Medicine - IJGMP - Text Messaging Interventions For Promoting - Henrietta Lee - Newland

International Journal of General
Medicine and Pharmacy (IJGMP)

ISSN(P): 2319-3999; ISSN(E): 2319-4006
Vol. 3, Issue 2, Mar 2014, 79-100
© IASET
TEXT MESSAGING INTERVENTIONS FOR PROMOTING MEDICATION ADHERENCE:

A REVIEW OF THE LITERATURE
LEE HENRIETTA1, D’SOUZA RALSTON2, BEAUTRAIS ANNETTE L3, LARKIN GREGORY LUKE4
1,2,4
Department of Surgery, School of Medicine, University of Auckland, Auckland, New Zealand
3
School of Health Sciences, University of Canterbury, Christchurch, New Zealand
ABSTRACT
We reviewed the literature on the use of mobile phone text messaging as an intervention to promote medication
adherence. A literature search was conducted of the PubMed, CINAHL, PsycInfo, EMBASE and Cochrane databases,
supplemented by grey literature hand searches. These searches returned 1752 results and the final selection consisted of
14 studies published between 2009 to 2013 with a randomized controlled trial (RCT) design, which were reviewed in
detail. The studies were conducted in 10 different countries, on patients suffering from five different diseases,
or discharged from the emergency department, or healthy volunteers seeking preventive treatment. There was great
variability in intervention design and the adherence outcome was defined and measured in multiple ways, making direct
comparison amongst studies difficult. Overall, the effectiveness of text messaging interventions on medication adherence
was mixed and some methodological flaws were identified which need to be addressed in future research. We also
conducted a preliminary exploration of other types of interventions available to promote medication adherence. We pooled
data from these studies to examine the average improvement in medication adherence across different types of
interventions. Comparisons of average improvements across the eight types of interventions suggested that text messaging
was the second most effective type of intervention (after behavioural interventions). While this finding augurs well for the
development of text messaging as an intervention to enhance medication adherence, it must be interpreted with caution due
to methodological limitations
KEYWORDS: Biomedical Technology, Medication Adherence, Randomized Controlled Trial, Review, Text Messaging
INTRODUCTION
Medication adherence can be defined as ‘the extent to which the patient follows medical instructions’
(World Health Organization, 2003). This definition, however, is limited due to its inadequacy in describing the different
therapeutic behaviors that reflect adherence to the range of interventions to treat chronic diseases (e.g. seeking medical
attention, filling prescriptions, taking medication appropriately) and its connotation that the patient is a passive recipient of
expert advice (WHO, 2003). Traditionally, other terms used interchangeably with adherence include, most commonly,
‘compliance’, but also ‘persistence’ and ‘concordance’ (Hugtenburg et al, 2013; National Council on Patient Information
and Education, 2007). The differences between these terms were increasingly acknowledged. Hugtenburg and
colleagues (2013) have provided a comprehensive discussion on the topic, and offered an improved definition of adherence
that many researchers have adopted: ‘the extent to which patients’ medication intake behaviour corresponds with the
recommendations of the health care provider’ (Hugtenburg et al, 2013). However, there continues to be a lack of
consensus on the definition and operationalization of adherence. Similarly, as “adherence” has multiple definitions,
www.iaset.us [email protected]
80 Lee Henrietta, D’Souza Ralston, Beautrais Annette L & Larkin Gregory Luke
‘adherence behavior’ is assessed in multiple ways. Depending on the type of disease and intervention, adherence behaviors
may include attending follow-up appointments and executing behavioural modification strategies (WHO, 2003). For the
purposes of this study, we have focused on patients’ adherence to taking prescribed pharmaceuticals only.
Each of the different methods used to measure adherence has strengths and weaknesses (Balkrishnan et al, 2007;
WHO, 2003). As with any self-reported measures, individual ratings, interviews and questionnaires used to assess
adherence are prone to subjectivity and social desirability biases. More objective measures, such as pill counts, may suffer
from counting inaccuracies and/or patients may throw away pills to show adherence. A recent innovation is a range of
electronic monitoring devices which record the time and date of each instance of opening and closing of the medication
container. However, opening and closing the medication containers may not reflect the actual taking of medication.
Rates of patient medication adherence are often reported to be poor. Poor medication adherence may compromise
the quality of treatment outcome, or worsen disease progression leading to complications, re-hospitalizations, emergency
department visits, or death (Balkrishnan et al, 2007; Castano et al, 2012). Adherence rates for patients with chronic
diseases in developed countries have been estimated to be 50% (WHO, 2003; Balkrishnan et al, 2007). These rates are
likely to be even lower in developing countries (WHO, 2003).
Poor adherence may be due to a number of factors. It may be unintentional, for example, patients forgetting to
take medication or not knowing how to take medication; it may also be intentional, for example, patients refusing to take
medication because of side effects, drug dependency, masking of other diseases, reduced long-term efficacy, stigma
associated with certain medications, or lack of knowledge and trust in the medication and its effects
(Hugtenburg et al, 2013).
Interventions to improve adherence may include ways to make taking medication easier, or to increase patients’
motivation to take medication (Van Dulmen et al, 2008). Ideally, interventions need to be tailored to the potential causes of
non-adherence (Balkrishnan et al, 2007; Hugtenburg et al, 2013). For example, a reminder system might be used for
patients who forget (Van Dulmen et al, 2008).
Technological solutions may represent one way to remove barriers to medical adherence, by using technology to
provide reminders to patients to take their medication (Van Dulmen et al, 2008). Different reminder methods have been
reported, including by telephone (Rinfret et al, 2013), pager (Safren et al, 2003; Simoni et al, 1999), and most recently, text
messaging through mobile phones. Text messaging is low cost, instant, and has become an increasingly popular way of
communicating health messages (Pop-Eleches et al, 2011; Suffoletto et al, 2012).
A search of the Cochrane database identified two recent reviews: The first review examined the use of mobile
phone text messaging in promoting adherence to antiretroviral therapy in patients with Human Immunodeficiency
Virus (HIV) (Horvath et al 2012). The second review explored mobile phone text messaging for preventive health care
(Vodopivec-Jamsek et al, 2012). However, neither review specifically examined the use of text messaging in improving
medication adherence in general, which is the aim of the present study.
METHOD
Search Method
We conducted a search of the PubMed, CINAHL, PsycInfo, EMBASE and Cochrane databases for published
Text Messaging Interventions for Promoting Medication Adherence: A Review of the Literature 81
papers, supplemented by grey literature hand searches, for the period from 2009 until the second week of December 2013.
We used the following search terms: (“text messaging” OR “mobile phone” OR “health technology”) AND
(“medication” OR “treatment” OR “discharge instruction” OR “prescription”) AND (“adherence” OR “compliance”
OR “reminder”). The filters applied include human participants, English language, and availability of at least an abstract.
A hand search of the reference lists of the review articles returned from the search above was also conducted.
A separate search was also conducted of PubMed with the search terms “interventions” AND “medication
adherence” to determine other types of interventions designed to promote medication adherence. However, the purpose of
this latter search was only to explore the different types of interventions available and their associated effect sizes.
This additional search was not the primary focus of the current review and will not be discussed in detail in this paper.
Inclusion Criteria
We included all studies with a randomized controlled trial (RCT) design where the primary focus of the study was
to examine the effectiveness of an intervention using text messaging as a way of promoting or enhancing patients’
medication adherence. Studies were selected if the primary outcome was adherence in some form of medication or drug
treatment. We included studies conducted in all clinical settings, with all types of diseases, all types of medications or drug
treatment. Studies conducted in all countries, with participants from all ethnicities were included.
Exclusion Criteria
Studies were excluded if: (1) they had a non-RCT design, (2) text messaging was one of many types of
interventions investigated as part of a large multi-modal intervention study, (3) medication adherence was not the primary
outcome of the study, (4) adherence was for a non-pharmaceutical form of treatment, for example, physiotherapy
appointment, cognitive behavioral therapy; (5) the intervention was designed for health care staff, not patients; (6) pediatric
participants, (7) the article had no abstract or full text available, (8) article was not in English.
Review Procedure
The search of text message intervention RCTs and the hand search of reference lists were done by HL, while the
search of other types of interventions available was done by RD. Selection of articles was done by firstly reviewing the
relevance of the articles’ titles; if relevance was not clear from the title, the abstract was examined. A list of abstracts from
shortlisted articles was compiled by HL, then passed to GL to review and determine the final list of articles. The full text of
each article on the final list was then downloaded and read in detail to extract information for the present review.
Analysis
Data were pooled (i) from the final selected list of text message intervention studies and (ii) from the list of other
types of adherence-building interventions available. We calculated average effect sizes of the intervention and control
groups of the text message intervention studies, and algebraically estimated mean effect sizes based on sample sizes of
each study to calculate mean adherence improvements.
A meta-analysis was not conducted due to the heterogeneity of methodologies employed across studies.
RESULTS AND DISCUSSIONS

