By Matt Englund -- Matthew.Englund@unchealth.unc.edu
Something like 70 percent of new mothers get the baby blues. Environmental factors such as stress and sleep deprivation combine with the enormous hormonal changes of childbirth to create symptoms such as mood swings, anxiety and restlessness. These symptoms normally subside after a few weeks.
But for 10 to 15 percent of women these feelings can persist, resulting in postpartum depression (PPD), and other perinatal mood and anxiety disorders.
“I think what’s remarkable is that more people don’t get sick, given the stress and the hormonal changes of childbirth” said Samantha Meltzer-Brody, MD, associate professor of psychiatry and director of the perinatal psychiatry program of the UNC Center for Women’s Mood Disorders.
In March, Meltzer-Brody and Patrick Sullivan, MD, Distinguished Professor of Genetics, led the launch of PPD ACT, an iOS app that will help researchers identify women with a history of PPD — a group that includes both those who are currently suffering from PPD and those who may have struggled with it decades ago. The app asks them a series of validated clinical screening questions and gives them a score based on their answers.
Some respondents are then asked to submit a DNA sample via a saliva collection kit. These samples are sequenced and compared against the DNA of women who have not suffered from PPD, with an eye toward understanding the genetic underpinning of perinatal mood disorders.
Sullivan, who is also the founding director of the UNC Psychiatric Genomics Consortium (PGC), will coordinate the genetic analysis of samples collected through the PPD ACT app. According to Sullivan, using a mobile app to recruit participants and gather data is a novel approach in psychiatric genetics.
“It’s completely new. There has never been something like this done in my field before,” said Sullivan.
The PGC studies a wide variety of psychiatric disorders – including schizophrenia, bipolar disorder and major depressive disorder. The recruitment goal for each type of disorder is 50,000 DNA samples.
“The problem is that we need really large samples, so quick recruitment is important,” said Sullivan. “If we get a hundred samples a week, it’s going to take a long time to get 50,000 samples.”
The pace of recruitment facilitated by the PPD ACT app has been remarkable. In just the first week, the app collected data from more than 8,000 participants and at the end of its first month that number has climbed to 10,000. So far, approximately 5,000 women are eligible and have consented for DNA collection via the collection kits.
A New Approach to an Old Problem
Instances of the baby blues have been described for centuries – even making an appearance in the work of Hippocrates. But it is only in the last few decades that the medical community has begun to recognize the clinical significance of postpartum depression and other perinatal mood disorders.
“All of the grandmothers and grandfathers of the field were British psychologists who noticed that in a group of vulnerable women childbirth was a trigger for psychiatric illness. It was unclear why this was the case but they recognized that it was critical to understand and treat it. They were also of the belief that if you treat the mother it’s better for the baby,” said Meltzer-Brody. “And the US did a really good job of ignoring it.”
When Meltzer-Brody began UNC’s Perinatal Mood and Anxiety Disorders Program in 2004 perceptions of PPD had begun to change. Meltzer-Brody’s efforts placed UNC in the vanguard of PPD treatment in the US.
The appointment of David Rubinow, MD, as chair of the department of psychiatry in 2006 encouraged Meltzer-Brody to further expand the program. Rubinow is an internationally renowned researcher in the pathophysiology of women’s reproductive mood disorders and served as chief of the Behavioral Endocrinology Branch of the National Institute of Mental Health for many years before coming to UNC in 2006.
“When Dr. Rubinow became chair of psychiatry at UNC, he created the UNC Center for Women’s Mood Disorders and we were able to expand and hire more people,” recalls Meltzer-Brody. “I wanted very much to be the first mother-baby unit in the country. And we were in 2011. I think UNC is uniquely collaborative and our success over the last 10 years building this program is because of the uniquely collaborative nature of this place.”
It was in this spirit that, with the encouragement of Rubinow, Meltzer-Brody first began collaborating with Sullivan on developing psychogenetic models of PPD. Their collaboration has lasted more than a decade.
During that time, the American medical community’s approach to perinatal mood disorders has shifted. In the clinic, increased screening is the norm. In the lab, researchers are searching for physical causes beyond the dramatic hormonal shifts associated with the perinatal period.
“There has been a lot of research that has started from the idea that it is not just the hormones themselves—because these changes can be found across populations. If that’s the case, what is unique about these women who get PPD? Is there something unique about their receptors? Is there something about their genetics that makes them uniquely vulnerable?” said Meltzer-Brody.
To answer these questions, she led the effort to create an international research consortium called Postpartum Depression: Action Towards Causes and Treatment (PACT), modeled after Sullivan’s PGC.
The relative homogeneity of the population that typically suffers from PPD makes this disorder well-suited to this kind of large-scale psychogenetic study.
