Table 1. Characteristics of included studies (n = 47).

Author	Year	Setting	Condition	Subjects	Length of study	Completion rate	Type of EAM	Intervention	Control
Andrade et al.	2005	The Johns Hopkins Moore HIV Clinic, Baltimore, MD	HIV	64 HIV-infected males and females ≥ 18 years attending HIV clinic	24 weeks	58/64	Electronic verbal prompting device—Disease Management Assistance System (DMAS). Device produces a timed, programmed voice message that prompts subjects to take medication. Records dosing times and dates when response button is pressed. Data can be uploaded and printed.	Same as control plus the DMAS device programmed with reminder messages and dosing times for each medication in the HAART regimen. Adherence results reviewed with participants.	Monthly 30-min adherence counselling session with education about barriers to adherence, hazards of non-adherence and their HAART regimen.
Artinian et al.	2003	Congestive Heart Failure (CHF) clinic of the Detroit Veterans Affairs Medical Centre	CHF	18 outpatients from the CHF clinic aged 50 to 87 years	3 months	18/18 (drop outs not discussed so assume 100% completion)	Medication compliance device—Med-eMonitor linked to a Web-based monitoring system via the patient’s telephone line. Stores up to 5 medicines and has an alarm to remind patients when to take their medication, which to take, and how many. Daily reminders given about healthy lifestyle, other medicines and questions about symptoms, blood pressure, weight. Records date and time stamp of pill compartment opening and patient responses to questions. Data uploaded daily to central server—accessible by clinicians.	Same as control plus a Med-eMonitor	Educational booklet on CHF self-care behaviours, cardiologist follow-up as per usual care, education from pharmacist about medication-taking and provision of written medication information
Brath et al.	2013	Diabetes outpatient clinic	Risk of cardiovascular conditions (≥2 of T2DM, Hypertension or hypercholesterolaemia)	150 patients with a defined risk for cardiovascular conditions	52 weeks	53/77	Medication adherence management system (mAMS): Electronic monitoring medication blister which connected with mobile phone app to send information about adherence to a web based telehealth system which was then used to remind patients automatically via text message	Electronic blister pack transmit user information via mobile phone and automatically sent feedback back to user via text message. Participants with < 70% adherence were also called once a week to try and increase adherence	Given a standard medication blister and returned to study group after control phase completed
Burgess et al.	2010	Paediatric asthma clinic within an outer metropolitan general hospital, Australia	Asthma	26 children aged between 6 and 14 years	4 months	26/26	Smartinhaler—validated EAM.	Same as control but had the measured adherence with preventive medication was fed back to the child, parent and physician then incorporated into the management plan for the next month.	Personalized asthma education and management plan, plus generic written information. Devices replaced monthly but adherence data not shared.
Chan et al.	2015	Regional hospital emergency department, Auckland, NZ	Asthma	220 children aged between 6 and 15 years	6 months	213/220	SmartTrack EAM with audiovisual reminder function for use with preventive medication. Records date, time and number of actuations used. Has 14 different reminders that ring twice daily, stopping after the correct dose is taken or after 15 min. Reminder only goes if the correct dose is not taken within 6 hours of the set reminder time.	SmartTrack EAM with audiovisual reminder function enabled (twice daily reminders)	SmartTrack EAM with audiovisual reminder function disabled
Charles et al.	2007	P3 research clinical trials facility, Wellington, NZ. Recruitment was from research volunteer databases, newspaper advertisements, informal contacts	Asthma	110 patients aged 12 to 65 years	2 week run-in period, 24 weeks after	90/110	Smartinhaler—EAM for use with pMDIs. It has an audiovisual reminder function, that emits an audible reminder (beep) at set times plus a visual cue to show patients whether they have taken their inhaler during a designated period or not (green before inhaler use; red once dose taken). Alarm stops after dose is taken or after 60 min. An electronic covert adherence log is included to record medication use, which can be uploaded to the study centre.	Fluticasone propionate 250 micrograms twice daily via the Smartinhaler device with covert adherence monitoring and an audiovisual reminder function (twice daily alarms)	Fluticasone propionate 250 micrograms twice daily via the Smartinhaler device with covert adherence monitoring
Christensen et al.	2010	Included by physicians in private practice or hospital ambulatories across Poland	Hypertension	784 patients aged ≥ 18 years	1 year (6 months with either the device or standard therapy, then crossed over to other arm for 6 months)	398/784 in final analysis	Helping Hand Data Capture device. Device has tablet blister cards and has an audiovisual reminder to remind patients once daily to take their medication. It records compliance by recording the date/time of each blister card removal.	Patients received medical treatment with telmisartan once daily (40 or 80mg) and received electronic compliance monitoring with the audiovisual reminder	Patients received medical treatment with telmisartan once daily (40 or 80mg) without the electronic monitor (standard therapy)
de Bruin et al.	2010	HIV outpatient clinic of the Academic Medical Centre in Amsterdam, The Netherlands	HIV	133 patients ≥ 18 years	9 months (2 months baseline measurement, 3 months intervention, 4 months follow-up)	116/133	Medication Event Monitoring System (MEMS) caps—electronic caps that fit on standard pill bottles and register the date/time of each pill bottle opening. Data can be downloaded and printed from the MEMS cap to provide a detailed but comprehensible overview of medication-taking behaviour. The MEMS-view cap used to feedback data to patients as it has a display on top.	Theory- and evidence-based behaviour change intervention—Adherence intervention (AIMS). HIV nurses delivered adherence strategies based on their adherence; those scoring >95% at baseline received "adherence sustaining" intervention; those <95% received "adherence improving" intervention. The "adherence sustaining" involved feedback of MEMS reports, reinforcement, brief discussion of any difficulties. The "adherence improving" included adherence information verbally and using graphs, discussion of patients’ MEMS-reports and comparing this to desired adherence as a motivation for change. Causes of nonadherent events discussed, tailored solutions identified, and patients asked to self-monitor adherence using the MEMS-view cap. At the next visit, patient difficulties discussed, MEMS reports examined, action plans adapted.	No intervention—only usual care consisting of verbal and written information about treatment and consequences of nonadherence; tailoring of the medication regimen and intake schedule to the patients’ daily life; monitoring of side effects; promotion of use of adherence assisting devices; discussion of adherence problems and solutions; feedback about viral load and CD4 count. The MEMS reports were discussed at the end of the study with the control group and intervention delivered if adherence was suboptimal.
De Geest et al.	2006	The University Hospital Basel, or Cantonal Hospital, Aarau, Switzerland	Renal transplant	18 patients ≥ 18 years	9 months (3 months intervention, 6 months follow-up)	13/18	Electronic bottle cap that registers data/time of bottle opening; data downloaded to a computer which generates lists and graphics of medication-taking habits.	Usual care plus 1 home visit after study inclusion, 3 follow-up phone calls at the end of each month. EM printouts were sent to patients before each phone session to enable discussions. Intervention involved behavioural, educational and social support interventions aimed to increase patient self-efficacy in taking medications consistently. Electronic monitoring printouts used for problem detection, proxy goal setting, regular targeted feedback. Interventions were made based on the assessment of reasons for non-adherence identified from the home visit or after discussion with patients about the EAM data. Possible solutions were identified with the patient, nurse and family. Improvements in adherence were rewarded; non-adherence addressed with adherence strategies.	Usual care but physicians were notified if patients were non-adherent; and depression scores suggested moderate/ severe depression or suicidal ideation. Any interventions made by physicians in response to non-adherence were noted.
Dobbels et al.	2017	University Hospitals of Leuven, Belgium	Heart, liver and lung transplant	205 patients ≥ 18 years	6 months (plus 3 month run-in phase)	149/205	Helping Hand Data Capture device. Device has tablet blister cards and a reminder to remind patients to take their medication. It records compliance by recording the date / time of each blister card removal.	Theory-based multicomponent staged tailored medication adherence intervention—with intervention manual, algorithm and scripts to highlight which behaviour change techniques to apply. Multicomponent tailored behavioural interventions (visits2–4) building on social cognitive theory and trans-theoretical model (e.g. electronic monitoring feedback, motivational interviewing).	Usual care—asked to use Helping hand throughout study and complete all visits to control for attention bias (research talked about non-medication topics for 20–30 mins)
Duncan et al.	2013	Rural, university-based hospital in the North-Eastern United States and an urban-based children’s hospital in the Midwest	Asthma	55 participants aged 9 to 15 years	4 sessions of treatment over 2 months with a 3-month follow-up. Total time was 6 sessions (recruitment; 2,4,6,8 weeks; follow-up) across ~5 months	48/55	MDILog-II electronic recording device—captures date/time of inhaler dispensing and whether the participant had inhaled the medication. Attaches to the ICS canister.	Teamwork intervention. Importance of parents and youth sharing responsibility for asthma management emphasized and learning methods for addressing conflicts. Families trained in a standardized level of parental supervision of medication use. As youth improved with adherence and reaching adherence goals, supervision reduced based on EM adherence information. Asthma Education arm served as an attention control condition—where they received similar education from therapists but did not have access to adherence data from the MDILog-II and parental involvement was not graded based on adherence information.	Standard care; on completion of follow-up, families were provided feedback on child’s adherence and offered an opportunity to receive either of the 2 interventions.
Elixhauser et al.	1990	Outpatient psychiatric clinic of the St. Louis Veterans Administration Medical Centre	Bipolar affective disorder	93 enrolled adult patients	4–8 months depending on frequency of visits set by the patient’s provider (visits could be 2- or 4-monthly)	67/93	Electronic medication monitor for use with oral medication. Involves two plastic blister sheets, each containing 21 blisters holding patient’s medication. The position of the blisters is updated every 15 minutes; if a blister is opened, the time is recorded. Data can be collected with a microcomputer. Printouts provide data on the timing of blister openings with a resolution of 15 minutes.	Monitoring of medication taking only between visit 1 and 2, then feedback of adherence based on electronic monitoring and lithium levels at visit 2, then follow-up at last visit 3. If non-compliance was evident, information was provided about approaches to improve medication taking. Medication received in the medication monitor.	Standard care (no monitoring) between visit 1 and 2, then feedback of adherence based on lithium levels alone at visit 2, then follow-up at last visit 3. Discussions about reasons for out-of-range values and suggestions for improving compliance provided. Medication received in a usual cylindrical vial.
Erickson et al.	2005	Hypertension specialty outpatient clinic within a large university-affiliated medical centre	Hypertension	42 subjects aged ≥ 21 years	3 months	37/42	Medication management system (MMS). The MMS uses patient-specific information to tailor the interactive technology to each patient, and aims to enhance adherence and communication between the patient and health provider. It includes the MedManager device which stores medication, provides reminder signals through an audio alarm and visual text message to alert patients to take a dose or enter data, and collects date/ time of opening of a medication well, blood pressure readings and potential symptoms of adverse effects of medication. Data are transmitted nightly to the central computer. Reports of patient medication use and clinical parameters were generated monthly and sent to the patient and physician.	Medication management system with standard medical care. This MMS was customized for the patient’s medication regimen, daily activities and any special instructions for administration based on patient, pharmacist and physician review.	Standard medical care alone
Forni Ogna et al.	2016	Cardiology intervention unit of the Lausanne university hospital	PCI with stent	123 adults	6 months	117/123	MEMS cap	Standard of care (SOC) + adherence electronic monitoring (EAM) group, in which drug intake was recorded but kept blinded until the study end, versus another intervention group—an integrated care group, with regular feedback on recorded adherence. Integrated care group = downloaded data every 6/52 and fed back at follow up in semi-structured motivational interviews with nurse, or pharmacist, and patients. SOC+EAM—recorded data electronically but patient and study staff blinded until study end.	Standard care
Foster et al.	2014	General practices in Greater Sydney, Australia	Asthma	143 patients aged 14 to 65 years with moderate–severe asthma	6 months with just 2 study visits (enrolment, then follow-up) with telephone data collection at BL, 2,4 and 6 months	129/143	SmartTrack EAM with audiovisual reminder function for use with preventive medication. Records date, time and number of actuations used and uploads data monthly to a secure website. Asks 3 onscreen questions about asthma control each month.	Twice daily customizable audiovisual inhaler reminders and feedback (IRF) delivered by GPs vs personalized adherence discussions between the GP and patient about adherence (PADS) vs IRF + PADs. PAD involved a short questionnaire about barriers to inhaler use with a discussion about the key barriers with goal setting and achievement strategies	Usual care based on "Asthma Cycle of Care" including one-off checking and teaching of inhaler technique and writing asthma action plans. Patients were offered a record of their adherence after the study.
Frick et al.	2001	Sexually Transmitted Disease and Family Planning Clinics, Coast Provincial General Hospital in Mombasa—the government referral hospital for the coastal region of Kenya	HIV model but tested with multivitamins	140 women aged between 18 and 45 years	1 month	120/140 in final analysis	RemindRx®—Microelectronic alarmed medication vial with programmable dosage administration times that records date/time when a button on the vial is depressed. The button also serves to silence the alarm. Data could be downloaded into a computer.	Electronic medication vial with alarm	Electronic medication vial with no alarm
Gregoriano et al.	2019	In- and outpatients from several hospitals in the Basel region	Asthma, COPD	169 adults	6 months	149/169	Smartinhaler and Polymedication Electronic Monitoring System (POEMS)	Acoustic reminder for inhalation and receives support calls when the medication is not taken as prescribed;—The reminder was automatically set at the time, when the patients had to inhale their dose (every day) and not only when a dose was missed. The support calls occurred only if the patients had not inhaled their medication as prescribed for more than 2 consecutive days. Recorded data with the Smartinhalers were uploaded daily at 00:00 to a web-based database via a wireless connection, so the health provider was able to control daily the performed inhalations and to intervene when necessary, by controlling the data on the database.	No reminder nor additional assistance or feedback regarding their medication adherence behaviour.
Hardstaff et al.	2003	Renal and Liver Transplant Unit, Freeman Hospital, Newcastle-upon-Tyne, UK	Renal transplant	75 renal transplant adult patients	12 months	48/75	Smart Top bottle. These bottles have specialised lids containing a microprocessor that records date/time of bottle opening/closing. Information downloadable into a computer database.	Received adherence feedback at first outpatient clinic appointment but then no further feedback	Received no feedback throughout the trial
Henriksson et al.	2016	Karolinska University hospital in Stockholm, Sweden	Renal transplant	80 adults	12 months	74/80	EAM with cellular capabilities (tracking device via Global System for Mobile Communications)	The patients loaded the EAM with a week’s worth of medication at a time. At the prescribed time for taking the medication, the EAM gave visual and audible signals. If the patient did not take their medication, the audible signal was repeated with increasing frequency for 120 minutes. After this (or after the medication was taken), the EAM sent an SMS message to the web-based software, thus providing information about patient compliance.	Standard care, no EAM
Hermann et al.	2011	Glaucoma clinic at the University Hospital in Athens, Greece	Glaucoma	37 patients ≥ 18 years with glaucoma or ocular hypertension	4 weeks	36/37	EAM for use with brimonidine eye drops 0.2%. Records time/date of use by measuring bottle motion and squeezing. Device not able to be separated from the bottle.	Open adherence monitoring with brimonidine twice (BD) or three times daily (TDS)	Masked adherence monitoring with brimonidine twice or three times daily
Joost et al.	2014	Erlangen University Hospital, Germany	Renal transplant	74 renal transplant patients ≥ 18 years	1 year	67/74	Medication Event Monitory System (MEMS) caps	Same as control plus an intensified pharmaceutical care programme targeting daily drug adherence in the year after transplant. This included additional inpatient and outpatient pharmaceutical care and counselling and a structured adherence management module focusing on adherence support. This was delivered during days 6–20 post-transplant (3 x ~30 min sessions), then consultations with the clinical pharmacist with oral & graphical feedback on adherence data occurred after discharge at least once per quarter up to a maximum of once monthly in the year post-transplant	Standardized drug and transplant training including 15-page written information on medications, rejections, tumour risks, infections and a 1 hour training session from the transplant physician on medications, and another 1 hour from nurses regarding practical application of their medication. Follow-up visits with the transplant centre continued as per usual, but there was no additional contact with the clinical pharmacist beyond regular MEMS refill and adherence data collection.
Kozuki et al.	2006	Community mental health centres in the Pacific North-west	Psychotic disorders	30 adult patients	3 months	28/30	eDEM—electronic monitoring cap that records daily execution of the regimen and produces a chronology of the time the medication was taken each day. Information can be downloaded and presented on a computer screen.	Visual-feedback therapy with structured psychodynamic therapy and visual feedback via monitoring cap to increase insight about medication behaviours and work on affect dimensions. Delivered every 2 weeks for 3 months. Focus is on both behaviours and emotional needs of persons with psychotic disorders and has both a behavioural (insight into pill-taking and acceptance of medications with aim to improve affective reactions) and psychodynamic component (encouraged to express concerns about medications / illness/ issues). Information from the cap was shown to patients on a computer screen at each session, then related this to patients’ appraisals of the behaviours.	Supportive counselling group—attentive listening only therapy technique used. Issues related to the medications not discussed. Delivered every 2 weeks for 3 months for 20–30 min per session. Used to control for confounding of attention from therapists.
Matteson-Kome et al.	2014	Mid-western outpatient Inflammatory bowel disease (IBD) clinic	IBD	6 adults ≥ 18 years	3 months intervention phase with a 60-day screening phase to identify non-adherent patients (< 85%) = total 5 months	5/6	Medication Event Monitoring System (MEMS Track Cap -) electronic bottle cap that monitors dosing (not timing) of medication.	Continuous self-improvement intervention (CSI) involving data evaluation and system refinement to help change behaviour by focusing on the patients’ personal systems rather than on their motivation/ intention. It fosters ritualistic and habitual health behaviours and requires less effort, motivation, and intention to maintain changes. This involved a face to face intervention assessing MEMS data after education on brief personal system theory. Patterns of adherence were analysed from MEMS data for patterns of non-adherence and potential personal system changes discussed for patients to implement during the 3-month study	Attention control intervention—face to face educational session presenting information on IBD education topics such as medical therapy, side effects, extra-intestinal manifestations of IBD and surgical modalities. MEMS also received.
