Perceptions, Attitudes, and Experience Regarding mHealth Among Homeless People in New York City Shelters

Abstract

Mobile health may be an effective means of providing access and education to the millions of homeless Americans. We conducted semi-structured interviews with 50 homeless people from different shelters in New York City to evaluate their perceptions, attitudes and experiences regarding mobile health. Participants’ average age was 51.66 (SD±11.34) years; duration of homelessness was 2.0 (SD±3.10) years. The majority had a mobile phone with the ability to receive and send text messages. Most participants attempted to maintain the same phone number over time. The homeless were welcoming and supportive of text messaging regarding healthcare issues including appointment reminders, health education, or management of diseases considering their barriers and mobility, and believed it would help them access necessary healthcare. Overwhelmingly they preferred text reminders that were short, positively framed, and directive in nature compared to lengthy or motivational texts. The majority believed that free cell phone plans would improve their engagement with, help them navigate, and ultimately improve their access to care. These positive attitudes and experience could be effectively used to improve the homeless healthcare. Policies to improve access to mobile health and adapted text messaging strategies regarding healthcare needs of this mobile population should be considered.

The homeless population of the United States is estimated to be approximately 3.5 million annually (National Coalition for Homeless, 2013; Link, Susser, Stueve et al, 1994). The homeless are more likely to be of a racial and ethnic minority background and men in their early to middle age periods are at higher risk (Fargo, Metraux, Byrne et al, 2012; Culhane, Metraux, Byrne et al, 2013). Close to a quarter of the homeless population are chronically homeless and a significant number stay in the shelter system for more than 1 year or are frequently homeless.

Most homeless individuals in the US were born during the latter part of the baby boom era and are in their early fifties; thus, they are more at risk of developing chronic diseases and need more preventive care (Culhane et al, 2013; Kushel, Vittinghoff, & Hass, 2001). Hypertension (HTN) is one of the most common conditions among the homeless (Gelberg, Linn, 1989; Szerlip & Szerlip, 2002; Moczygemba, Kennedy, Markus et al 2013; Kim, Daskalakis, Plumb et al, 2008; Savage, Lindsell, Gillespie et al 2006; Kleinman, Freeman, Perlman et al 1996), and rates of smoking and substance abuse are high among the homeless (Szerlip & Szerlip, 2002; Kim, Daskalakis, Plumb et al 2008; Lee, Hanlon, Ben-David et al, 2005). Compared to the general population, the homeless die from cancer twice as often, but their rate of cancer screening is lower (NYC Departments of Health and Homeless Services, 2005; Asgary, Garland, Sckell, 2014; Asgary, Garland, Jakubowski et al, 2014). In general, chronic diseases among the homeless are often not well controlled (Szerlip & Szerlip, 2002; Lee, Hanlon, Ben-David et al, 2005), and they face barriers to therapeutic life style changes (Moczygemba, Kennedy, Markus et al, 2013). They disproportionately suffer from mental illness, which complicates management of their medical conditions.

The homeless encounter multi-level barriers to accessing healthcare including lack of a primary care physician or consistent primary healthcare, inadequate preventive counseling, lack of insurance, fatalistic views regarding health issues, and history of discrimination in the health system (Asgary et al, 2014; Asgary et al 2014; Khandor, Mason, Chambers et al, 2011; Chau, Chin, Chang et al, 2002; Lebrun-Harris, Badgette, Jenkins et al, 2013; Wen, Hudak, & Hwang, 2007; Zlotnick & Zerger, 2008). There are biases against health needs and priorities of the homeless among providers who often lack training to address social issues that affect the healthcare of these individuals (Wen, Hudak, & Hwang, 2007). The current health system at best focuses on addressing acute health issues of the homeless and neglects their chronic disease management, health education or preventive care. The homeless populations are mobile, often lack an established relationship with health care providers, and are more likely to miss their medical appointments and follow ups due to multiple social conditions including unemployment, lower level of education, lack of social support, and substance abuse (Khandor, Mason, Chambers et al, 2011; Zlotnick & Zerger, 2008).

