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 JMIR mhealth and uhealth, is properly cited. The complete bibliographic information, a link to the original publication on http://mhealth.jmir.org/, as well as this copyright and license information must be included.
Rather than providing participants with study-specific data collection devices, their personal mobile phones are increasingly being used as a means for collecting geolocation and ecological momentary assessment (EMA) data in public health research.
The purpose of this study was to (1) describe the sociodemographic characteristics of respondents to an online survey screener assessing eligibility to participate in a mixed methods study collecting geolocation and EMA data via the participants’ personal mobile phones, and (2) examine how eligibility criteria requiring mobile phone ownership and an unlimited text messaging plan affected participant inclusion.
Adult (≥18 years) daily smokers were recruited via public advertisements, free weekly newspapers, printed flyers, and word of mouth. An online survey screener was used as the initial method of determining eligibility for study participation. The survey screened for twenty-eight inclusion criteria grouped into three categories, which included (1) cell phone use, (2) tobacco use, and (3) additional criteria
A total of 1003 individuals completed the online screener. Respondents were predominantly African American (605/1003, 60.3%) (60.4%), male (514/1003, 51.3%), and had a median age of 35 years (IQR 26-50). Nearly 50% (496/1003, 49.5%) were unemployed. Most smoked menthol cigarettes (699/1003, 69.7%), and had a median smoking history of 11 years (IQR 5-21). The majority owned a mobile phone (739/1003, 73.7%), could install apps (86.8%), used their mobile phone daily (89.5%), and had an unlimited text messaging plan (871/1003, 86.8%). Of those who completed the online screener, 302 were eligible to participate in the study; 163 were eligible after rescreening, and 117 were enrolled in the study. Compared to employed individuals, a significantly greater proportion of those who were unemployed were ineligible for the study based on mobile phone inclusion criteria (
Inclusion criteria requiring participants to use their personal mobile phones for data collection was not a major barrier to study participation for most respondents who completed the online screener, including those who were unemployed.
ClinicalTrials.gov NCT02261363; https://clinicaltrials.gov/ct2/show/NCT02261363 (Archived by WebCite at http://www.webcitation.org/6wOmDluSt)
Ecological momentary assessment (EMA), a real-time data capture method, has been increasingly used as a viable method to collect data from research participants over time and in various contexts outside of a research lab[
The use of mobile phones has become a method for EMA and geotracking data collection over the past ten years [
Instead of providing participants with study-specific handheld electronic devices, personal mobile phones are increasingly being used as a means for collecting EMA and geotracking data, providing a noncoercive (eg, no free phone for study participation), low-cost tool to collect data in real-time [
Data come from a parent study called the “Moment Study”, a mixed method longitudinal study that examined factors influencing e-cigarette initiation using a convenience sample of adult daily smokers residing in Washington, DC [
We recruited a convenience sample of adult smokers via public advertisements, free weekly newspapers, printed flyers, and word of mouth. The advertisements included a link to an online survey, which was the initial screening tool. The online survey screened for twenty-eight inclusion criteria grouped into three categories, which included (1) mobile phone use, (2) tobacco use, and (3) additional inclusion criteria.
The inclusion criteria for mobile phone use derived from the technological needs for EMA and geotracking data collection. We required daily mobile phone use and a preexisting unlimited text messaging plan to ensure that participants were comfortable sending and receiving texts and the number of text messages sent and received over the course of three weeks did not pose a financial burden on the participants. We also required the participants to have an Android or iPhone mobile phone because the geotracking app was only available on these operating systems. Device requirements were not expected to be restrictive, as Android and iPhone are the leading mobile phone operating systems in the United States with Android holding 56.4% and iPhone accounting for 42% of the market as of January 2017 [
In addition to the mobile phone inclusion criteria, we also required that participants be adult (≥18 years) daily cigarette smokers with restricted past 30-day use of other tobacco products. Additional inclusion criteria included age, pregnancy status, and both physical and mental health status (
Descriptive statistics were used to summarize the sociodemographic characteristics of individuals who completed the online screener, overall and eligibility by virtue of mobile phone ownership, and to describe the frequency of respondents meeting the study criteria. Observations with missing data were listwise deleted. Chi square tests and Mood’s median tests (a special case of chi square tests) were used to test the equality of proportions or medians. Statistical significance was set to a
A total of 1003 individuals took the online screener. Most respondents were African American (605/1003, 60.3%), male (514/1003, 51.3%), and had a median age of 35 years (IQR 26-50). Nearly all the respondents lived in Washington, DC, Virginia or Maryland (976/1003, 97.3%), and approximately half (496/1003, 49.5%) were unemployed. Most smoked menthol cigarettes (699/1003, 69.7%) and had smoked for a median of 11 years (IQR 5-21) (
Of the 1003 respondents to the online screener, 28.5% (286/1003) were ineligible because of mobile phone inclusion criteria. Differences by sociodemographic characteristics and mobile phone inclusion criteria were evident. African Americans (57.4% vs 70.5%,
Among the other inclusion criteria, 44.0% of the unemployed individuals met all tobacco use inclusion criteria, and 93.0% of unemployed individuals met all the additional inclusion criteria; there was no difference in eligibility by employment status for the tobacco use or all other inclusion criteria (results not shown).
