Prevention in Focus

Fall 2012 

Views from the front lines: Communicating risk

Interviews by James Wilton

We asked three people about the challenges of explaining risk to their clients and patients.

  • Jody Jollimore—Program Manager, Health Initiative for Men, Vancouver
  • Riyas Fadel—M.A. candidate, Sexology, Université du Québec à Montréal
  • Mona Loutfy—Infectious Diseases Specialist, Women's College Hospital; Research Director, Maple Leaf Medical Clinic; Associate professor, University of Toronto

Jody Jollimore

What are the major challenges in communicating risk?

The overarching challenge is how to communicate risk without confusing or discouraging people.

One challenge is trying to communicate risk without putting a number on it. The public often wants risk to be communicated as a percentage (such as 2%) or a ratio (such as a 1 in 50 probability of HIV transmission). However, it’s very difficult to communicate risk as a number and these ratios can easily be misunderstood. For example, a person could understand this ratio to mean that they can have anal sex without a condom 50 times before they get HIV when, in fact, they could be the guy who does it once and gets HIV or the guy who does it 600 times and doesn’t get HIV.

Another challenge is how to communicate the many variables that need to be considered, such as the viral load of the HIV-positive partner, whether either partner has sexually transmitted infections (STIs), the window period, how often someone is getting tested, and how much one partner trusts the other(s). All these things can affect transmission significantly but are not easily measured or quantified. The information is complicated and can discourage someone from trying to reduce their risk.

What are the limitations of current approaches to communicating risk?

Most current models for communicating risk group activities into “low-risk” and “high-risk” categories. For example, unprotected oral sex and protected anal sex are normally considered low-risk and all unprotected anal sex is assigned to the high-risk category. However, at the Health Initiative for Men, we don’t think this approach is nuanced enough to reflect the recent research or the complex and exciting sex lives of gay men. Gay guys know that risk is more than just “low” or “high” and that other variables can influence their risk of HIV transmission. We feel that this approach doesn’t give guys a lot of options to choose from to reduce their risk. Not all guys are able, or want, to reduce their risk by engaging in “low-risk” activities and not all unprotected anal sex is the same.  

What are your solutions?

At HIM we have developed a risk communication model that includes a risk calculator—it can be found on our “Do the Math: Calculate Your Risk” website. We opted to expand the “low-risk” and “high-risk” model to include more categories, including “no or very low,” “low,” “moderate,” “high” and “very high” risk. The model uses several factors to assess risk, including the type of sex they are having (oral or anal), the position they are assuming (top or bottom), the HIV status of both partners (poz, neg or unknown) and whether condoms are used. Using the statistics from the ManCount survey, we felt that these were the variables that gay men were most commonly using to make decisions and also those that people know the most about.

This approach gives gay men more options to reduce their risk. Instead of simply telling people that all unprotected anal sex is high-risk, we opted to say that it is more risky than oral sex and protected anal sex, but less risky if you are a top or with a same-status partner.

Our model doesn’t provide numbers, but it is based on absolute risk percentages calculated in a mathematical modelling study titled “Reducing the risk of sexual HIV transmission: quantifying the per-act risk for HIV on the basis of choice of partner, sex act, and condom use” (Varghese et al 2001), published in the journal Sexually Transmitted Diseases. The numbers in this study informed our “Do the Math” model and were used to assign different activities to a risk category.

We didn’t provide the numbers from this study in our model because we don’t feel that communicating risk as percentages and ratios is the most effective way of translating probability information to the average gay guy. Even those of us who are well versed in statistics can struggle to understand them, specifically the difference between absolute risk and relative risk and how the two can interact. You can look at the numbers but the more you look into all the factors used to calculate them, the more you realize how far from reality they are and how they have little meaning in the “real world.”

What are the limitations of your approach?

The main weakness of our model is its simplicity. Although it is more nuanced than other models, it doesn’t consider factors such as antiretroviral treatment, acute HIV infection, STIs, dates of testing, and trust. These factors were difficult to integrate into the calculator for various reasons.

In the case of viral load, there are still gaps in the research, particularly among gay men, and the messages are different depending on the context (for example for poz guys vs. neg guys, gay guys who go the bathhouse vs. guys in monogamous serodiscordant relationships). Since these factors need a more detailed explanation and there is no simple message that applies to everyone, they are discussed at length in the text on the website but were not integrated into our risk calculator. 