McKenney et al	1992	Residence in a retirement community or attending a primary care centre, Virginia	Hypertension	70 ambulant patients ≥ 50 years	2 x 12-week phases	Phase I: 69/70 Phase II: 59/70	Prescript TimeCap—an electronic compliance aid consisting of a medication vial with a cap displaying the last time the cap was removed.	Timepiece cap alone (vs control) for Phase I of study; then timepiece cap alone vs cap + cards for recording BP readings at each clinic visit vs cap + BP recording cards + home BP monitoring and documentation in the cards	Standard medication vial
Mehta et al.	2019	University of Pennsylvania general internal medicine practices	Hypertension	151 adults aged 18 to 75 years	4 months	126/151	Electronic pill bottle (arm 1) and bidirectional text messages (arm 2)	Arm 1: electronic pill bottle (AdhereTech): electronically monitor openings and transmit them to online platform. Participants received one of two daily feedback messages, depending on their adherence the day prior. Arm 2: bidirectional texting arm received text messages via the online platform, prompting the participant to reply via text with his/her adherence for that day. Mirroring the pill bottle arm, the subsequent days the feedback.	Usual care provided by clinical practice
Morton et al.	2017	Hospital clinics in Sheffield or Rotherham (UK)	Asthma	90 children aged 6 to 16 years	12 months	79/90	Smartinhalers’ and ‘Smartturbos’	Smartinhalers’ and ‘Smartturbos’ that delivered reminders when forgotten and was upload to clinician and discussed at 3 month check up	Same smart inhaler but no alarms and no review by clinician
Murray et al.	2007	University-affiliated, inner-city, ambulatory care practice—general medicine and cardiology practices of Wishard Health Services, Indianapolis, Indiana and Wishard Memorial Hospital	CHF	314 low-income patients ≥ 50 years with heart failure	12 months (9-month multi-level intervention with 3-month poststudy phase)	270/314	Medication Event Monitoring System (MEMS) V prescription container lids that recorded the time/date of each opening and closing. The MEMS cap was labelled with the same icon as the container body to allow correct matching of medicines.	Pharmacist intervention to improve adherence and health outcomes. Involved a baseline medication history, assessment of patient medication knowledge and skills, patient-centred verbal instructions and written materials about the medications, icon-based labelling of medications and a timeline to remind patients when to take their medications. The pharmacist monitored medication use, body weight, healthcare encounters and fed back information as needed to other health professionals.	Usual care which did not include patient-centred materials or any further contact with the intervention pharmacist besides an initial medication history
Nides et al.	1993	University of California at Los Angeles and John Hopkins University	COPD	251 patients aged 35 to 60 years	4 months	205/251	Nebulizer Chronolog—a microprocessor device recording time and date of actuation that can be downloaded into an IBM-compatible computer.	Patients informed about the function of the device and received printed copies of their EAM record of inhaler use at end of weeks 1 and 7 of the 12-week smoking cessation program. The health educator and participant jointly reviewed the pattern of inhaler use—praise given if usage satisfactory, and behavioural strategies developed for problem areas. These feedback sessions continued at each 4-month follow-up visit	Patients only told the device recorded amount of inhaled drug use—no information given that it was able to record patterns of use. No feedback given.
Okeke et al.	2009	Glaucoma services of the Scheie or Wilmer Eye Institutes	Glaucoma	66 patients ≥ 18 years	6 months (initial 3-month observational period of which 2 months of data from week 2 to 10 were used, plus 3-month intervention period)	Not stated	Dosing aid bottle—squeezes the drop from the bottle and records the time and date of delivery.	Educational video stressing importance of adherence, rationale, effects, and how to maximize adherence, a structured review of current barriers to drop taking and discussion of possible solutions with a study coordinator. Regular phone call reminders discussing administration, side effects, difficulties with drops—weekly for first follow-up month then every other week for next 2 months; plus audible and visible reminders on the dosing aid device used	No additional intervention beyond being told it is important to take your eyedrops as prescribed.
Onyirimba et al.	2003	Asthma Centre at Saint Francis Hospital and Medical Centre	Asthma	30 adult patients	10 weeks	19/30	MDI Chronologs	Standard asthma care plus direct, non-judgmental clinician-to-patient feedback discussion on their inhaled steroid and beta-agonist use (date/time of use) on all visits using electronic print-outs. This was fed back at days 7, 14, 21 and 42.	Standard asthma care including asthma education and development of a management plan (BL, days 7, 14, 21 and 42). Actuation data blinded to patient, clinician and other caregivers
Reddy et al.	2016	Medical centre in Philadelphia, US	CAD	125 veterans with known CAD and poor adherence, aged 30 to 75	13 weeks	117/126	"GlowCap. The bottle has a computer chip in the lid that communicates with a cellular connected plug-in nightlight. When all features are activated, the GlowCap monitor changes colour 1 h before the scheduled time to take the medication. If the medication is taken during this period, the pill bottle does not sound an alarm. If the medication is not taken within the designated period, the bottle flashes and sounds an alarm"	The individual feedback participants received a bottle with a daily alarm and a weekly adherence feedback report. Weekly feedback reports displayed participants’ medication adherence and assigned a value for weekly performance based on the number of days that they had opened the bottle. Participants in the partner feedback also had a copy of the report sent to their designated family member, friend, or peer. All participants and partners were trained on the interpretation of the weekly adherence report.	All patients received educational material on the importance of adherence to statin medication. The control group received this device, but none of the patient features were activated (no alarm or notification).
