Apr. 17 2017

Adolescent Health Comes of Age

With growing public health interest in studies of teenagers, Mailman faculty weigh-in on questions from research ethics to hidden bias

It’s a chestnut of contemporary life: the parent frustrated to understand their teenager. Public health has also struggled to come to terms with adolescence, not due to any teenage antics, but because the youngest ages were always the top research priority. But in recent years, as infant mortality has declined, the field of adolescent health has experienced a growth spurt, culminating in a 2016 Lancet Commission report that made the case for adolescence as a cornerstone for lifelong health.

“There is increasing recognition that if you take care of adolescents today, they will do better through their adult years, and their children will do better too,” says John Santelli, professor of Population and Family Health and contributor to the Lancet report. 

Santelli is the editor of a new series of UNICEF briefs that support continued research into the record 1.8 billion teenagers alive today, one of the goals set by the Lancet Commission. The seven briefs examine a variety of challenges facing researchers, from practical considerations about data sources to ethnical concerns.

A brief written by Santelli and co-authors Terry McGovern, acting chair of Population and Family Health, and researcher Sonia Haerizadeh suggests another reason for why scientists shy away from studies of adolescents: informed consent. To meet ethical standards, researchers follow precepts of informed consent to ensure that study participants understand research goals and voluntary ascent to taking part. But many researchers maintain the distorted view that adolescents have a limited ability to make reasoned choices, underscored by a low social status and susceptibility to adult coercion.

In studies of HIV in Sub-Saharan Africa, adolescents are frequently allowed in trials of educational interventions, but excluded from trials of HIV vaccines and pre-exposure prophylaxis (PrEP), an antiviral therapy used as prevention, even though these interventions would be helpful to this age group. As a result, clinicians must extrapolate findings from adult studies that may not apply to teenagers.

The UNIFEF brief argues that researchers should do more to consider children’s rights and capacities inscribed by the United Nations. In many countries, teenagers may already be legally emancipated through marriage or military service; parents may also be unavailable or misaligned with the interests of their child. Moreover, research from neuroscience and psychology suggests that even early adolescents whose brain and social development is still in process are capable of making informed choices about whether or not to participate in research. 

“I have too often seen adolescents excluded from specific research projects,” Santelli wrote in a UNICEF blog post. “Either the investigators or the ethics committee could not figure out an ethical way to include adolescents in studies that promote adolescent health and well-being. We can and should do better.”

Disclosure Bias

The specific methods researchers choose to study adolescents can make a big difference in the answers they get. In a new study published in the journal PLOS ONE, Lindsay Stark, associate professor of Population and Family Health, compared two techniques for measuring gender-based violence in a study of 165 teenage girls living in refugee camps in Ethiopia and conflict-affected communities in the Democratic Republic of Congo.

When researchers interviewed the refugee girls in a group setting, they heard about violence perpetrated by strangers or distant members of the community. By contrast, when they used tablet computers that afforded greater privacy, girls described violence predominantly perpetrated by family members and intimate partners. As one example, 26 of the 36 girls who reported unwanted sexual touching named an intimate partner or caregiver as a perpetrator in the computer survey, yet there was no mention of these people as perpetrators during the group interviews.

According to Stark, there are several explanations for these discrepancies. So-called “gate questions” used in the group interviews to open discussion may unintentionally keep discussion within the bounds of community norms about violence. Some cultures condone intimate partner violence by men against women in certain circumstances. Researchers and aid workers may also play a role in reinforcing the idea that violence is perpetrated by strangers by focusing disproportionately on public safety interventions like lit pathways and gender-specific bathrooms, while ignoring more hidden forms of violence in the private sphere.

The tablet method likely provided a truer picture of the high levels of violence experienced by these girls at the hands of both strangers and intimate partners. At the same time, says Stark, group discussions provided important truths about how violence is understood and expressed. “You want to know how people talk about these issues, so you can design a program that both addresses violence and is sensitive to those norms.”