Aman clutches his chest and collapses on the sidewalk. Call 911. Surelywewant themostadvanced life support (ALS) available—paramedics capable of intubating, administering intravenousmedications, andmanaging complex arrhythmias, as well as transport to the hospital. Or do we? Is it possible that he might do better with less-intensive interventions, especially if theyweredonemore skillfully and safely (facemaskbagging insteadof intubation) aswell asmore quickly, so hewould arrive faster to the hospital for definitive care? The value of advanced interventionsmight be especially small in urban areas where they prolong the normally short travel times to hospital. In sum, is it possible that basic life support (BLS)—using automatic defibrillators, cardiopulmonary resuscitation, and airway management without intubation—could be as good or better? (The term BLS in this commentary is used to describe units with basic emergency medical technicians who are also trained to use semiautomatic defibrillators but donot use advanced airway techniques or administer intravenous drugs.) Sanghavi et al1 provide us with provocative data in answer to this question in this issue of the JAMA Internal Medicine. They report on 39 000Medicare patients treated for outof-hospital cardiac arrest using billing data to determine whether patients receivedALSorBLS. The authors found that patients receivingALShad43% lower survival to hospital discharge, 51%hadmore poor neurologic function, and 30%had higher costs. The study by Sanghavi et al1 uses a different methodology thanmost previous studies of prehospital care; the population and analysis were based on billings for level of emergencymedical system(EMS) response, rather thanfromclinical records of presentation and treatment at the scene. Thismethodologyhasbothstrengthsandweaknesses.The strength is the sample size and generalizability. The weakness is that the analysis requires complex assumptions (including comparable variables known to affect cardiac arrest outcomes in both patient cohorts) and imputation of many missing values. The authors conducted multiple sensitivity analyses; the supporting arguments seem conservative and supportive of their assumptions. Their approach is similar to comparing theoutcomesofacutemyocardial infarction treated bycardiologistsvsprimarycarephysicians; it tests thecaresystem as a whole rather than each individual treatment. Certainvalid overall conclusions couldbe reachedwithout knowinganythingabout theexact clinicalmanagementused ineach case, as long as patient characteristics are the same.Whether the quality and validity of administrative data used this way is better or worse than the typical EMS database is unknown, but the likelihood that their findings are true is strengthened by its consistency with a growing body of research demonstrating the ineffectiveness of many ALS therapies. To understand why the value of ALS is still being debated some45yearsafter its inception,onemust trace itshistory.The studies that launchedresuscitation in theearly 1960swerecase reportsor smalluncontrolledcaseseriesofasphyxiatedhealthy youngdogs. Thedrugs used then (andnow) preceded the 1962 USFoodandDrugAdministration requirement that efficacybe demonstratedfortheir intendeduse.Researchmethodologywas far less sophisticated, and few or none of these studies would be published in anymajor journal today. These first reports appeared to fulfil adreamof centuries— thereversalofdeath itself.Defibrillationand intravenousdrugs spread quickly from their origins in the operating room to the first coronary care units and then experimentally to out-ofhospital use. Initially, treatment was by physicians in ambulances, targeted only at populations of patients having acute infarction. Soon ALS was in use by nascent EMS systems (focused primarily on improving trauma care) and the newly inventedparamedics, for anyandall comers.Thescientific foundationwasweakandanecdotal; therewereno largeprospective studiesor controlled trials, qualityoversightwasminimal, and bias for positive outcomes was common. By the 1990s,muchbetterquality studiesbegan toemerge. The largeOntario Prehospital AdvancedLife Support (OPALS) multicenter prospective study in 2004 was rigorously constructed and conducted to quality control standards seldom seen inEMSsystems.2 (This studywas intended tobe randomize clinical trial [RCT] but could not be owing to the belief of paramedics that itwas unethical towithholdALS, despite absence of proof of its efficacy). Adding ALS to a rapiddefibrillation systemdidnotproduceany improvement in survival or neurologic or health utility outcome in any subgroup. The results were replicated in a large, similar study in Taiwan. In2009,anRCTof 1183patientswithcardiacarrest inNorway randomized half of them to receive no intravenous ALS drugs; there was no difference in survival to hospital discharge or neurological outcome.3 A 2009 Cochrane summary found no difference between endotracheal intubation and other airway management strategies in emergency situations, including out-of-hospital cardiac arrest.4 And now we add the findings of Sanghavi et al.1 This negative trend is all too familiar inmedical research. Initial studies of a treatment are small, poorly controlled, and mostly positive, but then larger more rigorous studies follow that are all negative. This Related article page 196 Outcomes After Out-of-Hospital Cardiac Arrest Original Investigation Research