Piloting Health Text Messages for Rural Low-Income Mothers

Abstract

Approximately 85% of people living in rural settings own a cell phone, and of those, 76% send or receive text messages. Thus, text messaging may be an effective way to reach rural low-income mothers with important information and resources that will improve their health and well-being. This exploratory study examined the utility of using text messaging for disseminating health information to rural low-income mothers. Researchers tested messages that were personalized and sent from a source similar to participants, a mom named Ms. Peg. Ms. Peg sent text messages about physical activity, oral health, nutrition, and health insurance—facts and tips that addressed barriers and lifestyles in rural low-income contexts—over a 4-week period. Participants completed a prequestionnaire assessing healthy behaviors, text message use and preference, and demographic information. Participants then received two text messages for 4 weeks. Findings indicated that participants were satisfied receiving information through text messages, with half finding it “easy” and “convenient.” The majority reported being highly motivated by the messages, and for 41% the messages were associated with self-efficacy and intention to act. Close to half of the participants found the use of personalization effective and perceived the messages as “relatable.”

Keywords

social marketing/health communication rural health women’s health

Introduction

Mothers living in rural communities face barriers to good health due to poor health care infrastructure, weak organizational capacity, geographic distance, technological constraints, and cultural differences (Aldoory, Braun, Maring, Duggal, & Briones, 2015; Varkey, Kureshi, & Lesnick, 2010). Hartley (2004) stated that “rural culture” is a determinant of health care and more interventions should identify and address environmental and cultural factors unique to rural communities. This study explores the feasibility of using a text-based message system to decrease risk factors for chronic disease among low-income mothers living in a rural community.

Text messaging has been identified as a successful approach to affecting health knowledge, intent, and behavior change (Bock, Heron, Jennings, Magee, & Morrow, 2012; Cole-Lewis & Kershaw, 2010; Fjeldsoe, Marshall, & Miller, 2009). In rural communities in particular, text messaging is highly appealing due to the communication barriers that exist in rural areas. While rural Internet use has grown, it has lagged behind Internet use of suburban and urban communities (Fox, 2012). Thus a text message system for rural low-income mothers could potentially have significant effects on health risks. Approximately 85% of those living in rural settings own a cell phone, and of those, 76% send or receive text messages (Brenner, 2012; Duggan, 2013). Due to prevalence and potential effectiveness, the U.S. Department of Health and Human Services called for more text message interventions that address chronic disease and maternal child health (Greene, 2011). However, there is a dearth of research on text messaging and its use by and effects on rural low-income mothers.

The current study piloted text messages tailored to rural low-income mothers in order to measure preferences for and use of the channel and messages regarding physical activity, nutrition, dental health, and health insurance. The authors used a theoretical framework about source credibility and principles regarding messaging to rural mothers to link research and theory to the practice of designing health-related text messages.

Background/Literature Review

Text Messaging as a Part of Daily Lives

Text messaging is one of the most prevalent and popular cell phone activities today, with 76% of cell phone owners sending and receiving text messages (Head, Noar, Iannarino, & Grant Harrington, 2013; Smith, 2011). “Texting” is part of daily lives and allows for easy access to audiences who can readily obtain messages on their own time (Bock et al., 2012; Cole-Lewis & Kershaw, 2010; Kharbanda et al., 2011). While the body of knowledge on text messaging is large and growing, very little research can be found on rural populations and on rural mothers or low-income women in particular. One study addressed breast cancer prevention information seeking among rural women and found that 87% of participants used cell phones and 20% had an interest in receiving text messages about breast cancer prevention (Kratzke, Wilson, & Vilchis, 2013). No study was found that examined personalization of sources in text messages for rural low-income mothers.

Texting Low-Income Mothers

By reaching populations of mothers through a channel that they are already using on a daily basis, health text messages reduce the burden on mothers to alter their routine to obtain health information (Fjeldsoe et al., 2009, 2010, 2013; Smith, 2011). In other words, mothers favor receiving health-related text messages because they can read them when it is convenient (Fjeldsoe, Miller, O’Brien, & Marshall, 2012). Many low-income women do not have access to online resources, and rely on text messages (Stockwell et al., 2014). Kratzke and colleagues (Kratzke et al., 2013; Kratzke & Wilson, 2014) found that receiving health information via text message was more favorable among low-income, rural women than among higher income populations (Kratzke et al., 2013).

