Download a PDF of the Current Issue 2015 Volume 12 Number 3 July- September

The July Effect

Michael E. Mahla, MD Professor of Anesthesiology and Neurosurgery Associate Dean for Graduate Medical Education

The ‘July Effect’. Everybody knows it exists, right? In July with all the new housestaff in a teaching hospital, patients admitted to the teaching hospital or cared for in its clinics are much more likely to be exposed to physicians who don’t possess the appropriate skills or experience to provide optimal care. Thus (intuitively speaking), serious medical errors and worse outcomes are much more likely in July than at other times during the academic year. Nurses talk about the July experience all the time. Walking down the hall, I have commonly heard,”Oh, boy – another set of housestaff that we have to break in.”  Attending physicians remark “just when I get these housestaff to the point where I don’t have to pay attention to every little detail, they leave and I have to start all over again.” Even patients know about the July effect – especially when CBS News and similar organizations put up sensational headlines detailing this July effect.

Quoting from a July 12, 2011 article from CBS News 1:  

‘July effect’ in teaching hospitals increases odds patients will die’

(CBS) It’s summertime, and the living is easy ‐ unless you need treatment from a teaching hospital. Then you might be lucky just to get out.  

A new study reports that more patients receive worse‐ quality care or die at teaching hospitals during July because experienced residents shuffle off to greener pastures, leaving untrained ʺnewbiesʺ to take their spots and learn the ropes.


The ʹJuly Effectʹ occurs when these experienced physicians are replaced by new trainees who have little clinical experience, may be inadequately supervised in their new roles, and do not yet have a working knowledge of the hospital system, Dr. John Q. Young, associate pro‐ gram director for the Residency Training Program at the University of California ‐ San Francisco School of Medicine said in a written statement. ʺIt’s a perfect storm.ʺ