In the end, the risk calculator was a bit of a compromise. We asked ourselves “Do we want to get something out there or produce nothing because it’s too complicated?” We opted to take on the challenge and start a community dialogue about risk.


Riyas Fadel

In your experience, how do different people understand risk differently and how does this affect the choices they make?

People understand risk subjectively and this makes it a challenge to explain and communicate risk.

Frontline workers tend to use categories such as “high-risk,” “low-risk” and “negligible risk” to assess a person’s risk of HIV transmission. However, people understand these categories in different ways. For one person “high-risk” might be unprotected oral sex and for another person “high-risk” is unprotected anal sex. It’s all subjective and often depends on whether the person is more cautious in general or more of a risk-taker.

What I find interesting is that a person’s view of risk is not set in stone, it can change based on their personal experiences with risk. For example, if a person remains uninfected after engaging in activities they consider high-risk, their idea of what is risky may change. On the other hand, if a person engages in what they consider a low-risk activity and becomes infected, then their perception may change in the opposite way. This works the same with other types of “risk,” for example, driving fast.

How a person feels about the consequences of taking risks also plays a big role in how a person evaluates risk. Some people may not be concerned about becoming infected with a sexually transmitted infection because they figure that treatment is available, whereas others are very concerned.

Also, some people place a greater value on taking risks than others. We live in a society where it’s not uncommon for people to have a “no risk, no gain” attitude. For a long time research used the term “sensation-seeking” to characterize some gay men’s risk-taking profiles. These are men who go to bathhouses, raves and parties, do drugs and have multiple partners—all stuff based on sensation. Some of them seek out certain types of activities, which can include unprotected sex, because it has a certain meaning for them.  

However, I have met a lot of “sensation seekers” who always use condoms or other prevention strategies and have a very good “sexual health” approach to their safer sex practices, so these categories aren’t always useful and can be stigmatizing. We need to figure out what’s going on with someone instead of simply labelling them. As frontline workers, our approach should be focused on a person’s overall sexual health and how they can adopt, and maintain, strategies that help them reach their goals.

We now have a much better understanding of the biology of HIV transmission. How has new research made communicating risk more challenging?

The new research has complicated things in some ways. We have learned a lot about the dynamics of HIV transmission and, at the same time, about how we need to do a better job of explaining the research to clients. Communicating risk was already complicated and it’s only getting more complicated with all the relative risk numbers and statistics emerging from the research on treatment as prevention, post-exposure prophylaxis, pre-exposure prophylaxis and microbicides.

The new research can make frontline workers uneasy because the condom message was so simple and now the messages are so much more complex. Clients are learning about the research from a variety of sources and asking very specific questions, such as: “If being on treatment and having an undetectable viral load reduces the risk of HIV transmission by 96% and condoms reduce the risk by 98%, then is being on treatment only 2% worse than condoms?” Frontline workers are having to unpack this information with clients and it’s challenging.

While things are getting more complicated, it’s also a very exciting time to be working in HIV prevention. We now have a much larger box of prevention tools to discuss with clients.

How do you help a client assess their risk and develop safer sex strategies?

When frontline workers use statistics, they need to be very clear about what the numbers mean. Numbers can be a great tool to help people conceptualize risk but can also be very confusing. Every time I give a number or a statistic from a study, such as a 96% risk reduction or a 3% risk of transmission per act, I always contextualize it and ensure that people understand how these numbers were calculated and what they mean. When I do this, many clients realize that numbers are not really the answer they are looking for.

I also find it interesting and useful to explore the client’s perceptions of risk. I start by asking a client about a sexual activity and ask them to place it somewhere on a risk continuum. Next, I ask the client about the sexual activities they engage in and whether they consider them more or less risky than those already on the continuum. I find that this approach works well because it gives the client a clearer picture of what relative risk is and how different activities relate to each other.

Also, as much as a person understands the risks and doesn’t want to become infected, there is a point before or during a sex act when people sometimes don’t put into practice what they know or what they planned to do. We focus too much on saying “use a condom” or “you have the data, you should know better” and don’t talk enough about the context or the structures someone has to navigate to put what they know into practice.

When I did outreach work helping people evaluate risks and develop a strategy for safer sex practices, I would ask clients to make a list of things they like to do sexually and the context in which they would happen. We would then explore how they try to reduce their risk and how that changes depending on the context. I would also ask them about things they really want to do but have never done and things they have never done and would never do. The goal of these exercises is to prepare clients to negotiate safer sex in different contexts.