Text Messaging on Health Topics

There have been numerous studies that have examined text messages for health behavior changes, and many of those have addressed two of the four health topics in this pilot, nutrition, and physical activity. A few studies were found that examined text messaging for oral health, while no study was found that studied text messaging effects related to health insurance literacy. One study used qualitative research to develop nutrition text messages for low-income mothers, though the mothers were not from rural geographies (White et al., 2011). Researchers found that the most preferred messages addressed role modeling, eating together, having patience with food, and allowing children to serve themselves. The participating mothers also preferred messages that emphasized benefit over deficits in nutrition.

Theoretical Framework: Source Similarity

The theoretical framework used to develop text messages with rural mothers stems from source credibility literature and the concept of homophily (Hu & Sundar, 2010) and personal involvement (Kreuter & Haughton, 2006). This body of knowledge is sizable and has focused on the source characteristics that influence reception and response to a message. A source in this pilot study is the perceived person “speaking” to the audience. Perceived similarity by an audience to a message source can be defined as “homophily” and can include manifest traits such as race and sex or more subtle similarities such as with health beliefs and social status (Lu, 2013; Rogers, 2004). When the source in a message is perceived as similar in certain traits to the audience, research has shown that the audience will find the source likeable and credible (Hu & Sundar, 2010). Demographic similarity has increased the liking of and trust in a source (Spence, Lachlan, Westerman, & Spates, 2013). Homophily has also been shown to increase identification with the source while also reducing resistance to the sources’ message (Kalichman, Kelly, Hunter, Murphy, & Tyler, 1993; Kreuter & Haughton, 2006; Silvia, 2005).

While not a new area of study, source homophily in health messages has not been addressed frequently in studies of rural populations of women. Yet, importantly, related studies found that identifying with a source can result in greater attention and recall of health messages while also stimulating parasocial interaction and role modeling of behavior change (Hu & Sundar, 2010; Young, 2015). Researchers found increased empathy and behavioral intent from homophilious sources (Hu & Sundar, 2010; McQueen, Kreuter, Kalesan, & Alcaraz, 2011). Among populations of women in particular, research has shown that source credibility, likeability, and similarity of age and race/ethnicity have been associated with increased personal attention and salience with message content (Aldoory, 2001; Vardeman & Aldoory, 2008).

Message Content as Small, Realistic Action Steps

Health messages that include explicit action are more likely to be effective if the actions suggested are sensitive to everyday life constraints. Small, realistic action steps take into consideration everyday life constraints to health. Moreover, realistic action steps increase perceived self-efficacy that allows women to believe they are capable of performing the requested action. Self-efficacy is a well-tested factor in the relationship between health messages and intent to change behavior. Egbert and Parrott (2001) found that among rural women, self-efficacy influenced cancer detection practices. There have been very few studies that have tested self-efficacy and its relationship to text messaging (Evans, Wallace, & Snider, 2012; Horodynski, Silk, Hsieh, Hoffman, & Robson, 2015), though text messaging is a popular intervention for sharing reminders and facts that support health behavior change.

The current pilot study examined the use of small action steps in text messages about physical activity, nutrition, oral health, and health insurance. Four different health topics were included in this pilot for two reasons. First, researchers wanted to tease out main text messaging and source principles from a variety of health topics to assist practitioners in developing text messaging interventions across topics. Second, the different health topics may reveal differing opinions from participants regarding the usefulness of the source and the behavioral steps requested.

Research Questions

Research questions were used to guide the pilot study:

Research Question 1 [RQ1]: What is the perceived feasibility of using text messaging for disseminating health information to rural, low-income mothers?

RQ2: How do rural low-income mothers make meaning of source similarity in text messages about health?

RQ3: How do rural low-income mothers respond to simple, realistic health actions in text messages?

