RESEARCH METHODS: All panelists will receive a unique numeric participant ID from MTurk. Once the panel is confirmed, all 300 panelists will complete a health questionnaire that assesses demographics, general health, physical activity, health tracking processes, and consumer technology. Participants will be asked to enter their unique participant ID into a text box at the start of the questionnaire. At the end of the questionnaire, participants will be queried for their willingness to allow researchers to download their Fitbit data. All varieties and models of Fitbit will be allowed in this study.

Upon consent to Fitbit data access, participants will be routed to a third-party data service provider called Fitabase LLC (San Diego, California). Using the Fitbit application programming interface, third-party services such as Fitabase can access and aggregate self-tracker data. Fitabase provides researchers with a connection to the Fitbit infrastructure to support data collection. The research team will generate unique Fitabase links for each participant. When respondents reach the end of the self-administered survey, they will click on the link to Fitabase that corresponds to their participant ID. Upon completion of the Fitabase sign-up, participants will be given $10 via MTurk Prime's customer service team. Participants who complete the questionnaire but do not sign up with Fitabase will not receive the $1 incentive. Thus, participants who successfully complete the survey and donate their Fitbit data will receive an incentive totaling $11. We will contract with Fitabase to provide the research team with 30 days of retrospective data from the date of sign-up. Fitabase will be used to extract daily and intraday data from the linked Fitbit accounts. These variables include daily-level data on total steps, distance, calories burned, total sleep time, and daily active versus sedentary time. We will also obtain hourly-level data on calories burned, active versus sedentary time, heart rate, sleep, and step counts. The most granular output, intraday data, will include minute-level step counts. These data will be downloaded as both raw and aggregated files by day, per person. Participants can refuse to participate or cancel registration at any time. Cancellation via MTurk will remove their survey data from the study. In cases of cancellation after data collection is complete, participants will be provided with instructions on how to contact RTI staff to remove their Fitbit data.

We have established an extensive protocol to protect participants’ privacy. We will keep informed consent statements and participant data in separate locked files cabinets so that individuals are not easily connected to the study results. All sensor data will be stored on a firewall and password-protected project server at the University of California, San Diego. To reduce the risk to (and fear of) confidentiality, no identifying data will be collected in the field, only in a secure research laboratory. All subject records and data will be stripped of individual identifiers following data collection. We will assign each person a study ID and all records will be coded with the study ID rather than personal identifiers. The code that links the study ID and the name will be stored in a separate place than the data file until all of the measurements are complete. At that point, the personal information will be destroyed and all analyses will be conducted with the data set that has no personal identifiers. All data will be kept in locked cabinets at the study office, accessible only by investigators and project staff. Participants will also be told about the confidentiality procedures and that they have the right to refuse to answer questions or to terminate their participation in the study at any time without prejudice. If a participant decides to terminate participation all data collected on that individual will be destroyed. In order to protect the confidentiality of third parties, we treat all data with the same data security measures as PHI data. As image data is annotated by research staff, images that show places where a reasonable expectation of privacy is accepted (ie. banks, restrooms, locker rooms, hospitals, and medical and treatment centers) will be deleted. After image data is completely annotated, the image files will be destroyed and only the annotations will be used in data analysis Finally, participant data will not be sold or exchanged with anyone and data sharing as per NIH requirements will be performed within strict protocol-driven procedural guidelines. The PI will assure that all human protections standards are met.

Specific Aims:

• Determine the relationship between loneliness and social media use.
• Determine the relationship between personality measures and social media use.
• Determine the relationship between self-reported social media use and actual social media use.

