Online sleep diaries: considerations for system development and recommendations for data management.

STUDY OBJECTIVES: To present development considerations for online sleep diary systems that result in robust, interpretable, and reliable data; furthermore, to describe data management procedures to address common data entry errors that occur despite those considerations. METHODS: The online sleep diary capture component of the Sleep Healthy Using the Internet (SHUTi) intervention has been designed to promote data integrity. Features include diary entry restrictions to limit retrospective bias, reminder prompts and data visualizations to support user engagement, and data validation checks to reduce data entry errors. Despite these features, data entry errors still occur. Data management procedures relying largely on programming syntax to minimize researcher effort and maximize reliability and replicability. Presumed data entry errors are identified where users are believed to have incorrectly selected a date or AM versus PM on the 12-hour clock. Following these corrections, diaries are identified that have unresolvable errors, like negative total sleep time. RESULTS: Using the example of one of our fully-powered, U.S. national SHUTi randomized controlled trials, we demonstrate the application of these procedures: of 45,598 total submitted diaries, 487 diaries (0.01%) required modification due to date and/or AM/PM errors and 27 diaries (<0.001%) were eliminated due to unresolvable errors. CONCLUSION: To secure the most complete and valid data from online sleep diary systems, it is critical to consider the design of the data collection system and to develop replicable processes to manage data. CLINICAL TRIAL REGISTRATION: Sleep Healthy Using The Internet for Older Adult Sufferers of Insomnia and Sleeplessness (SHUTiOASIS); https://clinicaltrials.gov/ct2/show/NCT03213132; ClinicalTrials.gov ID: NCT03213132.

Development of a Digital Behavioral Intervention to Reduce the Consumption of Sugar-Sweetened Beverages Among Rural Appalachian Adults: Multiphased, Human-Centered Design Approach.

BACKGROUND: To avoid the low engagement and limited efficacy of digital behavioral health interventions, robust human-centered design (HCD) processes are needed. OBJECTIVE: The primary objective of this study was to describe a flexible, step-by-step HCD process to develop digital behavioral health interventions by illustrating iSIPsmarter as an example. iSIPsmarter is a digital intervention for reducing the consumption of sugar-sweetened beverages (SSBs) that comprises 6 internet-based cores metered out over time to deliver the program content, an integrated SMS text message strategy to engage users in reporting SSB behaviors, and an electronic cellular-enabled scale for in-home weighing. The secondary objective is to illustrate the key components and characteristics of iSIPsmarter that resulted from the HCD process. METHODS: The methods were guided by the Model for Internet Interventions and by best practices in HCD and instructional design processes (eg, rapid prototype development and think-aloud protocol). The 3-phased (ie, contextual, prototype testing, end user testing phases) process followed in this study included a series of 13 semistructured one-on-one interviews with 7 advisory team participants from the targeted Appalachian user group. The interviews were content coded by 2 researchers and then deductively coded to the suggested areas of digital behavioral health interventions. RESULTS: The participants provided rich perspectives pertaining to iSIPsmarter's appearance, behavioral prescriptions, burdens, content, delivery, message, participation, and assessment. These inputs included requests for built-in flexibility to account for varying internet and SMS text message accessibility among users; ideas to resolve the issues and problems encountered when using the prototypes, including those related to navigation and comprehension of content; ideas to enhance personalized feedback to support motivation and goal setting for SSB consumption and weight; and feedback to refine the development of realistic and relatable vignettes. The participants were able to interact with multiple prototype drafts, allowing researchers to capture and incorporate feedback related to the iSIPsmarter dashboard, daily SSB and weight diaries, action planning, core content, interactions, and vignettes. CONCLUSIONS: Using scientific models and established processes is critical for building robust and efficacious interventions. By applying an existing model and HCD and instructional design processes, we were able to identify assumptions and address the key areas of the iSIPsmarter intervention that were hypothesized to support users' engagement and promote behavior change. As evidenced by the rich feedback received from the advisory team members and the resulting iSIPsmarter product, the HCD methodology was instrumental in the development process. Although the final iSIPsmarter content is specific to improving SSB consumption behaviors among adults in rural areas, the intent is that this HCD process will have wide applications in the development of digital behavioral health interventions across multiple geographic and behavioral contexts.

