The Evidence Speaks Series is a recurring feature highlighting the latest in CHÉOS research. This series features summaries of select publications as well as in-depth features on the latest work from our investigators. The Series is designed to keep media and the research community up-to-date with CHÉOS’ current research results in the health outcomes field.

Hill A, Ramsey C, Dodek P, Kozek J, Fransoo R, Fowler R, Doupe M, Wong H, Scales D, Garland A. Examining mechanisms for gender differences in admission to intensive care units. Health Serv Res. 2019 Nov 10 epub ahead of print.

CHÉOS Scientists Drs. Peter Dodek and Hubert Wong are co-authors on the first study to explore the effect of multiple mechanisms on the gender imbalance in intensive care unit (ICU) admissions. Despite women outnumbering men in older age groups, they are much less likely to be admitted to the ICU. The reason for this imbalance is unknown, but several mechanisms have been proposed: (1) women have less family, social, and decisional support due to higher rates of widowhood; (2) gender differences in the presentation of some illnesses lead to different intensities of health care use; and (3) gender-based triage decisions made by medical personnel or patient preference for aggressive care. To understand the potential role of these three mechanisms, the study team linked several databases to form three cohorts: a general population cohort in Manitoba and two Canadian cohorts, one for patients receiving home care and one for nursing home care. As expected, the overall analysis showed that men were more likely to be admitted to the ICU than women. To analyze mechanism 1, the researchers controlled for number of children as well as marital status. This did not change the predominance of males in ICU admissions. As cardiovascular-related illnesses are responsible for approximately half of ICU admissions in men, the researchers tested the effect of mechanism 2 by removing cases related to heart attack or cardiac surgery. This narrowed the gender gap slightly, reducing the difference in likelihood of ICU admissions by 13.5%. Finally, the researchers investigated mechanism 3 by only comparing patients who had a previous ICU admission. This analysis eliminated the gender difference, suggesting that there was not a strong triage bias in favour of men. The primary weakness of this study is that its use of administrative data did not allow for in-depth exploration of some factors that could affect ICU admissions, including patient preferences and subjectivity in care provider decision making. Overall, this study does not support the role of these three mechanisms in the gender differences in ICU admissions.

Sanaee M, Pan K, Lee T, Koenig NA, Geoffrion R. Urinary tract infection after clean-contaminated pelvic surgery: a retrospective cohort study and prediction model. Int Urogynecol J. 2019 Oct 31 epub ahead of print.

CHÉOS Scientist Dr. Roxana Geoffrion led a study, co-authored by CHÉOS Statistician Dr. Terry Lee, to identify the rate of urinary tract infection (UTI) following gynecologic surgery and develop a risk prediction model for postoperative UTI. UTI is a common side effect of pelvic surgery and interventions to reduce UTI risk specific to female urology and gynecology have not been identified. Dr. Geoffrion and her team sought to develop a prediction model for UTI to aid the development of preventative strategies. The study, published in the International Urogynecology Journal, was a retrospective cohort analysis of 310 patients who underwent urogynecologic surgery or other gynecologic surgeries over a 16-month period. UTI was defined as a positive urine test and related symptoms within 6 weeks of surgery. Of the 310 patients, 66% were admitted to hospital following surgery and 71% required a trial of void, an assessment of the ability to empty the bladder. Forty women (13%) had a confirmed postoperative UTI, while 21 had symptoms indicative of a UTI but no laboratory confirmation. Risk factors for UTI development included prolapse repair, anti-incontinence surgery, and significant blood loss. In the predictive model, significant predictors of UTI development were older age, undergoing a greater number of surgical procedures, and prolonged difficulties in bladder voiding. This predictive model will assist surgical teams in taking preventative measures to avoid UTI, such as carefully screening older patients or providing counselling to reduce UTI risk. The model and identified risk factors will also help teams reduce unnecessary use of antibiotics by allowing them to provide these medications only to patients who have a high predicted risk of UTI and using alternative medications for those in lower risk categories.

Maas ET, Koehoorn M, McLeod CB. Descriptive Epidemiology of Gradual Return to Work for Workers with a Work-acquired Musculoskeletal Disorder in British Columbia, Canada. J Occup Environ Med. 2019 Nov 15 epub ahead of print.

A new study led by CHÉOS Scientist Dr. Mieke Koehoorn describes the factors that contribute to gradual return to work after a work-related musculoskeletal injury in British Columbia. Musculoskeletal injuries suffered at work are the leading cause of workers’ compensation claims and sickness absences in B.C. The current approach to rehabilitation from these injuries incorporates a gradual return to work, where patients increase work hours and load, with modified tasks, until full return to work is achieved. This approach has been shown to improve both physical and mental health, and increase the likelihood of full return compared with traditional approaches where workers are rehabilitated away from work until they are ready to return. However, the factors that contribute to the likelihood of employees being offered a gradual return to work after injury are not well known. To answer this question, Dr. Koehoorn, alongside colleagues from UBC’s Partnership for Work, Health, and Safety, used administrative WorkSafeBC and return to work data for musculoskeletal injury claims over a 5-year period. Of the 141,490 workers in the study cohort, back strains were the most common injury type (44%) and 41% of people in the cohort were involved in a gradual return to work. There was an increase in gradual return to work over time, from 35% of claims in 2010 to 47% of claims in 2015. The main worker determinants that increased the likelihood of participating in a gradual return to work were being female, having a serious injury, older age, higher wage, longer duration of sickness absence, and a recent previous work claim. Business size was the strongest workplace-related determinant of gradual return to work. These findings may inform the targeting of return to work interventions and will help to reduce the inequalities in who is provided with a gradual return to work. The results suggest that it may be particularly beneficial to target smaller businesses that may be unable to provide gradual return to work programs for their employees.