Method

To answer the RQs posed, a pilot text message intervention was designed and disseminated to a sample of rural low-income mothers, who responded to a series of questions via self-administered questionnaires and phone interviews. For each RQ, the majority of data derived from qualitative, open-ended questions posed to participants who were probed for preferences and behavioral intent. However, to gather some summary data on text message viewing behaviors and perceived text usefulness, nominal-level measures were also included. The measures are described below following the description of the sample and research procedures.

Sample Participants

Participants in the study were self-selected from a population of rural low-income mothers who lived in federally designated rural counties in a mid-Atlantic state. Criteria for inclusion in the study were the following: (1) the participant must have household income below the federal poverty level; (2) participant must have regularly used a cellular phone and text messaging over the past year; and (3) participant must be the mother of at least one child living in the home with her.

The study was approved by the university institutional review board for the ethical treatment of human subjects. Recruitment of participants was conducted through university extension agents located throughout the targeted rural counties. The agents posted fliers in their offices and talked to mothers in their communities. When an agent knew that a mother met the screening criteria she asked if the mother would like to participate in the pilot. All the mothers who were asked to participate agreed and voluntarily signed consent forms. A total of 40 participants began the study, but by the end of the study period, 3 participants dropped out, making the total 37 who completed the pilot.

The average age of participants was 30 years (range 20-52 years). The majority of participants identified as being never married (41.2%) or married (38.2%), with 20.6% reporting being divorced, separated, or widowed. Participants self-identified as White/Caucasian (67.6%), African American/Black (26.5%), and Hispanic/Latino (5.9%). One fifth of the sample had completed a college degree, and 23.5% had a high school diploma or GED. Almost 60% of the participants were not employed.

Procedures

To maintain confidentiality, participants were identified with a self-selected code number. The participants first completed a self-administered prequestionnaire, and after completing the questionnaire, they sent a text from their cell phones to opt in to the study. Then participants were told that they would be receiving text messages for 4 weeks from “Ms. Peg,” a mother who lived in their neighborhood and who had two young kids. Ms. Peg was described as someone who was “just like you,” which allowed participants to visually construct Ms. Peg’s age, race, and other visual traits according to their own imagination. Each participant then received a “welcome” text message from Ms. Peg: Ms. Peg was created out of the theoretical framework on source similarity that guided the creation of the pilot study. Based on the principles of homophily, researchers designed a source (Ms. Peg) that would be perceived as similar to participants in gender, age, motherhood, and neighborhood (which indicates socioeconomic status).

During the next 4 weeks, participants received eight text messages, two per week. Four weeks was the time period because this pilot focused on preferences and short-term recall of text content. A longer time period would reduce participants’ ability to express detailed preferences, and a shorter time period would not have allowed enough time between texts and for the four health topics to be addressed in two different ways. The two messages per week was determined based on former research on texting and rural mothers that indicated that rural mothers feel inundated with messages if they are delivered more than twice per week (Aldoory et al., 2015). The first of each week’s messages was a research-based health fact. An example of a text message health fact is “Research shows that when kids join you at the grocery store, more money is spent and more ‘junk’ food is bought.” The second of each week’s messages was one specific action step directly related to the previous health fact texted. These were sent from Ms. Peg. For example, Ms. Peg writes, “To save money and buy less junk food, I watch my friend’s kids so she can grocery shop alone, then she watches my kids when I shop.” Each week was a different topic: physical activity, nutrition, health insurance, and oral health. The topics were counterbalanced across participants so as not to result in bias.

The messages, health topics, and the name “Ms. Peg” were designed as part of a former study using a community-based participatory approach where a different group of low-income rural mothers decided on message factors and the personalization approach (Aldoory et al., 2015). The different health topics were included because in this formative research, all four health issues emerged as primary concerns for the mothers that they wanted to have addressed through future health text messaging.

On each Friday of the study period, a researcher interviewed participants by phone, to assess use, favorability, and behavioral intent. The questionnaires used during these interviews asked about message recall, usefulness of the messages, and any intent to perform the action requested of that week. Two weeks after texting was completed, each participant completed a postinterview over the phone with a researcher. Participants received an incentive for completing the study.

