Reducing the Length of Stay: How it Affects Patients and Their Families

Key Implications for Decision Makers

Reform of healthcare services by the Montreal-Centre Health and Social Services Board was introduced to reduce the length of short-term stays in hospital and expand the use of homecare services. This study covers the effects of these measures on the health and psychological well being of patients and their families.

The study, which compares the status of two groups of patients in 1996 and 1999, reveals the following.

  • Patients hospitalized with medical conditions (pneumonia, chronic obstructive lung disease, cellulitis) clearly did not experience a reduction in length of hospital stay.

  • Patients with medical conditions did not recover as well in 1999 as in 1996 in terms of physical health, but did recover better in terms of mental health. They also made greater use of homecare services.

  • Patients hospitalized with surgical conditions (hip fracture, hip and knee arthroplasty) posted a reduction in average length of hospital stay, but this reduction was not statistically significant and might be linked to the presence of a few cases with very long hospital stays.

  • A large drop in medical consultations, similar use of other institutional services, and a small reduction in services outside an institution (support services, health care), accompanied this potential reduction. Only the decline in medical consultations was statistically significant.

  • At a time when patients with heavy demands are increasingly kept at home, planning of any measure designed to reduce length of hospital stays should factor in the severity of patients' clinical condition and the effect on their health, which is already fragile.

  • The well being of caregivers did not change between 1996 and 1999.

  • Overall, there was no systematic negative effect on the well being of patients or caregivers.

Executive Summary


One objective of reforming Montreal's healthcare system was to reduce the length of hospital stays by substituting homecare services for hospital services. Given the major measures introduced by this reform, the Director of Public Health commissioned our research team to assess the benefits of the ambulatory shift for the Montreal region. The study was designed to measure the effect of the reduction in length of stay associated with the shift on the health and psychological well-being of patients and their families and on the cost for the individuals involved.

A survey of the literature on substitution of homecare services for hospital services identified 14 projects, four of which met criteria of internal validity. This survey reveals that homecare services have an effect similar to hospital services on the health of patients or caregivers. A single study identified a positive, although small, effect on the quality of life of patients who have suffered a hip fracture. The three studies that assessed the social costs in turn produced contradictory results. The studies available do not provide valid, accurate data on the effect of substituting homecare services for hospital services. It therefore is impossible to determine with certainty the nature of these repercussions on patient and caregiver health and on costs.

Our project differs from earlier studies on the following points:

  • Patients were identified at the start of the care episode.
  • Two groups of patients were compared at two times. Changes in the length of hospital stay and in service use separated the two comparison times.

The study had patients who were more suited to having homecare substituted for hospital care.


The team studied five target conditions to measure the effects of a reduction in length of hospital stays. The study shows that the length of stay did not change for medical conditions (pneumonia, chronic obstructive lung disease, cellulitis) but may have decreased for surgical cases (hip fracture, hip and knee arthroplasty). In these latter cases, the decrease is not statistically significant and might be linked to the presence of a few cases with very long stays. This reduction was accompanied by a decline in use of medical consultations in an institution with no significant effect on patients' physical or mental health.

Length of stay in medical cases did not decline during the period. We also found in medical cases an increase in the use of institutional services (especially laboratory tests) and services outside an institution.

The results for physical health suggest that medical patients did not recover as well in 1999 as in 1996, but those for mental health suggest that they did recover better in 1999. The effects observed are still minor and difficult to interpret in light of the overall effect on the patients involved.

The results for health and activities of caregivers suggest that the changes observed had no significant effect on them.

Overall, the shift to homecare had no systematic negative effect on the well being of patients or caregivers.

The interpretation of these results must consider the fact that the rate of refusal to participate in the study was much larger in 1999 than in 1996 and that the patients who declined to take part might well represent


This is a comparative study of two groups of patients hospitalized for one of the five target conditions for the ambulatory shift (one group in 1996 and the other in 1999). The patients recruited and their caregivers answered six questionnaires spread over more than a month. Patients also granted the research team access to their medical files in all institutions that provided them with services during this same period. These sources were used to measure the health of patients and caregivers, and utilization of services by patients.

The cases were identified in four Montreal hospitals. The patients involved were those admitted with one of the five target conditions: pneumonia, chronic obstructive lung disease, cellulitis, hip fracture, and hip and knee arthroplasty. They were identified in advance by having the greatest potential for reduction in length of hospital stay and because development of homecare services was planned for these conditions. After determining the eligibility of potential candidates, the research nurse met with the patients to invite them to take part in the project.

Data gathering was conducted for the first group in 1996 (time 1) and for the second group in 1999 (time 2). The information on patients was obtained from records in the hospital, rehabilitation hospital or CLSC and from interviews conducted in hospital and by telephone. The following information was gathered:

  • socio-demographic information
  • primary diagnosis and secondary diagnoses
  • the presence of co-existing or additional diseases (comorbidity)
  • length of stay in hospital
  • use of homecare services
  • all services and treatments received regardless of origin (including those provided by natural caregivers)
  • quality of life
  • financial impact
  • psychosocial aspects of health

The information on caregivers was gathered through questionnaires on the nature of the care provided to patients and the impact of that care on their other activities.