  This news article was reporting on a September, 2011 article published by Dr. Sumant Ranji and colleagues in the Annals of Internal Medicine2. The article, entitled “July Effect”: Impact of Academic Year‐End Changeover on Patient Outcomes: A Systematic Review, reviewed 39 studies involving inpatients only that looked at mortality, efficiency (length of stay, duration of procedures, hospital charges), morbidity, and medical error outcomes. Nearly all data came from US teaching hospitals. The authors noted great variability in the quality of the studies included in their meta‐analysis.  Higher quality studies that included larger sample sizes tended to show increased mortality and decreased efficiency around the time of changeover. Studies with smaller sample sizes looking at just morbidity and medical errors produced inconsistent results. The authors concluded “Mortality increases and efficiency decreases in hospitals because of year‐ end changeovers, although heterogeneity in the existing literature does not permit firm conclusions about the degree of risk posed, how changeover affects morbidity and rates of medical errors, or whether particular models are more or less problematic.”   As Ranji and colleagues point out, not all articles were able to clearly document a July effect.  Articles focusing on patients at very high risk for medical errors, for example, commonly did NOT document significantly different outcomes in July.  In an article from the Department of Surgery at Baylor University 3, Dr. Bakaeen and colleagues looked at outcomes from cardiac surgery in July versus the remainder of the academic year in 70,000+ cardiac surgical patients.  They did demonstrate that surgical procedure lengths were “slightly” longer, but risk adjusted outcomes were not related to the month of the year that the surgery was performed. Similarly, in a study published in 2011 by a group from the Faculty of Medicine at the University of Ottawa in Canada 4, no July effect was able to be documented.  This study looked at 259,748 encounters with 164,318 patients admitted to the medical, surgical and obstetrical services at a teaching hospital in Ottawa.  The overall mortality rate was 3% and was not related to housestaff experience level. In addition, this lack of association of mortality with housestaff experience level did not vary by service or by whether the admission was elective, urgent or emergent. A study by Schroeppel et al published in the Journal of the American College of Surgeons focused on the trauma patient population, which is also considered at high risk for medical errors due to the need for rapid decision making. In this study 5 conducted in a Level I trauma center, the investigators were unable to demonstrate a relationship between the month or quarter of the year and overall mortality, ICU days, or minutes in the resuscitation room.  In another study focusing on the acutely ill medical patient with acute coronary syndromes or decompensated heart failure 6, the investigators found no evidence of less optimal medical therapy for these patients in the early part of the academic year.   Finally, in a study published in the Journal of Perinatology 7, the authors compared 26,546 women who delivered their babies in July to 272,584 women who delivered their babies in the remainder of the academic year.  There were no statistically significant differences in the rates of cesarean delivery, urethral/bladder injury, third or fourth degree lacerations, wound complications, postpartum hemorrhage, transfusion, shoulder dystocia, chorioamnionitis, anesthesia‐related complications, brachial plexus injury, or birth asphyxia between the two groups.   Well, where do these studies leave us? Is there a July effect, or not?  Science suggests that there might be a July effect, but the available literature leaves many with that question still unanswered. I would suggest a more “spiritual” approach to this question and look at the question from the stand‐ point of the new trainee, the attending physicians, and the patient.   Without doubt, the July effect clearly exists in the minds of new trainees – particularly the PGY‐1 residents fresh from medical school. As our Dean and Associate Dean for Clinical Affairs pointed out clearly in this year’s housestaff orientation, the Doctor of Medicine degree is nearly the only terminal degree where the recipients are clearly not ready to begin an independent career.  Housestaff facing their first day on the inpatient wards or in the outpatient clinics have no idea what they will be facing – what diseases or disorders they will encounter, and worry whether their training has prepared them for their upcoming experiences. On top of that concern, new housestaff often know nothing about their community, have few or no friends in the area, and their personal support systems are often many hundreds of miles away.  We are asking our new housestaff to take care of some of the sickest patients they have ever encountered in a strange and new work environment in a com‐ munity where they have little or no support.  Thus, there is, by definition and even by design, a July effect for new PGY‐1 housestaff.   This July effect may also be seen in our more experienced housestaff who are now asked to take on more supervisory and more independent activities. While these housestaff know their work environment and their community and have hopefully developed personal support systems over the year (s) They have been here, taking on supervisory and more independent roles may certainly create anxiety and another type of July effect in our more experienced trainees.  These more experienced trainees will also be expected to help their new PGY‐1 col‐ leagues overcome their ‘July effect’.   Whether or not this July effect translates into less optimal patient experiences and outcomes depends on the attending physicians’ response to their own ‘July effect’.  As noted previously, attending physicians sometimes bemoan the fact that their experienced housestaff leave and less experienced, as well as brand new trainees, appear on the scene. Fortunately, though, the University of Florida College of Medicine is blessed with many faculty who know that this process of growth, development, and “moving on” is why they are here. They rise to the challenge of making sure that the July effect, which clearly exists for our housestaff, does not translate into more frequent errors and bad patient outcomes.   Certainly careful, direct supervision is the corner‐ stone to preventing impacts of the July Effect. However, there is another important aspect to preventing the July effect from translating into bad experiences and outcome.  Attending physicians, more senior housestaff, their nursing colleagues, and the hospital administration must strongly endeavor to make the learning environment safe for new trainees.  Our newest housestaff members, regardless of level, must feel safe to ask questions. They must not feel that they must do everything possible to avoid asking questions. I remember years ago during my intern training that it was very poor style to be constantly asking questions. Housestaff who did ask a lot of questions were quickly identified as weak and questions were commonly asked about “where did that person go to medical school”? So, I had to figure out a lot of things for myself – and commonly came upon answers and solutions that usually worked, but in retrospect were not optimal. At least, however, I didn’t have to reveal my ignorance and endure ridicule from nurses, my housestaff colleagues, and my attending physicians.  That approach to learning is clearly not optimal for either the patient or the learner.  Sure enough, some housestaff do have deficits in knowledge and experience.  The faculty and program directors are responsible for identifying those deficits and developing remediation pro‐ grams to correct them.  We are all (faculty, col‐ leagues, nurses, administrators) responsible for making sure that our trainees feel safe while they grow in knowledge and experience.  The best way, in my opinion, to lessen any impacts of the July effect, is to make the learning environment safe for our trainees – at all levels.   From the patient’s standpoint, the media has made sure that they believe there is a July effect.  The best way, in my view, to alleviate any patient anxiety about the July effect is communication.  Housestaff must let their patients know who they are, their experience level, and what their role in the patient’s care is going to be.  One of our patients’ greatest fears is that they have no way of knowing the experience level and skills of the doctors taking care of them. To the chronologically gifted patient (yes, I am one of them), all our housestaff tend to look very young.  I certainly cannot tell the difference between a PGY‐1 resident and a PGY‐4 resident unless I know them personally. If patients and families know that a physician caring for them is a PGY‐1 resident, I have found in many cases that they will bond closely with her/him and look to her/him as the “boots on the ground” advocate for them and their care. Equally important for the patient and the family is clear evidence that faculty and senior level housestaff are closely involved with the new trainees and are supervising them and providing them with the benefits of their knowledge, skills, and experience.  If patients know their doctors and their skill and experience levels, they will come to realize that more sets of eyes looking at their health problems and needs are better than one set of eyes, and that in fact, the care they receive is as good as it can possibly be. They will feel like they are part of their own healthcare team.  When we fail to communicate who we are, the July effect becomes very real to the patient not only in July, but in December and June as well.   To conclude, I would like to leave you with 3 major points to remember: There is a July effect in the minds of all our housestaff but most importantly for our PGY‐1 and PGY‐2 housestaff just beginning their career in medicine or in their specialties. The design of our graduate medical education system, with its incumbent moves around the country for training, adds further stress to the July effect. Faculty and more senior housestaff have an obligation to make the learning environment safe for all trainees – especially for the PGY‐1 and PGY‐2 trainees.  A safe learning environment leads to the asking of important questions – that are asked with the intent of providing good care to our patients. Patients must know who their doctors are – their experience level and their role in their care. The patient can then become part of their healthcare team, and when they are part of their own team, fear is greatly lessened.   If we do all these things to the best of our ability, the July effect will have no effect on how well we care for our patients.   References: 1. Jaslow R. ‘July Effect’ in Teaching Hospitals Increases Odds Patients will Die. CBS News Report, July 12, 2011. 2. Ranji SR et al. “July Effect”: Impact of Academic Year‐ End Changeover on Patient Outcomes: A Systematic Review. Ann Int Med 2011; 155 (5), 309‐315. 3. Bakaeen FG et al. The July Effect: impact of the beginning of the academic cycle on cardiac surgical out‐ comes in a cohort of 76,616 patients. Ann Thorac Surg 2009; 88: 70‐75. 4. Van Walraven C et al. Influence of house‐staff experience on teaching‐hospital mortality: the “July Phenomenon” revisited. J Hosp Med 2011; 6: 389‐94. 5. Schroeppel TJ et al. The “July Phenomenon”: is trauma the exception? J Am Coll Surg 2009; 209: 378‐84. 6. Garcia S, Canoniero M, Young L. The effect of July ad‐ mission in the process of care of patients with acute cardiovascular conditions. South Med J. 2009; 102: 602‐7. 7. Ford AA et al. Nationwide data confirms absence of ‘July Phenomenon’ in obstetrics: it’s safe to deliver in July. J Perinatol 2007; 27: 73‐6.