Think-Aloud Testing of a Novel Safer Drinking App for College Students During COVID-19: Usability Study.

BACKGROUND: Hazardous alcohol consumption, and binge drinking in particular, continues to be common among college students, posing the greatest risk for their health and safety. Despite widespread exposure to evidence-based preventive interventions among US undergraduates, only modest and temporary effects on risky drinking occur. Formative studies have demonstrated that students want a more engaging intervention tool for risky drinking that can be used just in time. OBJECTIVE: The purpose of this study is to test the appeal, relevance, and perceived utility of a draft mobile app for safer student drinking at a public university in Virginia. METHODS: Undergraduate student participants tested the draft mobile app via a web-based prototype that tailors to individual feedback with hot spots that responded to their taps to mimic app functionality. They narrated their impressions, navigation, and comments in a standardized think-aloud procedure. After each round of think-aloud interviews, researchers debriefed the investigators and developers to discuss findings and brainstorm app modifications. RESULTS: Minor changes to the functionality and aesthetics would improve usability of the app (eg, option for light mode in app settings). Student testers recommended tailoring the app to the needs of college students and to aspects of the local university's drinking culture. CONCLUSIONS: Findings from this study will be synthesized with information gained from other formative work to determine the final app features. We will test the app in a pilot randomized trial to assess app use and the impact of the app on college student drinking behavior over several months.

Seabed classification and source localization with Gaussian processes and machine learning.

Workshop '97 data are employed for seabed classification and source range estimation. The data are acoustic fields computed at vertically separated receivers for various ranges and different environments. Gaussian processes are applied for denoising the data and predicting the field at virtual receivers, sampling the water column densely within the array aperture. The enhanced fields are used in combination with machine learning to map the signals to one of 15 sediment-range classes (corresponding to three environments and five ranges). The classification results after using Gaussian processes for denoising are superior to those when noisy workshop data are employed.

Study protocol for iSIPsmarter: A randomized-controlled trial to evaluate the efficacy, reach, and engagement of a technology-based behavioral intervention to reduce sugary beverages among rural Appalachian adults.

Sugar-sweetened beverage (SSB) consumption is disproportionately high among rural Appalachian adults, with intakes double the national average and nearly four times the recommended amount. This trial targets this major dietary risk factor and addresses notable gaps in the rural digital health intervention literature. iSIPsmarter is a technology-based behavior and health literacy intervention aimed at improving SSB behaviors. It is comprised of six Internet-delivered, interactive Cores delivered weekly, an integrated short message service (SMS) strategy to engage users in tracking and reporting SSB behaviors, and a cellular-enabled scale for in-home weighing. iSIPsmarter is adapted from an evidence-based intervention and is grounded by the Theory of Planned Behavior and health literacy, numeracy, and media literacy concepts. The RCT is guided by the RE-AIM framework and targets 244 rural Appalachian adults. The goal is to examine the efficacy of iSIPsmarter to reduce SSB in a two-group design [iSIPsmarter vs. static Participant Education website] with four assessment points. Changes in secondary outcomes (e.g., diet quality, weight, quality of life) and maintenance of outcomes will also be evaluated. Additional secondary aims are to examine reach and representativeness, patterns of user engagement, and cost. Two tertiary aims are exploratory mediation analyses and a systems-level, participatory evaluation to understand context for future organizational-level adoption of iSIPsmarter. The long-term goal is to sustain an effective, scalable, and high reach behavioral intervention to reduce SSB-related health inequities and related chronic conditions (i.e., obesity, diabetes, some obesity-related cancers, heart disease, hypertension, dental decay) in rural Appalachia and beyond. ClinicalTrial registry: NCT05030753.

Liar! Liar! Identifying eligibility fraud by applicants in digital health research.