Measures

In total, participants completed six interviews during the study time period: the prequestionnaire interview, four brief interviews by phone during the 4 weeks, and a postinterview. Measures mainly comprised qualitative, open-ended items, but there were also some nominal-level quantitative items included. Both types of data were used to obtain a more comprehensive understanding of the experiences with the text messaging. The quantitative measures were not intended to collect probability statistics or to develop scales; since RQs were focused on meaning making and preferences, the addition of quantitative measures was merely to gather summary percentages of participant use of and preferences for text messages in order to contextualize the more in-depth responses provided.

The self-administered prequestionnaire assessed communication use, health attitudes, and current health behaviors. For example, questions included,

Thinking for a minute about your daily and evening activities, consider when you have looked for information about your health or your family’s health. First thinking of your health, where do you get most of your information about your health?

The mothers were also asked about their frequency of texting. For example, a question was, “On average, how many times per day would you say you send a text message to someone?”

During each week of the 4 weeks, participants responded to an interview where researchers asked questions about self-efficacy, engagement with messages, and behavioral intent. Several questions were used to measure engagement: when did participant view the message (immediately or later), how did the participant view the message (alone or with others), how often did the participant view message (once or more), was the message useful, and did the participant act on the message. The interview also included open-ended questions asking for opinions about texting, message content, and timing.

In the postinterview, researchers measured favorability, recall of, and overall usefulness of the health information from the text messages. Participants were also asked whether they acted on the messages or had intentions to do so. While these measures were quantitative, there were also open-ended questions asking participants to describe their favorite messages and why.

Data Analysis

There was both quantitative and qualitative data to analyze. Quantitative data were inputted into SPSS statistical software. The data set was checked by two researchers for accuracy after initial data entry. Since this was an exploratory pilot with a small nonprobability sample, percentages and frequencies were used to summarize quantitative responses about text message viewing, recall, and preference. In the Results section, percentages are provided for the relevant data that help answer the RQs.

For qualitative data, responses were categorized by RQ. For each RQ, responses were coded into themes that helped answer the RQs. Then researchers went back to the data and counted responses to assess proportion of respondents that supported each identified theme. The themes were shared with other researchers for confirmation and then with some of the rural mothers, for member checks.

Results

RQ1: Feasibility of Using Text Messaging

Engagement

Regardless of topic, nearly 90% of the participants reported viewing the texts immediately during the first week of receiving texts. That percentage decreased to 75% by the 4th week. The majority of participants read texts more than once and found the information provided to be “useful.” By Week 4, percentages dropped significantly in terms of whether participants acted on the information provided. In Week 1, almost 60% of participants acted on the information, in Week 2, 73% of participants acted on the information, in Week 3, 64% acted on the info, but in Week 4, only 27% acted on the information (see Table 1).

Table 1

Engagement Measures by Percentage of Participants, by Week

Item	Week 1	Week 2	Week 3	Week 4
Viewed immediately rather than later	89	89	81	75
Viewed with others rather than alone	40	35	47	44
Read >1	59.5	54	61	62
Found info useful	89	89	94.4	75.7
Acted on info	59.5	73	64	27

By topic, some interesting patterns emerged. Perceived usefulness of the information was rated high for all topics, but lowest for health insurance (78%). Two thirds participants acted on the information about oral health and exercise. Only 19% of participants followed through on the health insurance tip. The greatest percentage of participants read the health insurance texts more than once on receipt (67.6%; see Table 2).

Table 2

Engagement Measures by Percentage of Participants, by Topic

Item	Nutrition	Oral Health	Physical Activity	Health Insurance
Viewed immediately rather than later	89	91	92	70
Viewed with others rather than alone	45	40	39	52
Read >1	57	51	61	68
Found info useful	89	89	92	78
Acted on info	51	78	75	19

Balanced View of Texting as Channel

Half of the participants reported positive feelings toward texting as a channel of communication. These participants found texting to be easy for them, convenient, and a good way to reach people like them. One participant explained how she used her phone for all types of communication: “. . . actually it was nice to have the information sent directly to my phone. It saved me time from having to look it up. It was definitely helpful.” However, about half the participants did not favor the use of texting as a channel for health information. They used texting for personal relationships and did not wish to cross over to using texting as an educational channel.

