J Med Syst (2016)4:9 DOI 10.1007/s10916-016-0456-y
MOBILE SYSTEMS
Feasibility and Preliminary Outcomes of a Web and Smartphone–Based Medication Self-Management Platform for Chronically Ill Patients Helena Anglada-Martínez 1 & Maite Martin-Conde 1 & Marina Rovira-Illamola 1 & Jose Miguel Sotoca-Momblona 1 & Ethel Sequeira 2 & Valentin Aragunde 2 & M. Angels Moreno 2 & Marta Catalan 2 & Carles Codina-Jané 1
Received: 1 November 2015 / Accepted: 29 January 2016 # Springer Science+Business Media New York 2016
Abstract Objective The aim of this study was to assess the feasibility and preliminary outcomes of a medication self-management platform for chronically ill patients, Medplan. Methods We performed a 6-month single-arm prospective pre-post intervention study of patients receiving treatment for hypertension and/or dyslipidemia and/or heart failure and/or human immunodeficiency virus infection. During the pre-intervention phase, participants were followed according to their usual care; during the intervention phase, they used Medplan. We evaluated adherence, health outcomes, healthcare resources and measured the satisfaction of patients and health care professionals. Results The study population comprised 42 patients. No differences were found in adherence to medication measured by proportion of days covered with medication (PDC). However, when adherence was measured using the SMAQ, the percent-
age of adherent patients improved during the intervention phase (p < 0.05), and the number of days with missed doses decreased (p < 0.05). Adherence measured using the Medplan app showed poor concordance with PDC. No differences were found in health outcomes or in the use of health care resources during the study period. The mean satisfaction score for Medplan was 7.2 ± 2.7 out of 10 among patients and 7.3 ± 1.7 among health care professionals. In fact, 71.4 % of participants said they would recommend the app to a friend, and 88.1 % wanted to continue using it. Conclusion The Medplan platform proved to be feasible and was well accepted by its users. However, its impact on adherence differed depending on the assessment method. The lack of effect on PDC is mainly because patients were already good adherers at baseline. The study enabled us to validate the platform in real patients using many different mobile devices and to identify potential barriers to scaling up the platform.
This article is part of the Topical Collection on Mobile Systems * Helena Anglada-Martínez [email protected]
M. Angels Moreno [email protected]
Maite Martin-Conde [email protected]
Marta Catalan [email protected]
Marina Rovira-Illamola [email protected]
Carles Codina-Jané [email protected]
Jose Miguel Sotoca-Momblona [email protected]
1
Ethel Sequeira [email protected]
Pharmacy Service, Hospital Clínic Barcelona, Villarroel, 170 08036 Barcelona, Spain
2
Consorci d’Atenció Primària de l’Eixample Casanova, GrupTransversal de Recerca en Atenció Primària, IDIBAPS, University of Barcelona, Barcelona, Spain
Valentin Aragunde [email protected]
9
Page 2 of 14
Keywords Mobile phones . Chronic patients . Medication adherence . Applications
Introduction Chronic diseases, including human immunodeficiency virus (HIV) infection, are associated with long-lasting treatments and polypharmacy. However, poor adherence to medication is both prevalent and significant, and its consequences include increased hospital admissions and health care costs [1, 2]. Improving adherence is essential when attempting to improve long-term outcomes. As C. Everett Koop, MD, said BDrugs don’t work in patients who don’t take them^. In addition, one of the most commonly reported reasons for non adherence among chronically ill patients is forgetfulness [1, 3]. The numerous barriers to adherence include complexity of the treatment plan, expense, lack of understanding, lack of perceived benefit of medication (in patients who do not feel ill), side effects, embarrassment because the patient does not want other people to know he/she is taking medication, depression, poor health literacy, and lack of trust in the health care system [2]. Therefore, interventions to improve adherence must be customized to the patient’s needs and beliefs, and more than one intervention might be necessary [1]. However, no gold standard for measuring adherence is available, as all methods have their advantages and their drawbacks [1, 4]. Direct methods (e.g. biomarkers and electronic devices) are more objective but more expensive than indirect methods, which are easier to apply but less objective (e.g. pharmacy refills, self-reporting). Consequently, it is extensively recommended