Search Results
The PubMed search for text message intervention studies to promote medication adherence returned 1752 results.
HL reviewed the titles and/or abstract and selected 92 articles. Of these, 19 review articles were identified and were used to
ensure completeness. Only papers with prospectively collected primary source data were used. The most common reason
for exclusion was that the study did not use a RCT design. Following review by GL, the final selection consisted of
11 articles. Three more articles from the hand search of the reference lists of reviews were selected. A total of 14 RCTs
were reviewed in detail.
Study Characteristics
The 14 selected RCTs are summarized in Tables 1 and 2. All studies were published between 2009 and 2013.
The studies were conducted in 10 different countries: United States (4), Africa (1), Kenya (2), Nigeria (1),
New Zealand (1), The Netherlands (1), Spain (1), Denmark (1), Canada (1), France (1).
The 14 RCTs targeted a combined total of 2663 adult participants, and included different diseases and contexts:
patients discharged from the Emergency Department (1 study), and patients with human immunodeficiency virus
(5 studies), asthma (2 studies), type II diabetes (1 study), acne vulgaris with facial involvement (1 study) and schizophrenia
(1 study). In addition, three studies involved preventive interventions in healthy adult participants. The total sample size of
each study ranged from 23 patients to 538 patients, with approximately equal numbers of patients in the intervention and
control groups.
Types of Text Messaging Interventions
All of the text messages in the studies (see Table 1) were reminders for the patient to take their medications,
except for one study in which the text messages were statements to promote adherence by counteracting the beliefs
associated with non-adherence, for example, ‘taking your preventer every day protects you from asthma symptoms’
(Petrie et al, 2012). The medication reminder texts were sent along with a positive message in some studies
(Mbuagbaw et al, 2012, Pop-Eleches et al, 2011). Some reminders were interactive (Suffoletto et al, 2012;
Cocosila et al, 2009) or required the patient to respond (Lester et al, 2010; Hardy et al, 2011; Boker et al, 2012;
Maduka et al, 2013); others did not require interaction (Pop-Eleches et al, 2011). The frequency with which the text
messages were delivered varied, from once a week (Mbuagbaw et al, 2012; Lester et al, 2012) to a few times a day
(Suffoletto et al, 2012) or matching patients’ dosing frequencies (Hardy et al, 2011). The duration of the text messaging
interventions also varied, from a few days (Suffoletto et al, 2012) to 48 weeks (Pop-Eleches et al, 2011) or 12 months
(Lester et al, 2010).
The control groups in all studies but one (Hardy et al, 2011) had no exposure to text messages as part of the
intervention. In the study by Hardy and colleagues, the control group used a beeper. In some studies, the control groups
were exposed to ‘standard care’ (Lester et al, 2012; Montes et al, 2012) or ‘usual care’ (Mbuagbaw et al, 2012;
Petrie et al, 2012), but these terms were not always clearly defined.
Definition and Measurement of Adherence
The outcome ‘adherence’ was defined in multiple ways, including: ‘the extent to which patients take medications
as prescribed by their health care provider’ (Strandbygaard et al, 2010); ‘how well patients follow their prescribed
regimen’ (Vervloet et al, 2012); or in study-specific ways: ‘that patient had picked up prescription within 24 hours and
had no pills left on the day of intended completion of prescription’ (Suffoletto et al, 2012). The variability of how
adherence was defined in each study made comparisons across multiple studies difficult. In seven out of 14 studies
reviewed, no explicit definition of adherence was found anywhere in the article. Our review reinforced the point made by
Balkrishnan and colleagues (2007), and Strandbygaard and colleagues (2010), that studies to date have not had a consensus
on how adherence should be defined nor what constitutes an adequate level of adherence.
Some studies defined adherence using a cut-off percentage of the minimum proportion of participants who took
the prescribed dosage at a given time, for example, at least 80% (Strandbygaard et al, 2010; Petrie et al, 2012), 90%
(Pop-Eleches et al, 2011), 95% (Lester et al, 2010, Maduka et al, 2013). This setting of thresholds to determine
dichotomous ‘good’ versus ‘poor adherence’, or ‘adherent’ versus ‘non-adherent’ patients creates unnecessary challenges
as adherence measures the extent of medication taking behaviour which is best measured using a continuous scale
(Balkrishnan et al, 2007; WHO, 2003).
Poor adherence may include taking too much or too little of the prescribed medication, discontinuing medication
prematurely, refusing to fill pharmacy prescriptions, taking medication at the wrong time or in an ineffective way
(Van Dulmen et al, 2008; Hugtenburg et al, 2013). However, the studies reviewed typically assumed that non-adherence
was related only to taking an inadequate amount of medication (e.g. missed pills) and therefore did not address the other
types of prescription deviation.
In addition to being defined in different ways, adherence was also measured in multiple ways, including: picking
up the prescription; number or percentage of doses taken within a defined time period; percentage of participants who took
the prescribed dosage; calculating a percentage of the actual dosage taken divided by the expected dosage taken in the time
period according to prescription; number or percentage of pills missed; number of pharmacy refills. The most common
way adherence was examined was by the percentage of the actual versus expected dosage taken in a time period according
to prescription; this measure was used in half of the studies reviewed. Some studies used a mixture of the criteria above to
measure adherence.
These adherence outcomes were measured most commonly using self-reported measures (11 studies), pill count at
follow-up visits (2 studies), electronic monitoring systems (6 studies) such as Medication Event Monitoring System
(MEMS) or Real Time Medication Monitoring system (RTMM), or a combination of these methods (4 studies).
Intervention Outcomes
Overall, results were mixed regarding the effectiveness of the use of text messaging to enhance medication
adherence (Table 2). Eight studies reported a clear positive outcome in adherence, with intervention participants achieving
significantly higher adherence (Maduka et al, 2013; Petrie et al, 2012; Vervloet et al, 2012; Cocosila et al, 2009;
Strandbygaard et al, 2010; Hardy et al, 2011) and/or a significantly higher percentage of ‘adherent’ participants
(Lester et al, 2012; Petrie et al, 2012) or a significant improvement in Morisky Green Adherence Questionnaire (MAQ)
score (Montes et al, 2012) compared to controls. In five of these eight studies , the positive outcomes were based solely on
self-reported measures (Maduka et al, 2013; Petrie et al, 2012; Lester et al, 2012; Cocosila et al, 2009; Montes et al, 2012).
However, in Hardy and colleagues’ (2011) study, the positive outcomes were found only in electronically-monitored data,
arguably a more objective, robust measure than self-report.
Five studies reported a clear negative outcome in adherence (Suffoletto et al, 2012; Boker et al, 2012;
Ollivier et al, 2009; Hou et al, 2010; Mbuagbaw et al, 2012). In these studies there was no significant difference between
intervention and control participants in all measures of adherence employed, which were self-reported measures of pills
taken or missed, and electronically-monitored opening/closing of medication containers. In two of these five studies the
negative outcomes were based solely on electronically-monitored data (Boker et al, 2012; Ollivier et al, 2009).
Two studies that compared methods of measuring adherence found that adherent outcomes differed according to
the type of measurement method used (Hardy et al, 2011; Hou et al, 2010). In the study by Hardy and colleagues, the
percentage of increase in adherence in the intervention group over the control group was significant only when measured
by electronically-monitored data, The same outcome did not reach statistical significance when measured by self-reported
data. Hou and colleagues reported that the number of missed pills per cycle was significantly higher in
electronically-monitored data compared to participant diary data.
The study by Pop-Eleches and colleagues (2011) reported mixed findings. They found a significantly higher
percentage of ‘adherent’ participants in the intervention group relative to the control group, but only when the text
messages were delivered weekly, and not when they were delivered daily (Pop-Eleches et al, 2011). However, in all other
studies, no clear pattern was observed between frequency of text messages delivered and medication adherence.
Duration of Intervention and Follow-up
The duration of intervention in the 14 studies ranged from a few days post-discharge (Suffoletto et al, 2012) to
12 months (Lester et al, 2012) or 48 weeks (Pop-Eleches et al, 2011). No clear pattern of adherence emerged when
comparing shorter and longer interventions.
However, one important issue to consider is whether (and if so, for how long) the participants had been on the
medication prior to the intervention. In our reviewed studies, participants’ medication history varied (Table 1). In four of
the 14 studies reviewed, participants had not been taking the medication for which adherence was measured. Of the
remaining studies, in which participants had prior exposure to the medication, this exposure ranged from less than three
months to more than one year. Two studies with healthy volunteers reported, respectively, that 42%, and 62%, of the
participants had prior exposure to the medication concerned. Some studies focused on ‘non-adherent’ participants but the
criteria for non-adherence varied among studies. Overall, no consistent pattern emerged between prior exposure to
medication and adherence outcomes.
The length of the follow-up period after completion of the intervention was rarely reported. It is possible that most
of these studies did not have a follow-up phase after the intervention had been completed and that this explains the lack of
follow-up data. Three studies did report a follow-up period: approximately one week (Suffoletto et al, 2012), three months
(Montes et al, 2012), and five months (Petrie et al, 2012). Suffoletto and colleagues’ study obtained a negative outcome.
Petrie and colleagues’ study yielded a positive outcome, however, it was based on data at the endpoint of the intervention
rather than at subsequent follow-up. In Montes and colleagues’ study, the intervention group sustained significant
improvements in scores on an adherence questionnaire, compared to controls, at both three and six months follow-up.
Receiving reminder texts was not a significant predictor in adherence scores at six months, but it was at three months
(Montes et al, 2012).
Attrition
Another important methodological issue in these studies is attrition. Eysenbach and colleagues (2005) described
two types of attrition: (1) participants who dropped out of the study or who were lost to follow-up; and (2) participants who
remained in the study but stopped using the intervention. The 14 studies we reviewed had an average drop-out rate of
