Table 1.

Example of a target trial protocol for an observational study aiming to estimate the causal effect of renin-angiotensin system inhibitors versus calcium channel blockers on outcomes in patients with advanced CKD

Protocol Element	Description	Target Trial	Emulation with Observational Data from the Swedish Renal Registry
Eligibility criteria	Who will be included in this study?	Individuals 18 yr or older under nephrologist care with CKD G4 (i.e., eGFR <30 ml/min per 1.73 m2), no history of kidney transplantation, and no use of RASi or CCB in previous 180 d between January 2007 and December 2016	Same as target trial
Treatment strategies	Which precise treatment strategies or interventions will eligible individuals receive?	1. Initiate RASi (ACEi or ARB) only 2. Initiate CCB only	Same as target trial
Treatment assignment	How will eligible individuals be assigned to the treatment strategies?	Randomization, no blinding	Eligible individuals are assigned at baseline to the treatment strategy that their data are consistent with. To emulate randomization, we adjust for the following baseline confounders: age, sex, eGFR, systolic and diastolic blood pressure, medical history (heart failure, arrhythmia, peripheral vascular disease, cerebrovascular disease, ischemic heart disease, diabetes mellitus, hyperkalemia, AKI), medication use (β-blocker, thiazide diuretic, potassium-sparing diuretic, statin), and health care use (the total number of hospitalizations in previous year)
Outcomes	What outcomes will be measured during follow-up?	1. Kidney replacement therapy (dialysis or kidney transplantation) 2. All-cause mortality 3. Major adverse cardiovascular events (composite of cardiovascular death, nonfatal myocardial infarction, nonfatal stroke)	Same as target trial. Kidney replacement therapy is registered in the Swedish renal registry; all-cause/cardiovascular mortality is identified from the Swedish death registry; hospitalizations for myocardial infarction or stroke are identified through ICD-10 codes in the national patient registry
Causal estimand	Which causal estimand will be estimated with the observational data?	Intention-to-treat effect (effect of being randomized to treatment) Per protocol effect (effect of receiving treatment strategy as specified in protocol)	Per protocol effect (effect of receiving treatment strategy as specified in protocol)
Start and end of follow-up	When does follow-up start and when does it end?	Starts at randomization and ends at occurrence of end point, administrative censoring or 5 yr of follow-up	Starts at medication initiation (filled prescription) and ends at occurrence of end point, administrative censoring or 5 yr of follow-up
Statistical analysis	Which statistical analyses will be used to estimate the causal estimand?	Intention-to-treat analysis, non-naïve per protocol analysis	Per protocol analysis: Hazard ratios are estimated using Cox regression while adjusting for baseline confounders with inverse probability of treatment weighting. Weighted cumulative incidence curves are estimated using the Aalen–Johansen estimatora

RASi, renin-angiotensin system inhibitor; CCB, calcium channel blocker; ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; G, KDIGO G category.

^aA more elaborate description of the statistical analysis can be found in the corresponding paper.