Technical Convenience

Along with convenience of getting messages on the phone they carry with them, a few participants found it convenient to be able to store the messages on the phone for later retrieval. One mother said, “I remember thinking it was something I need to go back and look at again later. I still have it on my phone; I haven’t deleted it.” Another mother said, “But I do still have [the texts] on my phone and I plan to go back and look at them.” When asked what their opinion was on receiving two messages per week, most participants responded that they felt two per week was not too many but enough to offer cues to action.

Technical Challenge

Over the course of the 4 weeks, three participants identified problems with receiving information on their cell phones. These women cited poor reception, dropped texts, and inoperable devices. One explained, “My phone company said they had a couple of their towers down this week.” The challenges cited seemed to be a result of the rural location and cell tower operability.

Need for Interactivity

Some participants said that one-way messages from a similar persona was not enough to appeal to them, because they needed interactivity and control over sending messages as well. One woman commented that “it was weird” for her not to be able to text back this other mother, Ms. Peg, who was texting her. She said, “It was a random text with information, because it wasn’t a conversation talking back and forth.” Another participant complained that “there was no communication back and forth.”

RQ2: Perceptions of Source Similarity

Ms. Peg Is Identifiable

When participants were asked to recall the text messages from the previous week, many participants were able to identify Ms. Peg as the source. For example, one participant said, “She sent a text about free or low-cost prescriptions or health care appointments.” Another said, “Ms. Peg sent me two of them about making sure that your kids’ toothpaste has fluoride . . .”

Ms. Peg Is Relatable

Almost half the participants favored getting text messages from “another mom” like them because they related to the similar persona. Those in favor found Ms. Peg to offer a personal touch. They said Ms. Peg reminded them that someone else is going through what they were going through as a mother. One participant said, “I thought they understood what I was pretty much looking for all the time.” Another said, “It was good to know that somebody else does some of the things that I do with their kids, who actually cares about their health and stuff like that.” Another described how she related to Ms. Peg:

She was thrown into the same types of things, having busy lives, bearing children, so her messages seemed like things other moms could definitely relate to and that other moms have in common.

Ms. Peg Offers Social Support

Participants noted the value of having virtual social support. One mother said, “The text messages actually helped me out, someone who understood my situation.” Another mother compared it to a parent group but “without the commitment or the gas money or the cost of having people around, having someone who thinks the same things are important. Like they were thinking of you. It’s kind of nice.”

Ms. Peg Is Not a Neighbor

The lack of actual personal relationship reduced Ms. Peg’s ability to connect with some participants. However, this did not mean for the participants that the information was not useful. One mother explained, “I don’t know her. . . . so it wasn’t really personal. . . . I think as long as the information was useful I don’t think it mattered who it was coming from.” Another participants said that having just a couple of similar demographic characteristics was not enough for a source to connect to audience.

I didn’t feel as close, you know, because I felt like it was somebody who was more like a teacher. All she told me about herself was that she had two kids and her name was Ms. Peg. I felt if she was my age and lived in my area I would have hoped . . .

RQ3: Responses to Simple, Realistic Messages

Motivation to Act

Over half the participants reported being “highly motivated” by the simple and doable suggestions for increasing physical activity and nutrition in their daily lives. One participant gave an example: “Like the walking more, I’ve done more since getting that text. . . . I definitely would not be doing that had I not gotten that.” Another participant compared her past behaviors to her change due to the text messages:

Actually I have stopped using a lot of the elevators and started drinking more water. I went from drinking no water to four bottles a day, and taking the stairs instead of the elevator at school.

Trial Behavior

About half of the participants tried to change their behavior and said that they appreciated the texts and the changes. One woman said, “Dialing 211 on my phone is very simple.” Another woman said, “[I appreciated] the one about probably the grocery shopping, because I think it’s completely true. I’ve tried that since then and saved money and definitely bought healthier foods.”