Online studies enable researchers to recruit large, diverse samples, but the nature of these studies provides an opportunity for applicants to misrepresent themselves to increase the likelihood of meeting eligibility criteria for a trial, particularly those that provide financial incentives. This study describes rates of fraudulent applications to an online intervention trial of an Internet intervention for insomnia among older adults (ages ≥55). Applicants were recruited using traditional (e.g., flyers, health providers), online (e.g., Craigslist, Internet searches), and social media (e.g., Facebook) recruitment methods. Applicants first submitted an interest form that included identifying information (name, date of birth, address). This data was then queried against a national database (TransUnion's TLOxp) to determine the application's verification status. Applications were determined to be verified (i.e., information from interest form matched TLOxp report), potentially fraudulent (i.e., potential discrepancy in provided information on interest form versus TLOxp report), or fraudulent (i.e., confirmed discrepancy). Of 1766 total interest forms received, 125 (7.08%) were determined to be fraudulent. Enrollment attempts that were fraudulent were detected among 12.22% of applicants who reported learning of the study through online, 7.04% through social media, 4.58% through traditional, and 4.27% through other methods. Researchers conducting online trials should take precautions, as applicants may provide fraudulent information to gain access to their studies. Reviewing all applications and verifying the identities and eligibility of participants is critical to the integrity of online research trials.

Flow Rectification in Loopy Network Models of Bird Lungs.

We demonstrate flow rectification, valveless pumping, or alternating to direct current (AC-to-DC) conversion in macroscale fluidic networks with loops. Inspired by the unique anatomy of bird lungs and the phenomenon of directed airflow throughout the respiration cycle, we hypothesize, test, and validate that multiloop networks exhibit persistent circulation or DC flows when subject to oscillatory or AC forcing at high Reynolds numbers. Experiments reveal that disproportionately stronger circulation is generated for higher frequencies and amplitudes of the imposed oscillations, and this nonlinear response is corroborated by numerical simulations. Visualizations show that flow separation and vortex shedding at network junctions serve the valving function of directing current with appropriate timing in the oscillation cycle. These findings suggest strategies for controlling inertial flows through network topology and junction connectivity.

Computational genomics of Torque teno sus virus and Porcine circovirus in swine samples from Canada.

Emerging viral diseases include pathogens that can threaten the health of the Canadian swineherd. Anelloviruses and Circoviruses comprise of pathogens with veterinary significance. The aim of this study was to determine the genomic organization and phylogenetic relationships of Torque teno sus virus (TTsusV) and Porcine circovirus (PCV) from Canadian pig samples. Fecal and tissue specimens were collected during the winter, spring and summer of 2018. We utilized either virus- or genus-specific PCR assays to characterize the occurrence and genetic diversity of TTsusV and PCV in Canadian pigs. Pairwise comparison of all partial sequences and identity calculation was performed using MAFFT algorithm implemented in Sequence Demarcation Tool (SDT). The obtained full-length sequences were aligned using ClustalW, and phylogeny was inferred using a Maximum likelihood (ML) method by Geneious software. The PCR detection results revealed that the overall positive rate of TTsusV type-1 and type-2 was 45.6% and 32.6%, respectively. The TTsusV isolate MK990454 from Canada clustered in the subtype TTsusV1b, while the TTsusV isolate MK872392 fell in the subtype TTsusV2c, and all showed similarity to known American and Chinese isolates. In addition, our screening PCR showed that 2.7% of stool samples were positive for PCV1. Phylogenetic analysis using the full-length sequence demonstrated that PCV1 (MK872393) isolated from Quebec clustered with other Chinese PCV1 strains. Despite the far geographical distance between Canada and China, the close similarity between Canadian and Chinese TTsusV1 and 2, and PCV1 sequences may be explained by a considerable amount of pig trade between these two nations.

Modeling host-feeding preference and molecular systematics of mosquitoes in different ecological niches in Canada.