16.98% of eligible participants after randomization (range 1.18% - 39.02%). A high drop-out rate produces selection bias,
as unmotivated participants are under-represented in the sample. The intention-to-treat (ITT) analysis is one way of
overcoming this bias. In ITT, drop-out participants are included in the analysis so that the results reflect the total number of
participants at randomization. Ten of the 14 studies adopted the ITT approach.
The high rate of attrition is likely to be due to a variety of reasons. One possibility could be that the adherence
interventions were not tailored to the needs of participants (Verbrugghe et al, 2013; Van Dulmen et al, 2008;
Hugtenburg et al, 2013). Eysenback and colleagues (2005) have suggested that rate of attrition is likely to be higher if
participants perceive that the intervention is not producing benefits to themselves, is too complex, is not consistent with
personal values or needs, is an experiment on a limited basis, or if its effects are not visible to others. Future research
should address these factors when designing an intervention (e.g. by conducting focus groups, and pilot studies of trial
interventions) in efforts to develop interventions which are meaningful to patients and might, thereby, reduce attrition rates.
Other Types of Intervention to Promote Medication Adherence
Apart from text messaging interventions, we also explored other types of interventions reported in the literature to
enhance medication adherence. Our PubMed search returned 748 results and 48 studies were examined (Table 3).
The studies were a mixture of RCT and non-RCT designs. Other types of interventions designed to promote medication
adherence include the use of electronic reminders, pagers, Interactive Voice Response (IVR) systems, a support person or
group, financial incentives, educational interventions and behavioural interventions (for example, teaching cognitive
behavioural skills and offering counselling to patients with depression). Table 3 summarizes the average, and the range, of
improvement in medication adherence for each type of intervention. A comparison of the estimated effect sizes suggests
that behavioural interventions appear to be the most effective in improving medication adherence, followed by text
messaging interventions. The use of a support person (or group) and financial incentives also appear to be effective;
however, the estimates were based on a small number of studies.
Limitations
Our systematic review has some limitations. Firstly, we were unable to conduct a meta-analysis on the data
available due to the varied (and in some cases, lack of) definitions of adherence and different measurements of the concept
across studies. Secondly, the majority of the studies reviewed did not report measures of effect sizes; the effect sizes we
reported here are estimates calculated based on the information available in the articles reviewed. Thirdly, we did not
conduct formal assessments of the methodological quality of the studies reviewed. Finally, a small number of studies were
reviewed. Our search terms included ‘adherence’ and ‘compliance’. Since adherence had also been used interchangeably
with ‘persistence’ and ‘concordance’ (although ‘compliance’ was more common), we could perhaps expand our search
terms to include ‘persistence’ and ‘concordance’ in order to be more comprehensive.
CONCLUSIONS AND FUTURE RESEARCH
In summary, we reviewed the literature on the use of text messaging interventions to promote medication
adherence. This is a rapidly expanding field of research, partly attributable to the increasingly common use and ownership
of mobile phones in many countries. We identified and reviewed 14 RCTs, all of which were published in the last four
years (2009-2013). The studies were conducted in 10 different countries with healthy adult participants, discharged
emergency department patients, or patients suffering from five different types of diseases. Although we reviewed only
14 studies, there was a lot of variability in the intervention design and measurement of outcomes. The text messages varied
in their content and frequency of delivery. Adherence was not always clearly defined, and where it was defined, it was
defined differently and measured in one or multiple ways across studies, at variable time points. Not surprisingly, these
studies have reported mixed results for medication adherence outcomes. The heterogeneous design and measurement of
outcomes made it difficult to compare across studies and to determine the effectiveness of the interventions.
A number of methodological weaknesses are evident in the reviewed studies. Self-reported measures were a
popular way to measure adherence, despite evidence that significant differences were observed between self-reported data
and electronically-monitored data when measuring exactly the same variable (Hardy et al, 2011; Hou et al, 2010).
Self-reported measures tend to carry a social desirability bias. For example, in Hou and colleagues study (2010), the
number of missed pills was significantly lower when self-reported than when electronically-monitored. Self-reported
measures have the potential to make adherence levels appear to be higher than they actually may be. A majority of the
positive outcomes in the studies reviewed were based on self-reported measures, whereas the negative outcomes were more
evenly split between electronically-monitored and self-reported outcomes. A detailed investigation of the advantages and
disadvantages of using different methods to measure adherence is needed in future research. Given our findings, and given
the burgeoning number of studies of text message interventions to enhance medication adherence, we strongly recommend
that future intervention studies employ multiple measures of adherence, including both self-reported and objective and/or
monitoring methods, and compare and report results between the two. Other common flaws in the current studies include
poor description of the control group and of ‘usual care’, and contaminated exposure to the intervention material in the
control group. In addition, not all studies reported follow-up data, so the long-term efficacy of text messaging interventions
remains unclear. Although ITT analysis was used to reduce bias in the studies with high attrition rates, future studies needs
to consider ways of reducing attrition.
Text message reminders have the potential to improve medication adherence by providing cheap, instant, tailored
prompts to patients. However, until various methodological limitations are addressed in better designed and more carefully
controlled studies, the extent to which the promising findings from current studies can be generalized to other patient
groups and to other disease categories remains unclear.
REFERENCES
1. Adler DA, Bungay KM, Wilson IB, Pei Y, Supran S, Peckham E, et al. The impact of a pharmacist intervention
on 6-month outcomes in depressed primary care patients. General Hospital Psychiatry. 2004;26(3):199-209.
2. Al-Eidan FA, McElnay JC, Scott MG, McConnell JB. Management of Helicobacter pylori eradication--the
influence of structured counselling and follow-up. British Journal of Clinical Pharmacology. 2002;53(2):163-71.
3. Akerblad AC, Bengtsson F, Ekselius L, von Knorring L. Effects of an educational compliance enhancement
programme and therapeutic drug monitoring on treatment adherence in depressed patients managed by general
practitioners. International Clinical Psychopharmacology. 2003;18(6):347-54.
4. Andrade ASA, McGruder HF, Wu AW, Celano SA, Skolasky RL, Selnes OA, et al. A Programmable Prompting
Device Improves Adherence to Highly Active Antiretroviral Therapy in HIV-Infected Subjects with Memory
Impairment. Clinical Infectious Diseases. 2005;41(6):875-82.
5. Atherton-Naji A, Hamilton R, Riddle W, Naji S. Improving adherence to antidepressant drug treatment in primary
care: a feasibility study for a randomized controlled trial of educational intervention. Primary Care Psychiatry.
2001;7(2):61-7.
6. Balkrishnan R, Jayawant SS. Medication adherence research in populations: Measurement issues and other
challenges. Clinical therapeutics, 2007; 29(6): 1180-1183.
7. Boker A, Feetham HJ, Armstrong A, Purcell P, Jacobe H. Do automated text message increase adherence to acne
therapy? Results of a randomized, controlled trial. Journal of the American Academy of Dermatology, 2012;
67:1136-42.
8. Brook OH, van Hout H, Stalman W, Nieuwenhuyse H, Bakker B, Heerdink E, et al. A pharmacy-based coaching
program to improve adherence to antidepressant treatment among primary care patients. Psychiatric Services
(Washington, DC). 2005;56(4):487-9.
9. Capoccia KL, Boudreau DM, Blough DK, Ellsworth AJ, Clark DR, Stevens NG, et al. Randomized trial of
pharmacist interventions to improve depression care and outcomes in primary care. American Journal of
Health-System Pharmacy, 2004;61(4):364-72.
10. Castano PM, Bynum JY, Andres R, Lara M, Westhoff C. Effect of daily text messages on oral contraceptive
continuation: a randomized controlled trial. Obstetrics and Gynecology. 2012;119(1):14-20.
11. Castle T, Cunningham MA, Marsh GM. Antidepressant medication adherence via interactive voice response
telephone calls. The American Journal of Managed Care. 2012;18(9):e346-55.
12. Charles T, Quinn D, Weatherall M, Aldington S, Beasley R, Holt S. An audiovisual reminder function improves
adherence with inhaled corticosteroid therapy in asthma. The Journal of Allergy and Clinical Immunology.
2007;119(4):811-6.
13. Christensen A, Christrup LL, Fabricius PE, Chrostowska M, Wronka M, Narkiewicz K, et al. The impact of an
electronic monitoring and reminder device on patient compliance with antihypertensive therapy: a randomized
controlled trial. Journal of Hypertension. 2010;28(1):194-200.
14. Cocosila M, Archer N, Haynes RB, Yuan Y. Can wireless text messaging improve adherence to preventive
activities? Results of a randomised controlled trial. International Journal of Medical Informatics, 2009; 78:230-38
15. Costa FA, Guerreiro JP, Melo MN, Miranda AdC, Martins AP, Garçāo J, et al. Effect of reminder cards on
compliance with antihypertensive medication. International Journal of Pharmacy Practice. 2005;13(3):205-11.
16. Dietrich AJ, Oxman TE, John W Williams J, Schulberg HC, Bruce ML, Lee PW, et al. Re-engineering systems
for the treatment of depression in primary care: cluster randomised controlled trial. BMJ. 2004;329(7466):602.
17. Dobscha SK, Corson K, Hickam DH, Perrin NA, Kraemer DF, Gerrity MS. Depression decision support in
primary care: a cluster randomized trial. Annals of Internal Medicine. 2006; 145(7):477-87.
18. Eussen SR, van der Elst ME, Klungel OH, Rompelberg CJ, Garssen J, Oosterveld MH, et al. A pharmaceutical
care program to improve adherence to statin therapy: a randomized controlled trial. The Annals of
Pharmacotherapy. 2010; 44(12):1905-13.
19. Eysenbach G. The law of attrition. J Med Internet Res, 2005; 7(1): e11.
20. Finley PR, Rens HR, Pont JT, Gess SL, Louie C, Bull SA, et al. Impact of a collaborative care model on
depression in a primary care setting: a randomized controlled trial. Pharmacotherapy. 2003; 23(9):1175-85.
21. Fortney JC, Pyne JM, Edlund MJ, Williams DK, Robinson DE, Mittal D, et al. A randomized trial of
telemedicine-based collaborative care for depression. Journal of General Internal Medicine. 2007; 22(8):1086-93.
22. Gazmararian J, Jacobson KL, Pan Y, Schmotzer B, Kripalani S. Effect of a pharmacy-based health literacy
