This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Youth experiencing homelessness are at a risk for a variety of adverse outcomes. Given the widespread use of the internet and social media, these new technologies may be used to address their needs and for outreach purposes. However, little is known about how this group uses these resources.
This study investigated how homeless adolescents use these technologies for general and health-related purposes, whether the scope of their use changes with housing status, and their interest in a website dedicated to youth experiencing homelessness.
A convenience sample of youth aged 18 to 21 years was recruited from a youth-specific homeless shelter. All participants completed a 47-item survey, with 10 individuals completing a semistructured interview. Descriptive statistics, exact testing, logistic regression, and generalized estimating equation modeling was performed for quantitative data analysis. Interviews were transcribed verbatim, and NVivo 10 (QSR International) was employed to facilitate double coding and thematic analysis.
A total of 87 participants completed the survey with a mean age of 19.4 (SD 1.1) years. While experiencing homelessness, 56% (49/87) accessed the internet at least once a day, with 86% (75/87) accessing once a week. Access to a smartphone was associated with a 3.03 greater odds of accessing the internet and was the most frequently used device (66% of participants, 57/87). While experiencing homelessness, subjects reported a 68% decreased odds in internet access frequency (odds ratio [OR] 0.32,
While homeless youth experience changes in the frequency, amount of time, and specific uses of the internet and social media, study participants were able to access the internet regularly. The internet was used to search health-related topics. Given the importance of smartphones in accessing the internet, mobile-optimized websites may be an effective method for reaching this group.
An estimated 1.6 to 2 million youth in the United States experience homelessness yearly and often suffer adverse outcomes [
Numerous interventions have been designed to improve health and well-being of YEH including housing programs, substance use, mental health issues, and basic daily needs [
Given the ubiquity of internet and social media use among adolescents in the United States, mHealth technologies represent a unique opportunity for interventions to improve health outcomes among YEH [
Although these studies provide support for the use of the internet and various forms of social media as a viable method of connecting with YEH, several questions remain. More information is needed to determine if there are differences in youths’ rates of access to the internet and social media before a homeless episode as compared with when they are in the midst of an episode. It is possible that those who face homelessness may have already been disenfranchised regarding access to these technologies. Additionally, understanding differential access methods between the 2 time periods (prehomelessness and during homelessness) may provide insight on how to best reach these youth during episodes of homelessness as well as help determine preferred contact and outreach methods. Furthermore, understanding what information this group is seeking and how they seek it can inform interventions to improve health outcomes. Therefore, this study utilized a mixed-methods approach to examine rates of internet access, duration of internet use, devices used to access the internet, and characterize specific internet behaviors among a sample of youth before and during an episode of homelessness. We hypothesized that participants would have less access and less frequent use of the aforementioned technologies while experiencing homelessness. We also sought to identify health-related behaviors YEH engage in on the internet. Finally, using qualitative semistructured interviews, we explored how technology could be utilized to reach YEH to accurately and efficiently provide resources appropriate for YEH.
We engaged participants living at a nonprofit, youth homeless shelter in Philadelphia that annually serves approximately 350 youth aged 18 to 21 years. The facility provides services such as education and vocational advisement, an on-site medical clinic, and case management to the youth residing in the shelter. Study participants were 18-21 years old, housed at the shelter at the time of the study, and English speaking. Individuals were excluded if they were visibly intoxicated or unable to participate due to medical or psychosocial distress at the time of the interview. Only 1 subject was excluded for intoxication. Enrollment occurred between July 14 and September 12, 2014, Monday-Friday, 9 AM-5 PM. The team employed a convenience sampling strategy, and residents of the shelter were invited to participate in the study at on-site meal-time gatherings. All surveys and interviews were conducted in a private room in the shelter. This study was approved by the University of Pennsylvania Institutional Review Board.
All participants were considered homeless during this study by the investigators. Throughout the survey and interview, participants were asked to compare their internet, social media, and mobile health app usage during this period of homelessness to a time they did not consider themselves homeless and had stable housing.