“If we are studying major depression, we have males and females from 15 to 65 years old, who may also be dealing with things like addiction or the break-up of a close relationship; how many diseases are we really looking at? PPD only affects women in a similar age band. PPD is also a more genetic subtype of major depression. We published a paper that showed the heritability of PPD is on the order of 50 percent while the heritability major depression is 37 percent,” Sullivan said.
Launching the App
A web-based application was how Meltzer-Brody and Sullivan originally conceived of the large-scale DNA collection study that eventually became PPD ACT.
When Apple released its ResearchKit platform in 2015, they began to imagine how they could use it to their advantage, and to investigate cost-effective ways to acquire and distribute DNA spit kits.
In October of that year, the National Institute of Mental Health agreed to provide PACT with up to 25,000 spit kits to help facilitate genetic analysis of patients. Then in January of 2016, the U.S. Services Preventive Task Force (USSPTF) released a recommendation that all new mothers be screened for PPD.
But there were challenges beyond resources and technology; the findings also had to be scientifically rigorous and ethically sound.
“It’s the job of UNC’s Institutional Review Board to protect research subjects and getting their approval for a project like this was difficult because it had never been done before,” said Meltzer-Brody. “The difficulty wasn’t so much using an app to ask clinical questions as it was finding a way to model informed consent for DNA collection. At the same time, the science had to be great too. We had to be confident that the way we were phenotyping study participants would hold up to peer review and that the way we were doing all of it made sense.”
After receiving IRB approval, Meltzer-Brody, Sullivan and the team began to think about how best to get the word out about the app. The UNC team reached out to Postpartum Progress®, an international nonprofit organization that supports women with maternal mental illness.
Throughout the development process Meltzer-Brody and Sullivan had been in touch with representatives at Apple. After the USSPTF recommendations were released, Apple contacted Meltzer-Brody and told her that if she could expand the app globally, it could be part of an Apple announcement, so she contacted PACT members in the UK and Australia who began a mad dash to localize the code for each country and get ethics board approval in time.
In Chapel Hill, critical development support was provided by the UNC Health Care and School of Medicine Center for Innovation. Carol Lewis, Center for Innovation associate director, has worked closely with Meltzer-Brody and Sullivan to provide critical expertise and project management, including finding a local app developer, Little Green Software. Jerry Guintivano, PhD, a postdoctoral fellow funded by the UNC T32 in Women’s Reproductive Mood Disorders and a member of Sullivan’s lab, also played a vital role in the development of the PPD ACT app.
Following the release, several other countries have expressed interest in the app.
“There are many PACT members in other countries working towards localizing the app—Sweden, Denmark, Canada, South Africa. Each country has to figure out how to pay for app development costs for localization and their own DNA collection processes. I think some of them are interested in the app solely for help to phenotype women with PPD. And that’s actually ok with me if it screens people and helps them get treatment,” Meltzer-Brody said.
Reaching Patients in New Ways
Using mobile technology as a recruitment tool for this kind of large-scale genetic study has helped PACT researchers cast a wide net. So far the app has made it possible to reach women who currently live in areas that do not screen for PPD as well as those who may have experienced symptoms decades ago.
“I’ve been getting amazing emails from women in their 60s and 70s saying, ‘I suffered terribly 40 years ago,’ and telling their stories. They’re so glad that someone is now doing something about this,” said Meltzer-Brody. “Colleagues who are clinicians are calling me saying they are being contacted by people who wouldn’t have reached out otherwise, who have used the app and are reaching out for help. The impact in just these first few weeks has been amazing.”
The ease of use and availability of the app has also put it in the hands of the patients who need it most. PPD ACT evaluates the severity of a patient’s symptoms and gives them a score ranging from 0-30. The threshold score for sending in a DNA sample is 13. The average score for respondents thus far has been 23. “This means that the app is working even better than we expected,” said Sullivan, “and this is due in part to the self-selection of the population.”
Reaching women who do not have access to an iPhone will be the next obstacle for PACT, and to overcome it they have begun distributing iTouch devices to clinics that serve low-income populations – both in urban centers like Baltimore and Philadelphia, and in more out-of-the-way locations like rural Arkansas. The clinics that receive these iTouch devices will also have a supply of spit kits available to collect DNA from women who exceed the symptom threshold during the same visit.
Meltzer-Brody, Sullivan and their team are currently exploring new ways to make the app available to an even wider audience – perhaps by releasing it on other mobile platforms or adding new features to the app currently available. The version of the app that was released in March is only the first iteration of what Meltzer-Brody hopes is an ongoing refinement.
“We want to make this a clinically useful tool and we are looking for new ways to incorporate features like clinical tracking of symptoms and interactivity. We have all sorts of plans to find ways for the app to do everything but make you dinner.”
The PPD ACT app is sponsored by the UNC School of Medicine, the National Institute of Mental Health, Postpartum Progress® and the Foundation of Hope. The app is free to download from iTunes.