Methods, Materials and Analysis
Method 1: Survey will assess a) personality and loneliness variables, and b) social media use.
Method 2: Phone-tracking (trace) data will record how long participants spend using social media platforms for one month. The company that runs the software is RescueTime (https://www.rescuetime.com/)

1. First, in class I will introduce the study and explain that it consists of a survey and phone tracking.
2. I will walk them through the procedure, in class, for how they download the phone tracking software and enter the code assigned with my study, which will be determined by RescueTime the week of. The software will record how many seconds the participant spends on 5 social media apps (Facebook, Snapchat, Twitter, Instagram, and SMS texting) for 1 month. The data will all be anonymous and the phone tracking application will run in the background, so the participants will not receive any notifications or updates until the month is over.
3. Then that evening I will email the participants a link that evening and they will take the survey on Qualtrics, which should take between 10 and 15 minutes.
4. I will analyze the survey data using SPSS at some time during the month.
5. When the month is over, I will email the participants telling them they can delete the RescueTime software from their phones.
6. I will then analyze the phone-tracking data in SPSS, and compare it to the survey data.

All scales are standardized and currently used in social science literature. They are in the Survey Instrument document.

The data will all be anonymized and quantitatively analyzed in SPSS. I will run correlations, regressions, ANOVAS, and test for moderation and mediation with various dependent variables. The data will be used in my dissertation. The data will be paired between qualtrics and rescuetime by participant last name and randomly assigned a number. Once the two data sets are combined into one, the columns that contain any identifying information (such as last name or email) will be deleted from the data, leaving only the information I want. The emaill addresses will also be cut out and put into a new solo document, which I will use to email participants when the month is over. All participants (not just ones who completed) will be emailed that the study is over and they may delete the application from their phone.

Overview:

Our overall objective is to identify the contextual and dynamic risk factors associated with meth use and Antiretroviral Therapy (ART) nonadherence and to develop a machine learning algorithm to help define windows of opportunity to deliver interventions promoting relapse prevention and ART adherence in later studies. In order to identify the risk factors to ART nonadherence and meth use, we shall use a mixed methods approach. In particular, 50 People Living with HIV (PLWH) with comorbid meth use shall participate in individual interviews, complete baseline assessment questionnaires, provide daily responses to brief smartphone surveys over a period of 4 weeks, and provide location and movement data passively collected through electronic GPS devices over the 4-week survey period.

One obstacle to establishing a full understanding of the temporal associations and potential mechanisms between variables, such as meth use, ART adherence, and daily events, has been limitations in observing daily life “as it is lived”. As such, ecological momentary assessment (EMA) has been proposed as an ambulatory data collection technique that, through querying present moment experience multiple times throughout the day, researchers are more closely able to assess temporal relationships between daily life experiences, thoughts, feelings, and behaviors. Consequently, we propose to use EMA via smartphone-delivered surveys to more accurately record the daily events of HIV+ individuals, their experience of stressors, mood, and cravings for substance use and determine their associations with risk for meth use and nonadherence.

Furthermore, with the use of electronic GPS devices, we propose to use contextual data, such as location and activity, to understand how location and spatial patterns of movement are associated with risk for meth use and nonadherence. The novel use of global positioning system (GPS) technology overcomes conceptual and logistical challenges faced by relying on static representations of context, such as a residential address, or inaccurate self-reporting of daily movements when examining the relationship between context and health. Indeed, examinations of GPS information has already begun to elucidate important relationships in studies of HIV and drug use [4, 5].

In this study, variables, such as the number of walking trips taken and time spent at home shall be derived from GPS data. These environmental measures created from GPS analysis will be used in statistical modeling at the group level only, and will not be individually mapped or visualized.

Study Components:

I. Screening Interview

Recruitment staff will conduct a brief phone screen to determine preliminary eligibility. A phone screen is detailed in the attached documents in this IRB application for participants who have previously participated in a research study at the HIV Neurobehavioral Research Program (HNRP) (IRB #081375, #091062, #110089, #160075, #171024). These participants indicated in their signed consent document that they wished to be considered for future research studies.

