Hollander Logo
National Evaluation Logo
Centre on Aging Logo

Goals and outcomes
Full texts, fact sheets, and summaries
Media releases, news coverage, and announcements
#300 - 895 Fort Street
Victoria, BC, Canada
V8W 1H7
Tel: (250) 384-2776

Fax: (250) 389-0105

Substudy 2 Fact Sheet

Care Trajectories: The Natural History of Clients Moving Through the Continuing Care System (NA101-02)

Sponsor Organization:
Hollander Analytical Services Ltd.


This project is one of fifteen sub-studies of the National Evaluation of the Cost-Effectiveness of Home Care Project (NA101). As there has been very little research on how clients move through the continuing care system, sub-study 2 sought to document patterns of client movement and to determine if care patterns are predictable. The University of British Columbia maintains a linkable longitudinal database of B.C. data for hospitals, physicians, drugs, continuing care, mental health, and some aspects of vital statistics. This database was used to track the movement of clients over a 10-year period through the British Columbia continuing care system.

Following an extensive process of cleaning, each data set was analyzed for potential problems by inspecting the ranges of each variable to be used in the analysis based on a "Care Advice" form used to authorize access to all community and residential long term care services in B.C. Researchers found that the database for direct care services (nursing, PT/OT) was generally reliable but that it had the most anomalies. A total sample number of 6,384 clients was used for this study, which has three major sections. The first sets the context. The second describes the most common patterns of movement through the British Columbia continuing care system for the 1987/88 to 1996/97 fiscal years, for a cohort of clients admitted in the 1987/88 fiscal year. The third compares 15 different statistical models to predict how people move through the system of care.

Key Findings:

The project leaders identified the following outcomes:

  • Contrary to the assumption that there would be a small number of common patterns of movement, there was a wide variety of care trajectories, none with a large percentage of the clients.
  • Within the ten years of data available, the most common pattern was for clients to enter the system at a given level and type of care and die without any change in the level or type of care.
  • Markovian modeling provided a high degree of predictability of future resource utilization.
  • Given the lack of common care trajectories, costing was not undertaken.


Researchers initially assumed there would be four to six common trajectories or patterns of movement which would account for a significant proportion of all patterns. Predictable care patterns have implications for clinicians who could then prepare in advance for possible changes in care status. However, the data suggest that these assumptions cannot be supported empirically. The authors suggest that the findings represent significant new information about the natural history of client movement within the continuing care system, and that their analysis has implications, as yet unclear, for both clinical practice and resource planning.

Evaluation Methodology:

As this project was analytic/evaluative in nature, please see the "Activities" section.

HTF Contribution to the Project:

$1,505,000 (divided among the 15 sub-studies)

Language of Report: