Denyer et al. BMC Psychiatry (2022) 22:813 https://doi.org/10.1186/s12888-022-04429-6 STUDY PROTOCOL Open Access ADHD Remote Technology study of cardiometabolic risk factors and medication adherence (ART‑CARMA): a multi‑centre prospective cohort study protocol Hayley Denyer1* , J Antoni Ramos‑Quiroga2,3,4,5, Amos Folarin6,7,8,9,10, Carolina Ramos3, Petra Nemeth11, Andrea Bilbow12, Euan Woodward13, Susannah Whitwell14, Ulrich Müller‑Sedgwick15,16, Henrik Larsson17,18, Richard JB Dobson6,7,8,9,10   and Jonna Kuntsi1    Abstract Background: Emerging evidence points at substantial comorbidity between adult attention deficit hyperactivity dis‑ order (ADHD) and cardiometabolic diseases, but our understanding of the comorbidity and how to manage cardio‑ metabolic disease in adults with ADHD is limited. The ADHD Remote Technology study of cardiometabolic risk factors and medication adherence (ART-CARMA) project uses remote measurement technology to obtain real-world data from daily life to assess the extent to which ADHD medication treatment and physical activity, individually and jointly, may influence cardiometabolic risks in adults with ADHD. Our second main aim is to obtain valuable real-world data on adherence to pharmacological treatment and its predictors and correlates during daily life from adults with ADHD. Methods: ART-CARMA is a multi-site prospective cohort study within the EU-funded collaboration ‘TIMESPAN’ (Man‑ agement of chronic cardiometabolic disease and treatment discontinuity in adult ADHD patients) that will recruit 300 adults from adult ADHD waiting lists. The participants will be monitored remotely over a period of 12 months that starts from pre-treatment initiation. Passive monitoring, which involves the participants wearing a wrist-worn device (EmbracePlus) and downloading the RADAR-base Passive App and the Empatica Care App on their smartphone, provides ongoing data collection on a wide range of variables, such as physical activity, sleep, pulse rate (PR) and pulse rate variability (PRV), systolic peaks, electrodermal activity (EDA), oxygen saturation (SpO2), peripheral tempera‑ ture, smartphone usage including social connectivity, and the environment (e.g. ambient noise, light levels, relative location). By combining data across these variables measured, processes such as physical activity, sleep, autonomic arousal, and indicators of cardiovascular health can be captured. Active remote monitoring involves the participant completing tasks using a smartphone app (such as completing clinical questionnaires or speech tasks), measuring their blood pressure and weight, or using a PC/laptop (cognitive tasks). The ART system is built on the RADAR-base mobile-health platform. *Correspondence: hayley.denyer@kcl.ac.uk 1 Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, 16 De Crespigny Park, SE5 8AF London, UK Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Denyer et al. BMC Psychiatry (2022) 22:813 Page 2 of 13 Discussion: The long-term goal is to use these data to improve the management of cardiometabolic disease in adults with ADHD, and to improve ADHD medication treatment adherence and the personalisation of treatment. Keywords: Attention deficit hyperactivity disorder, ADHD, Cardiovascular disease, Medication adherence, mHealth, Remote measurement technology, Remote monitoring, Digital phenotyping Background report on adherence and discontinuation results [25]. A Attention deficit hyperactivity disorder (ADHD) is a systematic review found that of 91 observational stud- common psychiatric disorder, with a prevalence among ies only eleven studies reported adherence or discon- adults of 2.5% [1]. The disorder is diagnosed based on tinuation results in adults with ADHD [25]. Side effects impairing levels of inattentive, hyperactive and impulsive reported to ADHD medication include changes to sleep behaviours. Most adults with ADHD present with addi- behaviours (such as reduced total sleep time and delayed tional serious mental health problems that can include onset of sleep [26]), reduction in appetite and weight [27, anxiety, depression, emotional instability, antisocial 28]) and hypertension [10, 13, 28]. RCTs or observational behaviour, substance use and autism spectrum disor- studies that rely exclusively on parent/caregiver surveys der [2]. Emerging evidence further points at substantial often do not provide longer-term information about how comorbidity and shared genetics between adult ADHD side effects may influence ADHD medication treatment and cardiometabolic diseases (obesity, type-2 diabetes discontinuity [10, 29]. To improve our understanding of mellitus and cardiovascular disease). However, detailed the reasons for treatment discontinuity in adults with knowledge about the screening, diagnosis and clinical ADHD, moment-by-moment quantification of real-world management of adults with ADHD and co-occurring car- data from the individuals’ daily lives including both active diometabolic disease is lacking. (e.g., completing questionnaires) and passive (e.g., wear- The prevalence of ADHD medication use for adults has able device) monitoring is needed. increased markedly during the last two decades [3–5]. Remote measurement technology (RMT) offers new Randomised control trials (RCTs) on ADHD have shown opportunities for collecting detailed, frequent, long-term, short-term beneficial effects of ADHD medication on real-world data on clinical symptoms, functional and the core symptoms of ADHD [6–8]. However, concerns cognitive impairments, physiological measures, health about the safety of ADHD medications for serious cardi- behaviours (such as exercise and sleep) and measures ometabolic outcomes have been raised due to reports of of the environment on large sample sizes. The ADHD cardiovascular disease (CVD) events, including sudden Remote Technology (‘ART’) assessment and monitoring cardiac death, myocardial infarction and stroke in indi- system for adults with ADHD, developed by Kuntsi, Dob- viduals using ADHD medications [9–11]. ADHD medica- son and colleagues [30, 31] utilises this technology and tions may also elevate heart rate (HR) and blood pressure consists of both active and passive monitoring. Active [12, 13]. Despite the evidence from the RCTs on short- remote monitoring involves the participant completing term beneficial effects, there is a paucity of longer-term, tasks using a smartphone app (such as completing clini- real-world data on the impact of ADHD medication on a cal questionnaires or speech tasks) or using a PC/laptop wider range of outcomes, as well as of such effects when (cognitive tasks). Passive remote monitoring consists of combined with a non-pharmacological intervention, ongoing data collection using smartphone sensors and such as physical activity. The therapeutic effects of physi- a wearable device without active input from the partici- cal activity have been described extensively for several pant; data can be collected on a wide range of variables, psychiatric and cardiometabolic conditions, including such as physical activity, sleep, pulse rate (PR) and pulse depression, sleep problems, obesity, type-2 diabetes mel- rate variability (PRV), systolic peaks, electrodermal activ- litus and CVD [14–18]. There is also emerging evidence ity, oxygen saturation (SpO2), peripheral temperature, for positive effects of physical activity on ADHD [19–24], smartphone usage including social connectivity, and but long-term real-world data are needed, in particular the environment (e.g. ambient noise, light levels, rela- among adults. tive location). The ART system is underpinned by the Possible reasons for treatment discontinuity in adults RADAR-base open-source mobile-health platform [32, with ADHD include lack of treatment response, adverse 33]. side effects and poor adherence