GRAPEVINE, TEX. – Researchers found no evidence that the length of resident-covered inpatient rotations – whether 2 or 4 weeks – significantly detracts from hospitalists’ quality of life, or from the quality of care they provide.
However, although nonhospitalists’ patients did just as well as did those of hospitalists, their work/life balance suffered significantly when they worked 4-week inpatient shifts, Dr. Brian Lucas said at the annual meeting of the Society of Hospital Medicine.
Dr. Lucas’s randomized controlled trial on a general medicine teaching ward service was not powered to detect a difference in outcomes between hospitalists and nonhospitalists. Nonetheless, his findings suggest that hospitalists and nonhospitalists experience rotation duration differently, Dr. Lucas said in an interview. It simply may have to do with being accustomed to the rhythm of a hospitalist’s professional life.
"My sense is that hospitalists have the chance to return to a more ‘normal’ schedule after their rotations are over and regain their composure," said Dr. Lucas, chief of hospital medicine at the John H. Stroger Hospital of Cook County, Chicago. "The nonhospitalists don’t experience that. They work at the hospital for 2 or 4 weeks, put their clinic on hold, and at the end, go back to a clinic that has become overburdened. There’s a lot of stress dealing with that."
Dr. Lucas presented the overall findings of the same study at the Society of General Internal Medicine meeting in Phoenix the week earlier. The subanalysis explored at SHM examined work/life balance between the two rotation durations and among the different providers.
The study randomized 62 physicians – 18 hospitalists and 44 nonhospitalists – to sequences of 2- and 4-week rotations, with a total of 130 2-week shifts and 76 4-week shifts. The investigators examined 30-day outcomes in 12,352 patients discharged during these rotations, as well as physicians’ life stressors.
At the end of the study, physicians completed a survey containing portions of three validated life stress measures. Items from the Human Services Survey of the Maslach Burnout Inventory assessed emotional exhaustion; items from the Cohen Perceived Stress Scale measured stress; and items from the Physicians Worklife Study II (Med. Care 1999;37:1140-54) assessed feelings of control in the workplace.
The median number of rotations per physician was three; there was a median of 10 weeks between rotations.
Rotation length did not affect 30-day unplanned patient revisits for either schedule; the rate was 25% for both 2- and 4-week rotations. Residents, who were asked to evaluate their attending physician at the end of the study, did not give significantly different scores according to whether their supervisor worked the 2- or 4-week schedule.
However, Dr. Lucas said, physicians overall did report a significant worsening of work/life balance during the 4-week rotation as opposed to the 2-week rotation. To explore this finding, the investigators first added five characteristics to the model: number of children, being a hospitalist, being an international medical graduate, being a woman, and years of experience.
Three factors significantly affected the overall model. For each child, the attending physician’s work/life became significantly worse, which Dr. Lucas, who has three children of his own, said "is certainly intuitive." Being a hospitalist or an international medical graduate also was significantly protective of work/life balance overall. But whether being a hospitalist limits the imbalance of 4-week rotations depended on which facet of work/life balance was examined.
When considering the entire cohort and both rotation durations combined, the median score in the emotional exhaustion measure was 22. In this measure, the higher the score, the more emotional exhaustion is present. The median life-stress score – which increases as stress increases – was 6. The median score of perceived control at work was 19; this score goes down as perceived control decreases.
All of the scores improved significantly when the physicians worked 2-week shifts rather than 4-week shifts. The difference appeared to be driven by nonhospitalists, however. Their summary scores – which included reactions to both shift schedules – were worse than those of the overall cohort. Working a 2-week shift significantly improved all of these scores, compared with a 4-week shift.
Among hospitalists, however, the summary scores for both schedules were better: 11.8 for emotional exhaustion, 4.6 for life stress, and 21.5 for perceived control. Working a 2-week shift did not significantly improve any of these scores, compared with a 4-week shift, showing, Dr. Lucas said, that hospitalists coped equally well with both rotation lengths.
"In a multivariate model, however, the only significant interaction was perceived control," Dr. Lucas said. "In other words, we found that being a hospitalist protects against loss of control during 4-week rotation, but not that being a hospitalist protects against worsened emotional exhaustion or life stress.