to combine more than 1 method. The most common methods used in clinical practice are adherence measured by pharmacy refills and questionnaires. Pharmacy refills requires prescription claims data, although the fact that the patient refills his/her medications does not mean that they are being taken or they are taken properly [1, 5, 6]. Nevertheless, using pharmacy refills is cheap, objective, and quantifiable. Measurement by questionnaires, on the other hand, tends to overestimate adherence, although this approach is usually reliable if the patient admits to being a poor adherer [1, 5]. Although both methods are widely used in clinical practice, a recent review including mHealth intervention studies aimed at enhancing adherence to medication showed that only 5 of the 20 studies (25 %) included used more than 1 method [7]. The effects of the use in mobile phone short messages service (SMS) on medication adherence have been investigated in various studies including different patient’s profiles yielding mixed results [7, 8]. In two studies performed in Kenya, sending SMS improved adherence to antiretroviral treatment (ART) and viral load suppression [9, 10]. In contrast, in the study of Mbuagbaw et al., sending SMS did not
J Med Syst9 :04 )6102(
improved ART adherence among naïve patients after 6 months [11]. Several studies conducted in patients with asthma [12, 13] and heart failure [14], also showed that those patients receiving medication reminders had better adherence than those who don’t. However, in the study of MarquezContreras et al. [15] including antihypertensive patients, this type of intervention did not showed positive outcomes. Similarly, results on the effectiveness of mHealth in improving adherence in clinical outcomes are mixed. A recent review of studies that evaluated clinical outcomes in mHealth adherence interventions showed that of the 41 randomized controlled trials, significant improvements between groups were reported in 16 studies (39 %) [8]. On the other hand, the significant advances in telecommunication have led an extensive use of mobile phones, specially smartphones [16]. Mobile applications (apps) have the potential to be used as a supportive health tool, given the increasingly common use of smartphones, the powerful onboard computing capacity of the latest devices, and the unique relationship individuals have with their mobile phones [17]. However, the number of mHealth interventions including apps to track adherence are not as large as in SMS. In a recent review, including articles aimed to track medication adherence in patients with chronic diseases, only 23.4 % (25/107) of studies used an app or a more specialized software [8]. On the contrary, the use of SMS exclusively was the most commonly tool in 40.2 % (43/107) of studies. This is a striking result considering that in April 2012, 680 apps to enhance adherence were commercially available [18]. Still, mixed results are also found. Different medication reminder apps improved adherence to medication in hypertensive [19] or diabetic [20] patients, while no significant differences were shown in another study conducted with asthma patients [21]. Nevertheless, not all chronic patients will be willing to use a smartphone app to track adherence [22]. In an exploratory study (35 patients, mean age 67 years) assessing the usability and usefulness of existing medication management apps for older adults, participants stated that they would consider using an app to monitor adherence only if their health or memory was declining or they needed to manage medications for a relative [23]. Moreover, in order to assess the usability of the mHealth app, screen size [24], font size, comprehensive texts or the vocabulary used in the app, should correspond to the target users [25]. Currently, adherence can be monitored and medication reminders sent using several commercially available apps, although few are available in Spanish or Catalan and few incorporate a bidirectional communication tool. The objective of the present study was to evaluate our web and smartphone–based medication self-management platform, Medplan. Medplan incorporates medication reminders and healthy lifestyle advice and facilitates bidirectional communication between health care professionals and patients. In
J Med Syst (2016)4:9
the first phase of our study, we designed and developed Medplan. Here, we report the results of the second phase, in which a pilot study was conducted to assess the feasibility and effectiveness of the platform in terms of improvement in adherence, patient–professional communication, and health outcomes.
