Interview: Measuring Ontario’s performance on the HIV Care Cascade

Published 31, Aug, 2017
Author // Bob Leahy - Publisher

The Ontario HIV Treatment Network’s epidemiology unit has just released a report on how Ontario people living with HIV are doing on engagement in care, on treatment and viral suppression respectively. Bob Leahy interviews James Wilton and Abigail Kroch .

Interview: Measuring Ontario’s performance on the HIV Care Cascade

Read the full report “HIV care cascade in Ontario: Linkage to care, in care, on antiretroviral treatment, and virally suppressed” 

Bob Leahy: Thanks for talking to PositiveLite.com. Tell us about the good news that’s contained in the report.

OHTN: There’s lots of good news in the report!

First of all, the report represents the first large knowledge exchange product released by the Ontario HIV Surveillance and Epidemiology Initiative (OHESI). This is a new collaboration responsible for HIV surveillance in the province and is made up of five organizations: Public Health Ontario, the AIDS Bureau of the Ontario Ministry of Long-term Health and Care, the Public Health Agency of Canada and the Ontario HIV Treatment Network.

Second, the report presents the first data to emerge from the Ontario HIV Laboratory Cohort - a new data source developed using diagnostic and viral load testing databases at Public Health Ontario. The cohort fills an important gap in our understanding of Ontario’s HIV cascade, as previous cascade studies have relied on smaller sources, such as the OHTN Cohort Study (OCS), which captures about 25-30% of people engaged in care and may not be representative of all diagnosed people living with HIV in the province.

Finally, the data summarized in this report shows that engagement in Ontario’s HIV cascade has improved over time. In particular, between 2000 and 2015 the proportion of people with diagnosed HIV who are virally suppressed doubled from 41% to 80%.

In your prologue you say that the data supports Ontario’s ongoing ability to assess the impact of policy directions and program initiatives. But service and program delivery isn’t homogenous – it’s largely targeted towards specific populations. How does the report address disparities in progress in those specific populations so that, for example the ACB (African, Caribbean and Black) community can relate to it? Or agencies with an MSM (men who have sex with men) focus like ACT? Don’t we need information similar to the Ontario-wide data but segmented by factors like age, sex, population and region, for example, to be truly effective?

I agree that we need cascade data broken down by age, sex, population, region, and in other ways that are meaningful for service providers. This work is ongoing, so stay tuned for the release of new knowledge exchange products containing this information, including a new OHESI website. In the meantime, a lot of work has gone into analyzing cascade engagement among participants in the OCS. This work has identified populations that are more likely to encounter difficulties staying engaged in the care cascade.

Will we ever be in a place where, for example, an AIDS Service Organization (ASO) could see how people living with HIV in their service area are collectively placed on the cascade?

As part of our ongoing work, we are looking at cascade engagement broken down by the catchment areas of the 36 Public Health Units. While these areas may not align perfectly with specific ASOs, we hope it will be useful to these organizations.

It seems to me that the report’s usefulness would be enhanced by comparisons with the progress of other jurisdictions, including Canada as a whole, other provinces and globally. That’s not part of this report. Anything planned to address this?

Comparisons to other jurisdictions need to be done with a lot of caution. While the cascade is simple in concept, its measurement is not easy. Each jurisdiction has its own unique set of measurement challenges, which are often related to available data sources and their inherent limitations. These challenges can limit the usefulness of comparing between jurisdictions, as any observed differences may be due to methodological issues rather than reality. For these reasons, we have not included comparisons to other cascades in this report.

Overall, we feel that the most informative use of cascade data is to make comparisons within jurisdictions – for example, trends over time and by population – as these are less likely to be influenced by methodological challenges and limitations.

You have said “A full understanding of how close Ontario is to the combined 90-90-90 requires a reliable estimate of the first 90. That’s true of every jurisdiction, isn’t it?

It is true that estimating the first 90 (the percent of all people living with HIV who are diagnosed) is difficult for every jurisdiction. Calculating this percent requires an estimate of the total number of people living with HIV (denominator), as well as the total number of diagnosed people living with HIV (numerator). As it is not possible to measure the denominator directly, it is instead estimated through mathematical modelling and there are unavoidable uncertainties in these models. That is why the numbers produced by these models are normally accompanied by a range of possible values. In addition, the numerator can be difficult to measure directly and is often estimated via modeling.

Can we look at the report in the context of 90-90-90? What’s your estimate of where we in Ontario stand in relation to that measure right now?  I think elsewhere it’s reported as 81-81-94 in Ontario? How are we doing in relation to providing reliable 90-90-90 data for Canada, again recognizing the first 90 is always problematic to measure.  I understand you are working with PHAC and the other provinces on that? What’s that project’s current status?

In the cascade report, we only present the second and third 90 because that is what is measureable in the cohort. However, it is important to view the three UNAIDS targets as a combined target and not in isolation of each other. The overall goal of the UNAIDS strategy is that all three targets be met simultaneously by 2020 – meaning that 90% of all people living with HIV are diagnosed, 81% of all people living with HIV are on treatment, and 73% of all people living with HIV are virally suppressed. So you could actually look at the UNAIDS target as being 90-81-73 instead of 90-90-90!