behaviours (such as for- This protocol describes the ADHD Remote Technol- getting to take the medication). Although sub-optimal ogy study of cardiometabolic risk factors and medication response is one of the most common reasons reported adherence (‘ART-CARMA’), which uses the ART system. for ADHD medication discontinuation, few studies ART-CARMA is a clinical study within the EU-funded Denyer et al. BMC Psychiatry (2022) 22:813 Page 3 of 13 collaboration ‘TIMESPAN’ (Management of chronic car- TIMESPAN partners Andrea Bilbow and Euan Wood- diometabolic disease and treatment discontinuity in adult ward. Andrea Bilbow, a patient advocate for people with ADHD patients), which unites a multidisciplinary team ADHD, founder and Chief Executive of Attention Defi- of 17 partners from academia, small and medium-sized cit Disorder Information Services (ADDISS), has pro- enterprises (SMEs), patients and care providers. The vided input to the ADHD remote measure development main objective of TIMESPAN is to advance the manage- via the wider ART research programme (on which she is ment of adult ADHD and co-occurring cardiometabolic an external collaborator). She has provided further input disease by improving the identification and treatment of into the ART-CARMA study protocol specifically during individuals with these comorbidities. The wearable device the initial planning stage. Mr Euan Woodward, execu- included in ART-CARMA is the EmbracePlus developed tive director of the European Association for the Study of by the SME partner Empatica [34]. ART-CARMA will Obesity (EASO) has been involved in the project outline. use RMT in adults with ADHD to carry out unobtrusive, real-time data collection over a period of 12 months. By recruiting 300 adults from adult ADHD clinic waiting Study objectives lists and monitoring them remotely, we will obtain objec- The first main aim (objectives 1–3 below) is to obtain tively measured data relevant to cardiometabolic risk real-world data from the patient’s daily life on the extent profiles from the patient’s daily life, both on and off phar- to which ADHD medication treatment and physical macotherapy. By targeting the informative period from activity, individually and jointly, may influence cardio- pre-treatment initiation through to treatment initiation, metabolic risks in adults with ADHD. This will provide titration and the subsequent period, up to 12 months in new insights into disease patterns and help improve the total, we will obtain real-time data on multiple param- safety and effectiveness of pharmacological (i.e., ADHD eters, including side effects, that aims to inform future medication treatment) and non-pharmacological (i.e., personalisation of treatment. physical activity) interventions for patients with ADHD and co-occurring cardiometabolic disease. The second Patient involvement in the ART‑CARMA study protocol main aim (objectives 4–6 below) is to obtain in vivo, real- The ADHD Remote Technology (ART) system benefits world data from the patient’s daily life on adherence to from procedures developed in the RADAR-MDD study pharmacological treatment and its predictors and corre- protocol [35]; a RMT study that recruited over 600 indi- lates, over a remote monitoring period of 12 months that viduals with major depression disorder (MDD), over a starts from pre-treatment initiation. The long-term goal 24-month remote monitoring period. RADAR-MDD is to use these data to improve the management of cardi- underwent extensive development in collaboration with ometabolic disease in adults with ADHD, and to improve service-users to promote the design of acceptable and ADHD medication treatment adherence and the person- feasible RMT tools [36]. In addition, study participants alisation of treatment. from our initial ART pilot study provided feedback on ART-CARMA has six specific objectives: the acceptability of measures and procedures piloted Objective 1: Use RMT to identify real-world conse- in that study [31]. In ART-pilot, 95% of the participants quences of ADHD medication treatment on the fol- agreed that there is value in gathering data through RMT, lowing cardiometabolic risk (or protective) factors in with some participants explaining passive monitoring adults with ADHD: changes in HR, blood pressure, provides a unique opportunity to collect objective data. weight, smoking, alcohol use, diet, sleep and physi- All participants with and without ADHD (100%) agreed ological stress response. that the study measures blended into their daily life, and Objective 2: Use RMT to measure the potential con- 82% of participants were positive about using the system tribution of physical activity to a reduction in these for at least one year. We have incorporated improve- risk factors, independently and in combination with ments to the ART-CARMA study design that were sug- ADHD medication treatment. gested by the participants in ART-pilot, such as longer Objective 3: Use machine learning, applied to the intervals between the administration of the cognitive full set of multimodal remote monitoring data from tasks during the remote monitoring period. We also ran the 12-month period, to build prediction models of a service-user focus group that further informed this the cascade of effects that follow from the initiation study design regarding the frequency of questionnaire of ADHD medication treatment. This allows us to administration and the acceptability of additional meas- go beyond objectives 1 and 2, to obtain insight into ures included in ART-CARMA that were not included potential underlying mechanisms that are needed in ART-pilot. We have incorporated suggestions from for developing algorithms for personalised treat- Denyer et al. BMC Psychiatry (2022) 22:813 Page 4 of 13 ment. The prediction models allow us to assess com- scenarios, a sample size of approx. 250 is required for plex, temporal relationships between multiple meas- stable estimates [38]. These considerations, together with ures. the evidence that has accumulated from the RADAR- Objective 4: Use RMT to explore ADHD medication MDD remote monitoring study regarding potential chal- use patterns in adults with ADHD in the first year lenges with RMT protocols (e.g., attrition, missing data) of starting ADHD medication treatment (including ([39]), have led to the proposed sample size of 300. The treatment initiation, titration and the subsequent eligibility criteria for inclusion in this study are summa- period) and to quantify the extent of treatment rised in Table 1. adherence, non-adherence and treatment gaps. Objective 5: Use RMT to identify potential rea- Study procedures sons for non-adherence, with a focus on adverse Participants will be recruited over twenty-four months side effects (e.g. daytime sleepiness and changes in across two data collection sites of ART-CARMA at King’s electrodermal activity indexed arousal/alertness, College London (UK) and Vall d’Hebron Research Insti- changes in heart rate or blood pressure, changes in tute (Spain). In total, across the two sites, we will recruit appetite/weight, increased anxiety/depressive symp- 300 adults from adult ADHD clinic waiting lists, which toms or irritability, not feeling themselves/feeling enables an initial off-medication remote monitoring their personality changes when on medication, tics, period and subsequent continuous remote monitor- dizziness, sweating, headaches), lack of treatment ing following initiation of ADHD medication treatment. response (limited or no effect on ADHD symptoms The schedule of events is available in Table 2. The flow and impairments) and poor adherence behaviours of participants through the study, including reasons for (e.g. forgetting to take or difficulties taking ADHD not participating in the study, will be standardised across medication). We capture