Rigsby et al.	2000	Department of Veterans Affairs HIV clinic and the University of Connecticut Infectious Diseases Study Center—a community-based HIV clinical trials site in the City of Hartford Health Department in Hartford, Connecticut	HIV	55 HIV-infected adult subjects	12 weeks (intervention 0–4 weeks, then follow-up at weeks 8 and 12).	46/55	MEMS caps—fixed to the medication with the lowest baseline adherence in the 1-week baseline period.	Weekly sessions for four weeks of cue dose training with MEMS feedback (CD), or cue dose training with cash reinforcement for correct bottle openings (CD + CR). Cue-dose training linked medication taking to daily habits as cues and used MEMS data to reveal missed doses and suggest alternative cues. Contingency reinforcement using graduated cash payments at each weekly meeting for 4 weeks based on consecutive correctly timed bottle openings formed the base of the CD + CR intervention. The reinforcement began at $2 per correct dose and increased with each consecutive correct dose to a maximum of $10 per day. If the dose was not taken on time, it reset to $2.	Non-directive inquiries about adherence—asked about adherence in the week preceding the visit and encouraged to improve adherence. MEMS data not fed back.
Rosen et al.	2004	Primary care clinic at the VA Connecticut Healthcare System	Diabetes	79 adult patients enrolled but only 33 had lower than 80% baseline adherence and were randomised	4 months intervention + 3 months follow-up (no intervention, assessment only)	33/33	MEMS caps/Smart Caps	Cue-dose training with Smart Caps that display the number of hours since last bottle opening—programmable to beep at pre-determined times. Patients instructed to consider cues to remind them to take the medication with opportunities to discuss barriers to adherence. MEMS data given to health providers each month—and if patients had scheduled appointments, the MEMS data would be discussed with the patients.	Supportive counselling for first 5 patients based on self-reported (not MEMS) data but the supportive counselling had elements of the active intervention as the same people gave the counselling, so the next 12 patients had assessments only with no active counselling and no presentation of MEMS data to providers
Rosen et al.	2007	HIV clinics in the greater Hartford, Connecticut area	HIV	56 adult participants	32 weeks	36/56	MEMS caps with downloaded data to a computer. Print out shows date and time of each bottle opening over the preceding weeks and the list of doses taken.	Weekly contingency management-based counselling for 16 weeks then 16 weeks of additional data collection. At counselling, data from the MEMS caps were reviewed with patients to identify circumstances surrounding missed doses and identified cues to remind them to take the dose. Responses to the medications, routines for medication taking and efforts to cope with HIV also reviewed. Brief substance abuse counselling conducted. Participants were reinforced for MEMS measured adherence (within 3 hours of agreed times for dosing) with drawings from a bowl for prizes and bonus drawings for consecutive days of perfect adherence, and for consecutive weeks. There was a 26.7% chance to earn per $1.00 card, a 7.6% chance for $20.00, and a 0.2% chance of earning $100.00. Potential total earnings averaged $800. The bonus draws reset if perfect adherence did not occur. In order to be certain that participants sampled the reinforcement, participants received two draws for attending each of the first two counselling sessions. In addition, for the first two weeks, participants were reinforced for having taken any doses on the designated day, whereas afterwards, reinforcement was only provided when all a day’s doses had been taken on time. The providers all received monthly letters of the proportion of doses taken from the MEMS throughout the 32 weeks, but this was not actively followed up with the provider.	Weekly supportive counselling for 16 weeks as the "attention control" condition. Participants were asked about their adherence and offered support for efforts to improve adherence. MEMS data was not reviewed with the participants though and urine toxicology testing not conducted. Only an initial review of substance abuse was done and referrals made for treatment. Monthly letters on adherence (self-reported not from MEMS) were sent to providers
Ruppar	2010	Senior centres, senior living facilities, churches in two Midwestern US cities	Hypertension	15 subjects aged 60 years or older	28 weeks (8-week run-in period + 8-week intervention + 12-week follow-up)	15/15	MEMS electronic medication bottle cap with a digital display that provided daily adherence feedback of date and time of opening of the bottle during the 8-week intervention.	Behavioural feedback intervention with biweekly medication adherence and BP feedback (participants were informed of their adherence rate since the last visit and were shown a graphical display of their adherence behaviour to date; degree of change in the participants’ BP discussed and how it could have been impacted by improvements in adherence), habit counselling, review of medication-taking skills, medication and disease education, medication instruction card.	Received no adherence feedback and was seen by the investigator at weeks 12 and 20 only. Educational materials on managing arthritis pain were provided.