Viewing the UNAIDS target in this way emphasizes the importance of an accurate and reliable measure of the first 90, as the total number of people living with HIV is used in the denominator for all three of the 90-81-73 targets. Therefore, an unreliable estimate of this number can introduce uncertainty throughout the other measurements.

We are currently working to develop a more accurate understanding of the first 90 in Ontario. Historically, PHAC has calculated this estimate for Ontario and it has normally been in the region of 80%. We are collaborating with mathematical modelers in Ontario and at PHAC to take advantage of new data from the Ontario HIV Laboratory Cohort and leverage additional resources in Ontario, including the OCS and the Institute of Clinical Evaluative Sciences (ICES), to improve upon PHAC’s model. This work is ongoing and we have to have some preliminary estimates by early next year.

Does Ontario face special difficulties in arriving at that (first 90) estimate?

Ontario does face special difficulties estimating the first 90.

The total number of individuals ever diagnosed in a jurisdiction is a key measurement that is used to estimate the first 90. However, there are complications in measuring this number in Ontario because of the large number of individuals who are diagnosed through non-nominal types of testing (i.e. testing that does not involve providing a name, such as anonymous testing). Since many individuals diagnosed non-nominally also receive a nominal HIV test (i.e. testing that does involve providing a name) when entering care – there is double counting of these individuals. This double counting can lead to inflated estimates of the total number of people living with HIV in the province.

I see, I want to look at data not in this report but detailed elsewhere. The number of new HIV diagnoses  per annum in Ontario has risen in the last three years (2013 - 797, 2014 – 828, 2015 - 839, 2016 - 881; that's a 10.5 % increase over four years.). That’s not good. Are we able to say how much of that is due to an increase in the number of HIV tests? What other factors might be at play, do you think?

It is difficult to know whether the increase in new HIV diagnoses is due to an increase in HIV infections or other factors. As you mention, increased HIV testing could be playing a role – between 2013 and 2015 the number of HIV tests in Ontario increased by 10%.

The trend could also be influenced by migration to Ontario, both from within Canada and abroad. This is partly because the number of new HIV diagnoses reported for Ontario each year includes people who were diagnosed outside the province, moved to Ontario, and tested again. 

The population in Ontario has also been increasing overall. When this population increase is taken into account, the increase in diagnosis rate is more moderate (the diagnosis rate was 5.9 diagnoses per 100,000 of the Ontario population in 2013 and 6.3 per 100,000 in 2016 - a 7% increase). 

Despite the possible increasing trend in recent years, the number of diagnoses in 2016 was still lower than in 2011 and earlier when there were approximately 1,000 diagnoses or more each year (the number of diagnoses reached a high of about 2,000 in 1990).

How does that trend – annual rates of new diagnoses - compare with the rest of Canada in the same period? How about B.C.?

The most recent national surveillance data published by the Public Health Agency of Canada includes information up to 2015. Similar to Ontario, the number of diagnoses nationally has decreased since the mid-2000s but has been relatively stable since 2012. According to the latest data, Canada’s national HIV diagnosis rate was 5.8 per 100,000 in 2015 - slightly lower than the rate of 6.1 in Ontario for that year. Compared to other provinces, the Ontario rate in 2015 was lower than Saskatchewan (14.4) and Manitoba (8.1) but slightly higher than BC (5.1), Alberta (5.6), and Quebec (5.7).

About the interviewees: James is an Epidemiologist in the Applied Epidemiology Unit (AEU) at the Ontario HIV Treatment Network (OHTN) and Abigail is the Director of the AEU. Both are part of the Ontario HIV Epidemiology and Surveillance Initiative (OHESI) – a collaboration involving Public Health Ontario, the AIDS Bureau of the Ontario Ministry of Health and Long-Term care, the Public Health Agency of Canada, and the OHTN AEU.

 

About the Author

Bob Leahy - Publisher

Bob Leahy - Publisher

Award-winning blogger Bob Leahy first made his social media mark a decade ago on LiveJournal.com where there are still to this day almost 3,000 entries of his available to be read. He was a featured blogger on Ontario’s HIVStigma.com campaign, along with PositiveLite.com founder Brian Finch. He joined PositiveLite.com at its inception in 2009 and became it's Editor a year later.

Born in the UK, Bob’s background is in corporate banking, which he gladly left in 1994, after being diagnosed with HIV the previous year.  He has chaired the board of PARN (Peterborough AIDS Resource Network) and has been an executive board member of both the Ontario HIV Treatment Network (OHTN) and the Canadian AIDS Society (CAS).  He was inducted in to the Ontario AIDS Network’s Honour Roll in 2005.  Bob is currently a member of Ontario’s GMSH (Gay Men’s Sexual Health Alliance). He also writes for TheBody.com.

In 2012, Bob was honoured with the Queen Elizabeth II Diamond Jubilee medal for his work and commitment to HIV/AIDS in Canada.

Bob continues to write for this site while in the Positivelite.Com editor’s seat, with a particular interest  in HIV prevention, theatre and the arts in general. He is accredited media for a number of Toronto theatres. He lives in Warkworth, Ontario with his partner of thirty-two years and three dogs.