these time series data sites and documented in a flowchart (Fig. 1). across the periods of pre-treatment, treatment ini- Recruitment procedures will vary slightly across the tiation, titration and subsequent monitoring. two sites. Adults will be recruited who are on the wait- Objective 6: Identify psychiatric comorbidities that ing list and are not yet receiving ADHD medication. In predict treatment non-adherence or increase adverse the UK/British Isles, the clinical team will contact the effects. We will measure symptoms of autism spec- individuals by their preferred method of contact (post, trum disorder, antisocial behaviour, anxiety, depres- email, or telephone) and will provide them with informa- sion and substance use. tion about the study. If the patient is interested in taking part, the first baseline assessment with the research team will be scheduled. The participant information sheet and Methods informed consent form will be sent to patients a mini- Study design mum of 24 h before the baseline assessment. In Spain, ART-CARMA is a clinical study being undertaken as part individuals will be provided with information about the of a larger body of work by the Management of chronic study on the day of their ADHD assessment. If the patient cardiometabolic disease and treatment discontinuity in is interested in taking part, the baseline assessments will adult ADHD patients (TIMESPAN [37]) research con- be completed on the same day. Written informed consent sortium. Below is a summary of the methods used in will be obtained. ART-CARMA. The baseline assessments will happen either in-person ART-CARMA is a prospective observational non-ran- or virtually using Microsoft Teams depending on the site domised, non-interventional study, using wearable tech- they have been recruited from. In the UK/British Isles, nology and smartphone sensors, representing no change participants will be paid £30 at the point of enrolling in to the usual care or treatments of participants due to par- the study, and a further £5 for every completed outcome ticipation. There is no control comparison group, or ran- measure (paid at months 6 and 12). Participants will be domised group allocation. paid an additional £30 at the end of the study period (month 12). In Spain, participants will not receive any Study population financial compensation for their participation. Power calculations (GPower 3.1) for a regression analy- The baseline assessments will take approximately sis with five predictors indicate that a sample size of 280 four hours to complete. In Spain, the two in-person has 80% power (alpha 0.01) to detect a small effect size baseline assessment sessions will be completed in one (f2) of 0.07. Published Monte-Carlo simulations that day. In the UK/British Isles, the virtual baseline assess- aimed to determine the necessary sample size to achieve ment sessions will be completed across two days, with stable estimates for correlations suggest that, in typical approximately a week in between each session. Both Denyer et al. BMC Psychiatry (2022) 22:813 Page 5 of 13 Table 1 Eligibility criteria for participation in ART-CARMA ADHD group inclusion criteria Diagnosis of DSM-5 ADHD Aged 18–60 Able to give informed consent for participation Fluent in English Willing and able to complete self-reported assessments via smartphone Willing to use either their own Android phone or a study Android phone as their only smartphone during the data collection period Willing to wear the wearable device (EmbracePlus) during the data collection period Not on ADHD medication at the time of recruitment ADHD group exclusion criteria Psychosis, currently experiencing a major depressive episode, mania, drug dependence over the last six months or a major neurological disorder Recent contact with psychiatric acute care (admission, crisis team or liaison team (A&E)) in the last six months Any other major medical disease which might impact upon the patient’s ability to participate in normal daily activities (e.g., due to hospitalisations) Does not start ADHD medication following ADHD diagnosis (either due to personal choice or psychiatrist deciding not to prescribe ADHD medication) Pregnancy IQ < 70 Partner, family member or A parent, partner or a close friend, as chosen by the participant with ADHD close friend group inclusion Aged 18 or over criteria Willing and able to complete web-based questionnaires regarding the participant with ADHD DSM-5 The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, IQ intelligent quotient the in-person and virtual baseline sessions will consist telephone) and invite him/her to complete the question- of (1) the Diagnostic Interview for ADHD in Adults naires every four weeks. (DIVA [40]) to confirm an ADHD diagnosis, (2) a brief IQ and short-term working memory test [41, Remote data collection 42], (3) Mini International Neuropsychiatric Inter- The remote monitoring starts immediately following the view (MINI [43]) to establish psychiatric comorbidi- research baseline assessment and will last for 12 months. ties, (4) two baseline questionnaires: a demographic The remote monitoring consists of both active and pas- About You questionnaire and the Autism Spectrum sive components, following procedures established for Quotient questionnaire (AQ-10 [44]), (5) two cognitive the wider ART research programme, and is linked to the tasks of the Combined cued continuous performance RADAR-base mobile-health platform [32]. Below is a test and Go/NoGo (CPT/GNG) task and the Fast task summary of the methods of remote data collection used (while researcher present), (6) completion of question- in the ART-CARMA study (Fig. 2). naires and physical measurements of weight, waist cir- cumference and blood pressure, (7) training session Active App on the smartphone, wearable device and blood pres- Participants will be asked to download the RADAR- sure machine, and (8) guidance on how to collect DNA base Active App that requests the participant to com- saliva sample and microbiota sample (optional). Partici- plete questionnaires at specific intervals. Every day pants will be able to take breaks during the session if participants will be asked to complete a short question- required. naire for medication use on whether the participant During the baseline session, the participants with took their medication that day, with follow-on ques- ADHD will also be asked to identify a partner, a family tions on type and dose of medication or why medi- member or a close friend who could complete informant- cation was not taken (e.g., experienced side effects, report versions of the active monitoring questionnaires forgot to take their medication, or chose not to take on ADHD symptoms and impairment, and irritability, their ADHD medication that day). Every week partici- using the Research Electronic Data Capture (REDCap), pants will be asked to report their medication use side a web-based platform. The research team will contact effects using the CADDRA Patient ADHD Medication the partner, family member or close friend (via email or form [45]. This form contains four items on changes experienced since starting medication and 28 items Denyer et al. BMC Psychiatry (2022) 22:813 Page 6 of 13 Table 2 Schedule of events for ART-CARMA Month 0 1 2 3 4 5 6 7 … 11 12 Assessment 1 2 n/a n/a n/a n/a n/a n/a n/a n/a n/a n/a Baseline assessments Study explanation (WR) X Informed consent (WR) X Introductory training (WR) X ADHD diagnostic interview DIVA-5 (WR) X Psychiatric comorbidities MINI (WR) X IQ and working memory WASI-II and WAIS-IV (WR) X Socio-demographics, medical history, service use, experience with X technology (R/WR) Autism Spectrum Disorder AQ-10 (R/WR) X Cognitive tasks (WR) X X X Physical measures of weight and waist circumference (WR) X Blood pressure (WR) X DNA saliva sample