Little information exists regarding strategies that use mHealth (i.e., mobile technology for providing health information or services) to mitigate barriers to healthcare access among the homeless. Exploring perceptions, attitudes and experiences of the homeless regarding potential mHealth methods may help design programs to mitigate some of these barriers and address health disparities among homeless individuals. Qualitative research is useful in eliciting the perspective of recipients of healthcare [Marshall & Rossman, 1989), and semi-structured interviews provide opportunity to explore factors that affect health seeking behavior (Krueger, 1994; Sim, 1998). We explored these perspectives and attitudes among homeless in the New York City shelters.

METHODS

This study was performed at 6 shelters and/or shelter-based clinics supported by the Community Medicine Program of Lutheran Family Health Centers, New York City during 2014. We used both random and criteria sampling to enroll 50 non-domicile adult participants. We used criteria sampling to include women, age ranges below and above 50 years, and both chronic and recent homeless individuals. One research assistant approached patients in waiting rooms of the shelters or shelter-based clinics, discussed the study, assessed eligibility, obtained consents and performed interviews. Days of enrollment were selected randomly for each site. Criteria sampling was applied on an ongoing basis. Semi-structured interviews in English or Spanish (if needed) were performed in a private room in the shelters or shelter-based clinics. We also held formal discussions with key informants who possessed knowledge of particular relevance to the research themes, including staff and case managers of shelters, allied health workers and medical providers at shelter-based clinics. These discussions informed our interview tool and areas to explore further, but were not included in the analysis. This study received IRB approval from the Lutheran Family Health Centers, New York City.

Semi-structured individual interviews were performed rather than focus group discussions to assure a private and candid environment where participants could freely discuss their experience. We asked a series of open-ended questions, with directing probes or follow up clarifications if needed. Questions regarded possession of cell phones, attitudes and acceptance towards mHealth strategies such as text messaging on access to healthcare, and ideas on the content of text messages and the perceived effect of free of charge mobile services to improve preventive care and chronic disease management. We obtained oral consent. The interviews were recorded and documented verbatim and translated into English if needed. We then coded transcripts and analyzed them for major themes.

A qualitative descriptive approach was used for analysis by two authors (RA and AA). Content analysis was performed to identify core themes regarding knowledge, understanding, and perceptions of mobile technology and healthcare. We developed preliminary coding based on priority codes derived from theoretical framework and conceptual model guiding the study. Our theoretical framework was built around the following concepts; a) there are misconceptions and negative experiences regarding health care system among the homeless, largely due to barriers to accessing health care and lack of health education and proper counseling for preventive care by providers, b) the homeless are mobile populations and are often hard to reach, and b) the homeless have good experiences with mobile phones and technology, and have positive attitudes towards using mobile technology to improve their health care access. We performed critical deliberation about initial coding and reviewed coding for similarities and variations to achieve a high level of agreement. Two authors reviewed all codes independently, reviewed and discussed all codes, and discussed the specific categories and characterized and agreed upon major important themes. Coding was performed through open and selective coding. Codes fell into distinct but overarching categories. Codes were referenced back to the subject characterizations to evaluate responses based on age and gender. Emerging themes were compared across cases to explain commonality and variability of themes.

RESULTS

Fifty homeless men and women participated in the study. Demographics of participants are presented in table 1.

Table 1.

Demographics of participants

Indicators	N (%)
Average Age (SD; Range)	51.66 (±11.34; 25 to 79)
Female	29 (58)
Age above 50	33 (66)
Average years of homelessness (SD)	2.03 yrs (SD±3.10), max 14 yrs
History of Chronic Dis.	30 (60)
Active mental illness	10 (20)

^*History of chronic diseases included hypercholesterolemia, hypertension, diabetes, asthma/COPD, or seizure disorder.