This study employed a complementary qualitative and quantitative data collection approach, wherein quantitative and qualitative data were collected over the same period [
A 47-item survey was administered to all study participants. Questions and answer choices were read aloud to minimize literacy concerns. Study data were collected and managed using a secure electronic database [
Following quantitative data collection, approximately every fifth participant (n=10) was invited to complete a semistructured, approximately 45-min interview in a private office at the shelter with a member of the research team (LV). We employed a constant comparison approach, that is, we reviewed and analyzed the transcripts as data were collected, to determine if new information was emerging [
Descriptive statistics were used to characterize the sample. Means and standard deviations were used to describe continuous variables, and frequencies and percentages were used to describe categorical variables. Fisher exact tests were used to examine differences in demographics between participants who completed the survey and follow-up interview versus those who only completed the survey; two-sample
Interviews were transcribed verbatim and entered in NVivo 10 software (QSR International, Melbourne, Australia) to facilitate thematic coding and systematic analysis [
Participant demographics for the sample have been summarized in
While experiencing homelessness, 86% (75/87) participants were able to access the internet at least once a week and 56% (49/87) of youth were able to access the internet at least once a day. The majority of youth surveyed spent less than an hour using the internet each day, with smartphones (66%, 57/87) and public computers (59%, 51/87) being the most commonly used devices for access. Participants also used the internet for job searching (86%, 75/87), checking email (83%, 72/87), searching for housing (60%, 52/87), and browsing school websites (51%, 44/87). The majority of youth utilized social media (85%, 74/87). Significant differences in internet access frequency (
Adjusted GEE model results demonstrated that current homelessness was significantly associated with a lower odds of regular internet access frequency, more time spent using the internet, use of social media, and access to a smartphone (see
Furthermore, adjusted logistic regression models demonstrated that although smartphone access was significantly associated with greater odds of regular internet access while experiencing homelessness (
Participant demographics: youth living in a young adult homeless shelter in Philadelphia.
Characteristic | All participants (N=87) | Survey and follow-up interview participants (n=10) | Survey-only participants (n=77) | ||
19.4 (1.1) | 19.3 (1.2) | 19.4 (1.1) | .71 | ||
.12 | |||||
Male | 52 (60) | 5 (50) | 47 (61) | ||
Female | 34 (39) | 4 (40) | 30 (39) | ||
Transgender | 1 (1) | 1 (10) | 0 (0) | ||
African American or black | 65 (75) | 7 (70) | 58 (75) | .71 | |
White | 10 (12) | 1 (10) | 9 (12) | .99 | |
American Indian or Alaska Native | 5 (6) | 0 (0) | 5 (7) | .99 | |
Other | 19 (22) | 3 (30) | 16 (21) | .68 | |
.45 | |||||
Yes | 23 (26) | 4 (40) | 19 (25) | ||
No | 64 (74) | 6 (60) | 58 (75) | ||
.12 | |||||
Heterosexual | 74 (85) | 8 (80) | 66 (86) | ||
Gay | 3 (3) | 1 (10) | 2 (3) | ||
Lesbian | 1 (1) | 1 (10) | 0 (0) | ||
Bisexual | 8 (9) | 0 (0) | 8 (10) | ||
Prefer not to answer | 1 (1) | 0 (0) | 1 (1) |
a
Internet use characteristics of youth living in a Philadelphia youth homeless shelter comparing before experiencing homelessness to while experiencing homelessness.