Information Sharing

Importantly, some of the participants communicated to others about the action steps suggested (see Table 3). About 70% of participants agreed or strongly agreed that messages were worthy enough to be shared with others. One mother explained, “I did act on a couple of them and reflected on them and passed them on to parents at my job.”

Table 3

Percentage of Participants Willing to Share Information With Others, by Topic

Response	Nutrition	Dental	Exercise	Health Insurance
Disagree or strongly disagree	2.7	8	11	10.8
Neither agree nor disagree	24	19	14	8
Agree or strongly agree	70	73	72	78.4

Discussion

This was a small pilot study of a text-based communication intervention for a critical yet often ignored audience of rural low-income mothers. Using messages created through audience-centered design, researchers developed a text message system and explored (1) preference of texting as a channel for the mothers, (2) use of a “peer” mother as a source of information, and (3) perceived feasibility of small, realistic action steps as the message content.

Results indicated that texting was preferred by half the participants who found texting to be easy for them, convenient, and a good way to reach people like them. The others viewed texting as something that should be reserved for personal communication. However, almost all the participants found the messages they received via texting to be useful. The most useful messages were related to oral health. The texting was also a useful communication channel for those who wish to save the messages and return to them later.

The theoretical concept of homophily guided the creation of a “peer” source and had positive reactions from participants with regard to being able to identify with the source. Ms. Peg shared similar challenges as the rural women in the study, and the women felt she “understood” what they were going through. However, the identification was not enough to overcome the lack of parasocial interaction. Participants wanted to be able to interact with Ms. Peg, not just hear from her about health actions. In some ways, by making Ms. Peg as personal as she was, it created frustrations for the participants to not converse with her as a neighbor would.

The small, doable steps in messages were favored by participants for their feasibility. Participants were highly motivated to act on the suggested behaviors and were willing to try some actions. Furthermore, participants shared information from the messages with others. This information sharing is an indicator of message acceptance.

There were some limitations to the pilot research presented here. One challenge was with recruitment; researchers partnered with extension agents and thus relied on others to find rural mothers who fit the criteria. This plus the length of time for the study reduced final sample size. Future research should consider increasing incentives for participants. Another challenge was the inconsistent cell phone service reported by some participants. Technical problems with telephones, temporary account suspensions, and poor signal quality were barriers to participants.

Future research is recommended to overcome some limitations and extend the pilot data indicated here. For example, though the sample size here was appropriate for a pilot, future research should secure larger sample sizes and consider a randomized controlled trial of text messages for specific health topics. Future research should examine greater two-way interactions between the source and the audience, to enhance parasocial interaction. This would offer users the chance to get to know personas like Ms. Peg more, which was a recommendation by participants. While Ms. Peg was credible for many participants, some of the mothers still wanted to know additional personal details about Ms. Peg.

Conclusion

This study linked research to practice in its application of theory and literature to text messaging, using source similarity and feasible behavioral modifications. Principles and theory on channel, source, and message content guided this pilot study and were used to create text messages that can be applied to other contexts and for other health topics addressing rural low-income mothers. Findings presented here are described in a way that practitioners can design text messages for audiences of rural mothers who may be limited in their access to other health channels and sources.

The findings from this community-based pilot offer several applied recommendations and implications on practice. For example, several participants found that source similarity and personal relevance were important message factors for learning and behavioral change, though not everyone in the pilot gave these factors similar amount of credence. However, given the rural context, a source similar to mothers gave some participants a sense of community and social support. For rural areas where it might be difficult to create “mothers’ groups,” text messaging from a peer may be a welcomed substitute. With regard to timing for text messaging, practitioners have struggled with how many and when to send educational and motivational messages. In this pilot study, the participants responded that two text messages a week was the right balance of amount of health messages for them: Participants believed that it was not too many to be overburdened and repetitive but not too limited to be forgotten. With regard to the content of the text messages, information on how to design effective texts seemed to fall onto a continuum. The continuum has on one end “helpful reminder” and on the other “motivating behavior change.” Testing different types of text messages as they fall within the continuum would be beneficial to practitioners who are seeking a template for what messages to send as well as when and how.