Several mosquito-borne viruses (mobovirus) cause infections in Canada. Ecological data on mosquito species and host range in Canada remains elusive. The main aim of the current study is to determine the host range and molecular systematics of mosquito species in Canada. Mosquitoes were collected using BG-Sentinel traps and aspirators at 10 trapping sites in Canada during 2018 and 2019. Mosquitoes collected were identified via morphology and molecular techniques. Mosquito sequences were aligned by MUSCLE algorithm and evolutionary systematics were drawn using MEGA and SDT software. Moreover, the source of blood meals was identified using a DNA barcoding technique. A total of 5,708 female mosquitoes over 34 different taxa were collected. DNA barcodes and evolutionary tree analysis confirmed the identification of mosquito species in Canada. Of the total collected samples, 201 specimens were blood-fed female mosquitoes in 20 different taxa. Four mosquito species represented about half (51.47%) of all collected blood-fed specimens: Aede cinereus (39 specimens, 19.11%), Aedes triseriatus (23, 11.27%), Culex pipiens (22, 10.78%), and Anopheles punctipennis (21, 10.29%). The most common blood meal sources were humans (49 mosquito specimens, 24% of all blood-fed mosquito specimen), pigs (44, 21.5%), American red squirrels (28, 13.7%), white-tailed deers (28, 13.7%), and American crows (16, 7.8%). Here, we present the first analysis of the host-feeding preference of different mosquito species in Canada via molecular techniques. Our results on mosquito distribution and behavior will aid in the development of effective mitigation and control strategies to prevent or reduce human/animal health issues in regards to moboviruses.

Seabed classification using physics-based modeling and machine learning.

In this work, model-based methods are employed, along with machine learning techniques, to classify sediments in oceanic environments based on the geoacoustic properties of a two-layer seabed. Two different scenarios are investigated. First, a simple low-frequency case is set up, in which the acoustic field is modeled with normal modes. Four different hypotheses are made for seafloor sediment possibilities, and these are explored using both various machine learning techniques and a simple matched-field approach. For most noise levels, the latter has an inferior performance to the machine learning methods. Second, the high-frequency model of the scattering from a rough, two-layer seafloor is considered. Again, four different sediment possibilities are classified with machine learning. For higher accuracy, one-dimensional convolutional neural networks are employed. In both cases, the machine learning methods, both in simple and more complex formulations, lead to effective sediment characterization. The results assess the robustness to noise and model misspecification of different classifiers.

Mosquito Species Composition and Abundance in Quebec, Eastern Canada.

Given current and projected changes in the climate, the composition of mosquito species is predicted to shift geographically with implications for the transmission dynamics of vector-borne pathogens. Many mosquito species are rarely collected in Canada and their history is poorly understood; thus assessing their potential role as vectors for pathogenesis is difficult. Mosquitoes were collected from four trapping sites in Quebec Province, Canada, from June to September during 2018 and 2019 using BG sentinel traps. From all morphologically identified female mosquitoes, at least one specimen was selected for identification confirmation using the DNA-barcoding technique. Sequences were subjected to alignment and a Neighbor-Joining (NJ) tree was created using Geneious software. In total, 2,752 female mosquitoes belonging to 20 species over five genera: including Aedes (Ae.), Anopheles (An.), Culex (Cx.), Culiseta (Cu.), Coquillettidia (Cq.) were collected. The predominant mosquito was found to be Ae. cinereus. The highest number of mosquito species was captured in July, followed by August, September, and then June. Five genera were characterized by a distinctive set of cytochrome oxidase I (COI) sequences that formed well-supported clusters in the NJ-tree. The presence of Ae.japonicus in Quebec provides an initial look at the distribution of mosquito species in eastern Canada, which may put Canadians at risk of a wider range of arboviruses.

A Pilot RCT of an Internet Intervention to Reduce the Risk of Alcohol-Exposed Pregnancy.