Dr. Mona Loutfy

Can you tell me a bit about the situations in which you counsel patients about risk? And how do those situations differ from one another?

As a physician, I counsel individuals and couples who are sexually active and considering unprotected sex, for example, serodiscordant couples that want to conceive, or same sex male couples considering unprotected sex. I also counsel sexual assault victims who are considering taking HIV post-exposure prophylaxis (PEP).

Assessing HIV risk in the context of sexual assault can be more challenging than the consensual sex situation because there are often more unknown risk factors and variables. For example, the HIV status of the assailant may be unknown to the victim. If the assailant is known to be HIV-positive, it’s difficult to know if they were on treatment and had a fully suppressed viral load.

These unknowns need to be built into the risk assessment. If the HIV status is unknown, then we refer to the prevalence rate of HIV in the community of that area. For example, in the case of men between the ages of 19 and 50 in Toronto, the prevalence rate is 1%. The risk that the assailant was HIV-positive can then be multiplied by the risk posed by the kind of sexual act involved (receptive or insertive anal, vaginal or oral sex). Other risk factors that need to be considered include whether there were any STIs and whether there was any bleeding or tearing.

If the sex was consensual , then more of the risk factors are known. If I am counselling a serodiscordant couple that wants to conceive, I generally know the HIV status of both partners, whether there are any STIs, and the viral load of the HIV-positive partner. I frequently counsel serodiscordant couples on safer conception and how being on treatment and having a fully suppressed viral load can reduce the risk of HIV transmission through unprotected sex.

How do you communicate the risk of HIV transmission when the viral load is undetectable?

In the context of undetectable viral load, I go through the data and explain the Swiss Statement as well as the findings from the HPTN 052 study and from other cohort studies. I tell them that I can’t say there is zero risk when the viral load is undetectable, but that if you look at the studies among serodiscordant couples where the HIV-positive partner meets the conditions set out in the Swiss Statement, there have been no documented HIV transmissions.

Unfortunately, all these studies were done among heterosexual couples and therefore we don’t know how much these research findings apply to same-sex male couples. I tell same-sex couples that the data likely transfers to some degree, but trauma and tearing is more common with anal sex and can increase the risk.

When counselling, I use qualitative expressions to describe risk—for example, I say the risk is “very very low” when the viral load is undetectable and the couple is heterosexual. I don’t usually use numbers because we don’t necessarily know what the numbers mean, particularly for same-sex male couples. I also tend to avoid talking about relative risk because I don’t think the average person understands this concept very well.

We now have a much better understanding of the biology of HIV transmission. How has this research changed your approach to communicating risk?

I am fairly conservative about risk as a result of my personality and my role as a healthcare provider. However, with the new research on viral load and HIV transmission, I am relaxing a bit in my counselling. Instead of recommending that couples always use condoms, there are certain situations in which I can tell couples that the risk of HIV transmission is very low without condoms.

We have so much more data out there now, which is great and it has made things easier in some ways for counselling. However, with every additional study that comes out, we have more answers but also more questions. We will never know everything and there will always be some unknowns.

What are some of the tools you use to help someone assess their risk?

Regardless of the situation, I always start my counselling by saying that we are going to talk about probabilities and that it’s very hard to ever say that there is “no risk” of transmission. For most people, risk can be a hard concept to wrap their head around.

The two major things I do to help people understand their risk and make informed decisions are to make sure I take the time with the clients and make sure that I know the data. I try to book longer appointments for counselling sessions or make sure that I schedule them at the end of the day, so that we don’t have a restricted amount of time. It’s important to take the time to answer a client’s questions until they don’t have any more.

It’s also important to know the data very well so I can answer all these questions. I conduct some of the research myself so I can say I really know the data. For example, I recently led a research team in a systematic review of the evidence to better understand the risk of HIV transmission when the viral load of the HIV-positive partner is undetectable. This review has been submitted to a peer-reviewed journal for publication.

Another study we are currently conducting is called the Ontario HIV Pre-Conception Cohort Study. We are enrolling HIV-positive individuals who are interested in getting pregnant or parenting, with or without partners. Part of this study is aimed at gaining a better understanding of how people conceptualize risk and what risks they are willing to take.


Related article

To read the feature article on communicating risk, see Certainly uncertain: Challenges in communicating HIV risk.