X Microbiota sample X X Remote data collection Wearable sensors Continuous (months 1-12) Smartphone sensors (pRMT) Continuous (months 1-12) Medication use (aRMT) Continuous (months 1-12) Medication side effects CADDRA Patient ADHD Medication Form (aRMT) XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX ADHD BAARS-IV – self-report (aRMT) X X X X X X X X X X X Anxiety GAD-7 (aRMT) X X X X X X X X X X X Depression PHQ-8 (aRMT) X X X X X X X X X X X Aggression RPQ-A (aRMT) X X X X X X X X X X X Irritability ARI – self-report (aRMT) X X X X X X X X X X X Weight and waist circumference (aRMT) X X X X X X X X X X X Blood pressure (aRMT) X X X X X X X X X X X Smoking Test Fagerstrom (aRMT) X X X X X X X X X X X Alcohol use AUDIT (aRMT) X X X X X X X X X X X Diet 14-item Mediterranean diet adherence questionnaire (aRMT) X X X X X X X X X X X Speech sample task (aRMT) X X X X X X X X X X X Major life events LTE (aRMT) X X X ADHD BAARS-IV – other report (R) X X X X X X X X X X X Irritability (ARI – informant report) (R) X X X X X X X X X X X WR With researcher either in-person or virtually, R REDCap web-based platform, pRMT Passive App, aRMT Active App. X delivered once per month. XX delivered twice per month (every 2-weeks). XXXX delivered four times per month (every week) on side effects. Every four weeks, participants will current depression in the general population; Anxiety be asked to complete a longer set of questionnaires: will be measured using the 7-item Generalized Anxiety ADHD symptoms and impairments will be measured Disorder questionnaire; GAD-7 [48], which provides using the Barkley Adult ADHD Rating scale on cur- a dimensional score of anxiety symptoms; Aggression rent symptoms (BAARS-IV) [46] and Barkley ADHD will be measured using the 23-item Reactive-Proactive functional impairment scale [46]. This includes the 18 Aggression Questionnaire for Adults; RPQ-A [49]; diagnostic ADHD symptoms and 10 items on func- Irritability will be measured using the self-reported tional impairment; Depression will be measured using 7-item Affective Reactivity Index questionnaire; ARI-s the 8-item Personal Health Questionnaire Depression [50]; Smoking behaviours will be measured the 6-item Scale; PHQ-8 [47], which is a widely used measure of Test Fagerstrom questionnaire [51]; Alcohol use will Denyer et al. BMC Psychiatry (2022) 22:813 Page 7 of 13 Fig. 1 Participant flowchart be measured using the Alcohol Use Disorders Identifi- weighing scales and a tape measure; Major life events cation Test; AUDIT [52]; Diet will be measured using will be measured every 6 months using the 13-item the 14-item Mediterranean diet adherence question- List of Threatening Experiences; LTE [54]. The partici- naire [53]; Blood pressure and weight measurements pant will also be asked to complete a short speech task will also be taken. Participants will be provided with an monthly, on the Active App. Omron M2 Basic blood pressure monitoring machine, Denyer et al. BMC Psychiatry (2022) 22:813 Page 8 of 13 Fig. 2 ART-CARMA active and passive monitoring Participants will also self-administer the two cogni- peaks, electrodermal activity, oxygen saturation (SpO2), tive tasks of CPT/GNG and Fast task twice during the 12 peripheral temperature, and acceleration. By combining months (months 6 and 12) on their home PC or laptop, data across these variables measured, processes such as without the researcher present. In addition, ratings on physical activity, sleep, autonomic arousal, and indicators ADHD symptoms are obtained monthly also from a part- of cardiovascular health can be captured. ner, a family member or a close friend of the individual with ADHD using web-based Redcap. Empatica Care App The Care App collects data from the EmbracePlus and Passive App sends it to the Empatica Cloud, where it is stored. The Care App via Bluetooth®. The Care App then transmits Participants will be asked to download the RADAR-base data collected by the EmbracePlus is transferred to the passive remote monitoring app, configured for ART requirements [32], which will run in the background, the data to the Empatica Cloud via WiFi. To access and requiring no further input from participants. The Pas- use the Care App, the participant has to scan the QR sive App collects data on ambient noise, ambient light, code or type in the participant ID/password provided phone usage information (e.g., which apps were used and by the research team. The login credentials generated how long, when the phone was unlocked), passive audio, by the research team via the Care Portal are unique and Global Positioning System (GPS) location, Bluetooth de-identified. connectivity, battery life, gyroscope, steps and accelera- tion. Features of the passive audio, rather than raw audio Compatible Android smartphone itself, are extracted from the audio on the phone for Both the Passive App and the EmbracePlus require com- transmission. GPS location data will be randomised; that patible smartphones to be used in ART-CARMA. The EmbracePlus requires Bluetooth® 5.0. For participants is, providing relative location data, not absolute coordi- Passive App requires an Android smartphone and the nates. This prevents identification of an individual’s home address or precise geographical location. who do not own a smartphone, we will provide a compat- ible Android smartphone for them. For participants who Wearable sensors are using a non-Android or a non-compatible Android The wearable device in ART-CARMA is the Empatica smartphone, we will not recruit them unless they agree EmbracePlus [34]. Participants will be asked to wear the to switch to a compatible Android smartphone as their device for the duration of the study (12 months), which only phone for the duration of the study. We will provide will provide ongoing data collection of PR, PRV, systolic a compatible Android smartphone for them. Denyer et al. BMC Psychiatry (2022) 22:813 Page 9 of 13 Outcome measures will not influence their medical care, and that they The outcome measures are (i) the cardiometabolic risk are free to withdraw from the study at any point factors (HR, blood pressure, weight, smoking, alcohol without providing reasons. use, diet, sleep, physiological stress response) and (ii) 2. The research team may withdraw the participant in daily medication use (adherence), adverse side effects the event of inter-current illness, adverse event (AE), and other reasons for non-adherence. protocol violation, administrative or other reasons. Process evaluation In the case of participant self-withdrawal, all attempts To ensure a system which meets participants and will be made to follow up with the participant to estab- researcher needs, there will be continued monitor- lish cause of withdrawal, and to collect qualitative data ing of the ART system throughout the data collection regarding experience of participation. All data, includ- period, including participant adherence to wearable ing those from withdrawn participants, will be included devices and smartphone, and completion of measures. in the final analysis. If a participant withdraws from the At the study endpoint, we will offer post-participation study prematurely, we will consider replacing the partici- qualitative interviews to participants. Interviews will be pant if resources allow and if recruitment is ongoing for offered to all participants who completed to the study the study. end-point and to those who dropped out during their 12-month study period. Statistical and analysis plan Descriptive statistics for demographics, attrition rate Follow‑up procedure and data monitoring and number of participants using the remote assess- Participants will be contacted by telephone after one ment measurements without loss or damage and provid- month of the baseline assessment to address any con- ing