BACKGROUND: Preventing alcohol-exposed pregnancies (AEPs) could reduce the incidence of fetal alcohol spectrum disorders. Previous face-to-face interventions significantly reduced risk for AEP, but a scalable intervention is needed to reach more women at risk. METHODS: This study compared a 6 Core automated, interactive, and tailored Internet intervention, the Contraception and Alcohol Risk Reduction Internet Intervention (CARRII), to a static patient education (PE) website for its effect on AEP risk. Participants were recruited online to a pilot randomized clinical trial (RCT) with baseline, 9 weeks posttreatment, and 6-month (6-M) follow-up assessments. Seventy-one women completed online questionnaires and telephone interviews and were randomized to CARRII (n = 36) or PE (n = 35). Primary outcomes were rates of risky drinking, unprotected sex episodes, and AEP risk, collected from online prospective diaries. RESULTS: CARRII participants showed significant reductions in rate of unprotected sex from pretreatment (88.9%) to posttreatment (70.6%) (p < 0.04) and to 6-M follow-up (51.5%) (p = 0.001); rate of risky drinking from pretreatment (75.0%) to posttreatment (50.0%) (p < 0.02), but insignificant change from pretreatment to 6-M follow-up (57.6%) (p < 0.09); and rate of AEP risk from pretreatment (66.7%) to posttreatment (32.4%) (p = 0.001) and to 6-M follow-up (30.3%) (p = 0.005). PE participants demonstrated no significant changes on all 3 variables across all time points. Intent-to-treat group-by-time tests were not significant, but power was limited by missing diaries. Over 72% of CARRII participants completed all 6 Cores. Exploratory analyses suggest that higher program utilization is related to change. CONCLUSIONS: These data show that CARRII was acceptable, feasible, promising to reduce AEP risk, and merits further testing in a fully powered RCT.

Effect of a Web-Based Cognitive Behavior Therapy for Insomnia Intervention With 1-Year Follow-up: A Randomized Clinical Trial.

IMPORTANCE: Although cognitive behavior therapy for insomnia (CBT-I) has been established as the first-line recommendation for the millions of adults with chronic insomnia, there is a paucity of trained clinicians to deliver this much needed treatment. Internet-delivered CBT-I has shown promise as a method to overcome this obstacle; however, the long-term effectiveness has not been proven in a representative sample with chronic insomnia. OBJECTIVE: To evaluate a web-based, automated CBT-I intervention to improve insomnia in the short term (9 weeks) and long term (1 year). DESIGN, SETTING, AND PARTICIPANTS: A randomized clinical trial comparing the internet CBT-I with internet patient education at baseline, 9 weeks, 6 months, and 1 year. Altogether, 303 adults with chronic insomnia self-referred to participate, of whom 151 (49.8%) reported at least 1 medical or psychiatric comorbidity. INTERVENTIONS: The internet CBT-I (Sleep Healthy Using the Internet [SHUTi]) was a 6-week fully automated, interactive, and tailored web-based program that incorporated the primary tenets of face-to-face CBT-I. The online patient education program provided nontailored and fixed online information about insomnia. MAIN OUTCOMES AND MEASURES: The primary sleep outcomes were self-reported online ratings of insomnia severity (Insomnia Severity Index) and online sleep diary-derived values for sleep-onset latency and wake after sleep onset, collected prospectively for 10 days at each assessment period. The secondary sleep outcomes included sleep efficiency, number of awakenings, sleep quality, and total sleep time. RESULTS: Among 303 participants, the mean (SD) age was 43.28 (11.59) years, and 71.9% (218 of 303) were female. Of these, 151 were randomized to the SHUTi group and 152 to the online patient education group. Results of the 3 primary sleep outcomes showed that the overall group × time interaction was significant for all variables, favoring the SHUTi group (Insomnia Severity Index [F3,1063 = 20.65, P < .001], sleep-onset latency [F3,1042 = 6.01, P < .001], and wake after sleep onset [F3,1042 = 12.68, P < .001]). Within-group effect sizes demonstrated improvements from baseline to postassessment for the SHUTi participants (range, Cohen d = 0.79 [95% CI, 0.55-1.04] to d = 1.90 [95% CI, 1.62-2.18]). Treatment effects were maintained at the 1-year follow-up (SHUTi Insomnia Severity Index d = 2.32 [95% CI, 2.01-2.63], sleep-onset latency d = 1.41 [95% CI, 1.15-1.68], and wake after sleep onset d = 0.95 [95% CI, 0.70-1.21]), with 56.6% (69 of 122) achieving remission status and 69.7% (85 of 122) deemed treatment responders at 1 year based on Insomnia Severity Index data. All secondary sleep outcomes, except total sleep time, also showed significant overall group × time interactions, favoring the SHUTi group. CONCLUSIONS AND RELEVANCE: Given its efficacy and availability, internet-delivered CBT-I may have a key role in the dissemination of effective behavioral treatments for insomnia. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT01438697.

