Skip Ribbon Commands
Skip to main content
We are currently experiencing technical issues with registration and password reset for users with outlook.com, hotmail.com and live.com email accounts. We are working to resolve as quickly as possible and apologize for any inconvenience.
Research in action: Dr. Laura Rosella
Estimating diabetes population risk to inform community reduction strategies
​​
​​Hear Dr. Rosella discuss​ DPoRT.

The rate of people with Type 2 diabetes is growing at an alarming pace in Canada. According to the Canadian Diabetes Association, 2.7 million Canadians have diabetes and this is expected to rise to 4.2 million by 2020. PHO’s Dr. Laura Rosella is investigating ways to address this critical issue from a broad population health perspective. Rosella has developed tools that calculate the risk of diabetes in a given population, to help decision-makers and program planners create and implement diabetes reduction strategies in their communities.

Rosella’s current research builds on the Diabetes Population Risk Tool (DPoRT), a tool she developed with colleagues while doing her PhD. It estimates future diabetes risk based on routinely collected population data. DPoRT tries to do for a whole population what physicians do in assessing their patients’ risks: how to consider behaviour, risk factors, and broader context to determine likelihood of developing diabetes.

“A physician comes into a room, and tells the patient he or she has certain risk factors, and then proceeds to put those into a risk calculator. With that result, the physician can plan a course of treatment, such as prescribing a statin,” says Rosella. “But from the population health perspective, we didn’t have useful tools like that. At the community level, a medical officer of health or others who make decisions about programs didn’t have those same resources to ask, ‘What is the risk profile for my community and how can I take that information, estimate future burden, and then decide how to tackle it?’”  

image from DPoRT 
Absolute risk reduction in 10-year diabetes risk) as a result of (A) 5% weight reduction and
(B) a lifestyle intervention for different target populations
 
 (Rosella L.C., Lebenbaum, M. Estimating Population Benefits Using a Risk Tool: An Example of Diabetes in Canada. Annals of Epidemiology 2012; 22(9):667.)  
  
Rosella points out that decision-makers typically go to the literature or do an evidence synthesis to inform their local diabetes prevention strategies. DPoRT aims to do more: “It is about fitting in a different way of thinking — not just what intervention works, but how many people are at risk? Who are they and why is this relevant to my community?” says Rosella.
 
 
 

 

As Rosella notes, the interventions found in the literature rarely consider the distribution of the population. Rather, they focus on the interventions themselves — what works and what does not. DPoRT assists in determining which interventions will have the greatest impact to their communities.

DPoRT also allows decision-makers to evaluate the impact of different scenarios. “What if we targeted this population with this level of intervention, which we know about from the literature? What impact will this have versus another approach?  There are numerous diabetes strategies, so we run numbers in terms of expected interventions and cases, thinking about how many people need to be screened or targeted to get impact,” says Rosella. “The decision-makers provide the critical local context and input to determine what interventions are feasible and might succeed.”

DPoRT was developed and validated with local public health partners in Ontario. Peel Public Health participated in a very successful pilot, and Peel’s epidemiologists are now using it to inform delivery of services such as prevention programs.  

 
“In planning programs and policies aimed at reducing the prevalence of chronic diseases in the total population, public health decision-makers have hitherto had to rely on data generated in clinical settings, on high-risk populations,” said Dr. David Mowat, medical officer of health, Peel Region. “The advent of DPoRT enables us to understand risk within the entire population and to plan the most effective approaches to prevention for everyone.”
 

 

 

 

Rosella has received a Canadian Institutes of Health Research grant to further disseminate and implement the tool in health units in Ontario and in other jurisdictions in Canada. Her research is now focusing on the implementation of DPoRT. To do so, she’s been consulting with decision-makers in multiple Canadian provinces and jurisdictions and is continuing to adapt the tool based on user feedback.

User feedback continues to drive enhancements to DPoRT. For example, the original design predicted incidence (i.e. new cases in a given time period). However, decision-makers also wanted to understand the financial impact of those potential new cases. Rosella is now integrating estimates of attributable costs into the tools, adding new capabilities that enhance the measurement of population risk. “In this climate, this fiscal component needs to be in every conversation,” says Rosella.  

image from DPoRT 
Ten-year estimate of diabetes risk, % (diamonds) and number of incident cases in thousands (bars) according to baseline HbA1c and Fasting Plasma Glucose (FPG), in the Canadian Health Measures Survey collected 2007 to 2011
 
(Rosella L.C. 1, Rivera L, Lebenbaum M. Implications of Clinical Targets for Diabetes Prevention. Annals of Epidemiology 2013; 23(9):587-588.)
 
Decision-makers also wanted to know prevalence. “In terms of burden, it’s not just how many people are currently living with diabetes, but what new cases — the focus for prevention — are predicted. Decision-makers need the full picture — existing plus new cases — to get a total number and understand the burden,” says Rosella. “DPoRT puts this information in the hands of the people who need it to plan, prioritize, and develop a course of action.”
 
Rosella’s published research on DPoRT and its use can be found at:
 
Risk distribution and its influence on the population targets for diabetes prevention. Rosella L, Lenenbaum M, Li Y, Wang J, Manuel D. Preventative Medicine. October 2013. http://www.sciencedirect.com/science/article/pii/S0091743513003770
 
The role of ethnicity in predicting diabetes risk at the population level. Rosella LC, Mustard C, Stukel TA, Corey P, Hux J, Roos L, Manuel DG. Ethn Health. January 2012. http://www.ncbi.nlm.nih.gov/pubmed/22292745
 
A population-based risk algorithm for the development of diabetes: development and validation of the Diabetes Population Risk Tool (DPoRT). Rosella LC, Manuel DG, Burchill C, Stukel TA; PHIAT-DM team. J Epidemiol Community Health. July 2011. http://www.ncbi.nlm.nih.gov/pubmed/20515896
 
How many Canadians will be diagnosed with diabetes between 2007 and 2017? Assessing population risk. ICES Investigative Report. Manuel DG, Rosella LCA, Tuna M, Bennett C. Institute for Clinical Evaluative Sciences. 2010. http://www.ices.on.ca/file/Diabetes%20Risks%20June%2016%202010.pdf
 
​​
Page last reviewed:  
Page last updated: 12/01/2017 10:21 AM
Uncontrolled print copy. Valid only on day of Print: [date]
Page updated on [date/time] 12/01/2017 10:21 AM
© , Ontario Agency for Health Protection and Promotion