Hospital Standardized Mortality Ratio

On December 11, 2008, the Canadian Institute for Health Information (CIHI) publicly released Hospital Standardized Mortality Ratio (HSMR) results for eligible acute care hospitals and regions across Canada (excluding Quebec).

What is HSMR?

The Canadian Institute for Health Information (CIHI) has calculated the HSMR for eligible acute care hospitals and regions in Canada (excluding Quebec). HSMR is an overall quality indicator and measurement tool that compares a hospital's mortality rate with the overall average rate. It has been used by many hospitals worldwide to assess and analyze mortality rates and to identify areas for improvement.
More information is available on the CIHI web site.

How is HSMR calculated?

The HSMR is calculated as a ratio of:
The number of (actual) observed deaths x 100?The number of expected deaths
A ratio that is equal to 100 suggests that there is no difference between the hospital's mortality rate and the average national rate.

A ratio greater than 100 suggests that the hospital's mortality rate is higher than the average national rate.

A ratio less than 100 suggests that hospital's mortality rate is lower than the average national rate.

The HSMR is based on diagnosis groups that account for 80% of all deaths in acute care hospitals, and is adjusted for factors such as: diagnosis group, age, sex, length of stay, admission category, comorbidities, and transfers.

This year, CIHI will be reporting if the HSMR result is statistically different from the baseline of 2004-05 baseline of 100 (p<0.05).

Why was Quinte Health's result higher than 100 in previous years?

We were concerned about the increase in the Hospital Standardized Mortality Rate at Quinte Health. In response, we asked a physician to conduct a chart review to determine what could be causing the increase so that we could address any possible issues. The physician determined that there are no concerns from a quality of care perspective. The overall number of deaths at Quinte Health is not increasing.

We have determined that the increase is primarily caused by three factors:

1. There were a number of coding issues related to identifying patients as "palliative". These issues have now been resolved. (Palliative care is defined as a patient who has been diagnosed with a terminal illness and is receiving only comfort care, not aggressive treatment.)
2. We have experienced an increase in the number of alternate level of care patients who are staying on medical units much longer, rather than moving to Complex Continuing Care or to long-term care facilities. Patients on CCC units are not counted in HSMR rates.
3. There is a general increase in the number of palliative patients at Quinte Health because of the aging population.

In fact, when you exclude palliative cases, the Hospital Standardized Mortality Rate at Quinte Health drops to 89, well below the national baseline of 100.

In general, HSMR scores can be influenced by a number of factors, some of which are beyond the hospital's control. These factors include:?

• Infrastructure: the availability of physicians, long-term care homes, hospice, screening facilities, and public health programs may influence how and where patients are treated, and how and where patients die.
• Population characteristics: the incidence of specific illnesses in the population a hospital serves, as well as the communities in which patients live, can influence their health status and health outcomes.

Quinte Health will continue to review the CIHI report and work with health professionals to identify and implement system-level improvements - as we do when we receive reports on other safety and quality indicators, such as those in the OHA's Hospital Report, or in the course of our standard quality improvement programs.

Why is the HSMR an important measure?

HSMR has been used by hospitals in several countries to assess in-hospital mortality rates and to help organizations identify areas for improvement. As a quality indicator and a measure to help hospitals and health professionals follow trends in their hospital mortality rate, HSMR can be used as yet another tool to help improve quality of care and patient safety over time.

Can the results be used as a guide to choosing hospitals?

No. The HSMR results should not be used as a guide to choosing which hospital to seek care at. HSMR looks at mortality rates only and represents just a snapshot of one area of a hospital's performance.

A higher than average HSMR result does not necessarily mean that a hospital is "unsafe;" a lower than average HSMR does not necessarily mean a hospital is "safe." It is vital that HSMR results be viewed in the context of other performance indicators.

Hospital care is complicated and depends on many factors, not all of which are reflected or accounted for by HSMR. That is why many indicators must be examined in order to get a sense of how hospitals are performing - where they excel and where improvements could be made. It is important to look at all of these indicators in combination. To judge performance on only one indicator would be misleading.

What conclusions can patients draw from the HSMR indicator?

HSMR is another, helpful measure for us to ensure your care gets even better over time. It provides a snapshot or screening tool of what has happened with respect to one aspect of care - mortality - in a hospital/LHIN/Regional Health Authority over a certain period of time.

We must also remember that not all deaths are preventable. The purpose of the HSMR is to provide useful information, research and practice to support hospitals in their ongoing efforts toward improving hospital performance.

The analysis of HSMR, when coupled with system-level improvements based on that analysis, have led to significant performance and quality improvements in other jurisdictions.

We will continue to work with our health professionals to make those kinds of improvements.

How will hospitals and doctors use HSMR data to improve patient safety?

As a relatively new tool in Canada, HSMR is another measure we now have at our disposal to help improve patient safety and care. Like other patient safety indicators, it will be used to improve performance and safety over time.

Ontario hospitals are beginning to use the HSMR for internal benchmarking purposes. Beginning with the reporting that started last year, this year's reporting helps to show hospitals how their HSMR has changed, where they have made progress and where they can continue to improve.

This is similar to hospitals in other jurisdictions, including the United Kingdom, the United States, Sweden and Holland, where HSMR data has been used to monitor performance and reduce mortality rates.

For example, by tracking HSMR and implementing a range of improvements as a result of what they learned, the UK's Walsall hospital was able to reduce mortality by 40% in only four years through the establishment of seven clinical governance groups and the implementation of a range of improvements.

Through HSMR, hospitals have been able to learn more about the tool, more effectively examine their results, identify areas for improvement, implement strategies to lower mortality, and track results over time.