Influences of volitional and forced intentions on physical activity and effort within the theory of planned behaviour.

In the present study, we examined the utility of volitional and forced intentions in predicting participation in physical activities and effort within the theory of planned behaviour. Four hundred and forty-four participants (184 males, 260 females) aged 19.1 +/- 3.3 years (mean +/- s) completed self-report measures of intentions, attitudes, subjective norms, perceived behavioural control, volitional intentions, forced intentions, and past behaviour in a physical activity context. Six weeks later, they completed self-report measures of physical activity behaviour and effort. Results indicated that volitional intentions and forced intentions contributed to the prediction of effort over and above intentions, attitudes, subjective norms, perceptions of control, past behaviour, and the product terms of attitudesxintentions and subjective normsxintentions. Volitional intentions and forced intentions did not predict participation in physical activities over and above effort. We concluded that volitional intentions and forced intentions assist in the explanation of effort in the context of physical activity behaviour.

Color naming, lens aging, and grue: what the optics of the aging eye can teach us about color language.

Many languages without separate terms for green and blue are or were spoken in locations receiving above-average exposure to ultraviolet-B (UV-B) radiation. It has been proposed that this correlation is caused by premature lens aging. This conclusion was supported by an experiment in which younger observers used the term "blue" less often when they described simulated paint chips filtered through the equivalent of an older observer's lens-removing much short-wavelength light-than when they described the unfiltered versions of the same paint chips. Some stimuli that were called "blue" without simulated aging were called "green" when filtered. However, in the experiment reported here, we found that the proportion of "blue" color-name responses did not differ between younger subjects and older observers with known ocular media optical densities. Color naming for stimuli that were nominally green, blue-green, or blue was virtually identical for older and younger observers who viewed the same (unfiltered) stimuli. Our results are inconsistent with the lens-brunescence hypothesis.

Recognition memory for source and occurrence: the importance of recollection.

Previous recognition memory studies indicate that when both recollection and familiarity are expected to contribute to recognition performance (e.g., discriminating studied items from nonstudied items) the dual-process and the unequal-variance signal detection models provide comparable accounts of performance. When familiarity is not expected to be useful (e.g., when items from two equally familiar sources are discriminated between), the dual-process model provides a significantly better account of performance. In the present study, source recognition was tested under conditions in which familiarity could have been used to perform a list-discrimination task; participants were required to discriminate between strong studied items, weak studied items, and new items. The dual-process model provided a better account of performance than did the unequal-variance model. Moreover, the results indicated that the unequal-variance assumption in a single-process signal detection model was not a valid substitution for recollection and that recollection was used to make recognition judgments even when assessments of familiarity were useful.

Separating sensitivity from response bias: implications of comparisons of yes-no and forced-choice tests for models and measures of recognition memory.

A fundamental challenge to psychological research is the measurement of cognitive processes uncontaminated by response strategies resulting from different testing procedures. Test-free estimates of ability are vital when comparing the performance of different groups or different conditions. The current study applied several sets of measurement models to both forced-choice and yes-no recognition memory tests and concluded that the traditional signal-detection model resulted in distorted estimates of accuracy. Two-factor models were necessary to separate memory sensitivity from response bias. These models indicated that (a) memory accuracy did not differ across the tests and (b) the tests relied on the same underlying memory processes. The results illustrate the pitfalls of using a single-component model to measure accuracy in tasks that reflect 2 or more underlying processes.