intervention and patient characteristics on medication refill adherence in an urban health system. The Annals of
Pharmacotherapy. 2010;44(1):80-7.
23. Gensichen J, von Korff M, Peitz M, Muth C, Beyer M, Guthlin C, et al. Case management for depression by
health care assistants in small primary care practices: a cluster randomized trial. Annals of Internal Medicine.
2009; 151(6):369-78.
24. Hardy H, Kumar V, Doros G, Farmer E, Drainoni ML, Rybin D, et al. Randomized controlled trial of a
personalized cellular phone reminder system to enhance adherence to antiretroviral therapy. AIDS Patient Care
and STDs. 2011; 25(3):153-61.
25. Ho LY, Camejo L, Kahook MY, Noecker R. Effect of audible and visual reminders on adherence in glaucoma
patients using a commercially available dosing aid. Clinical Ophthalmology (Auckland, NZ). 2008; 2(4):769-72.
26. Horvath, T., Azman, H., Kennedy, G. E., & Rutherford, G. W. Mobile phone text messaging for promoting
adherence to antiretroviral therapy in patients with HIV infection. Cochrane Database Syst Rev, 2012, 3.
27. Hou MY, Hurwitz S, Kavanagh E, Fortin J, Goldberg AB. Using daily text-message reminders to improve
adherence with oral contraceptives: a randomized controlled trial. Obstetrics and Gynecology. 2010; 116(3):
633-40.
28. Hugtenburg JG, Timmers L, Elders PJM, Vervloet M, Van Dijk L. Definitions, variants, and causes of
nonadherence with medication: a challenge for tailored interventions. Patient Preference and Adherence, 2013;
7:675-82.
29. Hunkeler EM, Meresman JF, Hargreaves WA, Fireman B, Berman WH, Kirsch AJ, et al. Efficacy of nurse
telehealth care and peer support in augmenting treatment of depression in primary care. Archives of Family
Medicine. 2000; 9(8):700-8.
30. Katon W, Von Korff M, Lin E, Walker E, Simon GE, Bush T, et al. Collaborative management to achieve
treatment guidelines. Impact on depression in primary care. JAMA, 1995; 273(13):1026-31.
31. Katon W, Robinson P, Von Korff M, Lin E, Bush T, Ludman E, et al. A multifaceted intervention to improve
treatment of depression in primary care. Archives of General Psychiatry. 1996; 53(10):924-32.
32. Katzelnick DJ, Simon GE, Pearson SD, Manning WG, Helstad CP, Henk HJ, et al. Randomized trial of a
depression management program in high utilizers of medical care. Archives of Family Medicine. 2000; 9(4):
345-51.
33. Katon W, Von Korff M, Lin E, et al. Stepped collaborative care for primary care patients with persistent
symptoms of depression: A randomized trial. Archives of General Psychiatry. 1999; 56(12):1109-15.
34. Kutcher S, Leblanc J, Maclaren C, Hadrava V. A randomized trial of a specific adherence enhancement program
in sertraline-treated adults with major depressive disorder in a primary care setting. Progress in
Neuro-psychopharmacology & Biological Psychiatry. 2002; 26(3):591-6.
35. Laster SF, Martin JL, Fleming JB. The effect of a medication alarm device on patient compliance with topical
pilocarpine. Journal of the American Optometric Association. 1996; 67(11):654-8.
36. Lester RT, Ritvo P, Mills EJ, Kariri A, Karanja S, Chung MH, Jack W, Habyarimana J, Sadatsafavi M,
Najafzafeh M, Marra CA, Estambale B, Ngugi E, Blake Ball T, Thabane L, Gelmon LJ, Kimani J, Ackers M,
Plummer FA. Efects of a mobile phone short messag service on antiretroviral treatment adherence in Kenya
(WelTel Kenya1): A randomised trial. Lancet, 2010; 376:1838-45
37. Lawrence DB, Allison W, Chen JC, Demand M. Improving medication adherence with a targeted,
technology-driven disease management intervention. Disease Management, 2008;11(3):141-4.
38. Maduka O, Tobin-West CI. Adherence counseling and reminder text messages improve uptake of antiretroviral
therapy in a tertiary hospital in Nigeria. Nigerian Journal of Clinical Practice, 2013; 16(3):302-8.
39. Mbuagbaw L, Thabane, L, Ongolo-Zogo P, Lester RT, Mills EJ, Smieja M, Dolovich L, Kouanfack C.
The Cameroon Mobile Phone SMS (CAMPS) Trial: A Randomized Trial of Text Messaging versus Usual Care
for Adherence to Antiretroviral Therapy. PLOS ONE, 2012; 7(12): e46909
40. McCarthy ML, Ding R, Roderer NK, Steinwachs DM, Ortmann MJ, Pham JC, et al. Does providing prescription
information or services improve medication adherence among patients discharged from the emergency
department? A randomized controlled trial. Annals of Emergency Medicine. 2013;62(3):212-23
41. Montes JM, Medina E, Gomez-Beneyto M, Maurino J. A short message service (SMS)-based strategy for
enhancing adherence to antipsychotic medication in schizophrenia. Psychiatry Research, 2012; 200:89-95
42. Mundt JC, Clarke GN, Burroughs D, Brenneman DO, Griest JH. Effectiveness of antidepressant
pharmacotherapy: the impact of medication compliance and patient education. Depression and Anxiety.
2001;13(1):1-10.
43. National Council on Patient Information and Education (NCPIE). Enhancing prescription medicine adherence: a
national action plan, Aug 2007.
44. Ollivier L, Romand O, Marimoutou C, Michel R, Pognant C, Todesco A, Migliani R, Baudon D, Boutin J-P. Use
of short message service (SMS) to improve malaria chemoprophylaxis compliance after returning from a malaria
endemic area. Malaria Journal, 2009; 8:236.
45. Petrie KJ, Perry K, Broadbent E, Weinman J. A text message programme designed to modify patients’ illness and
treatment beliefs improves self-reported adherence to asthma preventer medication. British Journal of Health
Psychology, 2012; 17:74-84.
46. Pladevall M, Brotons C, Gabriel R, Arnau A, Suarez C, de la Figuera M, et al. Multicenter cluster-randomized
trial of a multifactorial intervention to improve antihypertensive medication adherence and blood pressure control
among patients at high cardiovascular risk (the COM99 study). Circulation. 2010;122(12):1183-91.
47. Pop-Eleches C, Thirumurthy H, Habyarimana JP, Zivin JG, Goldstein MP, de Walque D, et al. Mobile phone
technologies improve adherence to antiretroviral treatment in a resource-limited setting: a randomized controlled
trial of text message reminders. AIDS (London, England). 2011; 25(6):825-34.
48. Priebe S, Yeeles K, Bremner S, Lauber C, Eldridge S, Ashby D, et al. Effectiveness of financial incentives to
improve adherence to maintenance treatment with antipsychotics: cluster randomised controlled trial. BMJ. 2013;
347:f5847.
49. Reynolds NR, Testa MA, Su M, Chesney MA, Neidig JL, Frank I, et al. Telephone support to improve
antiretroviral medication adherence: a multisite, randomized controlled trial. J Acquir Immune Defic Syndr. 2008;
47(1):62-8.
50. Rickles NM, Svarstad BL, Statz-Paynter JL, Taylor LV, Kobak KA. Pharmacist telemonitoring of antidepressant
use: effects on pharmacist-patient collaboration. Journal of the American Pharmacists Association, 2005;
45(3):344-53.
51. Rinfret S, Rodes-Cabau J, Bagur R, Dery JP, Dorais M, Larose E, et al. Telephone contact to improve adherence
to dual antiplatelet therapy after drug-eluting stent implantation. Heart, 2013; 99(8):562-9.
52. Safren SA, Hendriksen ES, Desousa N, Boswell SL, Mayer KH. Use of an on-line pager system to increase
adherence to antiretroviral medications. AIDS Care. 2003;15(6):787-93.
53. Simon GE, VonKorff M, Rutter C, Wagner E. Randomised trial of monitoring, feedback, and management of care
by telephone to improve treatment of depression in primary care. BMJ. 2000;320(7234):550-4.
54. Simon GE, Ludman EJ, Tutty S, Operskalski B, Von Korff M. Telephone psychotherapy and telephone care
management for primary care patients starting antidepressant treatment: a randomized controlled trial. JAMA,
2004;292(8):935-42.
55. Simon GE, Ludman EJ, Operskalski BH. Randomized trial of a telephone care management program for
outpatients starting antidepressant treatment. Psychiatric Services (Washington, DC). 2006;57(10):1441-5.
56. Simoni JM, Huh D, Frick PA, Pearson CR, Andrasik MP, Dunbar PJ, et al. Peer support and pager messaging to
promote antiretroviral modifying therapy in Seattle: a randomized controlled trial. Journal of Acquired Immune
Deficiency Syndromes (1999). 2009; 52(4):465-73.
57. Sirey JA, Bruce ML, Kales HC. Improving antidepressant adherence and depression outcomes in primary care:
the treatment initiation and participation (TIP) program. The American Journal of Geriatric Psychiatry,
2010;18(6):554-62.
58. Solomon DH, Iversen MD, Avorn J, Gleeson T, Brookhart MA, Patrick AR, et al. Osteoporosis telephonic
intervention to improve medication regimen adherence: a large, pragmatic, randomized controlled trial. Arch
Intern Med. 2012; 172(6): 477-83.
59. Strandbygaard U, Thomsen SF, Backer V. A daily SMS reminder increases adherence to asthma treatment:
a three-month follow-up study. Respiratory Medicine. 2010; 104(2):166-71.
60. Suffoletto B, Calabria J, Ross A, Callaway C, Yealy DM. A mobile phone text message program to measure oral
antibiotic use and provide feedback on adherence to patients discharged from the emergency department.
Academic Emergency Medicine, 2012; 19:949-58.
61. Unutzer J, Katon W, Callahan CM, Williams JW, Jr., Hunkeler E, Harpole L, et al. Collaborative care
management of late-life depression in the primary care setting: a randomized controlled trial. JAMA, 2002;
288(22):2836-45.
62. Van Dulmen S, Sluijs E, Van Dijk L, et al. Furthering patient adherence: a position paper of the international
expert forum on patient adherence based on an internet forum discussion. BMC Health Services Research. 2008;
8:47.
63. Verbrugghe M, Verhaeghe S, Lauwaert K, Beeckman D & Van Hecke A. Determinants and associated factors
influencing medication adherence and persistence to oral anticancer drugs: a systematic review. Cancer Treat Res,
2013; 39(6): 610-21.
64. Vervloet M, van Dijk L, Santen-Reestman J, van Vlikmen B, van Wingerden P, Bouvy ML, de Bakker DH. SMS
reminders improve adherence to oral medication in type 2 diabetes patients who are real time electronically
monitored. International Journal of Medical Informatics, 2012; 81:594-604.
65. Vodopivec-Jamsek, V., de Jongh, T., Gurol-Urganci, I., Atun, R., & Car, J. (2012). Mobile phone messaging for
preventive health care. Cochrane Database Syst Rev, 3.
66. Wells KB, Sherbourne C, Schoenbaum M, Duan N, Meredith L, Unutzer J, et al. Impact of disseminating quality
improvement programs for depression in managed primary care: a randomized controlled trial. JAMA, 2000;
283(2):212-2
67. World Health Organization. Adherence to long-term therapies: Evidence for action. 2003.
http://www.webcitation.org/5FkqlCAQv. Accessed 26 Jan, 2009
APPENDIECS
Table 1: Methodological Features of 14 Reviewed RCT Studies
Definition and
Duration of
Study Setting Sample Intervention Measurement of
Intervention
Adherence
Intervention:
Interactive text
Not explicitly
reminders to take
defined
vitamin C, with
N=102 participants reinforcing or
Measured by self-
(aged 18 or above) correcting feedback,
reported total
from a Canadian depending on
number of Vitamin
An University acknowledgement.
C pills missed in
unidentified First two weeks, one
Cocosila et al the final week of
Canadian n=52 intervention reminder text per 1 month
(2009) study
University, n=50 control day, feedback every
Canada two days. Final two
Dosage was one
42% participants weeks, one reminder
pill per day
took vitamin C text every two days,
therefore the no. of
previously feedback every three
pills taken is
days
calculated by 7 –
no. of missed pills
Control: No exposure
to text messages
Defined as ‘taken
a doxycycline pill
on a given day’. If
not, they were
considered ‘non-
adherent’ on that
day
N=424 soldiers Intervention: Daily Measured by:

returning from text medication 1. Proportion of
Côte d'Ivoire reminder to take ‘fully adherent’
malaria participants who
Ollivier et al Military n=202 intervention chemoprophylaxis took one pill every
28 days
(2009) base, France n=222 control delivered at midday day for the
duration of the
No prior exposure Control: study
to malaria No exposure to text 2. Average no. of
chemoprophylaxis messages pills taken
3. Daily % of
adherent
participants
Measured using
electronically-
monitored data
(MEMS)
N=82 women Intervention: Daily Not explicitly
Planned taking the oral reminder text to take defined
Parenthood contraceptive pill, oral contraceptive
Hou et al (2010) Clinic, mean age 22 years pill sent at 3 months Measured by
Boston, participants’ chosen comparison of two
USA n=41 intervention time methods:
n=41 control 1.Rate of missed
Control: pills, measured

62% used No exposure to text using
contraceptive pill messages electronically-
previously monitored data
(SIMPill) 2.Self-
Currently on pill reported diary of
>1month excluded daily pill taking
Defined as ‘taken
N=538 HIV Intervention: >95% of provided
patients (aged 18 or Weekly interactive pills at both 6 and
above) on texts about health 12 month follow
Three HIV antiretroviral status and reminders up visits’
Lester et al (2010) Clinics, therapy for the first about phone-based 12 months
Kenya time support Measured by self-
reports of how
n=273 intervention Control: standard many pills the
n=265 control care patient missed in
the past 30 days.
Defined as ‘the
extent to which
patients take
medications as
prescribed by their
health care
provider’. ‘Non-
N=26 patients
Enrolled Intervention: adherence’ as
with asthma (aged
through Medication reminder ‘insufficient intake
18-45) on discos
advertiseme text sent daily at 8 weeks of the prescribed
Strandbygaard et Seretide for 4
nts in free 10am medicine’
al (2010) weeks
newspaper, Adherence rate
Copenhagen Control: no text cut-off at 80%
n=12 intervention
, Denmark reminders
n=14 control
Measured by
% medicine taken
via a medicine
dose count on the
discos Seretide
inhaler device at
clinical visits
Not explicitly
Intervention:
defined
personalized
interactive
Measured by
medication reminder
1. Self-report
N=23 HIV infected text delivered daily
(SR), 7 day recall
patients (aged 18 or matching their dosing
2. Difference
above) on frequency, beeps
between no. of
Boston antiretroviral every 15 mins until
prescribed doses
Medical therapy for at least text is acknowledged
Hardy et al (2011) 6 weeks and no. of missed
Center, 3 months and
doses, divided by
Boston <85% adherent Control: medication
no. of prescribed
reminder beeper
doses – measured
n=12 intervention which beep once at
by pill count (PC)
n=11 control each expected time
3. Electronically-
for a dose, beeping
monitored data
does not repeat and
(MEMS)
no acknowledgement
needed.
A composite
adherence score
(CAS) is
calculated based
on the above three
measures
Defined as ‘taking
twice-daily
Four Intervention medication at least
N=428 patients
Groups from 2x2 90% per 12-week
(aged 18 or above)
design of short period’
on antiretroviral
(reminder) vs long
therapy for less
texts (reminder plus Measured by
Chulaimbo than 3 months
support message), no. of actual
Rural Health
and frequent (daily) medication bottle
Center n=70 short daily
Pop-Eleches et al vs non-frequent opening divided
(CRHC), text 48 weeks
(2011) (weekly) texts. All by total no. of
Nyanza n=72 long daily
texts are one-way prescribed
Province, text
and non-interactive. medication bottle
Kenya n=73 short weekly
opening per 12-
text
Control: week period
n=74 long weekly
No exposure to text
text
messages. Measured using
n=139 control
electronically-
monitored data
(MEMS)
Not explicitly
defined
Two
university- Measured by:
Intervention:
affiliated 1. Actual no. of
personalized
dermatology N=40 patients opening/closing
interactive
clinics in (aged 12-35) with events for each
medication reminder
Dallas and mild to moderate medication tube
texts sent twice daily
Davis, USA. facial acne suitable divided by
at anticipated time of
Also from for treatment with expected no. of
Boker et al (2012) medication use or 12 weeks
advertiseme topical medications opening/closing
patient’s preferred
nt posted events over 12
time.
around the No prior weeks, using
medical experience with electronically-
Control: no text
campus and medication monitored data
reminders
on (MEMS)
Craigslist.co
m 2. Self-reports of
medication taking
(via returned texts)
Intervention group: Not explicitly
Weekly text sent defined
N=200 HIV every Wednesday at
positive patients 9am with a Measured during
HIV/AIDS (aged 21 or above) medication reminder, interviews:
Managemen on antiretroviral a positive message, 1. Visual
Mbuagbaw et al
t Clinic, therapy for at least and helpline number, 6 months Analogue Scale
(2012)
Cameroon, 1 month in addition to usual (VAS)
Africa care (antiretroviral 2.No. of missed
n=101 intervention therapy counselling doses
n=99 control and home visits). 3.No. of pharmacy
refills
Control group:
Usual care only. No