II. Initial Visit and Individual Interview (up to 3 hours):

The initial visit at the HNRP shall include the completion consent documents, collection of background information, and completion of standard psychosocial measures (many of which are standard at the HNRP and have previously been used in other studies. Specifically, the initial visit battery shall include some of the following measures: Demographics and social background, the MacArthur scale of subjective social status, Ira Wilson Questionnaire, Beliefs related to medications, Supports and barriers to adherence, Promis measures of self-efficacy for managing medications, self-efficacy for managing emotions, and social support, the Perceived stress scale, Meth Reasons for Using, Meth Outcome Expectancies, the Risk assessment battery, the DAST-10, the Functional Assessment of Health Inventory (FAHI), Beck anxiety index (BAI), Beck depression index (BDI), the Adult Hope Scale, the HIV Stigma Scale, and the Smartphone operability scale.

These assessments shall take 1 hour to complete and are included as appendices to this application. This visit shall also include participation in a recorded individual interview, covering topics such as supports and barriers to ART adherence and sobriety, lasting about 1 hour. The script for the individual interview is also included as an attachment to this application.

Additionally, participants shall be oriented to the daily survey phase of the study (which shall involve the completion of two daily smartphone surveys and location/activity tracking) and be familiarized with use of cellphones for daily surveys.

III. Daily EMA Surveys and Passive Collection of Location Data (4 weeks)

Participants will complete two brief daily EMA surveys using either their own mobile phones or provided study smartphones over a period of 4 weeks. Survey links shall be sent out to participants twice daily (morning and evening) at times set to the participants’ preferences. Surveys shall be delivered to the smartphones via a text message link which, when followed, directs to a secure online survey created and run on the SurveyMonkey platform.

These weblinks, delivered by text message, shall be generated to be specific to each participant and data collected through SurveyMonkey shall only have study ID as a participant identifier (participants will also have to enter a unique study id in order to access the surveys). As such, participant responses shall be stored without any identifying information and IP addresses shall not be collected.

The first daily survey shall be completed in the morning and the second in the evening. The morning survey shall take 3 minutes to complete and shall include questions on sleep quality, sleep duration, and mood. The evening survey shall take 5 minutes to complete and shall briefly assess mood, recent meth and other substance use, cravings, daily events, sexual risk behaviors, and ART adherence. As the daily surveys shall be completed online through a secure survey-site, participant responses (including their responses of substance use) shall not be stored locally on the mobile phones. The daily survey items are included as attachments to this IRB application.

In addition to daily surveys, we shall also collect information on activity and GPS location, measured by an electronic GPS device. Participants shall be instructed to wear a GPS device, similar in size and weight to a small cell phone (the QStarz BT-Q1000XT) on a belt around their waist during waking hours (except while bathing or swimming) and to recharge the devices while sleeping. This GPS device shall collect latitude and longitude coordinates as well as the date and time of measurements. To minimize intrusion into daily life and potential embarrassment, participants will be told that they can wear the GPS device under or over their clothing. Such GPS devices have been used by other researchers at UCSD interested in studying the effect of geographical context on physical activity (IRB Protocol #140510 and #150336) and by researchers at other institutions interested in exploring the role of location and movement on substance abuse risk [5].

The daily surveys shall enquire about illicit substance use and location data may capture where substances are obtained/used; however, as these two forms of data shall be collected using separate strategies, in the unlikely event that data are intercepted, these data cannot be linked to each other by third parties. Any potential risk of embarrassment or loss of privacy due to analysis with GPS data shall be mitigated by deriving variables from the GPS data, like the number of places visited or time spent at home. Additionally, environmental measures created from the GPS analysis will be used in statistical modeling at the group level only, and will not be individually mapped or visualized for publication purposes. Furthermore, GPS coordinate data shall not be analyzed at a level finer than that of the census tract. Study variables to be created from GPS data are included in the appendices to this application. The risks of loss of privacy posed by completion of daily surveys and location tracking are anticipated to be no greater than those inherent to daily use of the Internet and of use of GPS devices or devices that allow location services.

IV. Follow-up Visit (1 hour)

Participants will return for a second and final visit at the HNRP after the 4-week survey period. During this visit, participants will complete the feedback questionnaire and return the GPS devices as well as any loaned study phones.