Ownership of cell phones and using text messaging

The majority of women (25) and men (14) had a working cell phone with the ability to send and receive text messages; “I can get text messages? and voice messages?” [M53]. Most participants had their phones for at least some years; “[I have had it] for 4–5 years”[M52]; “two years” [F52]; “2 or 3 years, I have other back up phones…”[F25]; “This has been my number for 4 or 5 years”[F54]; “one year” [F52]; “Since 2009 I have not, but [then] I just got a new phone” [F53]. Others indicated a relatively shorter time period of ownership, including within months; “not long” [F49]; “ 4 months” [F27]. Most participants knew how to send or receive text messages. However, some older men had difficulty sending text messages due to small phones; “but I am not using text much, can’t see well” [M79].

Most of the participants who did not have cell phones had them before; “no [I don’t have now], it was stolen” [M52]; “no I don’t have cell phone, [I] use land line” [M58]; Not now but I had until weeks ago” [M60]. Additional quotes are presented in table 2.

Table 2.

Mobile phone ownership and experience among homeless, New York City, 2014

How long have you had cell phone	Men “I’ve had it for about 10 months” “I keep them all the time so family members can contact me”[M59] “4 years” [M55] “3 years, not often I have the same one for a while” [M53] “don’t know number, my wife uses it, I don’t like cell phones. I prefer landlines”[M55] Women “One month, first number” [F59] “One year” [F43]
How often do you change phone/number	“I don’t” [M53] “Same number it was given to me from Safe Link, I could text as well, they pay 250 min a month and rest is on you if you go over, if you forget to recertify (based on low income) then you will get new number, I had same number for some years” [M60] Women “3 times in 2 years” [F57] “I change it once a year” change it frequently” [F43] “I try not to [change it]” [F27]; “3 times in 2 years [I have changed it]” [F57]; “I have it for several years, it’s a 212 number I will never change it” [F42] “ One year” [F33]
How much credit/minutes you have, do you have enough	Men “it’s a free phone, they call it an obamaphone” “I only make calls when I need to more than when I want to”[M59] “usually yeah”[M52] “Yeah, but I turned it off this time because I didn’t” [M58] “I make sure my phone gets paid”[M53] “no”[M75] Women “Have free Obamaphone, but I add minutes” [F46] “Always, that was a part of a deal”[F59] “I keep the bill paid, that’s important for me to keep in touch with my family. Healthcare appointments’ [F52] “Obamaphone is basic phone you can get if you’re on public assistance. 60 minutes a month, rolls over. Don’t check e-mail, its hard to text. have to be conservative with it”[F42] “Yeah Ive never struggled on that”[F54] “yeah”[F42] “we’ll see” “its very important”[F59] “ sometime” [F28] “yeah unlimited plan” [F33]

Experience of using mobile technology

Participants usually kept the same phone numbers, however, men were more likely than women to change phone numbers over time. Older women changed their phone numbers less often than younger women. “Try not to, [I] used to change it often” [F53]; “Every 3 years [I] change it” [F42]; “I change it once a year. Change it frequently” [F43]; “[I have had] it for long time, got it from Google software, had smart phone and there were problem with them but I kept his internet Google phone” [M53]; “Two years same number, different phone. Don’t change the number” [F52]; “[I have it for] 3years ”[M61]; “ [I have it for] Almost 4 years”[M62]; “[I have it for] 3 months”[F42]; “I’ve had it for about 10 months” “I keep them all the time so family members can contact me”[M59]. A minority stated that their current phone was their first mobile number/phone;“4 months, I never had before in my life, this is the first time, Medicaid sent it in” [M55]; “First time I did get a cell” [F49].

Around a third of both men and women had an “Obamaphone” (provided and paid by insurances) and elaborated on being able to manage keeping enough phone plan minutes. “It’s a free phone, they call it an Obamaphone. I only make calls when I need to more than when I want to” [M59]; “Mine was paid, yes I could afford pay extra minutes” [M60] “I do [afford it]” [F27]; I try to keep at least 5 minutes on the phone”[F43]; “Yes, its only 50 dollars a month” [F45]; “This is from insurance”[F56]. Very few had unlimited minutes. Close to half stated they usually have money to pay for their phone and minutes; “Yes, I can pay the bill” [M53]; “yeah, usually” [F67]; “well, I have to save it, many pennies”[F57]. And minority got help from family for their plan. “No, I get 250 minutes, sometimes my friend will help me get the 30 dollar unlimited, most of my calls are to my doctors”[F53]; “no, my daughters are the ones helping me to pay”[F47]; “I never pay the bill for the phone, my family helps with that. They have to, it’s a must if they want to keep in touch”[F25]. Please see table 2 for additional quotes.