exposure to text
messages.
Not explicitly
defined
N=254 patients
(aged 18-65) with Measured by
schizophrenia on a scores on the
single oral Intervention: Morisky Green
56 antipsychotic medication reminder Adherence
outpatient medication, who texts sent daily at Questionnaire
Montes et al psychiatric are clinically stable either 11am or 2pm (MAQ), a self-
3 months
(2012) centres with change in report measure of
throughout severity or new Control: no text the failure of
Spain treatment in the last reminders, standard patients to take
6 months care prescriptions due
to forgetfulness,
n=100 intervention carelessness,
n=154 control stopping treatment
when they feel like
it.
Intervention: texts
Recruited
with messages that
through
counteract beliefs
asthma N=147 patients Not explicitly
associated with
medication (aged 16-45) with defined ‘Optimally
medication non-
flyers and asthma, currently adherent’ defined
adherence, sent twice
emails to ‘non-adherent’ (not as >80% adherent
Petrie et al (2012) daily (weeks 1-6), 18 weeks
members of defined) measured through
then once daily
a targeted self-reports over
(weeks 7-12), then
marketing n=73 intervention telephone
three per week
website, n=74 control questionnaire
(weeks 13-18).
New
Zealand
Control: Usual care
Defined as ‘a
Intervention group: patient who
Interactive text reported that s/he
messages about (1) picked up
prescriptive pickup prescription within
N=200 patients
and dosage taken. the first 24 hours
(aged 18 or above)
Feedback provided. of discharge; and
prescribed oral
ED, Western One to four texts per (2) had no pills left
Suffoletto et al antibiotics at 1-2 days post
Pennsylvani day, depending on on the day of
(2012) discharge discharge
a response. intended
completion of
n=100 intervention
Control group: prescription’
n=100 control
Printed discharge
instructions only. No Measured by self-
exposure to text reports over
messages. telephone
questionnaire
N=104 patients Intervention: Defined as ‘how
(aged 18-65) with medication reminder well patients
40 Mediq
type II diabetes on texts sent at each follow their
Vervloet et al pharmacies,
oral anti-diabetic time period they 6 months prescribed
(2012) The
medication with chose to take their regimen’
Netherlands
insulin for at least prescribed doses
1 year and <80% Measured by
adherent in Control: no text 1. No. of days

