Sex differences in admission to intensive care units: the role of social support factors

This study addressed issues of equity related to access to critical care.

Research Results

Findings: Contrary to our hypothesis, we found that the male predominance of admission to an intensive care unit (ICU) is not explained by widowhood, or other differences in social supports of men versus women towards the end-of-life. Indeed among elderly individuals, being married was associated with a lower likelihood of being admitted to an ICU. Our data also provides an indirect clue to the actual explanation for the male predominance of ICU admission in the elderly. Specifically, we found evidence highly consistent with the idea that older women, in general, are less willing than older men to undergo ICU care. This evidence does not suggest that there is a disparity between men and women in access to ICU care.

Impact of findings: This indicates that there is not a need for urgent efforts to remediate bias in ICU access for women. But it would be interesting and valuable to understand the motivation(s) behind a generally lower desire of woman for aggressive medical care. Do older women have an overly pessimistic view of the outcomes of such care, or do older men have an overly optimistic view? Unrealistic understanding of the range and likelihood of outcomes of treating critical illness could result in some individuals denying themselves a chance to survive with good functional outcomes, and other individuals going through futile and uncomfortable healthcare interventions.

About the Project

Researchers have found that for certain illnesses patients may be more likely to receive certain kinds of care depending on whether they are men or women. Previous studies by this team have suggested: that men receive more critical care treatments for almost all kinds of conditions; this male predominance of critical care is accentuated in elderly people; and that women are less likely to survive their critical illness, despite being about as sick as men when they are admitted to the hospital or ICU.

The investigators assessed whether the differences in admission to ICUs for men and women over the age of 65 years are due to differences in social supports (marital status, number of children), and prior personal or family member experience with critical care.

A better understanding of the factors that influence health care decisions and outcomes for critically ill older people and differences by gender is needed to develop action plans to ensure equitable delivery of care to these individuals.

Project Team

Principal Investigator:

Allan Garland, MD, MA, BSc — University of Manitoba


Peter Dodek, MD, MHSc — St. Paul’s Hospital

Malcolm Doupe, PhD, MSc, BPEd — University of Manitoba

Robert Fowler, MD, MSc — Sunnybrook Research Institute

Randy Fransoo, PhD — University of Manitoba

George Heckman, MD, MSc, FRCP(C) — University of Waterloo

Jean-François Kozak, PhD, MA — Providence Health Care, Vancouver

Clare Ramsey, MD, MSc, FRCPSC — University of Manitoba

Hubert Wong, PhD — Providence Health Care, Vancouver

Project Contact: Dr. Allan Garland —

CORE 2012-19

Key words: sex; social supports; ICU admission; ICU mortality; widowhood; elderly



Ramsey, C., Fowler, R., Dodek, P., Fransoo, R., Doupe, M., Wong, H., Kozak, J., Hill, A., Chen, H., Yogendran, M., Garland, A. (May 2015). The Effects of Widowhood on Sex Differences in Admission to Intensive Care Units. Presented at 2015 American Thoracic Society International Conference, Denver, CO. American Journal of Respiratory and Critical Care Medicine 191:A1393.

Ramsey, C.D., Hill, A., Dodek, P., Fowler, R., Kozak, J., Wong, H., Fransoo, R., Doupe, M., Yogendran, M., Garland, A. (May 2016). Male Predominance in Admission to Intensive Care Units (ICU) Disappears After Surviving One ICU Admission. Presented at 2016 American Thoracic Society International Conference, San Francisco, CA. American Journal of Respiratory and Critical Care Medicine 193:A2957.

Hill, A, Ramsey, C.D., Dodek, P., Kozak, J., Doupe, M., Fowler, R., Wong, H., Fransoo, R., Scales, D.C., Garland, A. (May 2016). Marital Status Does Not Explain the Male Predominance of ICU Admission Among Home Care Clients. Presented at 2016 American Thoracic Society International Conference, San Francisco, CA. American Journal of Respiratory and Critical Care Medicine. 193:A3642.

Hill A, Ramsey C, Dodek P, Kozak J, Fransoo R, Fowler R, Doupe M, Wong H, Scales D, Garland A. Examining mechanisms for gender differences in admission to intensive care units. Wiley Online Library: Health Services Research.

Rationale, Hypothesis, Objectives & Research Plan

Rationale: The rationale behind the current study begins with the assumption that among the elderly, the willingness to undergo aggressive care will be higher if ones’ spouse/partner is still alive.