The text messages designed for the study increased perceived personal connections with a Ms. Peg, which can be replicated in future text message programs accordingly. This pilot study was unique because of its creation of a Ms. Peg out of the voices of rural mothers themselves. This research was also unique in that it explored multiple health topics for purposes of devising thematic principles for text messaging. Findings contribute to the current body of literature by responding to an information need of an audience often ignored in health communication, that of rural low-income mothers. Ultimately, this pilot study helped advance state-of-the-art practice by focusing on the innovative channel of text messaging, the personalization of source, and the sharing of feasible behavioral actions.

Footnotes

Acknowledgements

The authors wish to acknowledge Dr. Bonnie Braun and Dr. Lis Maring for their support and assistance with this project.

This pilot study would not have been possible without funding from UMD Extension HealthSmart. The funds were used for participant incentives and graduate assistant stipends.

References

Aldoory

(2001). Making health communications meaningful for women: Factors that influence involvement. Journal of Public Relations Research, 13, 163-185. doi:10.1207/s1532754xjprr1302_3

Aldoory

Braun

Maring

E. F.

Duggal

Briones

R. L.

(2015). Empowerment in the process of health messaging for rural low-income mothers: An exploratory message design project. Women & Health, 55, 297-313.

Bock

B. C.

Heron

K. E.

Jennings

E. G.

Magee

J. C.

Morrow

K. M.

(2012). User preference for a text message-based smoking cessation intervention. Health Education & Behavior, 40, 152-159.

Brenner

(2012). Mobile technology fact sheet. Washington, DC: Pew Internet & American Life Project. Retrieved from http://pewinternet.org/Commentary/2012/February/Pew-Internet-Mobile.aspx

Cole-Lewis

Kershaw

(2010). Text messaging as a tool for behavior change in disease prevention and management. Epidemiologic Reviews, 32, 56-69. doi:10.1093/epirev/mxq004

Duggan

(2013). Cell phone activities 2013. Washington, DC: Pew Research Center. Retrieved from http://www.pewinternet.org/2013/09/19/cell-phone-activities-2013/

Egbert

Parrott

(2001). Self-efficacy and rural women’s performance of breast and cervical cancer detection practices. Journal of Health Communication, 6, 219-233.

Evans

W. D.

Wallace

J. L.

Snider

(2012). Pilot evaluation of the Text4baby mobile health program. BMC Public Health, 12, 1-10.

Fjeldsoe

B. S.

Miller

Y. D.

O’Brien

J. L.

Marshall

A. L.

(2012). Iterative development of MobileMums: A physical activity intervention for women with young children. International Journal of Behavioral Nutrition and Physical Activity, 9, 151-162. doi:10.1186/1479-5868-9-151

10.

Fjeldsoe

Marshall

Miller

(2009). Behavior change interventions delivered by mobile telephone short-message service. American Journal of Preventive Medicine, 36, 165-173.

11.

Fjeldsoe

Marshall

Miller

(2010). MobileMums: A randomized controlled trial of an SMS-based physical activity intervention. Annals of Behavioral Medicine, 39, 101-111.

12.

Fjeldsoe

Marshall

Miller

(2013). Social cognitive mediators of the effect of the MobileMums intervention on physical activity. Health Psychology, 32, 729-738. doi:10.1037/a0027548

13.

Fox

(2012). Rural e-patients face access challenges. Washington, DC: Pew Research Center. Retrieved from http://www.pewinternet.org/2012/11/26/rural-e-patients-face-access-challenges/

14.

Greene

A. H.

(2011). HHS Text4Health Task Force makes texting recommendations to secretary. Retrieved from http://www.privsecblog.com/2011/10/articles/healthcare/hhs-text4health-task-force-makes-texting-recommendations-to-secretary/

15.

Hartley

(2004). Rural health disparities, population health, and rural culture. American Journal of Public Health, 94, 1675-1678.

16.

Head

K. J.

Noar

S. M.

Iannarino

N. T.

Grant Harrington

(2013). Efficacy of text messaging-based interventions for health promotion: A meta-analysis. Social Science & Medicine, 97, 41-48. doi:10.1016/j.socscimed.2013.08.003

17.

