Spontaneous subarachnoid hemorrhage (sSAH) is still a challenging neurosurgical disease entity. Emergency and elective care of sSAH is critical given that cerebral vasospasm, rebleeding, increased intracranial pressure due to acute brain swelling, related intra-parenchymal hemorrhage, or acute hydrocephalus, hemodynamic instability, cerebral infarction, seizure, electrolytes imbalances and cardiac rhythm abnormalities due to myocardial injury are all possible life-threatening complications.5)10)17)

Surprisingly, the unnoticed sSAH remains as high as 12%, resulting in four-fold higher rates of morbidity and mortality.16) A common comment from a variety of studies suggests that early identification and treatment of sSAH and related complications are critical components to achievement of desirable outcomes.5)

Meanwhile, thorough reappraisal of educational programs for training of interns and residents in western countries has increased due to results suggesting an increment in unacceptable outcomes in the late summer (that falls during the spring in Korea), which is the unskilled season for beginner interns and residents.23) This tendency, termed a "July effect" in western countries has come from the July inrush of inexperienced interns and residents of the first year. Because the training hospital workforce is highly dependent on intern and resident doctors for duty, especially for night time and emergency room, they stand for the first line of care at these hospitals and may frequently be urged to make acute management decisions and provide care with limited guidance. Because reliable clinical evidence indicates that favorable outcomes in sSAH may be associated with early identification of the disease, rigid compliance to clinical management guidelines and experience and knowledge of corresponding physicians could play a considerable role in such outcomes.5) In Korea, a similar phenomenon might have occurred in March because of the difference in the period of educational systems between western countries and Korea. Some clinical studies on the "July effect" have been reported; on the contrary, no available study on the 'March effect' has been conducted in Korea, especially in the neurosurgical field. The authors attempted to identify a 'March effect', defined as a temporary increase in the frequency of adverse outcomes (mortality rate and unfavorable discharge rate), among a cohort of patients admitted to our hospital under the diagnosis of sSAH.

A total of 455 consecutive patients with sSAH between January 2008 and December 2010 were retrospectively enrolled in the present study. Their mean age was approximately 54 years (54.61) and sex ratio was approximately 4:6 (Male: Female = 41.3%: 58.7%). A summary of patients' clinical information is shown in Table 1 and more detailed information according to the admission month is shown in Table 2.

Four groups (quarter 1 to 4), divided according to their admission month, did not show statistically significant difference in their clinical characteristics, including age, Hunt-Hess and Fisher grades at admission and type of treatment procedures (Table 3).

The 'July phenomenon or effect' in western countries (called a 'March effect' in our study) has been studied across different medical specialties, including general surgery, obstetrics and gynecology and internal medicine.2)4)5)8)10)14)18)20)21)27) Evidence of or against the presence of a 'July effect' has been inconsistent. The principal causes for discordance among published studies are likely a function of several phenomena. First, many of these studies derived their data from single-center populations and thus may more accurately reflect the clinical and educational practice patterns of a particular hospital. Second, smaller studies may have failed to normalize data on changes in seasonal practice patterns, which may falsely manifest as relative changes in outcomes. Third, as in our study, complete information regarding long-term outcomes and prognosis were not available in the database and thus could not be used as outcome metrics. Fourth, differing end points and adverse outcome metrics were used among the different studies, including ours, making inter-study comparisons difficult.

The majority of studies on the 'July effect' have found no meaningful difference in patient outcomes in July compared to other months. Buchwald et al.,7) who evaluated neck and back problems, found no increase in the length of hospital stay for July admissions, compared with May. Banco et al.,2) who focused on the incidence of perioperative spinal infection rates at different times during the academic year. They found no association between infection rates in the month of July or other times of the year when residents or fellows were new on the service.

On the other hand, some studies have demonstrated a 'July effect'. In a recent study, after examining all available US death certificates from 1979 to 2006, Phillips and Barker23) reported a 10% increase in fatal medical errors in the inpatient, outpatient and emergency room setting. Englesbe et al.11) reported an increase in postoperative mortality of 41% among patients treated by interns within the American College of Surgeons National Surgical Quality Improvement Program. One study reported a slight increase in overall surgical mortality during the years 1991 to 1997 at teaching hospitals.4) Walling and Veremakis28) reported that first-year residents made more prescribing errors during early months of their training compared with later in the year. In addition, Inaba et al.14) demonstrated an increased risk of errors resulting in preventable or potentially preventable complications at the beginning of the academic year (July and August) compared with the end of the academic year (May and June) at an academic trauma center.

To the best of the author's knowledge, no neurosurgical studies on the 'July effect (March effect)' have been conducted in Korea. In this retrospective, cohort study of 455 consecutive sSAH admissions, we found no statistically significant increase in the mortality or unfavorable discharge rate for the month of March in our hospital compared with all other months.

In the present study, there are several limitations. First, this study is a retrospective study, which has the potential for analytic limitations in some clinical variables and situations, such as medical errors and adverse events. Second, possible limitation is the risks inherent to emergency cases. Given the circumstances of the tertiary emergency room, there are a few potentials for misdiagnosis at admission, which may result in inaccurate mortality and/or unfavorable discharge rates. Third, there is another limitation of the retrospective study. We were not able to analyze the rates of complications and near misses that were not associated with mortality or unfavorable discharges, committed by neurosurgery interns and residents, especially during management in the intensive care unit, which may be a more sensitive tool for evaluation of a 'March effect'. Fourth, one of the limitations to this study is that our data are limited to in-hospital observations. Therefore, mortality occurring after discharge would not be assessed. However, because complications and medical errors, which resulted in death of the patients, committed by neurosurgery interns and residents are most likely going to be in-hospital and not after discharge, we believe that this should not have a significant effect on the result of our study. Fifth, the temporal resolution is limited to monthly intervals and fluctuations in adverse outcomes over time periods shorter than one month would not be detected in the current study, especially if the rotation schedules of interns are shorter than one month. And, finally, we did not consider the clinical experience of the surgeons performing surgical clipping and/or endovascular interventions in our hospital, which might be the most important factor influencing the patients' outcome. However, cases of each surgeon were evenly allocated over one year, which might not have an effect on the monthly variation of patients' outcome.

Generally speaking, it is reasonable to assume that at the start of an academic year, a number of errors made by interns and residents would be higher compared with later months. Alternatively, the absence of a "March effect (July effect)" may be attributable to the increased supervision by more senior residents and faculties during the start of the academic year.

Findings of our study suggest that the 'March effect' was not found for neurosurgical mortality or unfavorable discharge rates in patients with sSAH in the author's hospital. Even with some limitations, our study adds to the limited knowledge on the 'March effect' in the field of neurosurgery. In order to establish a 'March effect' in the neurosurgical field, including sSAH, a meta-analysis from multi-institutional prospective study is anticipated.