By Shania Kennedy

Researchers at the Pennsylvania State University (Penn State) have revealed that mobile phone data increasingly used in public health efforts may underrepresent vulnerable populations and magnify health inequities.

These findings, described in a study published last week in PLOS Digital Health, highlight the adverse impact that data gaps in mobile phone data, and other information used as proxies for human movement and contacts, may have on disease outbreak response.

By failing to account for gaps and biases in mobile phone data, public health stakeholders may not get an accurate representation of disease spread or healthcare access in certain populations.

“Populations with limited access to health care are also often overlooked in other data sources, including censuses,” said associate professor of biology at the Penn State Eberly College of Science and leader of the research team, Nita Bharti, PhD, in the press release. “New, convenient data sources like mobile phones can offer important insights into these populations, but it is critical that we identify and measure their biases.”

To investigate the representativeness of phone data for vulnerable populations, the researchers evaluated phone ownership, human mobility, and healthcare access in a mobile, rural population in Namibia.

The research team noted that mobile phone data is often used to guide public health efforts to combat infectious diseases like malaria across Namibia, but mobile phone ownership and healthcare access vary significantly within the country’s population. Much of the population lives in urban areas with reliable healthcare access, but many people live in remote or rural locations.

The researchers surveyed over 250 people within two remote settlements in Namibia’s Kunene province. The population regularly experiences vaccine-preventable infectious diseases, but the distance to the nearest health clinic is significant, and residents are largely nomadic depending on the season.

Phone ownership among the individuals surveyed was low compared to those in urban areas, with 31 per cent of survey participants reporting that they owned a mobile phone compared to 95 per cent of individuals in urban areas in 2013.

Approximately 59 per cent of those surveyed indicated that they had used a phone before in their lifetime.

The research team also found that phone ownership and usage were skewed toward men in the cohort. Phone owners and users were also more likely to travel to more locations and have better access to healthcare, indicating potential disparities within the population that may impact public health efforts.

“We found that, within these already vulnerable populations, the most vulnerable people were underrepresented in these phone data because they didn’t own phones or have access to phones,” said a graduate student in Bharti’s lab at Penn State and first author of the paper Alexandre Blake. “One common way to make up for missing data is to simply scale it up and assume that missing data are the same as recorded data. But we distinctly found that the people missing in phone data are less mobile with reduced access to health care. And with respect to making public health decisions, these are very important differences.”

However, mobile phone reception also has a role to play in the availability and quality of these data. Phone owners in the cohort often traveled to locations that lacked phone reception, meaning that much of their mobility cannot be accurately captured.

“Even if you own a phone, you can only be tracked in locations where you get signal,” Blake explained. “So, phone data, especially from remote areas, will only capture a specific segment of the population and can record only some of their movements. If phone data were used to predict the potential spread of an infectious disease in a region like the one we studied, most movements and contacts would be missed. Without accounting for data biases, movements based on phone data would be misleading and ineffective for outbreak response efforts attempting to limit the spatial spread of a disease.”

To address the data gaps present in mobile phone data and avoid perpetuating health disparities, the researchers recommended that public health stakeholders must account for and measure biases in phone data. Additionally, using multiple types of data with non-overlapping biases can help ensure that public health interventions are effective and equitable.

“All data have biases but are still valuable resources, and phone data are no exception,” Bharti stated. “Acknowledging that data are not just under representative and showing that they are in fact biased helps our field move towards correctly interpreting data, measuring biases and looking for ways to measure what’s missing.”

“Equitable access to health care is a basic human right, and addressing health inequities in underrepresented populations is essential for public health progress,” she continued. “You don’t have to look to low- or middle-income nations to find underrepresentation in vulnerable groups. We would see the same absence of vulnerable groups in widely used data if we looked, for example, at a rural part of Pennsylvania or Mississippi or in urban areas, like New York City or Los Angeles. There are gaps and biases in all data that underrepresent the segments of populations most in need of improved health services. Failing to acknowledge these biases can direct resources away from these groups and lead to public health interventions that magnify inequities.”