pharmacy refills reminders without dosing
2. % missed doses
n=56 intervention 3. % doses taken
n=48 control within agreed time
period, and within
predefined
standardized time
windows (1-4
hours)
Measured using
electronically-
monitored data
(RTMM) and self-
reported
questionnaire
Intervention:
Twice per week
(Monday, Thursday
mornings) interactive
text messages with
Defined by ‘>95%
adherence-related
in no. of doses
information,
taken divided by
N= 104 HIV medication reminder,
no. of doses
positive patients on and telephone
Tertiary prescribed’
highly active numbers available for
Maduka et al health care
antiretroviral more information, 4 months
(2013) institution, Measured using
therapy for at least delivered over 4
Nigeria self-reported
3 months and months along with
measure of the no.
<95% adherent monthly adherence
of pills the patient
counselling session
missed in the past
7 days.
Control: standard
care – educational
messages, occasional
warnings/questioning
of adherence
Note: I = Intervention, C = Control, p<0.05 in bold.
Table 2: Main Outcomes of the 14 Reviewed RCT Studies

Net Effect Intention
Study Adherence Outcome Key Findings on Limitations to Treat
Adherence Analysis
Measurement time points: baseline,
1-month (endpoint)
 No. of vitamin pills taken for
intervention group increased from 1.3
(baseline) to 4.5 (1-month), control
group increased from 1.6 (baseline) to 1. Limited external
Cocosila et al
3.7 (1-month). Mean % increase I>C, Positive validity to outpatients Yes
(2009)
p=0.001 2. Self-reported measure
 % self-reported increase in adherence,
I(94%)>C(67%)
 No difference between no. of missed
pills at 1-month, I(2.5) vs C(3.3),
p=0.134
Measurement time points: baseline, day 28

1. High no. of
(endpoint)
participants lost to
 Daily % of adherent participants
follow-up (80%
decreased from 94.6% (baseline) to
remained)
Ollivier et al 67.6% (day 28) for intervention group; Negative No
2. Contamination of
(2009) and 95.2% (baseline) to 65.8% (day
intervention and control
28) for control.
group due to shared
 No differences between I and C in the
information
daily % of adherent participants at
baseline and day 28, p>0.05.
1. Use of alternative
reminders (e.g. alarm
clocks) in both groups,
Measurement time points: weeks 1, 2, 3 at with higher usage
each monthly cycles 1, 2, 3 among control
 Overall rate of missed pills per cycle participants, C 68% > I
measured electronically: I(4.9) vs 36%, p=0.003.
C(4.6), p>0.05; measured by self- 2. Rate of missed pills
Hou et al (2010) Negative Yes
reports: I(1.4) vs C(1.1) p>0.05 per cycle recorded in
 No. of missed pills increased over SIMPill was higher than
time, p=0.020, but did not differ diary, p<0.001 for both
between groups, p=0.580 I and C.
3. Possible under-
estimation of missed pill
rate due to participant
manipulation of SIMPill
Measurement time points: 6 months, 12
months 1. Intervention
 % ‘adherent’ participants: I (62%) > C participants forwarding
Lester et al
(50%), p=0.006, after adjusting for Positive their weekly text Yes
(2010)
baseline covariates p=0.003 messages to control
 Number needed to treat (NNT) for participants
adherence was 9, 95% CI 5.0-29.5
Measurement time points: week 0 baseline,
week 4 randomization, week 12 follow-up
1. Small sample size
 Intervention group adherence rate at 4
2.Short follow-up period
weeks (77.9%) and 12 weeks (81.5%),
3. Possibility of dose
Strandbygaard p=0.520
Positive dumping-participants Yes
et al (2010)  Control group adherence rate at 4
empty medication
weeks (84.2%) and 12 weeks (70.1%),
device prior to
p=0.010
assessment visits
 I-C difference in adherence rate at 12
weeks: 17.8%, p=0.019
Measurement time points: 3 weeks, 6
weeks
 Adherence % at 3 weeks, as measured 1. Small sample size
by SR (I-C mean difference = 12.2, 2. No clinical outcome
p=0.137), PC (I-C mean difference = data (e.g. CD4 count)
15.7, p=0.144), MEMS (I>C mean was collected
Hardy et al difference = 28.1, p=0.012), CAS 3. Short follow up
Positive No
(2011) (I>C mean difference = 24.8, period (6 weeks)
p=0.018). 4. No data on how many
 Adherence % at 6 weeks, as measured reminder beeps patients
by SR (I-C mean difference = 20.2, heard before attending
p=0.069), PC (I-C mean difference = their phone.
13.7, p=0.153), MEMS (I>C mean
difference = 33.4, p=0.002), CAS
(I>C mean difference = 27.1,