Characteristics of individuals who took the initial online screener, overall and by mobile phone inclusion criteria, in the Washington, DC area.
Characteristic | Mobile phone inclusion criteria | ||||
Overall (N=1003) | Met inclusion criteria (>N=717) | Did not meet inclusion criteria (N=286) | |||
<.001 | |||||
Female | 478 (47.7) | 346 (48.3) | 132 (46.2) | ||
Male | 514 (51.3) | 371 (51.7) | 143 (50.0) | ||
<.001 | |||||
White | 248 (24.7) | 193 (27.0) | 55 (20.0) | ||
African American | 605 (60.3) | 411 (57.4) | 194 (70.5) | ||
Asian | 39 (3.9) | 35 (4.9) | 4 (1.5) | ||
Native Hawaiian | 1 (0.1) | 1 (0.1) | 0 (0) | ||
American Indian | 2 (0.2) | 1 (0.1) | 1 (0.4) | ||
Other | 52 (5.2)) | 41 (5.7) | 11 (4.0) | ||
More than 1 race | 44 (4.4) | 34 (4.7) | 11 (3.6) | ||
Hispanic, n (%) | 77 (7.7) | 58 (8.1) | 19 (6.6) | .530 | |
Age, median (IQRb)c | 35 (26-50) | 32 (25-45) | 47 (32-55) | <.001 | |
Years smoking, median (IQR)c | 11 (5-21) | 10 (6-20) | 20 (11-35) | <.001 | |
Menthol smoker n (%) | 699 (69.7) | 493 (68.8) | 206 (72.0) | <.001 | |
Live in the DC metro area, n (%) | 976 (97.3) | 708 (98.9) | 268 (93.7) | .201 | |
Employed | 473 (47.2) | 384 (53.6) | 89 (31.3) | <.001 | |
Not employed | 496 (49.5) | 333 (46.4) | 197 (68.9) |
aThe following variables had missing data: gender (1%), race (1.2%), employment (3.3%).
bIQR: interquartile range.
cMood’s median tests were used to test for differences in median age and years smoked.
Percentage of screened individuals who satisfied mobile phone, tobacco use, or additional inclusion criteria in the online screener (N=1003).
