The American Psychiatric Association (APA) has updated its Privacy Policy and Terms of Use, including with new information specifically addressed to individuals in the European Economic Area. As described in the Privacy Policy and Terms of Use, this website utilizes cookies, including for the purpose of offering an optimal online experience and services tailored to your preferences.

Please read the entire Privacy Policy and Terms of Use. By closing this message, browsing this website, continuing the navigation, or otherwise continuing to use the APA's websites, you confirm that you understand and accept the terms of the Privacy Policy and Terms of Use, including the utilization of cookies.

×
Published Online:https://doi.org/10.1176/appi.ps.201800482

Abstract

Objective:

This study evaluated the feasibility, acceptability, and preliminary efficacy of psychoeducation plus an automated text messaging intervention (Individualized Texting for Adherence Building–Cardiovascular [iTAB-CV]) to improve adherence to antihypertensives and bipolar disorder medication.

Methods:

After a psychoeducation program, iTAB-CV was administered for 2 months. In month 1, participants received one educational-motivational and one mood rating text daily. In month 2, medication reminders were added.

Results:

The sample (N=38) was 74% African American and 53% women, with a mean±SD age of 51.53±9.06. Antihypertensive nonadherence decreased from a mean of 43%±23% to 21%±18% at 12 weeks (χ2=34.6, df=3, p<0.001). Systolic blood pressure decreased from a mean of 144.8±15.5 mmHg to 136.0±17.8 mmHg (χ2=17.6, df=3, p<.001). Retention was 100%.

Conclusions:

In this uncontrolled trial, participants were highly engaged and medication adherence and reduced systolic blood pressure were sustained after psychoeducation plus iTAB-CV. Because iTAB-CV is automated and delivered remotely, it has the potential to reach a large and challenging population.

HIGHLIGHTS

  • Daily contact via text messaging with high-risk patients with comorbid hypertension and bipolar disorder is both feasible and acceptable.

  • Text messaging in conjunction with psychoeducation appears to strengthen the automaticity of medication taking and has the potential to improve adherence and sustain lower systolic blood pressure and mood symptoms.

Between 25% and 45% of those living with bipolar disorder have hypertension. Poor adherence to antihypertension medication occurs among 50%−80% of patients with bipolar disorder, likely contributing to premature mortality and shortened life expectancy (1). No single intervention has strong evidence for improving medication adherence (2), and the few existing bipolar disorder adherence interventions do not address comorbid medical conditions (2). Therefore, novel approaches to improving hypertension medication adherence are needed for people with bipolar disorder.

Evidence is growing that mobile health tools (mHealth) have a positive impact on behavior change. A bidirectional text messaging system called Individualized Texting for Adherence Building (iTAB) has been used in complex populations in which medication adherence rates are both suboptimal and critical for health outcomes (3), but it has not yet been used with co-occurring bipolar disorder and hypertension.

This intervention was built on an expanded version of the Attitude-Social Influence-Efficacy (ASE) (4) model of behavioral intent and change (see the online supplement to this report), which posits that to achieve a consistent medication-taking habit, a successful intervention must remediate prospective memory deficits with reminders and include reinforcement.

The iTAB-Cardiovascular (iTAB-CV) was developed by using qualitative methodology (5). We evaluated psychoeducation plus the iTAB-CV for feasibility, acceptability, and preliminary efficacy.

Methods

The study used a prospective cohort design (see online supplement) and ran from April 2017 to July 2018. All eligible participants received psychoeducation followed by iTAB-CV and continued with their primary care and mental health treatment providers.

Participants were recruited at a major academic medical center located in an urban center. Inclusion criteria included a clinical diagnosis of bipolar disorder for 2 or more years as determined by a standardized diagnostic interview, the Mini-International Neuropsychiatric Interview (MINI; 6); a diagnosis of hypertension, per patient self-report, dating from at least 6 months before enrollment; systolic blood pressure of 130 or higher at screening; having been prescribed one or more regularly scheduled antihypertension medication for 3 or more months since diagnosis; missing 20% or more of dosages of one or more antihypertensive; and able to participate in psychiatric interviews. Participants also had to be willing to respond to text messages and use an electronic medication tracking device over the 90-day study period. Participants were excluded if they were younger than age 21 or at high immediate risk of suicide.

Participants received psychoeducation about the symptoms, risks, and important role of medication in treatment of both hypertension and bipolar disorder. The information was gathered from publicly available information disseminated by the American Heart Association on hypertension and by the National Institute of Mental Health on bipolar disorder. This information was reviewed for accuracy and understandability by experts in the area of internal medicine and psychiatry. The psychoeducation portion was delivered face-to-face individually and took approximately 15–20 minutes. The study was approved by the University Hospitals Cleveland Medical Center’s institutional review board, and all participants provided written informed consent.

In month 1 of iTAB-CV, participants received one text message daily with psychoeducational or motivational content. Participants responded as to whether they found the message helpful or not (see text categories, sample stems, and sample text exchanges in the online supplement). Participants also received one mood rating scale daily. In month 2, specific medication reminders and immediate reinforcement were added, and daily mood ratings continued. We used the percentage of answered texts as a proxy for engagement. Responses were monitored by an automated system.

Before the start of iTAB-CV, participants could decline text stems from a list of 42 prewritten reminders in seven categories derived from the ASE model, the qualitative study, and consultation with clinical experts. Participants then reviewed 16 reinforcement stems (e.g., “You’re doing wonderfully with taking your meds!”) with the option to remove and create their own adherence reminders and reinforcement messages.

Reminder messages were sent up to four times a day via a central server on the basis of how many medication dosages the participant was prescribed. The system had protections against multiple or inaccurate responses. After 3 consecutive days of unacknowledged messages, the automated system sent an outreach message. The research assistant also had real-time access to participants’ response logs to identify technical problems.

We used the Tablets Routine Questionnaire (TRQ), a self-report measure of the percentage of days with missed doses of a given medication in the past week and month (7), to assess adherence for each regularly scheduled antihypertension medication prescribed for 3 or more months. For participants taking more than one antihypertension medication, we calculated an average. We also assessed adherence for each evidence-based regularly scheduled maintenance bipolar disorder medication (lithium, anticonvulsant, antipsychotic) prescribed for 3 or more months; we also calculated an average for in cases of multiple medications. PRN medications were not included.

We used the eCAP, an objective measure of medication adherence, to record the number of bottle openings (8). The antihypertension medication missed most frequently in the past week went into the eCAP. If two medications were missed at the same frequency, the one dosed most frequently was chosen. We calculated the percentage of dosages taken and then reversed the direction of the eCAP data so that higher numbers indicated worse adherence and corresponded to the TRQ.

Blood pressure was measured by using an automated monitor. The average of three readings was recorded at each visit following a standardized protocol. Bipolar symptoms were measured with the Brief Psychiatric Symptom Scale (BPRS; 9), the Montgomery-Åsberg Depression Rating Scale (MADRS; 10), and the Young Mania Rating Scale (YMRS; 11).

The Self-Report Habit Index (SRHI) is a 12-item self-report measure of habit strength for taking medications (12). Higher scores indicate more automaticity and experience with a habit.

Treatment acceptability and satisfaction were measured via a self-report exit questionnaire, and feasibility was evaluated by determining the percentage of texts answered. All measurements were taken at screening, at baseline (4 weeks), and at visit 1 (V1; 8 weeks) and visit 2 (V2; 12 weeks) follow-ups (aside from eCAP, which was introduced after screening).We used Friedman’s tests to assess the significance of change in adherence for past-week TRQ and eCAP and change in systolic blood pressure over the course of the study. We ran nonparametric statistics when analyzing adherence and systolic blood pressure. Post hoc analyses used a Bonferroni correction. We computed change in eCAP by subtracting baseline values from V2. Change in habit strength (SRHI) and TRQ were computed by subtracting screen values from V2 values. One-way, repeated-measures analyses of variance were run for symptom measures and SRHI. We ran Spearman correlations between adherence, habit strength, and systolic blood pressure.

Results

Thirty-eight participants were enrolled, and 100% completed the study; there was no attrition (see CONSORT diagram in online supplement). Demographic and clinical variables are presented in the online supplement. The mean±SD age of the sample was 51.5±9.1 years; 53% (N=20) were women, and 74% (N=28) were African American. The mean systolic blood pressure at screening was 144.8±15.5 mmHg. The mean body mass index was 22.15±9.4, and 61% (N=23) of the sample were smokers. Five individuals (13%) had a history of stroke; six (16%), heart disease; 22 (58%), high cholesterol; and 15 (40%), diabetes. The mean number of medications at screening was 1.5±0.7 for hypertension and 1.3±0.5 for bipolar disorder. The mean treatment adherence at screening averaged 43%±23% and 45%±28% of days with missing doses in the past week for hypertension and bipolar medications, respectively. The mean BPRS score at screening was in the mild range, mean MADRS score was at the low end of the moderate depressive range, and mean YMRS score was in the normal range. We found a significant improvement in baseline TRQ for antihypertensives and bipolar medications, with a mean past-week TRQ of 21%±20% and 21%±26%, respectively

In stage 1, the mean number of educational-motivational texts sent was 31.52±4.9, and the mean percentage of valid responses was 66%±33%. Of those responses, 95%±7.1% of the messages were deemed helpful. In stage 2, the mean number of medication reminders was 50.2±15.4, and the mean percentage of valid responses was 67%±28%. Mean percentage of valid responses to mood messages was 62%±35% for stage 1 and 56%±29% for stage 2, respectively.

The proportion of missed medication decreased from screening to baseline and was then maintained throughout the study for both antihypertensives and bipolar medications (see online supplement). Antihypertension and bipolar drug adherence improvement was significant between screening and baseline, screening and V1, and screening and V2 for both the past week and the past month. We found no significant differences between any of the time points for eCAP in either past week or past month.

We found a statistically significant difference in systolic blood pressure at the different time points (χ2=17.61, df=3, p<.001). Post hoc analyses revealed a statistically significant reduction in systolic blood pressure from screening (mean=144.8±15.5; median=141.3) to baseline (mean=133.0±17.9; median=132.3) and from screening to V1 (mean=134.8±19.6; median=129.5). We found no significant correlations between change in eCAP use from baseline to V2 and change in hypertension SRHI score from screening to V2 or between change in eCAP use and change in TRQ from screening to V2. However, there was a significant correlation between change in hypertension SRHI score and change in past-week hypertension TRQ between screening and V2 (rs=−0.37, df=36, p=0.02).

We found a statistically significant difference in BPRS scores between screening and each follow-up time point. MADRS scores lowered significantly from screening to V2. YMRS scores did not significantly decrease. The results show that self-reported hypertension habit strength significantly increased between all combinations of time points except between V1 and V2.

All time points were significant or trending toward significance for correlation between hypertension SRHI and TRQ (screening: rs=−0.40, df=36, p=0.01; baseline: rs=−0.32, df=36, p=0.053; V1: rs=−0.43, df=36, p<0.01; V2: rs=−0.44, df=36, p<0.01).

Exit interviews indicated that 79% (N=30) of participants agreed or strongly agreed that the benefit of receiving texts outweighed the hassle, 87% (N=33) agreed or strongly agreed that iTAB-CV texts were useful, 100% (N=38) agreed or strongly agreed that they would recommend iTAB-CV to others, and 95% (N=36) either agreed or strongly agreed that they would continue iTAB-CV after the end of the study, if given the chance.

Discussion

Our results support the feasibility, acceptability, and preliminary potential efficacy of psychoeducation combined with a personalized text messaging intervention (iTAB-CV) in maintaining adherence with both antihypertensives and bipolar medications in a high-risk sample with comorbidity. Managing the treatment of patients with psychiatric and medical comorbidity is a significant challenge for clinicians. It is encouraging that patients with poorly controlled blood pressure and bipolar disorder in this pilot were highly engaged in this low-resource, remotely delivered intervention, responding to a majority of the texts and finding the educational-motivational texts to be helpful. Moreover, 100% of participants completed the study, which is particularly notable given their psychiatric diagnoses and acknowledged poor adherence to medical treatment at the start of the study. The large percentage of African Americans in our sample makes the results distinctive because African Americans tend to be underrepresented in clinical trials and have higher rates of uncontrolled hypertension (13). Although our findings need to be interpreted cautiously given the uncontrolled methodology, psychoeducation plus iTAB-CV has potential to be a practical and scalable approach to help high-risk patients engage in managing their hypertension.

Self-reported medication-taking behavior improved from screening to baseline and was then maintained for both types of medications. The habit of taking medication also became stronger throughout the study, and both systolic blood pressure and psychiatric symptom improvements were maintained, suggesting that habit formation is a foundational feature of medication adherence. These results are consistent with a recent meta-analysis showing mobile text messaging doubles the odds of medication taking (14). To our knowledge, no studies to date have evaluated such an intervention in individuals with comorbid hypertension and bipolar disorder.

As in previous iTAB trials with other patient populations (3), the iTAB-CV results support a high level of patient engagement, which may have resulted from personalizing message stems and reinforcers, removing unwanted content, and choosing windows of time in which to receive messages. There may be specific advantages to implementing personalized text messaging for medication adherence support in clinics that serve individuals with chronic mental illness because it makes these individuals feel cared about and recognizes the importance of both mental and overall health.

The iTAB-CV approach is consistent with the trend for psychiatric patients to seek personalized and technology-driven care. In the course of conducting the trial, we identified future refinements that could enhance its impact and acceptability. Future text systems could personalize the frequency of messages such that messages taper off when adherence increases and return with waning adherence. If iTAB-CV was integrated into routine clinical care, real-time adherence information could be directed to prescribing clinicians and inform medical decision making. Clinicians who are certain that medications are being taken regularly may use that information in their prescribing decisions. Adherence information might also make treatment more efficient by avoiding unnecessary drug switches and improving the clinician-patient treatment alliance by helping both parties see the link between medication taking and blood pressure response with a given medication or course of treatment.

Limitations to consider included that the study was not controlled and that we did not have eCAP screening data to compare with self-report data because eCAP was started at baseline. In addition, both self-reported medication adherence and systolic blood pressure were significantly reduced from screening to baseline, presumably as a function of regression to the mean or because patients were participating in an adherence study (Hawthorne effect) and using an eCAP, which has been shown to independently affect adherence. However, although medication monitoring alone may have an impact, it is unlikely to sustain long-term adherence or affect clinical outcomes. As such, it is promising that improvement in self-reported adherence was maintained throughout the course of the study and corresponded with lower systolic blood pressure. Implementing an intervention such as iTAB-CV on a large scale has few downsides; the text messaging component seems to keep patients engaged over longer time periods with minimal burden for the provider or patient. Despite the limitations, our pilot feasibility and acceptability data provide a strong basis for carrying out an RCT with the improvements to the design noted earlier.

Conclusions

mHealth interventions are becoming increasingly popular for self-management of chronic conditions and have the potential to affect key health behaviors, including medication adherence, particularly in individuals with multiple morbidities. This study suggests that daily contact via text messaging is both feasible and acceptable and that text messaging in conjunction with psychoeducation appears to strengthen the automaticity of medication taking and has the potential to improve adherence and sustain lower systolic blood pressure and mood symptoms. Larger-scale studies with rigorous control groups are necessary to determine the long-term efficacy of such interventions and to tease out the relative contribution of various elements. In sum, an approach that integrates mHealth with face-to-face care appears to improve medication adherence among those with psychiatric and medical comorbidity.

Department of Psychiatry, Case Western Reserve University School of Medicine, Cleveland (Levin, Sajatovic, Rahman, Aebi, Cassidy, Blixen, Klein, Fuentes-Casiano); Neurological and Behavioral Outcomes Center (Levin, Sajatovic, Tatsuoka, Blixen) and Department of Medicine (Rahman), University Hospitals Cleveland Medical Center, Cleveland; Department of Psychiatry, University of California, San Diego, La Jolla (Depp, Cushman, Johnston, Moore); Department of Psychology, Cleveland State University, Cleveland (Eskew).
Send correspondence to Dr. Levin ().