Methods Study design We performed a 6-month single-arm, prospective, pre-post intervention study. During the pre-intervention phase, patients were followed according to their usual care schedule; during the intervention phase, participants used Medplan. The study protocol was approved by the Clinical Research Ethics Committee of Hospital Clínic, Barcelona, Spain. Study participants Study participants were recruited from Casanova Primary Care Clinic and Hospital Clinic, which are located in the center of Barcelona, and were cared for by 4 physicians at the primary care clinic. The inclusion criteria were treatment for heart failure and/or hypertension and/or dyslipidemia for more than 1 month, age ≥18 years, ownership of an iOS (iPhone) or Android smartphone device, and registration in the Catalan Electronic Prescription System (CEPS). The study population also included HIV-infected patients participating in an integrated care program. These patients are well controlled HIV patients with undetectable viral loads and its follow-up is coordinated between primary and secondary care, as these patient’s profiles are similar to other chronic diseases. In fact, the aim of this program is that HIV infections may be considered like any other infection or chronic disease by patients, primary care physicians, as well as the rest of the population. The exclusion criteria were not having a smartphone, using a device that was not Android or iOS (e.g. Blackberry or Windows Phone), residence in a nursing home, cognitive disabilities or language barriers preventing use of electronic devices, or refusal to participate in the study. Study procedures Once the patients provided their consent at the initial visit, demographic and other baseline data were collected. Patients were asked to fill out the quality of life questionnaire EQ-5D [26] at baseline and at the third and sixth months. They also completed the Simplified Medication Adherence Questionnaire (SMAQ) [27] at baseline and every 6 weeks. The SMAQ is composed of 6 questions: 1) Do you ever forget
Page 3 of 14 9
to take your medicines? Yes/No; 2) Do you always take your medication at the indicated time? Yes/No; 3) Sometimes, if you feel worse, do you stop taking your medicines? Yes/No; 4) Did you not take any of your medicines over the past weekend? Yes/No 5) Last week, how often have you not taken your medicines? A: none; B: 1-2; C:3-5; D:6-10; E: more than 10; 6) Since the last visit, how many days have you not taken any medicine at all? Number of days. A patient is considered to be non adherent if he responded to question 1=yes and/or 2=no and/or 3=yes and/or 4= yes and/or 5= C, D or E and/or 6= > 2 days. Question 5 can be used as a semiquantitative measure: A: 95–100 % adherence; B: 85–94 %; C: 65–84 %; D: 30–64 %; E
MOBILE SYSTEMS
Feasibility and Preliminary Outcomes of a Web and Smartphone–Based Medication Self-Management Platform for Chronically Ill Patients Helena Anglada-Martínez 1 & Maite Martin-Conde 1 & Marina Rovira-Illamola 1 & Jose Miguel Sotoca-Momblona 1 & Ethel Sequeira 2 & Valentin Aragunde 2 & M. Angels Moreno 2 & Marta Catalan 2 & Carles Codina-Jané 1
Received: 1 November 2015 / Accepted: 29 January 2016 # Springer Science+Business Media New York 2016
Abstract Objective The aim of this study was to assess the feasibility and preliminary outcomes of a medication self-management platform for chronically ill patients, Medplan. Methods We performed a 6-month single-arm prospective pre-post intervention study of patients receiving treatment for hypertension and/or dyslipidemia and/or heart failure and/or human immunodeficiency virus infection. During the pre-intervention phase, participants were followed according to their usual care; during the intervention phase, they used Medplan. We evaluated adherence, health outcomes, healthcare resources and measured the satisfaction of patients and health care professionals. Results The study population comprised 42 patients. No differences were found in adherence to medication measured by proportion of days covered with medication (PDC). However, when adherence was measured using the SMAQ, the percent-
age of adherent patients improved during the intervention phase (p < 0.05), and the number of days with missed doses decreased (p < 0.05). Adherence measured using the Medplan app showed poor concordance with PDC. No differences were found in health outcomes or in the use of health care resources during the study period. The mean satisfaction score for Medplan was 7.2 ± 2.7 out of 10 among patients and 7.3 ± 1.7 among health care professionals. In fact, 71.4 % of participants said they would recommend the app to a friend, and 88.1 % wanted to continue using it. Conclusion The Medplan platform proved to be feasible and was well accepted by its users. However, its impact on adherence differed depending on the assessment method. The lack of effect on PDC is mainly because patients were already good adherers at baseline. The study enabled us to validate the platform in real patients using many different mobile devices and to identify potential barriers to scaling up the platform.
