresholds = c(m = 0.10)
)

Good measured balance does not establish the absence of unmeasured confounding.

16 16. Primary Outcome Analysis

The primary analysis method depends on outcome type.

16.1 16.1 Time-to-Event Outcome

Use Cox proportional hazards regression or another prespecified survival model.

library(survival)

fit <- coxph(
  Surv(followup_time, event) ~ treatment,
  data = analysis_data,
  weights = weight,
  robust = TRUE
)

Report:

  • HR,
  • 95% CI,
  • and P-value when appropriate.

16.2 16.2 Binary Outcome

Possible methods:

  • logistic regression,
  • log-binomial regression,
  • modified Poisson regression.

Report:

  • OR,
  • RR,
  • or RD, depending on the estimand.

16.3 16.3 Continuous Outcome

Possible methods:

  • linear regression,
  • weighted regression,
  • mixed models for repeated measurements.

16.4 16.4 Count Outcome

Possible methods:

  • Poisson regression,
  • negative binomial regression.

17 17. Conditional vs Marginal Effect

The analysis should distinguish between:

For example:

The estimand should be specified in advance.

18 18. Follow-up and Censoring

Follow-up should begin at the index date and end at the earliest of:

Censoring rules depend on the target estimand.

For example:

Potential informative censoring should be evaluated.

Inverse probability of censoring weighting may be considered when appropriate.

19 19. Missing Data

Summarize:

Possible methods include:

The primary method should be prespecified.

Sensitivity analyses should evaluate departures from missing-at-random assumptions when relevant.

20 20. Measurement Error and Misclassification

Evaluate the potential for:

When possible:

21 21. Model Diagnostics

Depending on the primary model, assess:

For Cox models:

ph_test <- cox.zph(fit)
print(ph_test)
plot(ph_test)

22 22. Competing Risks

When competing events are clinically important, consider:

Choice of method should reflect the scientific question.

23 23. Effect Modification and Subgroup Analyses

Prespecified subgroups may include:

Effect modification should preferably be tested with interaction terms.

Exploratory subgroup analyses should be labeled as such.

24 24. Sensitivity Analyses

Common RWE sensitivity analyses include:

  1. alternative exposure definitions;
  2. alternative outcome definitions;
  3. alternative baseline windows;
  4. alternative confounder sets;
  5. propensity-score matching;
  6. IPTW;
  7. overlap weighting;
  8. multivariable regression;
  9. weight trimming/truncation;
  10. alternative censoring rules;
  11. as-treated analysis;
  12. treatment-policy-like analysis;
  13. multiple imputation;
  14. negative-control outcomes or exposures;
  15. lagged exposure analyses;
  16. competing-risk analyses;
  17. inverse probability of censoring weighting;
  18. assessment of unmeasured confounding;
  19. quantitative bias analysis;
  20. E-value analysis when appropriate.

Results should be compared across methods for:

25 25. Negative Controls and Falsification Analyses

When feasible, use:

These analyses can help detect residual confounding, surveillance bias, or systematic error.

A null negative-control result does not prove that confounding is absent.

26 26. Unmeasured Confounding

Residual and unmeasured confounding should be explicitly acknowledged.

Possible approaches include:

No single method can prove the absence of unmeasured confounding.

27 27. Generalizability and Transportability

Describe the population to which results apply.

Assess:

The analytic population after matching or weighting may differ from the original source population.

28 28. Statistical Inference

Unless otherwise specified:

The primary analysis should be distinguished from:

Multiplicity should be acknowledged when many outcomes or subgroup analyses are conducted.

29 29. Reporting of Results

Recommended outputs include:

29.1 Table 1

Baseline characteristics before adjustment.

29.2 Table 2

Baseline characteristics after matching or weighting.

29.3 Table 3

Primary and sensitivity treatment-effect estimates.

Method Effect Estimate 95% CI P-value
Primary analysis
Propensity-score matching
IPTW
Multivariable regression
Sensitivity analysis

29.4 Figures

Recommended figures include:

  1. cohort flow diagram;
  2. propensity-score overlap plot;
  3. love plot;
  4. Kaplan-Meier or cumulative-incidence plot if relevant;
  5. forest plot comparing primary and sensitivity estimates.

30 30. Interpretation

Interpretation should focus on:

Observational RWE findings should generally be described as adjusted associations unless sufficient design and causal assumptions support a causal interpretation.

31 31. Key Limitations

Common RWE limitations include:

32 32. Quality Control and Reproducibility

All key analysis steps should be reproducible.

Key variables requiring validation include:

Whenever feasible:

33 33. Documentation of Deviations

Any material deviation from the prespecified protocol or SAP should be documented with:

34 Appendix A. Minimal RWE Analysis Workflow

library(dplyr)
library(MatchIt)
library(cobalt)
library(survival)

# 1. Define analytic cohort
analysis_data <- raw_data %>%
  filter(
    eligible == 1
  )

# 2. Propensity score
ps_model <- glm(
  treatment ~ age + sex + bmi + hba1c +
    prior_cvd + ckd + smoking,
  family = binomial(),
  data = analysis_data
)

analysis_data$ps <- predict(
  ps_model,
  type = "response"
)

# 3. IPTW
analysis_data <- analysis_data %>%
  mutate(
    treat_num = as.integer(treatment == "TreatmentA"),
    iptw = ifelse(
      treat_num == 1,
      1 / ps,
      1 / (1 - ps)
    )
  )

# 4. Weighted outcome model
fit_iptw <- coxph(
  Surv(followup_time, event) ~ treatment,
  data = analysis_data,
  weights = iptw,
  robust = TRUE
)

summary(fit_iptw)

35 Appendix B. Pre-Analysis Checklist

Before final analysis, confirm: