Much of what we know about syncope—its causes, investigation, outcome, and treatment—comes from selected populations in often quite atypical settings. These are usually academic, tertiary care referral centers, with local expertise and reputations in the assessment and treatment of syncope. Often the studies are led by cardiac electrophysiologists or other specialists with particular interests. One would not be surprised if the patients were equally atypical. Indeed, many studies focus on highly symptomatic patients or those with obvious and quite significant underlying comorbidities ((1),2), and the walking well are often excluded. A recent position paper of the Canadian Cardiovascular Society reviewed international data and concluded that the 1-month and 1-year risks of death after an emergency department admission for syncope are 0.7% and 10%, respectively (3). Although these studies are of considerable interest and usefulness for physicians treating patients who resemble those in the studies, they also provide a mild sense of free-floating unreality: just who are the bulk of syncope patients, and what happens to them? In particular, within an entire health care system, what is the outcome of apparently low-risk patients? This is of considerable importance to health administrators and planners who need to know about this large denominator to provide for their health care.

The paper by Ruwald et al. (4) from Denmark in this issue of the Journal makes a significant contribution here. The Danish health care system has nationwide databases. This allowed Ruwald et al. (4) to study every patient in Denmark with a first-time syncope admission to an emergency department or hospital from 2001 through 2009 and then select for further analysis the 40% of the patients who appeared to have no pre-existing comorbidity. They did this by screening for comorbidities based on diagnostic codes and separately by linking the health care database to the pharmacy database and screening for antihypertensive and hypoglycemic medications. Altogether they found 37,017 patients with no known related comorbidities at the time of presentation and then tracked their subsequent journeys through administrative databases for a mean of 4.5 years, along with those of 185,085 well-matched control subjects. Ruwald et al. compared the groups with outcomes such as death, recurrent syncope, cardiovascular morbidity, and cardiac device insertion.

Not surprisingly (5), the overall risk of death in this apparently well population was very low and not different from that of the control group. The yearly overall mortality was ∼1.4% in each group over a mean 4.5-year follow-up. This reassuringly agrees with older, much smaller studies that concluded that syncope on the whole has a benign prognosis (6). It also contrasts starkly with the relatively high mortality reported in many recent studies based in emergency departments ((3),(7),8), wherein the annual mortality rate averages 10% (3). These include patients with significant comorbidities. The authors did find excess mortality in patients 25 to 74 years of age (the age of 74 years was defined as the upper limit of middle age), and their hazard ratio was ∼1.2 to 2.2 that of age-matched control subjects. This excess mortality was mirrored by a substantial increase in the likelihood of a cardiovascular hospitalization or insertion of a cardiac rhythm control device. Interestingly, much older patients with syncope did not have an excess risk of death compared with the control subjects.

This paper provides solid grounding of our understanding of the numbers and long-term outcome of apparently well syncope patients in an entire nation, and this alone is a unique contribution. It is a comprehensive countrywide snapshot of the outcome of apparently low-risk syncope patients. Few other countries are able to mount such an effort, and the authors (and administrators) should be congratulated. It shows that even low-risk patients with apparently no comorbidities on presentation need to be reviewed carefully, although this is not a unique or novel insight. The latter highlights the difficulty faced by all physicians who assess syncope patients: how to risk-stratify accurately ((3),(7),8) and how to spend health care resources efficiently and wisely (9). We are still far from solving this problem; syncope consumed $2.4 billion in the United States in 2003 alone (10), about as much as asthma did. Finally, their paper highlights the precision around small point estimates that large administrative databases can provide.

This report provides a strong basis for related questions, but it also highlights a gap between administrative data at this level and data at the level of the clinic and the patient. The first gap is the eventual diagnosis and outcome. For example, we know almost nothing about the diagnosis of the patients in the study, and there are no electrocardiographic data. In many other studies, abnormal electrocardiographic findings provide significant prognostic value (3). We can infer that many had heretofore undeclared cardiovascular disease because they had an increased likelihood of subsequent cardiovascular admissions, stroke, and insertion of pacemakers and defibrillators. We have no idea of the cause of their deaths nor how their courses compared with those of syncope patients with previously declared comorbidities. There is no information on their course in hospital or subsequent clinics nor on their health care consumption and the efficiency of their workup. What is missing and would be of considerable interest is information on how physicians made their decisions to investigate ((11),12) and with which tools. Several risk-stratification tools based in the emergency department have appeared over the past 10 years, but they generally would have screened out this group as they usually feature markers such as known heart disease, heart failure, anemia, dyspnea, and advancing age ((3),8). Finally, there are the limitations based on studies from single countries. How do these patients and this health care system compare with those of other countries, particularly those without anything resembling a unitary system?

Ruwald et al. provide an important contribution that will set the context for more studies on how to provide syncope assessment. Given the lack of declared baseline comorbidities, what are the most prevalent outcomes? Where should clinical suspicion be directed, and how should we target resources at the most likely undeclared diagnosis? Where and when in the time after admission was the diagnosis made? What are the costs of assessing this low-risk group of patients? Given the overall low risk of the patients, these become nontrivial considerations. What is the most effective selection of tests and type of health care provider who could provide early assessment in a syncope clinic? Answers to these will emerge in the next few years and will be solidly grounded in countrywide studies such as the one by Ruwald et al. (4).

⁎Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.