Hypothesis: Based on the fact that women generally outlive men, the current study tests the hypothesis that the male predominance of ICU admission is mediated by a lower desire of unmarried/widowed individuals to undergo ICU care.

Objectives: For cohorts (1) and (2) we assessed the impact of marital status on ICU admission using multivariable Cox regression of time from cohort entry to ICU admission, or multivariable logistic regression of ICU admission during acute care hospitalization, adjusted for covariates. For cohort (3) we used logistic regression to evaluate admission to ICU among those admitted to hospital. Our primary analyses randomly selected one record for each subject included.

Research Plan: For this retrospective study, we studied three cohorts of individuals 65 years and older: (1) all Manitobans during the interval 2004-2012; and all Canadians [excepting Québec] during the interval 2007-2012, (2) living in nursing homes or (3) living at home and enrolled in homecare. These administrative datasets, which include information about marital status, were obtained from the Manitoba Centre for Health Policy, or from the Canadian Insitute for Health Information.

Communication to Family

Key Findings

  • We thought that being married and having more social support may explain why more men are admitted to the intensive care unit (ICU).
  • However, our study showed that marital status and widowhood status did not explain why men outnumber women in the ICU.
  • Our study also suggests that healthcare provider bias does not account for the higher number of men admitted to the ICU.

Evidence gap addressed by study

Compared to men, women are less likely to be admitted to the ICU. One reason for the different numbers of men and women may be a preference for healthcare providers to admit men, rather than women. Our research showed that men tend to be more severely ill, regardless of their age and overall risk for certain critical conditions. Although we don’t know why men are more likely to be admitted to ICU, it is possible that decisions about aggressive care differ between men and women, as patients, or as their substitute decision-makers, or healthcare providers who make the decisions in the hospital setting.

How these findings could impact older adults living with frailty and their family or friend caregivers, and how this impact might be measured

  • Our findings show that sex differences in social supports do not explain the higher number of men admitted to ICU.
  • Our study did not show that healthcare providers were preferably admitting men rather that women. However, our findings suggest that older women are, in general, less likely than older men to want aggressive care that is provided in ICUs.

Remaining knowledge/research gaps

  • To make sure we have correctly understood the sex differences in ICU admissions, we will interview older people and their substitute decision-makers in a variety of care settings.
Communication to Researchers

Key Findings

  • Contrary to our hypothesis, the male predominance of intensive care unit (ICU) admissions is not accounted for by differences in social supports between older men and women. Specifically, adjustment for marital/widowhood status did not alter the observed male predominance.
  • Our findings also argue against the male predominance in ICUs being explained by a bias among ICU gatekeepers, the healthcare providers.

Evidence gap addressed by study

Evidence that men outnumber women in ICUs has raised concern about sex-related disparities in ICU access.  Although we have demonstrated that in most age groups the excess of men populating ICUs is a consequence of their higher rates of critical illness, in the oldest cohorts men still outnumber women despite the fact that in those age groups women outnumber men in society, even accounting for differing rates of critical illness.  The reasons for this are unknown but could relate to sex-based differences in decisions for aggressive care by patients, their substitute decision-makers, or health care providers who serve as the gatekeepers for ICU admission.

Brief overview of the methodology

We hypothesized that rather than representing an effect of sex per se, the deficit of older women in ICUs is a consequence of a lower willingness for aggressive ICU care among those with less social support by virtue of not having a spouse/partner, particularly due to widowhood.  We evaluated this in a retrospective, observational, population-based study using three separate Canadian datasets of individuals ≥65 years old, from 2004-2012: all Manitoba residents, and nursing home residents and homecare patients throughout Canada.  Our outcome was admission to an ICU.  Our analysis used multivariable regression analysis to explore whether the male predominance of ICU admission was attenuated when marital situation and other covariates were included in the models.  In the Manitoba cohort we performed a secondary analysis assessing the sex ratio of ICU admission among the subset of men and women who had previously been admitted to an ICU.

Potential impact of findings on clinical practice/patient care and how this impact might be measured

  • Our findings indicate that sex differences in social supports do not underlie the male predominance of ICU admission.
  • By also providing evidence against the presence of a sex bias on the part of ICU gatekeepers, our findings instead suggest that older women are, in general, less desirous than older men in undergoing aggressive care provided in ICUs.

Remaining knowledge/research gaps

  • Confirming and extending our interpretation will require interviewing older people and their substitute decision-makers in a variety of settings to gain a better understanding of preference differences between men and women, and the reasons behind them.