This study was supported by a grant from the National Heart, Lung, and Blood Institute (1R21HL132364-01, Dr. Levin, principal investigator) and a Clinical and Translational Science Award (UL1TR 00043) for REDCap. NCT Clinicaltrials.gov ID: NCT02983877.

Dr. Sajatovic has research grants from Alkermes, Centers for Disease Control and Prevention, Janssen, Merck, National Institutes of Health, Pfizer, Reinberger Foundation, Reuter Foundation, and Woodruff Foundation. She is a consultant to Bracket, Health Analytics, Neurocrine, Otsuka, Sunovion, and Supernus and has received royalties from Johns Hopkins University Press, Oxford University Press, Springer Press, and UpToDate. Dr. Moore has research grants from the National Institutes of Health and California HIV/AIDS Research Program, and Gilead Sciences has provided study drug for work unrelated to this project. The other authors report no financial relationships with commercial interests.

References

1 Weiner M, Warren L, Fiedorowicz JG: Cardiovascular morbidity and mortality in bipolar disorder. Ann Clin Psychiatry 2011; 23:40–47MedlineGoogle Scholar

2 Levin JB, Krivenko A, Howland M, et al.: Medication adherence in patients with bipolar disorder: a comprehensive review. CNS Drugs 2016; 30:819–835Crossref, MedlineGoogle Scholar

3 Moore DJ, Poquette A, Casaletto KB, et al.: Individualized texting for adherence building (iTAB): improving antiretroviral dose timing among HIV-infected persons with co-occurring bipolar disorder. AIDS Behav 2015; 19:459–471Crossref, MedlineGoogle Scholar

4 De Vries H, Mudde AN: Predicting stage transitions for smoking cessation applying the attitude-social influence-efficacy model. Psychol Health 1998;13:369-385CrossrefGoogle Scholar

5 Blixen C, Sajatovic M, Moore DJ, et al.: Patient participation in the development of a customized m-Health intervention to improve medication adherence in poorly adherent individuals with bipolar disorder (BD) and hypertension (HTN). Int J Healthc 2018; 4:25–35Crossref, MedlineGoogle Scholar

6 Sheehan DV, Lecrubier Y, Sheehan KH, et al.: The Mini-International Neuropsychiatric Interview (MINI): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998; 59(suppl 20):22–33, quiz 34–57MedlineGoogle Scholar

7 Scott J, Pope M: Nonadherence with mood stabilizers: prevalence and predictors. J Clin Psychiatry 2002; 63:384–390Crossref, MedlineGoogle Scholar

8 Information Mediary Corporation: eCAP . . . the smart cap that enhances medication adherence. Ottawa, Information Mediary Corporation, 2015. www.informationmediary.com/ecap. Accessed.Google Scholar

9 Shafer A: Meta-analysis of the Brief Psychiatric Rating Scale factor structure. Psychol Assess 2005; 17:324–335Crossref, MedlineGoogle Scholar

10 Montgomery SA, Asberg M: A new depression scale designed to be sensitive to change. Br J Psychiatry 1979; 134:382–389Crossref, MedlineGoogle Scholar

11 Young RC, Biggs JT, Ziegler VE, et al.: A rating scale for mania: reliability, validity and sensitivity. Br J Psychiatry 1978; 133:429–435Crossref, MedlineGoogle Scholar

12 Verplanken B, Orbell S: Reflections on past behavior: a self-report index of habit strength. J Appl Soc Psychol 2003; 33:1313–1330CrossrefGoogle Scholar

13 Centers for Disease Control and Prevention: Racial/ethnic disparities in the awareness, treatment, and control of hypertension—United States, 2003–2010. MMWR Morb Mortal Wkly Rep 2013; 62:351–355MedlineGoogle Scholar

14 Thakkar J, Kurup R, Laba TL, et al.: Mobile telephone text messaging for medication adherence in chronic disease: a meta-analysis. JAMA Intern Med 2016; 176:340–349Crossref, MedlineGoogle Scholar