We designed and administered a cross-sectional survey of patients presenting to 3 EDs in southern Connecticut that are part of the Yale-New Haven Health System (YNHHS). Data were collected over 24 hours a day, 7 days a week for a total of 6 weeks. During the study period, the annual census for Yale-New Haven Hospital York Street Campus, an urban academic teaching hospital, was approximately 81,000 adult visits per year and serves a population that is 52% white, 28% black, and 18% Hispanic, with 40% receiving Medicaid. Bridgeport Hospital, an urban academic affiliate of YNHHS, receives approximately 45,000 adult visits per year and serves a population that is 44% white, 38% black, and 15% Hispanic, with 46% receiving Medicaid. The annual census for the Saint Raphael Campus ED, best described as an urban community ED, was approximately 45,000 visits per year and serves a population that is 36% white, 31% black, and 34% Hispanic, with 50% receiving Medicaid.

Research assistants (RAs) enrolled patients presenting to 1 of the 3 EDs. Twenty-two trained RAs enrolled patients on every ED shift, 24 hours a day, 7 days a week, during a 6-week period (July 12 to August 30, 2012). Patients were excluded if they were 17 years of age or younger; alcohol or drug impaired; had a condition that precluded interview; were in police custody; had active psychosis, suicidal, or homicidal ideation; or were unwilling to consent. RAs entered patient data into the electronic data capture system based on time of patient arrival (Figure 1). The institutional review board of each participating hospital approved all study procedures.

Our research consortium reviewed and selected questions from the information technology study conducted at Brown University’s ED [66], by the Department of Veterans Affairs [67], and some instruments from health communication literature [68]. Our multidisciplinary research group consisted of individuals with expertise in informatics, emergency medicine, bioinformatics, engineering, and social sciences who recommended validated questions to include on the survey instrument based on their specific areas, such as media usage [69,70], substance abuse [71,72], tobacco use [73,74], the elderly [75-81], public health records [82], veterans [83], and ethnic minorities [84-86]. The survey was derived from other validated survey or screening questionnaires and new media surveys in combination with original questions specific to the ED, health care, and patient populations. Participants were asked a series of questions representing a number of domains, such as: (1) new media technology ownership (eg, “Do you own a cell phone?”); (2) new media use (eg, “What do you use your cell phone for? Check ALL that apply.”); (3) type of technology owned (eg, “Is your cell phone a mobile phone (eg, iPhone, Blackberry, Android?”)); and (4) frequency of use (eg, “How often do you use your cell phone for text messaging?”). Contingent on answers to these prior questions, participants were asked about new media behaviors such as: (1) seeking health information (eg, “Do you use your cell phone to look up health information?”); and (2) tracking or managing one’s health (eg, “Do you use a software application on your phone to help you track or manage your health?”). The survey ended with the collection of the following demographic data: age, gender, ethnicity, race, preferred language, highest level of education completed, rural/urban status, and annual household income.

The survey was pilot tested over the course of 1 month (with observers) and tested for fourth grade Flesch-Kincaid readability. Some data regarding race were missing (<1%) due to confusion between “race” and “ethnicity.” Thus, participants who reported Latino/Hispanic as a racial category were corrected using hot deck imputation [87-98].

We compared ED patients’ new media use between 3 urban EDs in southern Connecticut. The survey was conducted in English and Spanish. We derived point estimates with 95% confidence intervals (CI) using the normal-theory method for a binomial parameter. Variables of interest include P-values based on the test for a binomial proportion. Analyses were performed using SPSS version 20 (IBM Corp, Armonk, NY).

While the conventional main focus of hospital EDs has been to provide immediate treatment to patients with acute conditions, the use of new media could extend the reach of the ED visit. These clinical encounters provide unique and important opportunities to the clinicians and system of health care to positively influence individual health behavior beyond the emergency department setting.

We sought to define and differentiate new media from cell phone ownership to bring health care operationalization of electronic devices consistent with the communication literature. Furthermore, because information technology is already playing an increasing role in improving health care, delivering interventions, navigating the health care system, and improving the public’s health at large, we wanted to determine (beyond the anecdotal) that sufficient numbers of ED patients own and use new media. Survey participants’ ownership of cell phones (4934/5788, 85.25%) and device usage for calling (4892/4934, 99.15%) and texting (3595/4935, 72.86%) were high. Among all cell phone owners, mobile phone ownership was moderate (2500/4934, 50.67%) with minorities reporting the highest rate of ownership. Benchmarked against the Pew-CHCF study [99], we observed similar prevalence figures for cell phone ownership and use for texting as well as mobile phone ownership (Table 3).

EDs are concerned with enhancing continuity of care throughout an entire health system and optimizing cost containment. As a result, they have generally heightened and expanded their attention to pre-hospital and post-discharge care implications of acute care. Finding new forms of effective communication facilitates expanding the scope of prevention, health promotion, health maintenance, and disease management services [100-102].

ED patients are segmented in this study to those most likely to own and use new media technology. Hence, we determined the characteristics of ED patients that own and use new media to tailor intervention strategies. Text messaging can be used to provide health information to most cell phone users (depending on their phone plan). We examined the relationship between ED patients’ use of text messaging and individual patient characteristics. A higher prevalence of text messaging was reported by ED patients who were female, younger, nonwhite, and more educated (Table 2). Text messaging was less common among ED patients regardless of gender, age group, race/ethnicity, or socioeconomic status.

While mobile phone ownership is not as ubiquitous as overall cell phone ownership, mobile phone technology is important for behavioral interventions (eg, mobile phone health apps). Thus, we examined the relationship between ED patients’ mobile phone ownership and individual patient characteristics. Participants who were female, younger, nonwhite, and had higher income reported greater ownership of mobile phone technology (Table 3). Notably, ED participants with lower household income, less formal education, and either urban or rural residency (data not shown) reported the highest ownership.

Female and younger ED patients who owned mobile phones, as well as those with greater educational attainment, reported higher online searching for health or medical information. Consistently, ED participants reported greater health information seeking than the general population as measured by Pew-CHCF.

We found similar patterns of usage of cell and mobile phones in both the ED patient population and the general population with the exception that ED patients are more likely to use desktop and laptop computers to seek health information than a mobile phone and the general population is more likely to rely on e-mail to communicate through a laptop or desktop computer.

We compared the prevalence of health information seeking by ED patients with that of the general population. ED participants invariably reported greater health information seeking than participants in the Pew-CHCF survey. Individuals presenting to the emergency department likely have health conditions that trigger new media use to manage disease and seek information on treatment and care.

Racial/ethnic minorities and persons of lower socioeconomic status were overrepresented in the EDs as compared to the general US catchment area. Compared to the benchmark Pew-CHCF survey, our ED sample was similar in terms of gender but (predictably) was made up of more nonwhite participants who were poorer and had less schooling.

Our study formally defines new media and disambiguates cell phones from mobile phones. We established a scientifically derived baseline of new media use for ED patients and determined that a critical mass of patients use new media and would perhaps benefit from new media technology to manage their health and seek information. Most importantly, we found that more marginalized populations—such as the poor, homeless [48], and minority patients—do not differ significantly in ownership or usage rates from the general population and that sufficient ownership exists to reach a significant portion of the population using new media. New media may be a health care equalizer to address health care disparities by reaching minorities and low income patients better. This research also suggests that potentially assisting ED patients without information technology is an option to extend services such as the Lifeline Program for Low-Income Consumers [103]. This study increases confidence in the utility of new media for health care services, interventions, and follow up [61,104 105].