Characteristic | Before experiencing homelessness (N=87) | While experiencing homelessness (N=87) | ||
<.001 | ||||
≥once/day | 69 (79) | 49 (56) | ||
<once/day | 18 (21) | 38 (44) | ||
<.001 | ||||
<20 min | 8 (9) | 23 (26) | ||
20 min-1 h | 20 (23) | 32 (37) | ||
1-2 h | 10 (12) | 11 (13) | ||
>2 h | 49 (56) | 21 (24) | ||
Access to smartphone, n (%) | 74 (85) | 57 (66) | <.001 | |
Job searching, n (%) | 81 (93) | 75 (86) | .11 | |
Use of social media (Facebook, Twitter, and Instagram), n (%) | 85 (98) | 74 (85) | .01 | |
Use of email, n (%) | 79 (91) | 72 (83) | .07 | |
Housing searches, n (%) | 55 (63) | 52 (60) | .56 | |
School website searches, n (%) | 55 (63) | 44 (51) | .06 |
a
Survey respondents were asked if they would use a website or app if one were designed to focus specifically on issues facing YEH. A total of 91% (79/87) of the participants indicated that they would use a website or app specifically designed to support YEH.
We conducted semistructured interviews with 10 study participants that were representative of the larger sample (
The majority of interviewed youth noted that there were striking differences in how they currently use the internet and social media compared with before experiencing homelessness. Consistent with survey findings, participants reported less frequent access of the internet and less time spent using the internet. However, the interviews revealed that the decrease in internet activity was nuanced. Specifically, participants experienced more substantial decreases in entertainment or leisure activities. internet activities became more goal oriented with participants using most of their time on the internet to look for basic needs such as housing, food, and employment. Finally, the youth indicated that time restrictions on public computers impacted their internet activities by making them more purpose driven:
It’s not so much for leisure or entertainment anymore. It’s like I'm going to there with a purpose now. I'm really looking for, or waiting for certain information to help me get where it is I'm trying to go.
Several participants spent time discussing how their perception of their housing security affected how they used the internet. When they considered their living situation to be more tenuous and unstable, the internet was regarded as an essential resource. Participants indicated they would use the internet to contact people who might help them with housing/or to find food resources. When their living situation became more secure, the internet could then be used more for entertainment purposes. Conversely, one participant indicated that now that she had found stable housing through the shelter, she rarely used the internet. For this youth, the internet was exclusively used as a resource during times of crisis:
It’s [the internet] really more important when you're struggling, because you need it. It’s like you need it, because you need to keep in contact and you need to make sure you write. You need it for a lot, than when you're not struggling [with housing].
Participants indicated they thought about their health frequently, with most (9/10) indicating they thought about their health daily. For many participants, health was a source of preoccupation and worry. The participants discussed being stressed about developing various acute illnesses or other diseases, worrying about current health conditions such as asthma and HIV, focusing on their physical fitness, and finding ways to stay healthy.
Participants used the internet as a resource to find information about acute illnesses, child development, alternative medicine and remedies, cancer, exercise, nutrition and calories, provider and hospital contact information, and STIs. Although the youth were able to easily find information on the internet about these various topics, many reported having trouble interpreting the results of their searches. Often, they found the information was presented using complicated terminology that was meant for health care professionals. Some described working around this limitation by looking up synonyms for the words they did not understand. Others expressed frustration at the sheer volume of information that was available and the difficulty they experienced when trying sorting through the results of an internet search on a topic:
Like I'll be thinking like if I do certain things, how it’s going to affect me on the inside, or like I'll be wondering if I'm sick or if I'm about to take a pill or something, I'll hurry up and Google it, to see what it’s going to do to my insides. I do it all day.
The youth were asked about using social media as a means to connect with health care providers or clinics. The group had mixed views; some opposed the idea, whereas others thought it would be a convenient way to reach their physicians. Email and phone conversations were mentioned as preferred methods of contact:
I don’t know. I don’t feel like it’s professional for me to talk to a doctor on Facebook. It’s not really my way of looking at things. I feel like they should be a business or somebody I should call, not on Facebook.