Perception and attitude towards text messaging to improve medical care

The majority of women, across age groups and more than men, welcomed and preferred receiving text messaging regarding health messages and navigating health system. “Hell yeah, yes I always like to be reminded. I’d like them to do it a week in advance as well as the night before” [F59]; “Yeah, that would be cool, that’s fine” [F53]; “That helps a lot because that’s something I can save or store” [F43]; “I think it is a good idea depending who sends it, if my doctor or for my health it is great, I love it” [M53]; “If it is my doctor it is OK and if it is for health, texting is good for health related” [M58]. Very few preferred solely phone calls or in-person encounters because they were not familiar with texting. “Even to call. Calling is better. Sometimes I don’t remember to check the phone…” [F56]; “Not really, I don’t know how to use it yet, calling is better” [F59]. Additional quotes are presented in the table 3.

Table 3.

Attitude towards text messaging and mobile technologies in regards to healthcare among homeless in New York City, 2014

How do you feel if someone sends you text messages regarding your healthcare	“They could call me for appointment reminders and ill answer and tell them to call her [my wife] and shell tell me” [M55] “It’s a free phone, they call it an Obamaphone” “I only make calls when I need to more than when I want to”[M59] “that’s no problem, that’s great” [M52] “Yes, that is good idea, I am not annoyed at all” [M60] Women “That’s good, that would be a nice thing” [F57] “That would be okay with me. Rather they call me into the office and give me information” (in person)’ [F52] “Uh, yeah, as long as it wasn’t monotonous” “wouldn’t want to get three texts” [F42] “It would be helpful. it might slip my mind, might be going through something that day, might need a reminder” [F49] “Yeah that would be very helpful” [F66] “Yeah I think so, its always good to get a reminder” [F59]
What/how do you suggest/prefer text message be like, the content of texts	“Reminder” [M75], “A reminder, I would like the text message the day before” [M59] “Time for your test…whatever” [M62] “Just reminders” [M58], “Tell me what needs to be done, information” [M53] “Information or asking about my health” [M53] “Whatever works for you/doctor, as far as I get info it is all right” [M60] “The information” [F46] “A reminder of the appointment and what it is for” [F52] “Reminders when to get screenings” “my memory isn’t all that, the seizures make me forget sometimes” [F66] “Time for a check up! And just say the appointment, GYN appt on this date. Very straightforward” [F59]
What could be advantages of free phones or plans for health related issues	“Yeah, I guess so, how free is it if you run out of minutes…so no, the plan I have now I pay 40 dollars a month” [M62] “I have one that is free, if they take it back then sure, it would help a lot” [M55] “Yeah, definitely, that’s one of the calls I don’t ignore” [M55] “yeah” [F27] “Oh yeah, definitely, I would love to have a free phone” [F57] “Yes” “could call that cell to remind about appointments” [F40] “It probably would”. Extra incentive for others’ [F49]

Overwhelmingly both men and women preferred a simple text message reminder that provides necessary information rather than lengthier texts and motivational messages. “To remind me and also to inform me if there’s been something wrong I can come in and take care of it immediately” [M52]; “Reminder with information of my appointment and location and time” [F49]; “I think it would be just a general reminder… [because] with questions I may not have enough minutes left” [F53]; “Little reminder” not that much info” [F49]; “Don’t forget, a reminder” [M61]; “Just say you have an appointment, date, time, address, and a number to call back to confirm. Something simple, nothing too elaborate. Less information, all the extra shit distracts me. Text message is easier for me because it’s something visual [F43]. Very few participants requested other health tips/recommendations and others suggested more interesting messages. “Pink [colored message]” [F43]; “Not repeating the same information verbatim. If there is new information, new machine…little tips. Things that would take the stress off of it. Not the same stuff you could get from a book” [F42]; “Information, good news, what I need to do” [M55].