Inclusion criteria | n (%) | ||
739 (73.7) | |||
Mobile phone allows app installation | 871 (86.8) | ||
Use mobile phone daily | 898 (89.5) | ||
Unlimited text message plan | 871 (86.8) | ||
Meet all the mobile phone inclusion criteria | 717 (71.5) | ||
890 (88.7) | |||
≥5 years of smoking | 772 (77.0) | ||
≥8 cigarettes a day | 689 (68.7) | ||
Use of little cigars no more than 5 times in the last 30 days | 814 (81.2) | ||
Use of cigars no more than 5 times in last the 30 days | 998 (99.5) | ||
Use of hookah no more than 5 times in the last 30 days | 996 (96.3) | ||
No use of pipe (with tobacco, not including hookah) in the last 30 days | 995 (95.2) | ||
No use of chewing tobacco in the last 30 days | 995 (99.2) | ||
No use of dip/snuff in the last 30 days | 997 (99.4) | ||
No use of snus in the last 30 days | 1000 (99.7) | ||
No use of nicotine products (like gum, patches) in the last 30 days | 961 (95.8) | ||
No use of e-cigarettes in the last 30 days | 833 (83.1) | ||
Interested in trying an e-cigarette | 689 (68.7) | ||
Meet all tobacco use inclusion criteria | 426 (42.5) | ||
Not on cessation medication | 947 (94.4) | ||
Not breastfeeding or planning to become pregnant | 975 (97.2) | ||
No heart disease/uncontrolled blood pressure | 941 (93.8) | ||
No psychosis | 943 (94.0) | ||
No suicidal thoughts | 949 (94.6) | ||
Not enrolled currently in an alcohol treatment program | 940 (93.7) | ||
Not out of town for more than 5 nights in the next 6 weeks | 933 (93.0) | ||
18 years or older | 1002 (99.9) | ||
Reside in the DC metro area | 976 (97.3) | ||
Willing to travel to data collection site four times in three weeks | 949 (94.6) | ||
Meet all additional inclusion criteria | 916 (91.3) |
The majority of the respondents owned an iPhone or Android mobile phone (739/1003, 73.7%) and could install apps on their mobile phone (871/1003, 86.8%). Many used their mobile phones daily (898/1003, 89.5%) and had an unlimited text messaging plan (871/1003, 86.8%). Most (717/1003, 71.5%) met the mobile phone inclusion criteria, and 91.3% met all the additional inclusion criteria; however, only 42.5% (426/1003) met all the tobacco inclusion criteria. Overall, 302 individuals were eligible to participate in the study with 163 eligible after being rescreened; 117 were ultimately enrolled in the study (
This study describes the sociodemographic characteristics of individuals responding to an online screener, and examines how eligibility criteria requiring mobile phone ownership affected study eligibility. A total of 1003 individuals completed the online screener of whom 73.7% (739/1003) owned a mobile phone; of the mobile phone owners, 86.8% (871/1003) could install apps, 89.5% (898/1003) used their mobile phones daily, and 86.8% (871/1003) had an unlimited text messaging plan. While our mobile phone inclusion criteria were strict, 46.4% (333/717) of the unemployed respondents still met the criteria, suggesting that mobile phone ownership is not a major barrier to study participation, even for unemployed individuals.
These findings also have broader relevance for the tobacco control field, which increasingly uses mobile phone technology to examine tobacco product use, surveil tobacco retail environments, and deliver smoking cessation interventions [
The study utilized an online screener, which may have been a barrier to individuals who lacked internet access. Had we initially screened potential participants via telephone, we may have found that a greater proportion of respondents were ineligible because of lack of mobile phone ownership. Additionally, given that this study was conducted in Washington, DC, our findings may not be generalizable to other settings. Moreover, it is important to consider that all data collected in this study were self-reported responses to the online screener questions. Therefore, the respondents’ mobile phone and tobacco use may have been under- or over-reported owing to social desirability and recall bias.
This research suggests that using the participants’ own mobile phones for data collection, including geotracking and text messaging, was not a barrier to study participation for the majority of the respondents who took the initial online screener. Employment was related to fulfilling the mobile phone inclusion criteria; however, nearly half of the eligible respondents were unemployed. As mobile phone ownership continues to grow, researchers should consider using the participants’ own mobile phones as feasible data collection devices given their study’s target population.
ecological momentary assessment
Global Positioning System
interquartile range
Research reported in this publication was supported by the National Institute on Drug Abuse of the National Institutes of Health and FDA Center for Tobacco Products under grant number 5R21DA036472. JP’s time was also supported by the National Institute on Drug Abuse of the National Institutes of Health (NIH) and FDA Center for Tobacco Products under grant K01DA037950. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the Food and Drug Administration. At the time of the study, SS was a scholar with the HIV/AIDS, Substance Abuse, and Trauma Training Program (HA-STTP) at the University of California, Los Angeles, supported by an award from the National Institute on Drug Abuse (R25DA035692).
EH conceived the study design and the overall manuscript, led the data collection and analysis team, wrote the first draft of the paper, and worked as a research associate in the parent study. HE and YZ conducted the data analyses. SS and LR contributed to the analysis, interpretation of results, and manuscript writing. LR was involved as a research assistant in the parent study; SS was project director in the parent study, and JP was principal investigator in the parent study.
None declared.