p=0.009).
Measurement time points: four 12-week 1. Bottle openings may
period, weeks 0-48 not reflect dose taking
 Compared to control (40%), % behaviour
‘adherent’ participants was higher in 2. Measured adherence
Pop-Eleches
weekly texts (53%, p=0.030), but Mixed in only one tablet not Yes
et al (2011)
similar in the entire regime
daily texts (41%, p=0.920), 3. No clinical outcome
long texts (47%, p=0.240) and measured (e.g. HIV-
short texts (47%, p=0.240) RNA determinations)
Measurement time points: 6 weeks, 12 1. Small sample size
weeks 2. Twice daily texts
 Electronically-monitored adherence were too frequent
Boker et al
rates over 12 weeks: I (33.9%) vs C Negative 3.Reported opening of No
(2012)
(36.5%), p=0.500 closing of tube may not
 Self-reported adherence rates for the accurately reflect
intervention group was 74.4% medication application
Measurement time points: baseline, 3
months, 6 months
 No. of participants with >95%
adherence (measured by VAS) at 3mo 1.Adherence rates may
(I<C p=0.029), 6mo (I vs C, p=0.542) not be the same over
Mbuagbaw et al  No. of participants with >90% longer periods
Negative Yes
(2012) adherence (measured by VAS) at 3mo 2.Possible exposure to
(I vs C, p=0.094), 6mo (I>C p=0.027) other reminder methods
 Missed doses reported, 3mo (I vs C, in control group
p=0.622), 6mo (I vs C, p>0.999)
 No. of pharmacy refills, 3mo (I vs C,
p=0.139), 6mo (I vs C, p=0.617)
Measurement time points: baseline, 3
months (endpoint), 6 months (follow-up) 1. May not generalize to
 Improvements in total MAQ score at 3 other types of patients,
months: I (25%) > C (17.5%), e.g. less stable, on
p=0.020 several drugs
2. Other factors (e.g.
 More patients from intervention (37%)
drug experience, side
achieved a ‘good adherence’ score
effects) associated with
Montes et al than control (22.7%).
Positive adherence not examined Yes
(2012)  Improvements in total MAQ score at 6
3. Self-reported measure
months was maintained p=0.040
4. No blinding was used
 Drop in affirmative responses to MAQ 5. Patients on
after 3 months medication early
 Receiving reminder texts was a morning or late evening
significant predictor in improved could not benefit from
MAQ score at 3 months, but not at 6 fixed-time texts
months
1. Large dropout rate,
Measurement time points: 6 weeks, 12 93 participants remained
weeks, 18 weeks (endpoint), 6 months and at final follow up
9 months (follow-up) 2. Participants limited to
Petrie et al
 Self-reported adherence: I(57.8%)>C Positive <45 years old, not No
(2012)
(43.2%) p=0.003 generalizable to older
 >80% adherent: I(25.9%)>C (10.6%) populations
p=0.034 3.Self-reported
measures
1.Self-report
Suffoletto et al Measurement time points: up to 3 days
Negative 2.28% missing data at Yes
(2012) post prescription completion
<14 days
 % Adherent, I(57%) vs C (45%), 3.Other factors (e.g.

p=0.160 health literacy,
 % Filled prescription within first 24 depression, cognitive
hours, I(78%) vs C (69%), p=0.260 impairment) associate
 % No pills left at completion, I(68%) with non-adherence not
vs C (59%), p=0.300 measured
4. Group size
imbalances in disease
categories
Measurement time points: RTMM data
daily plus questionnaire at 6 months
 No. of days without dosing: I(11.9) vs 1. Small sample size
C(13.8), p=0.283 2. Possible language
issues
 % missed doses: I(14.5) vs C(19.2),
3. Differences in
p=0.065
recruitment response
Vervloet et al  % doses taken within agreed time
Positive rates between Yes
(2012) period I(56.7) > C(43.2), p=0.003
pharmacies
 % doses taken within standardized 4. Did not compare
windows I>C, p=0.002 – p=0.007 against automated
 Within 15 min, 30 min, 60 min of text reminders
sent, extra 26.9%, 34.5%, 46.2%
forgotten doses were taken
respectively
Measurement time points: baseline,
monthly counselling visits, and at the end
of 4 months
1. Self-reported
 Self-reported adherence at 4 months
measures
follow-up: I (76.9% ‘adherent’) >
Maduka et al 2. Possible interactions
C(55.8%‘adherent’), p=0.022, Positive Yes
(2013) of intervention and
Cohen’s w = 0.224
control patients outside
 Median CD4+ cell count at baseline: I-
hospital
C, p>0.05, at 4-months I (578.0
cells/ml) > C (361.5 cells/ml),
p=0.007
Table 3: Interventions for Medication Adherence, Other than Text Message

Reminders, with Pooled Estimated Effect Sizes
Range of Average
Intervention to Improvement in Improvement in
Improve Number of Number of Follow-up Medication Medication
Medication RCT’s Participants Period Adherence (with Adherence (with
Adherence Respect to Control Respect to
Group) Control Group)
3 days – 12
Text Messaging 14 2663 0%-27.1% 8.7%
months
1 week – 12
Education 10 7215 0%-68.4% 4.8%
months
Electronic
Medication 3 weeks - 12
12 1523 0% -19% 8%
Reminder months
Messages
2 weeks – 9
Pager 2 294 0%-15% 3.6%
months
Behavioural
21 9483 1-12 months 0%-50.8% 17.7%
Intervention
Interactive Voice
1 39020 6 months 0% 0%
Response
Support (person 6 months-
3 1561 9% - 91% 53.0%
or group) 12 months
Financial
1 141 12 months 11.5% 11.5%
Incentives

Medicine - IJGMP - Text Messaging Interventions For Promoting - Henrietta Lee - Newland

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Medicine - IJGMP - Text Messaging Interventions For Promoting - Henrietta Lee - Newland

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Medicine - IJGMP - Text Messaging Interventions For Promoting - Henrietta Lee - Newland

Uploaded by

Copyright:

Available Formats

International Journal of General

Medicine and Pharmacy (IJGMP)

TEXT MESSAGING INTERVENTIONS FOR PROMOTING MEDICATION ADHERENCE:

RESULTS AND DISCUSSIONS

Types of Text Messaging Interventions

Definition and Measurement of Adherence

Duration of Intervention and Follow-up

Other Types of Intervention to Promote Medication Adherence

CONCLUSIONS AND FUTURE RESEARCH

N=424 soldiers Intervention: Daily Measured by:

Control: pills, measured

Usual care only. No

adherent in Control: no text 1. No. of days

Table 2: Main Outcomes of the 14 Reviewed RCT Studies

Measurement time points: baseline, day 28

(I>C mean difference = 27.1,

 % Adherent, I(57%) vs C (45%), 3.Other factors (e.g.

Table 3: Interventions for Medication Adherence, Other than Text Message

You might also like