The majority women and men elaborated that they would be more open to text messaging, more empowered to keep medical appointments, and better able to follow health recommendations if they were given free phones or phone plans. “I never thought of that, I think everyone would want that, if that’s possible I’d be 100 percent behind it” [F53]; “I would imagine it helps a lot for people who can’t afford it” [M53]; “That would be great!” [M53]; “That would help a lot of people. Communication means everything. In the shelter you should be able to contact your counselor and they give you the message. But it doesn’t happen” [F43]; “Yeah because I need to call my psychiatrist and I didn’t call” [F56]; “Hell yeah, who doesn’t want that” [F59]. A few men did not think free phones and plans would help them much regarding access to health information as they faced other difficulties or had distrust of the government and system. “I wouldn’t take one of those, I’d rather buy myself a plan. The government never did anything for me before why do they wanna give me a free phone?” [M55].

Discussion

The majority of homeless in NYC own mobile phones. While some homeless had difficulty maintaining their phone plans consistently, others either got free limited insurance phone plans or were supported by family or friends. Most homeless are familiar with and know how to use text messaging, however, older men may have more difficulty using text messaging due to lack of practice or poor eye sights. The homeless usually make every attempt to keep the same phone numbers over time to maintain their connections and communications as they are mobile and lack access to landlines. Limited previous research indicates that 70% of the homeless owned cell phones without significant difference in new media use, modality, or frequency compared to domicile populations (Post, Vaca, Doran et al, 2013). The homeless, however, have significantly lower rate of contract plans with unlimited minutes (Post, Vaca, Doran et al, 2013). The homeless appreciate receiving text messages from providers or clinics regarding their healthcare and better management of their medical concerns. Many believe that free cell phone plans will give them better control over managing their health problems and navigating the convoluted healthcare system. Prior studies have shown that patients experiencing homelessness welcome health information regarding substance abuse, smoking cessation, pregnancy, domestic violence and mental health (Post, Vaca, Doran et al, 2013; Eyrich-Garg, 2010).

There are millions of homeless Americans who need and want to have better access to care but face multi-levels barriers that are not addressed systematically (Asgary et al, 2014; Asgary et al, 2014; Khandor et al, 2011; Chau, et al 2002; Lebrun-Harris et al, 2013; Wen et al, 2007; Zlotnick & Zerger, 2008). The current complex health system poses difficulties to the homeless to navigate it effectively for their preventive care and chronic disease management. They also face discrimination in the health care system which dissuades them from seeking care (Wen et al 2007; Cooper, Roter, Carson et al 2012; Hausmann, Hannon, Kresevic et al, 2011). Providers may have prejudice and biases against the medical needs and priorities of minorities and focus largely on addressing their acute care needs rather than chronic disease management, risk reduction and risky behaviors and preventive care (Hausmann et al 2011; Teal, Shada, Gill et al 2010). Strategies to address biases among providers have been proposed (Devine, Forscher, Austin et al 2012; Peek, Wilson, Bussey-Jones et al 2012), and mHealth strategies may help reduce missed opportunities for health education and counseling by providing a more consistent approach and improve connectivity (Post et al 2013). The homeless in our sample acknowledged their barriers and social conditions and recommended and welcomed using mHealth technology to get reminders for their preventive care and medical appointments to improve adherence and receive health education.