All participants expressed interest in the development of a website or app that would aggregate resources and address topics to help YEH. They also indicated they would use this resource and recommend it to friends. The youth suggested many topics, including information about reliable housing, job opportunities, and women’s and men’s health issues. They also recommended that youth living at the shelter be involved in website development. A full list of potential website topics that were suggested by participants can be found in
Participants also suggested using social media, YouTube, and word of mouth as ways to ensure the website or app reached the people who would be most in need of the resources:
I think it’s [the website] cool, because, honestly, we can use all the help we can get.
Housing
Shelters and emergency housing
Legitimate housing and reliable landlords
Food pantries and kitchens
Classifieds and job opportunities
Health
Women’s and men’s health sections
Clinic information, planned parenthood locations
Sexually transmitted infection information
Substance use information
General Equivalency Diploma and education resources
Important phone numbers
Services
Serve as a Web-based repository for important information (eg, immunization records, school transcripts)
The results of this study contribute to a small volume but growing body of literature that investigate how to best reach and engage YEH. Consistent with previous research, this study indicated that YEH are able to access the internet and social media with relative frequency, though less often than when they were housed [
A significant body of research in this area has emerged from the work of Rice and colleagues in Los Angeles, CA [
The results of this study also highlight the importance of personal health among study participants. However, many homeless youth lack access to the health care system due to a lack of health insurance. Additionally, many do not utilize available services because they are addressing more pressing needs such as finding food and housing or due to perceived and actual barriers to accessing medical care [
In addition to a resource that focuses on the health needs of YEH, this study identifies the need and interest in an internet-based platform that addresses the multitude of challenges faced by homeless youth. The results from the qualitative portion of the study further elucidated that limited phone data plans and computer time limits at public libraries had a strong influence on their internet behaviors. As the prevalence of smartphone use is high in this population, any internet outreach efforts must be specifically designed to be compatible with mobile devices. Websites that do not work well on mobile devices may discourage use and reduce their desired outreach effect.
This research stresses the importance of digital inclusion, especially with regard to this vulnerable population. Digital inclusion refers to the ability of groups and individuals to gain access to the internet, identify appropriate material and resources, as well as have opportunities for training to obtain the necessary skills to effectively use available material [
There are several limitations to this exploratory study. Given that participants currently in shelter were asked to recall behavior before experiencing homelessness, responses may have been subject to recall bias. Additionally, because participants were all recruited from the same youth shelter in one city, results may have limited transferability and generalizability. Our sample size was relatively small with only 87 subjects, which led to wide confidence intervals throughout analysis. Future work can seek to examine internet use among larger samples and more diverse demographics of YEH to better establish specific types of interventions that would be the most useful. Despite these sample size limitations, among similar studies recruiting YEH, this study has a comparable sample size. Additionally, it is among one of the first and largest mixed methods studies to describe access to the internet and social media among a sample of YEH in a major Northeastern US city.
Although the overall time spent using the internet and utilizing social media sites was decreased compared with the time in which they were not struggling with homelessness, the young people in our sample were able to use these resources with relative frequency. Smartphones and public libraries were instrumental in providing access. Moving forward, mobile device friendly, adolescent-centered Web platforms, may serve as an important resource to providing services to and connecting with young people struggling with homelessness.
Internet use survey.
Internet and youth homelessness codebook and coding tree.
GEE model results.
generalized estimating equation
odds ratio
sexually transmitted infection
youth experiencing homelessness
The authors would like to thank Karen Hudson, MSW, and Melissa Berrios Johnson, MSW, for their assistance in developing the survey questions. The authors would also like to acknowledge Covenant House Pennsylvania for allowing them to conduct research and Kenneth Ginsberg, MD; Hugh Organ, MA, and Iva Bonaparte, MA, for their assistance in participant recruitment. This work has previously been presented in poster presentations at Pediatric Academic Societies 2015 Annual Meeting and the American Public Health Association Annual 2015 Annual Meeting. This work was supported by the Arnold P Gold Foundation Humanism Student Summer Fellowship Grant and NIMH K23MH102128 awarded to ND.
None declared.