Horodynski

M. A.

Silk

Hsieh

Hoffman

Robson

(2015). Tools for teen moms to reduce infant obesity: A randomized clinical trial. BMC Public Health, 15, 183-203. doi:10.1186/s12889-015-1345-x

18.

Sundar

S. S.

(2010). Effects of online health sources on credibility and behavioral intentions. Communication Research, 37, 105-132.

19.

Kalichman

S. C.

Kelly

J. A.

Hunter

T. L.

Murphy

D. A.

Tyler

(1993). Culturally tailored HIV-AIDS risk-reduction messages targeted to African-American urban women: Impact on risk sensitization and risk reduction. Journal of Consulting and Clinical Psychology, 61, 291-295.

20.

Kharbanda

E. O.

Stockwell

M. S.

Fox

H. W.

Andres

Lara

Rickert

V. I.

(2011). Text message reminders to promote human papillomavirus vaccination. Vaccine, 29, 2537-2541.

21.

Kratzke

Wilson

(2014). Ethnic differences in breast cancer prevention information-seeking among rural women: Will provider mobile messages work? Journal of Cancer Education, 29, 541-547.

22.

Kratzke

Wilson

Vilchis

(2013). Reaching rural women: Breast cancer prevention information seeking behaviors and interest in Internet, cell phone, and text use. Journal of Community Health, 38, 54-61. doi:10.1007/s10900-012-9579-3

23.

Kreuter

M. W.

Haughton

L. T.

(2006). Integrating culture into health information for African American women. American Behavioral Scientist, 49, 794-811.

24.

A. S.

(2013). An experimental test of the persuasive effect of source similarity in narrative and nonnarrative health blogs. Journal of Medical Internet Research, 17(7), e142.

25.

McQueen

Kreuter

M. W.

Kalesan

Alcaraz

K. I.

(2011). Understanding narrative effects: The impact of breast cancer survivor stories on message processing, attitudes, and beliefs among African American women. Health Psychology, 30, 674-682. doi:10.1037/a0025395

26.

Rogers

E. M.

(2004). A prospective and retrospective look at the diffusion model. Journal of Health Communication, 9, 13-19.

27.

Silvia

P. J.

(2005). Deflecting reactance: The role of similarity in increasing compliance and reducing resistance. Basic and Applied Social Psychology, 27, 277-284.

28.

Smith

(2011). Americans and text messaging. Washington, DC: Pew Internet & American Life Project. Retrieved from http://www.pewinternet.org/files/old-media//Files/Reports/2011/Americans%20and%20Text%20Messaging.pdf

29.

Spence

P. R.

Lachlan

K. A.

Westerman

Spates

S. A.

(2013). Where the gates matter less: Ethnicity and perceived source credibility in social media health messages. Howard Journal of Communications, 24, 1-16.

30.

Stockwell

M. S.

Westhoff

Kharbanda

E. O.

Vargas

C. Y.

Camargo

Vawdrey

D. K.

Castano

P. M.

(2014). Influenza vaccine text message reminders for urban, low-income pregnant women: A randomized controlled trial. American Journal of Public Health, 104, e7-e12. doi:10.2105/AJPH.2013.301620

31.

Vardeman

J. E.

Aldoory

(2008). A qualitative study of how women make meaning of contradictory media messages about the risks of eating fish. Health Communication, 23, 282-291.

32.

Varkey

Kureshi

Lesnick

(2010). Empowerment of women and its association with the health of the community. Journal of Women’s Health, 19, 71-76.

33.

White

A. H.

Wilson

J. F.

Burns

Blum-Kemelor

Singh

Race

P. O.

. . . Lockett

A. F.

(2011). Use of qualitative research to inform development of nutrition messages for low-income mothers of preschool children. Journal of Nutrition Education and Behavior, 43, 19-27.

34.

Young

(2015). Source similarity and social media health messages: Extending construal level theory to message sources. Cyberpsychology, Behavior, and Social Networking, 18, 547-551.