adequate data for the duration of the study period cerns or questions. Brief follow-up phone-calls will then will be estimated. Using a survival model approach [55] be offered every three months to maintain engagement and clustering analyses, we will investigate whether with participants throughout the course of the study any demographics and/or other numerical information and to remind participants of any upcoming assess- obtained during baseline of the study might serve as a ments. Participants will be encouraged to live their lives predictor for subject drop out/percentage of adequate as normal, responding to the Active App notifications data. Using mixed-effect models with participants as a when required. The research team will be able to review random factor, we will estimate whether relationships incoming data using the RADAR-base management between the amount of usable data during the weeks portal and Empatica Research Portal. The research team prior to the outcome assessment and obtained scale’s may contact participants if there is a loss of data from a measurements exist. device. All contact with participants will be recorded as Regression and machine learning approaches will be evidence of feasibility and acceptability of outcomes. used to investigate the changes in cardiometabolic factors measured by RMT (listed in Objective 1) following initia- Study debrief tion of ADHD treatment. Data obtained from the accel- At the end of the follow-up period, participants will erometer will be used to label days/part of the days with receive a ‘debrief ’ session, which will serve to collect different gradation levels of activity levels and the dura- end-point acceptability and usability outcomes, and to tion of such activities on a per day basis. For Objective 2, retrieve study materials (devices, chargers etc.). This will Activity and derived features will be used to statistically be done either over the phone or on Microsoft Teams test the effect of average activity level on cardiometabolic with a member of the research team and will be audio- factors defined in Objective 1 with/without treatment. recorded using Microsoft Teams. Participants will also As well as typical statistical modelling approaches, this have an opportunity to view and discuss their own data. analysis will utilise both conventional machine learning methods including support vector machines and random Adverse events and study withdrawal forest in combination with feature selection and fusion There may be several reasons for withdrawal from the approaches, as well as more contemporary deep learning study: methods (Objective 3). 1. Participant choosing to no longer participate: partici- Actigraphy read-outs will also be used to derive sleep pants will be informed both in writing and verbally characteristics using Empatica’s proprietary, 2-step algo- that participation is voluntary, that non-participation rithm that automatically identifies sleep periods and Denyer et al. BMC Psychiatry (2022) 22:813 Page 10 of 13 within the determined sleep period classifies epochs as restricted to system administrators. Data will be placed sleep or wake with a 1-minute granularity. Sleep data within the secure network of King’s College and Rivest– will be further analysed in relation to Active App data Shamir–Adleman (RSA) asymmetric public key encryp- and outcome scales. Utilizing EmbracePlus’ sensor suite, tion will be applied to access. All the data will be stored sleep analyses can further be refined by looking at sleep anonymised and highly restricted access will be given to and wake timing, wake bouts, sleep efficiency and frag- specific authorised researchers. Industry grade security mentation, activity level(s) over the day during wake will be used to secure data access. Authorised access to (based on 3-axis accelerometers), several features of EDA secure Storage Server using RSA encryption over Secure during wake (gated with motion and temperature) and Shell (SSH-RSA), sFTP access, Access Control Lists sleep (including sleep storm patterns). Empatica’s team is (ACLs) and user-groups will be used to control object- also able to post-process EmbracePlus’ sleep or rest data level permissions in the storage for each user. Embrace- to obtain estimates of respiration and PR, controlling Plus data collected by Empatica will be pseudonymised for several factors (movement artefact and compliance) (dummy information, such dummy name and email in its processing [56–59]. Raw, regularly sampled blood address will be entered into the Empatica accounts). volume pulse (BVP) data will be used to estimate PR and Information sheet and consent forms provide details PRV metrics and respiratory dynamics that result from about how participants personal data is handled. Par- the interplay of the sympathetic and parasympathetic ticipants will also have the opportunity before giving sign nervous system. The extraction of features from the consent to ask the research worker any questions regard- homogeneous sample of remotely monitored biosignals ing this. All the personal information will be handled will be done according to well-established and openly according to GDPR and this will be explained in detail to documented methods (e.g., Task Force of the European the participants. Society of Cardiology and the North American Society of Empatica Cloud is the cloud infrastructure where the Pacing and Electrophysiology). Artefacts will be detected, pseudonymised data recorded by the EmbracePlus is col- and the signal noise will be appropriately filtered when it lected and made accessible via web interfaces and data is possible. Spectral and time-domain features extracted repositories. Data is securely stored and encrypted within from continuous biosensor data will be aggregated across an Amazon Web Services (AWS) S3 bucket located in the a period of time using statistical characteristics, aligned United States (N. Virginia), and then pulled to secure and collated with features extracted from smartphones, sFTP storage placed in KCL (with the same authorisation assessments and Redcap, for statistical analysis. and access controls discussed above). Data is encrypted Medication use clustering will be analysed for details at rest and in transit. Access to this data is granted by a around behaviour and pattern of use in Objective 4 and set of unique, secret keys provided by Empatica. The extended in Objective 5 to look at specific predictors of Empatica Portal is GDPR compliant and a Data Process- use. Time-variant clustering will be used to define sub- ing Agreement has already been established between the groups of medication use classes across time, including research parties and the participant information sheet the probability of transitioning between classes using and consent form includes information about the wear- mixture latent Markov models. able sensor data. Machine learning will be implemented in order to build To maintain participant anonymity, their data will individualised prediction models for treatment response, be pseudonymised. Their personal information will be symptom classification and stratification, and medication stored separately in REDCap from information collected adherence (Objective 6). Based on previous work, differ- from the apps and wearable. Participant’s name will be ent prediction models including Neural Processes model replaced with a code, which will be randomly gener- personalization and meta-learning techniques, developed ated and a combination of a human readable ID (usu- for few-shot learning to individualize models, will be ally a number), project name, and site location, and the used [60]. information collected from the apps and wearable will only be associated with this code. Information from the Data protection wearable and smartphone apps will be encrypted (scram- We have ensured the highest standards of data security bled) so that only the research team can see it. Trans- are in place. All electronic data collected will be streamed fer of data between different components is secured to Amazon Elastic Kubernetes System (EKS) server using industry standards. Data extracted from the plat- where the RADAR-base platform will be deployed for form are also secure and private; only people who have data collection. Amazon EKS servers will be located in rights can access it. Pseudonymised data will be securely London (UK), adhering to General Data Protection Rules stored on servers managed by the Institute of Psychia- (GDPR). Access to RADAR-base stack on Amazon EKS is try, Psychology and Neuroscience, KCL, and also on 3rd Denyer et al. BMC Psychiatry (2022) 22:813 Page 11 of 13 party servers, which can provide additional security and questionnaire; AUDIT: Alcohol use disorders identification test questionnaire; AWS: Amazon Web Services; BAARS-IV: Barkley adult ADHD rating scale; BVP: backup. These data will not be linked back to participants Blood volume pulse; CEIm: Ethics committee of research with medicines; CPT: personal information. All details of how data is pseu- Continuous performance test; CVD: Cardiovascular disease; DIVA: Diagnostic donymised will be included in the Information Sheets interview for ADHD in adults; EASO: European Association for the Study of Obesity; EDA: Electrodermal activity; EKS: Elastic Kubernetes System; GAD-7: and Informed Consent Forms. 7-item generalized anxiety disorder questionnaire; GDPR: General data protec‑ tion rules; GNG: Go/NoGo; GPS: Global positioning system; HR: Heart rate; IQ: Intelligence quotient; LTE: 13-item list of threatening experiences question‑ Discussion naire; MINI: Mini international neuropsychiatric interview; pRMT: Passive app; Using RMT, ART-CARMA will obtain real-world data PHQ-8: 8-item patient health questionnaire; PR: Pulse rate; PRV: Pulse rate variability; RADAR-MDD: Remote assessment of disease and relapse in major on ADHD medication adherence and its predictors, and depressive disorder; RCTs: Randomised control trials; REC: Research ethics on the extent to which ADHD medication treatment and committee; REDCap: Research electronic data capture; RMT: Remote measure‑ physical activity may influence cardiometabolic risks in ment technology; RPQ-A: Reactive-proactive aggression questionnaire; RSA: Rivest–Shamir–Adleman; SMEs: Small and medium-sized enterprises; SpO2: adults with ADHD. Detailed digital phenotyping has not Oxygen saturation; SSH: Secure Shell; TIMESPAN: Management of chronic been captured in previous large real-world studies that cardiometabolic disease and treatment discontinuity in adult ADHD patients.. have relied on health record databases. Acknowledgements ART-CARMA benefits from the recent development Not applicable. of the ADHD Remote Technology (ART) system [30, 31], the RADAR-base platform [33] and the new EmbracePlus Authors’ contributions JK, RD, JARQ and AF conceived, designed and implemented the study. HD wearable device from our SME partner Empatica [34]. and JK were major contributors in writing the manuscript. RD and AF wrote The multi-parametric measurements over 12 months, the manuscript text on the statistical power and planned analyses. HL, CR, PN, starting from the pre-treatment initiation period, would SW, UMS, AB, EW and HD contributed to study design; HL further contributed to the study’s conception and HD to its implementation. All authors read and not be possible using other methods. The phenotyping approved the final manuscript. covers detailed physiological measures, health behav- iours (such as physical activity and sleep), self-reported Funding The ART-CARMA study has received funding from the European Union’s clinical symptoms, medication adherence and medica- Horizon 2020 research and innovation programme under grant agreement tion side effects, cognitive performance, speech samples No 965381. This communication reflects the views of the authors, and the and passive monitoring data on the environment and European Commission is not responsible for any use that may be made of the information it contains. HD is supported by the UK Medical Research Council digital behaviours. (MR/N013700/1) and King’s College London member of the MRC Doctoral The long-term aim for ART-CARMA is to improve Training Partnership in Biomedical Sciences. the management of cardiometabolic disease in adults The funding body have not been involved in the design of the study, the col‑ lection or analysis of data, or the interpretation of data. with ADHD, and to improve ADHD medication treat- ment adherence and the personalisation of treatment. Availability of data and materials RMT uniquely has the potential, in the future, to offer Not applicable – this manuscript does not contain any data. the clinician easy and time-saving access to detailed data on symptoms and impairments during the titration pro- Declarations cess and subsequent long-term monitoring of treatment Ethics approval and consent to participate effects, which could help improve patient outcomes in a ART-CARMA will be conducted per the Declaration of Helsinki and Good clini‑ cost-effective way. Future personalised treatment could cal practice, adhering to principles outlined in the NHS Research Governance Framework for Health and Social Care (2nd edition). Ethical approval has been further involve personalised feedback using a clinical app. obtained in London from the Camberwell St Giles Research Ethics Committee The project will improve our understanding of the fea- (REC reference: 21/LO/0825) and in Spain from the Vall d’Hebron University sibility and acceptability of long-term RMT data collec- Hospital Ethics Committee (CEIm code: EOM(AG)004/2022(5944)). All staff working on the study have received training in study conduct, informed tion in adults with ADHD. Feedback from participants consent and risk assessment. RMT data will be pseudonymised and stored in a and thorough follow-up procedures will help identify research database in accordance with the General Data Protection Regulation. challenges for long-term remote monitoring specifi- The informed consent process will ensure that participants understand the nature of the study, and the data being collected. Interested individuals will cally for individuals with ADHD and inform on how to be sent the study materials, including information sheet and consent form overcome barriers for future implementation of RMT to review. If, after reading, they wish to participate, they will be invited to an procedures. enrolment session which will involve the collection of written consent before the administration of any study procedures. They will understand that their privacy is protected and that they can withdraw at any time without giving a reason. Participants may not necessarily be in contact with healthcare services Abbreviations during the study, but for those that are, their relationships with their care ACLs: Access Control Lists; ADDISS: Attention Deficit Disorder Information Ser‑ teams will not be impacted by participation or withdrawal from the study. vices; ADHD: Attention deficit hyperactivity disorder; AE: Adverse event; ARI-s: Affective reactivity index questionnaire; aRMT: Active app; ART​: ADHD Remote Consent for publication Technology; ART-CARMA: ADHD Remote Technology study of cardiometabolic Not applicable. risk factors and medication adherence; AQ-10: Autism Spectrum Quotient Denyer et al. BMC Psychiatry (2022) 22:813 Page 12 of 13 Competing interests 7. Adler LA, Zimmerman B, Starr HL, Silber S, Palumbo J, Orman C, et al. JARQ was on the speakers’ bureau and/or acted as consultant for Janssen- Efficacy and safety of OROS methylphenidate in adults with attention- Cilag, Novartis, Shire, Takeda, Bial, Shionogi, Sincrolab, Novartis, BMS, Medice, deficit/hyperactivity disorder: a randomized, placebo-controlled, Rubió, Uriach, Technofarma and Raffo in the last 3 years. He also received double-blind, parallel group, dose-escalation study. J Clin Psychophar‑ travel awards (air tickets + hotel) for taking part in psychiatric meetings from macol. 