Health System Efficiency: Saving Money Can Save Lives Too

It turns out that having a more efficient health care system is not just about sustainability or bean counting – it also can save lives. 

A new study by the Canadian Institute for Health Information* tries to measure the efficiency of Canadian health systems across Canadian regions. The core project team was made up of Sara Allin, Diana Ridgeway, Li Wang, Erin Graves, Jean Harvey and Jeremy Veillard, in ongoing consultation with Michel Grignon, Director, Centre for Health Economics and Policy Analysis, and Associate Professor, McMaster University.

The results find that health regions across Canada vary substantially in their ability to reduce treatable causes of death even after controlling for environmental factors and socio-economic characteristics such as per capita income, proportion of population with post-secondary education and the proportion foreign-born.  If all health regions were to maximize their efficiency, the freeing up of additional resources would reduce deaths due to treatable causes by 18-35 percent.  This translates into the potential prevention of 12,600 to 24,500 premature deaths on Canada per year – without incurring additional costs.

The study defines a health system as a jurisdiction that includes health activities under the mandate of provincial and territorial ministries of health.  It defines the output of a health system as the reduction in the potential years of life lost (PYLL) from treatable causes of death. The resource measure is a monetary one – per capita spending on the main health sectors at the health region level. Efficiency is therefore the effectiveness with which health systems use resources to reduce the PYLL due to treatable causes of death.  In other words, how effectively is spending converted into the output of reducing the potential years of life lost. 

Due to data availability, the study includes 89 out of 101 Canadian health regions and then only for the 10 provinces. Efficiency at the health region level is estimated through DEA (Data Envelopment Analysis) by looking at the gap between a region’s actual outputs and the efficiency frontier based on a performance comparison.  A regression analysis then looked at some of the factors associated with this efficiency gap.

In constructing the regression estimates, external factors outside the control of the health system were accounted for such as: proportions of recent immigrants, non-aboriginal people and people with a post-secondary degree.  As well, an additional set of contextual, clinical factors and operational were included – the age and gender structure of the population, measures of socio-economic status and inequality, prevalence of smoking and obesity, chronic conditions and rates of hospital readmission, organizational and managerial variables such as how much care was provided in higher cost settings.  In total, the included factors account for about half of the variation in the efficiency estimates with the remaining unexplained variation attributable to clinical practice variation and population and patient characteristics that could not be measured.

The report is well worth a read.  It demonstrates that saving money in our health care system is not simply a mantra of bean-counter accountants, health administrators and economists but an exercise in trying to help improve outcomes through finding additional resources within the system.  There is room for improvement for all regions.  This will be the way of the future given that new money to buy “transformative change” is not on the horizon.  One thing I would have liked to see and hope a future study addresses is what the efficiency ranking of the individual regions is.  It would be interesting to see which provincial health care systems have been doing the best on the efficiency front and therefore can provide examples to the others.

 

*I should note here that I am a member of the CIHI NHEX Advisory Panel.

 

Comments

[Nick Rowe]

Those strike me as being pretty big numbers, Livio. Can't help but wonder though if there weren't omitted variables, that skewed the results. OK, they tried to adjust for as much as possible, but there must be things they didn't have data on, or we don't know about, that make healthcare spending look more efficient in some places than it really is.

[Jciconsult]

One of the issues is the analysis of value proposition of health care. As I am unfortunately no longer consuming the option value of the health care but rather an extensive user of its services I can speak from personal experience that some of the efficiency gains have clearly come at the expense of and stress to patient and caregivers. Personal expenses and costs as well as the value of increased stress and time costs for the patient are not sufficiently recognized in these studies.

[Shangwen]

@Nick: If by big numbers you're referring to the 18-35% potential reduction in preventable deaths, the analysis notes that most of these could be gained through improvements in primary care, population health, and reduction in preventable hospital readmissions. The problem is that, while true, many of these changes depend on voluntary behavioral changes in the patient population rather than process or technical improvements in the healthcare system itself.

[Nathanael]

I'm not entirely surprised, given that there's some evidence that health care, a field where terrible information quality is standard, is prone to *completely useless* spending and even *counterproductive* spending.

I wouldn't be surprised if the "inefficient" places have policies which are not merely inefficent, but just plain bad -- here in the US, many hospital administrations have set up perverse incentives set up to kick people out of hospital early, even when it means that they need to be readmitted at great expense later.

In short, I'm going to guess that "efficiency" is the wrong way to think about it: provide better health care and your efficiency will probably go up. Try for "efficiency" and your efficiency will probably go down. (As Jciconsult says, "some of the efficiency gains have clearly come at the expense of and stress to patient and caregivers.")