No Evidence of 'July Effect' in Cardiac Surgical Outcomes

Batya Swift Yasgur MA, LSW

August 01, 2019

July, when new medical graduates begin their in-hospital training, has often been perceived as a time of increased patient complications and medical errors — a phenomenon sometimes called the "July effect."

A new study analyzing over 470,000 cardiac procedures in more than 35 million hospital admissions between 2012 and 2014 dispels that assumption, finding no poorer outcomes from major cardiac surgeries in teaching hospitals during the first quartile of the academic year (Q1, July to September) than the last quartile (Q4, April to June).

"We found that the July effect does not exist for major cardiac surgery procedures on a national level," cofirst author Rohan Shah, MD, MPH, cardiac surgery research associate, Brigham and Women's Hospital, Boston, told theheart.org | Medscape Cardiology.

"This is an important finding with implications on overall resident training and education in the operating room, while maintaining excellent outcomes. What it means for patients is that they should not be fearful/concerned of having surgery in July, when the new residents/trainees are starting," he said.

The study was published online July 19, 2019 in theAnnals of Thoracic Surgery.

"Anecdotal" Perception

"The topic of resident education is very germane in the current reporting," cofirst author Samir Hirji, MD, MPH, clinical fellow in surgery, Brigham and Women's Hospital, told theheart.org | Medscape Cardiology.

"In the past, teaching residents came at the price of poor outcomes, but was considered a necessary cost," he said.

Moreover, together with the influx of new residents, the most senior residents are graduating from the program, "further adding to the anecdote that patient care may be inferior during the beginning of the academic year."

No previous studies have examined the so-called July effect, which "describes the supposed drop-off in patient care in teaching hospitals with the start of new residents with less experience," he said.

Previous studies in other surgical specialties did indeed show an increase in mortality and/or complications at the start of the academic/residency year, to the point that some patients "avoid getting surgery in July for this particular reason."

Therefore, "in the current era of public reported outcomes and introduction of new cardiac surgery training models, we felt [the July effect] was a topic worth pursuing," he said.

To investigate the question, the researchers used the National Inpatient Sample (NIS), the largest publicly available all-payer database in the United States, accounting for 20% of all hospital admissions in patients 18 years and older and then weighted to account for all admissions nationally.

The study included all adults who had undergone coronary artery bypass grafting (CABG), surgical aortic valve replacement (AVR), mitral valve repair or replacement (MV), or isolated thoracic aortic aneurysm (TAA) procedures between 2012 and 2014.

Patients with missing data, concomitant procedures during the index hospitalization, or who underwent transcatheter AVR were excluded.

The primary outcome was in-hospital mortality; secondary outcomes included inpatient costs, length of stay, complications, and patient disposition after surgery.

Month Irrelevant

During the study period, there were 301,105 CABG, 111,260 AVR, 54,985 MV, and 2,655 TAA procedures performed, of which 47,240 (48.9%), 54,540 (49.0%), 27,065 (49.2%), and 1,345 (50.7%) admissions, respectively, took place in Q1.

In general, there were no significant clinical differences in baseline patient demographics, comorbidities, and admission characteristics between patients admitted in Q1 and Q4; however, there were a few exceptions:

  • CABG patients admitted in Q1 were slightly younger (66.0 vs 66.2 years; P < .05) and statistically more likely to have diabetes with chronic complications and obesity (8.9% vs 8.4% and 23.6% vs 22.7%, respectively; P < .05 for both) compared with those admitted in Q4.

  • AVR patients admitted in Q1 were slightly younger (67.5 vs 67.9 years; P < .05) and more likely to have coagulopathy disorders or obesity (34.4% vs 32.5% and 21.1% vs 19.8%, respectively; P < .05 for both)

  • There was no significant difference in age between MV surgery patients admitted in Q1 vs Q4, although Q1 patients were more likely to be female (48.2% vs 46.0%; P < .05) and to have a high prevalence of coagulopathic disorders (32.9% vs 30.4%; P < .05).

  • Patient undergoing TAA procedure in Q1 vs Q4 were younger (57.8 vs 60.3 years), although this did not reach statistical significance (= .11), and were also less likely to have peripheral vascular disorders (19.7% vs 27.9%; P < .05).

There was little variation between unadjusted in-hospital outcomes, disposition, and hospital factors in Q1 vs Q4 admissions.

Although in-hospital mortality for CABG patients was similar between Q1 and Q4 (2.4% vs 2.2%), patients admitted in Q1 had slightly higher hospital inpatient costs tha  those admitted in Q4 ($54,310 vs $53,223; P < .01).

For AVR patients, those admitted in Q1 were more likely than those admitted in Q4 to experience postoperative cardiac complications (14.7% vs 13.5%; = .02).

After risk adjustment, in-hospital mortality was not found to be higher in Q1 than in Q4.

Table 1. In-hospital Mortality by Procedure
Procedure Sample Size (n) Odds Ratio 95% CI P Value
CABG 301,105 1.0 0.9–1.2 .57
AVR 111,260 1.0 0.8–1.3 .76
MV 54,985 1.0 0.8–1.3 .69
TAA 2,655 1.1 0.1–37.3 .98

Although overall unadjusted differences in in-hospital mortality by month for CABG and surgical AVR were significant, outcomes of patients admitted in July "were not notably better or worse, compared with patient admitted in other months."

"Our study showed that in cardiac surgery, you will receive the best care, regardless of the month that you get your operation," senior author Tsuyoshi Kaneko, MD, cardiac surgeon at Brigham and Women's Hospital, told theheart.org | Medscape Cardiology.

"I think this is a very important message for the patients and their loved ones," added Kaneko, who is also assistant professor of surgery at Harvard Medical School.

Vulnerable Period

Teaching status did not influence risk-adjusted mortality between Q1 and Q1 for CABG or isolated TAA, but teaching hospitals had significantly lower mortality than nonteaching hospitals for AVR and MV.

Table 2. In-Hospital Mortality in Teaching and Nonteaching Hospitals
Procedure and Quarter Teaching Hospitals, % Nonteaching Hospitals, % P Value
CABG
Q1 2.9 2.3 >.05
Q4 2.2 2.2 >.05
TAA (on postestimation analysis of regression models)
Q1 4.5 4.0 >.05
Q4 4.4 3.9 >.05
AVR
Q1 3.17 3.97 <.01
Q4 3.12 3.92 <.01
MV
Q1 4.40 5.78 <.01
Q4 4.24 5.58 <.01

Hirji noted that although the investigators did not analyze outcomes specifically at Brigham and Women's Hospital, and therefore could not provide data on the subject, "we have not seen any decline in patient care in the month of July."

"The entire cardiac surgery team recognizes that this is a vulnerable period and the team, including the nurses and the physician assistants, plays a critical role in maintain the high quality," he added.

Hirji described the balance between supervision by attending physicians and resident autonomy as "constantly shifting during teaching," with a greater focus on "direct and strict supervision" during the earlier months of the academic year, and a shift to greater autonomy as the year progresses — a process he described as "a fine balance."

He expressed hope that studies such as theirs "will help reinforce the fact that our clinical processes and hospital systems have multiple processes that help provide the highest level of care."

Study coauthor Cheryl K. Zogg, MSPH, MHS, was supported by a grant from the National Institutes of Health Medical Scientist Training Program. Kaneko disclosed a financial relationship with Edwards Lifesciences, Abbott, and Medtronic. The other authors disclosed no relevant financial relationships.

Ann Thorac Surg. Published online July 19, 2019. Abstract

For more theheart.org | Medscape Cardiology news, join us on Facebook and Twitter.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....