2009;29(3):239–47. Janssen-Cilag, Rubió, Shire, Takeda, Shionogi, Bial and Medice. The Depart‑ 8. A 14-month. Randomized clinical trial of treatment strategies for ment of Psychiatry chaired by JARQ received unrestricted educational and attention-deficit/hyperactivity disorder. The MTA Cooperative Group. research support from the following companies in the last 3 years: Janssen- Multimodal Treatment Study of children with ADHD. Arch Gen Psychia‑ Cilag, Shire, Oryzon, Roche, Psious, and Rubió. HL reports receiving grants from try. 1999;56(12):1073–86. Shire Pharmaceuticals; personal fees from and serving as a speaker for Medice, 9. Nissen SE. ADHD drugs and cardiovascular risk. N Engl J Med. Shire/Takeda Pharmaceuticals and Evolan Pharma AB; and sponsorship for 2006;354(14):1445–8. a conference on attention-deficit/hyperactivity disorder from Shire/Takeda 10. Hennissen L, Bakker MJ, Banaschewski T, Carucci S, Coghill D, Danck‑ Pharmaceuticals and Evolan Pharma AB, all outside the submitted work. HL aerts M, et al. Cardiovascular Effects of stimulant and non-stimulant is editor-in-chief of JCPP Advances. JK has given talks at educational events medication for children and adolescents with ADHD: a systematic sponsored by Medice; all funds are received by King’s College London and review and Meta-analysis of trials of Methylphenidate, Amphetamines used for studies of ADHD. and Atomoxetine. CNS Drugs. 2017;31(3):199–215. 11. Fay TB, Alpert MA. Cardiovascular Effects of Drugs used to treat Atten‑ Author details tion-Deficit/Hyperactivity disorder part 2: Impact on Cardiovascular 1 Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, events and recommendations for evaluation and monitoring. Cardiol Psychology and Neuroscience, King’s College London, 16 De Crespigny Park, Rev. 2019;27(4):173–8. SE5 8AF London, UK. 2 Department of Psychiatry, Hospital Universitari Vall 12. Mick E, McManus DD, Goldberg RJ. Meta-analysis of increased heart d’Hebron, Barcelona, Catalonia, Spain. 3 Group of Psychiatry, Mental Health rate and blood pressure associated with CNS stimulant treatment of and Addictions, Vall d’Hebron Research Institute (VHIR), Barcelona, Catalonia, ADHD in adults. Eur Neuropsychopharmacol. 2013;23(6):534–41. Spain. 4 Biomedical Network Research Centre on Mental Health (CIBER‑ 13. Dalsgaard S, Kvist AP, Leckman JF, Nielsen HS, Simonsen M. Cardiovas‑ SAM), Barcelona, Catalonia, Spain. 5 Department of Psychiatry and Forensic cular safety of stimulants in children with attention-deficit/hyperactiv‑ Medicine, Universitat Autònoma de Barcelona, Barcelona, Catalonia, Spain. ity disorder: a nationwide prospective cohort study. J Child Adolesc 6 Department of Biostatistics and Health Informatics, Institute of Psychiatry, Psychopharmacol. 2014;24(6):302–10. Psychology and Neuroscience, King’s College London, London, UK. 7 Institute 14. Dunn AL, Trivedi MH, Kampert JB, Clark CG, Chambliss HO. Exercise of Health Informatics, University College London, London, UK. 8 NIHR Biomedi‑ treatment for depression: efficacy and dose response. Am J Prev Med. cal Research Centre at South London and Maudsley NHS Foundation Trust 2005;28(1):1–8. and King’s College London, London, UK. 9 Health Data Research UK London, 15. Youngstedt SD. Effects of exercise on sleep. Clin Sports Med. University College London, London, UK. 10 NIHR Biomedical Research Centre 2005;24(2):355–65. xi. at University College London Hospitals NHS Foundation Trust, London, UK. 16. Haapanen N, Miilunpalo S, Vuori I, Oja P, Pasanen M. Association of 11 Empatica Srl, Milan, Italy. 12 The National Attention Deficit Disorder Informa‑ leisure time physical activity with the risk of coronary heart disease, tion and Support Service, ADDISS, Edgware, Middlesex, UK. 13 European hypertension and diabetes in middle-aged men and women. Int J Association for the Study of Obesity – Ireland, Dublin, Ireland. 14 South London Epidemiol. 1997;26(4):739–47. and Maudsley NHS Foundation Trust, London, UK. 15 Adult Neurodevelopmen‑ 17. Wei M, Gibbons LW, Kampert JB, Nichaman MZ, Blair SN. Low cardiores‑ tal Service, Health and Community Services, Government of Jersey, St Helier, piratory fitness and physical inactivity as predictors of mortality in men Jersey. 16 Department of Psychiatry, University of Cambridge, Cambridge, UK. with type 2 diabetes. Ann Intern Med. 2000;132(8):605–11. 17 Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, 18. Hu FB, Stampfer MJ, Solomon C, Liu S, Colditz GA, Speizer FE, et al. Stockholm, Sweden. 18 School of Medical Sciences, Örebro University, Örebro, Physical activity and risk for cardiovascular events in diabetic women. Sweden. Ann Intern Med. 2001;134(2):96–105. 19. Rommel AS, Halperin JM, Mill J, Asherson P, Kuntsi J. Protection from Received: 31 October 2022 Accepted: 25 November 2022 genetic diathesis in attention-deficit/hyperactivity disorder: possible complementary roles of exercise. J Am Acad Child Adolesc Psychiatry. 2013;52(9):900–10. 20. Cerrillo-Urbina AJ, García-Hermoso A, Sánchez-López M, Pardo-Guijarro MJ, Santos Gómez JL, Martínez-Vizcaíno V. The effects of physical References exercise in children with attention deficit hyperactivity disorder: a 1. Faraone SV, Banaschewski T, Coghill D, Zheng Y, Biederman J, Bellgrove systematic review and meta-analysis of randomized control trials. Child MA, et al. The World Federation of ADHD International Consensus State‑ Care Health Dev. 2015;41(6):779–88. ment: 208 evidence-based conclusions about the disorder. Neuroscience 21. Vysniauske R, Verburgh L, Oosterlaan J, Molendijk ML. The Effects of and Biobehavioral Reviews. 2021;128:789–818. Physical Exercise on Functional Outcomes in the treatment of ADHD: a 2. Asherson P, Buitelaar J, Faraone SV, Rohde LA. Adult attention-deficit Meta-analysis. J Atten Disord. 2020;24(5):644–54. hyperactivity disorder: key conceptual issues. Lancet Psychiatry. 22. Zang Y. Impact of physical exercise on children with attention deficit 2016;3(6):568–78. hyperactivity disorders: evidence through a meta-analysis. Med (Bal‑ 3. McCarthy S, Wilton L, Murray ML, Hodgkins P, Asherson P, Wong IC. The tim). 2019;98(46):e17980. epidemiology of pharmacologically treated attention deficit hyperactivity 23. Rommel AS, Lichtenstein P, Rydell M, Kuja-Halkola R, Asherson P, disorder (ADHD) in children, adolescents and adults in UK primary care. Kuntsi J, et al. Is physical activity causally Associated with symptoms BMC Pediatr. 2012;12:78. of Attention-Deficit/Hyperactivity disorder? J Am Acad Child Adolesc 4. Castle L, Aubert RE, Verbrugge RR, Khalid M, Epstein RS. Trends in medica‑ Psychiatry. 2015;54(7):565–70. tion treatment for ADHD. J Atten Disord. 2007;10(4):335–42. 24. Du Rietz E, Barker AR, Michelini G, Rommel AS, Vainieri I, Asherson P, 5. Raman SR, Man KKC, Bahmanyar S, Berard A, Bilder S, Boukhris T, et al. et al. Beneficial effects of acute high-intensity exercise on electrophysi‑ Trends in attention-deficit hyperactivity disorder medication use: a retro‑ ological indices of attention processes in young adult men. Behav spective observational study using population-based databases. Lancet Brain Res. 2019;359:474–84. Psychiatry. 2018;5(10):824–35. 25. Gajria K, Lu M, Sikirica V, Greven P, Zhong Y, Qin P, et al. Adher‑ 6. Banaschewski T, Coghill D, Santosh P, Zuddas A, Asherson P, Buitelaar ence, persistence, and medication discontinuation in patients with J, et al. Long-acting medications for the hyperkinetic disorders. A attention-deficit/hyperactivity disorder - a systematic literature review. systematic review and european treatment guideline. Eur Child Adolesc Neuropsychiatr Dis Treat. 