Mobile technologies are ubiquitous and provide a potentially excellent platform to provide targeted health services especially for mobile populations or those who are out of reach and socially marginalized without direct access to usual healthcare facilities or health education. Homeless have poor access to primary care and use emergency departments largely due to their difficulty to effectively communicate and connect with health providers or health system and maintain follow up visits (Post et al 2013). mHealth strategies have been used for the management of chronic diseases such as HTN and Diabetes with excellent results in improving uptake, adherence, and clinical outcomes of treatment modalities (Car, Gurol-Urganci, de Jongh et al 2012; Chen, Fang, Chen et al 2008; Leong, Chen, leong, et al 2006; Marquez Contreras, de la Figuera, von Wichmann, et al 2004; McGillicuddy, Gregoski, Weiland, et al 2013; Dick, Nundy, Solomon et al 2011; Guy, Hocking, Wand, et al 2012), that could be adapted to address the medical needs of the homeless (Eyrich-Garg, 2010).

The homeless overwhelmingly preferred simple reminders or short straightforward text messages regarding health issues and disliked long or motivational texts. Contrary to general assumption, they are enthusiastic about getting health education that targets their specifics risks and conditions and improves their preventive care, which has been largely ignored. Studies have shown that the content and format of text messaging are important and need to be adjusted to the specifics of populations and their needs (Car et al, 2012; Marquez Contreras et al 2004; Dick et al, 2011). Type and content of such mHealth communications and their frequency could be tailored toward the specifics of their age range, gender and cultural background. The homeless in our study were largely open to receiving straightforward health messages coming from their providers or people they recognize.

Existing opportunities in the shelter system with case workers and social services could be coupled with mHealth text messaging to more effectively help the homeless connect with the health system, reinforce peer education, facilitate making medical services referrals and bring the homeless and medical providers together and improve communication. They can provide an opportunity for reciprocal communication as messages and recommendations are recorded and stored for follow ups and support during counseling for changing unhealthy behaviors and addressing misconceptions. mHealth modalities could serve as patient navigators to counterbalance the multi-level barriers to accessing healthcare as they have shown sustained improvement in chronic diseases management in the general population.

Our study is among the very few qualitative studies regarding mHealth strategies among the homeless with direct input from patients and without preconceived notions from providers or the health systems which help better understand priorities and underpinning of their decision-making regarding healthcare. We collected data from multiple shelters in different neighborhoods and boroughs of New York City, a city with one of largest homeless populations in the country, and we included a good sample size of different age ranges and genders, which makes our data more generalizable.

Our study is not without limitations. We primarily enrolled participants from shelters and may have missed homeless individuals living exclusively on the street. This however constitutes less than 10% of the total homeless population in New York City. Due to difficulty gathering all participants in one location and considering their social conditions and constraints we preferred semi-structured interviews to focus group discussions. Focus groups may allow for more synergy and potentially improve sharing of experience and perception. We did not collect history of substance abuse and educational background and or race/ethnicity data and were unable to compare our responses across that spectrum. However, previous data from our population indicates that the majority of our population are Black and Hispanic, and more than half have high school or less of formal education (Asgary et al, 2014, Asgary et al 2014). We have collected data on mental health problems, which did not differ in regards to response themes among the homeless.

Despite adequate attitude and perceptions regarding mobile technologies and text messaging for healthcare issues, these strategies have not been generally evaluated or used for the homeless. Due to multiple system level barriers, the current health system fails to provide effective support for the homeless to have access to the same standards of care that average Americans have. The mHealth platforms could be very effective in improving knowledge and access to care for largely mobile populations of the homeless who face discrimination and prejudice within health system and generally avoid healthcare system (Wen et al, 2007), and therefore miss common opportunities for health education, regular care, or check ups. Health education and strategies to improve and promote healthy behaviors are particularly important among this unusually marginalized population. The attitude of the health system needs to change significantly from providing only basic care to more equal opportunities for accessing preventive care and management of chronic diseases, which may be achieved through the effective use of mobile technology in the health system. Aside from health system changes, there needs to be societal strategies to address homelessness and to prevent homelessness.

Conclusions

The homeless regularly use mobile technologies and welcome text messaging modalities to improve their health care. The significant positive attitude and experience of mobile technologies could be effectively used to improve their connection with health care system and providers, health education, and preventive care and their chronic disease management. Policies and plans to improve availability and access to mobile technologies along with targeted and adapted mHealth strategies should be considered for highly vulnerable and mobile homeless populations.