Russell et al.	2011	Tertiary care transplant centre located in the Midwestern United States	Renal transplant	15 adult renal transplant recipients aged 21 years or older	9 months (3-month screening phase + 6- month intervention)	15/15	MEMS Track cap—date and time of removal of the cap from the vial,	Continuous self-improvement intervention (CSI) involving collaboration between the participant and clinical nurse specialist on identification of the person’s life routines, important people, possible solutions to enhance medication taking. Monthly medication taking feedback was also delivered via a graphic printout of daily adherence from electronic monitoring. This was conducted monthly during the 6-month intervention.	Attention control—provided with educational brochures from the International Transplant Nurses Society to address healthy post-transplant behaviours. The first brochure was delivered via a home visit and subsequent brochures were mailed. Monthly phone calls to review the brochures and ask participants if they had any questions about the information were made to provide equal attention time and perceived benefits to the control group.
Sabin et al.	2010	Dali Second People’s Hospital in Dali, Yunnan province, China	HIV	80 enrolled, 68 subjects ≥ 18 years old randomised	12 months (pre-intervention phase months 1–6 of monitoring to identify high or low adherence for stratification; intervention period months 7–12)	64/68	Med-ic—Electronic drug monitor pill bottle.	Counselling with feedback from electronic drug monitors. Data from the monitor downloaded each month and the previous month’s data was reviewed with the patient. Those with less than 9% adherence were flagged for counselling with a physician or nurse using the monitor adherence data after the clinic. Data was provided to both the patient and the clinician as a printout with the percent of doses taken, percent taken on time, and a visual display of doses taken by time. Reasons for missing or off-time doses were discussed and problems/challenges identified at the counselling.	Standard care with no feedback of collected adherence data to the patient nor the clinician. However—those whose monthly written self-reports indicated < 95% adherence also received further counselling. The main difference with the control is that the ’flagging’ for counselling relies on self-report rather than electronic drug monitor data.
Smith et al.	2003	Hospital-based infection disease clinic at the University of North Carolina Hospitals in Chapel Hill, NC	HIV	43 individuals ≥ 18 years	3 months but clinical outcomes assessed within 1 year of randomisation	Not stated	MEMS electronic monitors on medication bottles.	Self-management intervention based on feedback of adherence performance and principles of social cognitive theory and self-regulation—3 components of information exchange, skills development and social support enlistment. The program included medication education as per the control group, skills training and development exercises, monthly visits for medication consultations and one-on-one counselling for 3 months, monthly feedback of adherence performance using diary notes, supportive feedback about how closely they adherence to the dosing schedule and with graphical dosing information from the electronic monitoring caps. Goal setting was also done.	Usual care with medication education—written and verbal—and assistance with scheduling of doses. Strategies to improve adherence were discussed. No follow-up visits
Sulaiman et al.	2018	University specialist asthma hospital clinic	Asthma	218 adults with stage 3 to 5 asthma	3 months	148/218	INhaler Compliance Assessment (INCA) attached to inhaler to make a digital audio recording each time the inhaler is used	Basing on information obtained directly from the INCA acoustic recording device, the group discusses patterns of adherence and training on technique of inhaler use as part of biofeedback-guided training.	Generalised strategies to improve adherence, while technique errors are corrected using checklists. Repeated training in inhaler use, adherence and disease management, no biofeedback
Sutton et al.	2014	Primary care clinics in Oxfordshire, Buckinghamshire, Suffolk, Essex, Huntingdonshire	Diabetes	226 adults ≥ 18 years	8 weeks	184/226 for adherence analysis; 193/226 for HbA1c analysis	TrackCap	Electronic container for medication	Standard medication packaging
Tashkin et al.	1991	John Hopkins University and UCLA	COPD	237 adults of a larger intervention group that received a group-based smoking cessation programme, education, counselling and NRT	12 weeks (4 months)	197/237 (40 forgot devices, missed the appointment or had malfunctioning devices)	Nebulizer Chronolog—small, portable electronic device housing a pMDI—records the date and time of each actuation and is read out by an IBM PC.	Informed of the function of the Chronolog and given feedback of the adherence information to enhance adherence. For the feedback participants, the Chronolog memory was read by the interventionist on several occasions over the 12-week program. If the feedback participants were not using the bronchodilator three times per day at appropriately spaced intervals or were not using 2 actuations per set, the information was given to them and the interventionist worked with them to improve adherence to the prescribed regimen. Those who had good compliance were congratulated and encouraged to continue proper inhaler use.	Blinded to adherence monitoring function of the device.