2014;10:1543–69. Psychiatry. 2006;15(8):476–95. Denyer et al. BMC Psychiatry (2022) 22:813 Page 13 of 13 26. Kidwell KM, Van Dyk TR, Lundahl A, Nelson TD. Stimulant medica‑ 47. Kroenke K, Strine TW, Spitzer RL, Williams JB, Berry JT, Mokdad AH. The tions and sleep for Youth with ADHD: a Meta-analysis. Pediatrics. PHQ-8 as a measure of current depression in the general population. J 2015;136(6):1144–53. Affect Disord. 2009;114(1–3):163–73. 27. Holmskov M, Storebø OJ, Moreira-Maia CR, Ramstad E, Magnusson FL, 48. Spitzer RL, Kroenke K, Williams JB, Löwe B. A brief measure for assess‑ Krogh HB, et al. Gastrointestinal adverse events during methylphenidate ing generalized anxiety disorder: the GAD-7. Arch Intern Med. treatment of children and adolescents with attention deficit hyperactiv‑ 2006;166(10):1092–7. ity disorder: a systematic review with meta-analysis and Trial Sequential 49. Raine A, Dodge K, Loeber R, Gatzke-Kopp L, Lynam D, Reynolds C, et al. Analysis of randomised clinical trials. PLoS ONE. 2017;12(6):e0178187. The reactive-proactive aggression questionnaire: Differential Correlates 28. Solmi M, Fornaro M, Ostinelli EG, Zangani C, Croatto G, Monaco F, et al. of reactive and proactive aggression in adolescent boys. Aggress Behav. Safety of 80 antidepressants, antipsychotics, anti-attention-deficit/hyper‑ 2006;32(2):159–71. activity medications and mood stabilizers in children and adolescents 50. Stringaris A, Goodman R, Ferdinando S, Razdan V, Muhrer E, Leibenluft E, with psychiatric disorders: a large scale systematic meta-review of 78 et al. The affective reactivity index: a concise irritability scale for clinical adverse effects. World Psychiatry. 2020;19(2):214–32. and research settings. J Child Psychol Psychiatry. 2012;53(11):1109–17. 29. Cortese S, Adamo N, Del Giovane C, Mohr-Jensen C, Hayes AJ, Carucci S, 51. Heatherton TF, Kozlowski LT, Frecker RC, Fagerström KO. The Fagerström et al. Comparative efficacy and tolerability of medications for attention- Test for Nicotine Dependence: a revision of the Fagerström Tolerance deficit hyperactivity disorder in children, adolescents, and adults: Questionnaire. Br J Addict. 1991;86(9):1119–27. a systematic review and network meta-analysis. Lancet Psychiatry. 52. Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG. The Alcohol Use 2018;5(9):727–38. Disorders Identification Test, Guidelines for Use in Primary Care, Second 30. Sun S, Denyer H, Sankesara H, Deng Q, Ranjan Y, Conde P, Rashid Z, Edition, Department of Mental Health and Substance Dependence, Bendayan R, Asherson P, Bilbow A, Hollis C, Groom M, Folarin A, Dobson R, World Health Organization, CH-1211 Geneva 27, Switzerland. 2001. Kuntsi J. Remote administration of ADHD-sensitive cognitive tasks: a pilot 53. Martínez-González MA, García-Arellano A, Toledo E, Salas-Salvadó J, Buil- study. Under review. Cosiales P, Corella D, et al. A 14-item Mediterranean diet assessment tool 31. Denyer H, Deng Q, Adanijo A, Asherson P, Bilbow A, Folarin A, Groom M, and obesity indexes among high-risk subjects: the PREDIMED trial. PLoS Hollis C, Wykes T, Dobson R, Kuntsi J, Simblett S. Barriers to and facilitators ONE. 2012;7(8):e43134. of using remote measurement technology in the long-term monitoring 54. Brugha TS, Cragg D. The list of threatening experiences: the reliability of individuals with ADHD: a qualitative analysis. Under review. and validity of a brief life events questionnaire. Acta Psychiatr Scand. 32. Ranjan Y, Rashid Z, Stewart C, Conde P, Begale M, Verbeeck D, et al. 1990;82(1):77–81. RADAR-Base: open source Mobile Health platform for Collecting, Moni‑ 55. Bewick V, Cheek L, Ball J. Statistics review 12: survival analysis. Crit Care. toring, and analyzing data using sensors, wearables, and Mobile Devices. 2004;8(5):389–94. JMIR Mhealth Uhealth. 2019;7(8):e11734. 56. Sano A, Picard RW, Stickgold R. Quantitative analysis of wrist electroder‑ 33. RADAR-base. http://​www.​radar-​base.​org. Accessed 8th June 2022. mal activity during sleep. Int J Psychophysiol. 2014;94(3):382–9. 34. Empatica. www.empatica.com/embraceplus. Accessed 8th June 2022. 57. Onorati F, Regalia G, Caborni C, Migliorini M, Bender D, Poh MZ, et al. Mul‑ 35. Matcham F, Barattieri di San Pietro C, Bulgari V, de Girolamo G, Dobson ticenter clinical assessment of improved wearable multimodal convulsive R, Eriksson H, et al. Remote assessment of disease and relapse in major seizure detectors. Epilepsia. 2017;58(11):1870–9. depressive disorder (RADAR-MDD): a multi-centre prospective cohort 58. Picard RW, Fedor S, Ayzenberg Y. Multiple arousal theory and daily-life study protocol. BMC Psychiatry. 2019;19(1):72. electrodermal activity asymmetry. Emot Rev. 2016;8(1):62–75. 36. Simblett S, Matcham F, Siddi S, Bulgari V, Barattieri di San Pietro C, Hortas 59. Boucsein W. Electrodermal activity, 2nd ed. New York, NY, US: Springer López J, et al. Barriers to and facilitators of Engagement with mHealth Science + Business Media; 2012. xviii, 618-xviii, p. Technology for Remote Measurement and Management of Depression: 60. Personalized. acute stress classification from physiological signals with qualitative analysis. JMIR Mhealth Uhealth. 2019;7(1):e11325. neural processes https://​arxiv.​org/​abs/​2002.​04176. Accessed 26th Sep‑ 37. Management of chronic cardiometabolic disease. and treatment dis‑ tember 2022. continuity in adult ADHD patients. https://​times​pan.​eu/#​coord​inati​on. Accessed 8th June 2022. 38. Schönbrodt FD, Perugini M. At what sample size do correlations stabilize? Publisher’s Note J Res Pers. 2013;47(5):609–12. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 39. Ranjan Y, Folarin AA, Stewart C, Conde P, Dobson R, Rashid Z. Challenges lished maps and institutional affiliations. & solutions in a hybrid mHealth mobile app. Adjunct Proceedings of the 2019 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2019 ACM International Symposium on Wearable Computers; London, United Kingdom: Association for Com‑ puting Machinery; 2019. p. 183–6. 40. Kooij J, Francken M. DIVA 2.0 diagnostic interview voor ADHD in adults bij volwassenen [DIVA 2.0 diagnostic interview for ADHD in adults]. DIVA Foundation. 2010. 41. Wechsler D. WASI-II: Wechsler abbreviated scale of intelligence: Psych‑ Corp. 2011. 42. Wechsler D. Wechsler adult intelligence scale–Fourth Edition (WAIS–IV). Ready to submit your research ? Choose BMC and benefit from: San Antonio, TX: NCS Pearson. 2008(22):498. 43. Momen NC, Plana-Ripoll O, Agerbo E, Benros ME, Børglum AD, Chris‑ • fast, convenient online submission tensen MK, et al. Association between Mental Disorders and subsequent medical conditions. N Engl J Med. 2020;382(18):1721–31. • thorough peer review by experienced researchers in your field 44. Allison C, Auyeung B, Baron-Cohen S. Toward brief “Red Flags” for autism • rapid publication on acceptance screening: the short Autism Spectrum Quotient and the short quantita‑ • support for research data, including large and complex data types tive checklist for Autism in toddlers in 1,000 cases and 3,000 controls [cor‑ rected]. J Am Acad Child Adolesc Psychiatry. 2012;51(2):202. :– 12.e7. • gold Open Access which fosters wider collaboration and increased citations 45. CADDRA Patient ADHD Medication Form. https://​www.​caddra.​ca/​ • maximum visibility for your research: over 100M website views per year wp-​conte​nt/​uploa​ds/​CADDRA-​Patie​nt-​ADHD-​Medic​ation-​Form.​pdf. Accessed 8th June 2022. At BMC, research is always in progress. 46. Barkley RA, Murphy KR. Attention-deficit hyperactivity disorder: A clinical workbook. New York: Guilford Press; 2006. Learn more biomedcentral.com/submissions