van Onzenoort et al.	2012	Maastricht University Hospital, Maastricht, The Netherlands and surrounding general practitioners’ practices	Hypertension	470 patients ≥18 years with mild–moderate hypertension as part of a larger HOMERUS trial	1 year with seven follow-up visits (a placebo run-in period of 4 weeks was also conducted before study initiation)	Not stated	MEMS cap V TrackCaps	Adherence monitoring with MEMS (but adherence was not fed back) and pill count	Adherence monitoring by pill count alone
Vasbinder et al.	2017	Outpatient clinics in the Netherlands	Asthma	219 children aged 4 to 11 years	12 months	213/219 (only analysed 209/219)	Real-time medication monitoring (RTMM) device, which was connected to the pressurised metered-dose inhaler (pMDI) and recorded the time and date of administered ICS doses. Immediately after each actuation of the pMDI, data were sent to the study database through the mobile telephone network	The intervention group received tailored SMS reminders, sent only when a dose was at risk of omission	No SMS reminders but still had RTMM device that recorded time and date of ICS dose
Velligan et al	2013	Community mental health centre from two counties in Texas	Schizophrenia	142 patients aged between 18 and 60	9 months after 1-month baseline assessment of adherence monitoring with the MM and pill count	132/142	Med-eMonitor (MM)—an electronic medication monitor that prompts use of medication, cues medication taking, warns patients when the wrong medication is taken or at the wrong time, records complaints and alerts staff of failures to take medication as prescribed.	PharmCAT (in-person)—supports medication taking with an array or environmental supports e.g. using pill containers, signs, alarms, checklists established in weekly home visits from a PharmCAT therapist vs Med-eMonitor (electronic) adherence intervention which is the only support and only contact is via phone if patient missed doses (adherence server checked every 3 days); phone contact addressed practical issues, or motivation issues. Both aim to bypass controlled processes in favour of automatic processes and habit formation—reinforcing adherence with electronic messages or social reinforcement.	Treatment as usual—case management and psychiatry appointments at the community mental health center
Wilson et al.	2010	Two academic medical centres, a community health centre, general medicine practice and private infectious diseases practice in the US	HIV	156 adult patients	6 study visits (BL, visits 1–4 before a provider visit, final (6–12 weeks after 4th provider visit)	106/156	MEMS cap	Cross over study. Three-page report of MEMS adherence data given to physician prior to a routine office visit. Self-report data on adherence, patients’ beliefs about therapy, reasons for missed doses, alcohol and drug use and depression also given to the physician. Group 1—received the report prior to the 2 consecutive visits followed by 2 visits with no report. Group 2—no report for first 2 visits, then report with the last 2 visits. The first and third visits were recorded (one intervention, one control)
Wu et al.	2006	John Hopkins Moore (HIV) clinic	HIV	64 patients ≥ 18 years	6 months	48/64	Disease Management Assistance System (DMAS)—a prompting device that verbally reminds patients at medication times and records doses when manually pushed; eDEM to measure adherence.	DMAS + monthly 30 minute adherence educational session	Education only
Yeh et al.	2017	Paediatric MS specialist hospital clinic, US	MS	71 children aged 10 to 18	6 months	49/71	MEMS cap + behavioural feedback	Subjects received a supplemental device which downloaded Adherence data from the MEMS cap for use by the behavioural interventionist during a telephone feedback session at 1,2 and 3 months post-enrolment. Phone called based on MI based on adherence (parents not involved)	Video related to the disease modifying therapy (DMT) in paediatric MS sent at 1, 2 and 3 months

HIV, Human Immunodeficiency Virus; HAART, Highly active antiretroviral combination therapy; DMAS, Disease Management Assistance System; CHF, Congestive Heart Failure; T2DM, Type 2 Diabetes Mellitus; pMDI, pressurised metered dose inhaler; mAMS, Medication adherence management system; EAM, Electronic medication monitor; NZ, New Zealand; MEMS, Medication Event Monitoring System; MMS, Medication Management System; SOC, Standard of Care; IRF, Inhaler reminders and feedback; PAD, Personalised adherence discussion; COPD, Chronic Obstructive Pulmonary Disease; POEMS, Polymedication Electronic Monitoring System; UK, United Kingdom; BD, twice daily; TDS, three times daily; CSI, Continuous self-improvement intervention; IBD, inflammatory bowel disease; CAD, coronary artery disease; CD, cue-dose training; CR, cue-dose training with reinforcement; INCA, INhaler Compliance Assessment; ICS, inhaled corticosteroids; SMS, short message service; RTMM, real-time medication monitoring system; MS, multiple sclerosis; DMT, disease modifying therapy.