This article is part of the Topical Collection on Mobile Systems * Helena Anglada-Martínez [email protected]
M. Angels Moreno [email protected]
Maite Martin-Conde [email protected]
Marta Catalan [email protected]
Marina Rovira-Illamola [email protected]
Carles Codina-Jané [email protected]
Jose Miguel Sotoca-Momblona [email protected]
1
Ethel Sequeira [email protected]
Pharmacy Service, Hospital Clínic Barcelona, Villarroel, 170 08036 Barcelona, Spain
2
Consorci d’Atenció Primària de l’Eixample Casanova, GrupTransversal de Recerca en Atenció Primària, IDIBAPS, University of Barcelona, Barcelona, Spain
Valentin Aragunde [email protected]
9
Page 2 of 14
Keywords Mobile phones . Chronic patients . Medication adherence . Applications
Introduction Chronic diseases, including human immunodeficiency virus (HIV) infection, are associated with long-lasting treatments and polypharmacy. However, poor adherence to medication is both prevalent and significant, and its consequences include increased hospital admissions and health care costs [1, 2]. Improving adherence is essential when attempting to improve long-term outcomes. As C. Everett Koop, MD, said BDrugs don’t work in patients who don’t take them^. In addition, one of the most commonly reported reasons for non adherence among chronically ill patients is forgetfulness [1, 3]. The numerous barriers to adherence include complexity of the treatment plan, expense, lack of understanding, lack of perceived benefit of medication (in patients who do not feel ill), side effects, embarrassment because the patient does not want other people to know he/she is taking medication, depression, poor health literacy, and lack of trust in the health care system [2]. Therefore, interventions to improve adherence must be customized to the patient’s needs and beliefs, and more than one intervention might be necessary [1]. However, no gold standard for measuring adherence is available, as all methods have their advantages and their drawbacks [1, 4]. Direct methods (e.g. biomarkers and electronic devices) are more objective but more expensive than indirect methods, which are easier to apply but less objective (e.g. pharmacy refills, self-reporting). Consequently, it is extensively recommended to combine more than 1 method. The most common methods used in clinical practice are adherence measured by pharmacy refills and questionnaires. Pharmacy refills requires prescription claims data, although the fact that the patient refills his/her medications does not mean that they are being taken or they are taken properly [1, 5, 6]. Nevertheless, using pharmacy refills is cheap, objective, and quantifiable. Measurement by questionnaires, on the other hand, tends to overestimate adherence, although this approach is usually reliable if the patient admits to being a poor adherer [1, 5]. Although both methods are widely used in clinical practice, a recent review including mHealth intervention studies aimed at enhancing adherence to medication showed that only 5 of the 20 studies (25 %) included used more than 1 method [7]. The effects of the use in mobile phone short messages service (SMS) on medication adherence have been investigated in various studies including different patient’s profiles yielding mixed results [7, 8]. In two studies performed in Kenya, sending SMS improved adherence to antiretroviral treatment (ART) and viral load suppression [9, 10]. In contrast, in the study of Mbuagbaw et al., sending SMS did not
J Med Syst9 :04 )6102(
improved ART adherence among naïve patients after 6 months [11]. Several studies conducted in patients with asthma [12, 13] and heart failure [14], also showed that those patients receiving medication reminders had better adherence than those who don’t. However, in the study of MarquezContreras et al. [15] including antihypertensive patients, this type of intervention did not showed positive outcomes. Similarly, results on the effectiveness of mHealth in improving adherence in clinical outcomes are mixed. A recent review of studies that evaluated clinical outcomes in mHealth adherence interventions showed that of the 41 randomized controlled trials, significant improvements between groups were reported in 16 studies (39 %) [8]. On the other hand, the significant advances in telecommunication have led an extensive use of mobile phones, specially smartphones [16]. Mobile applications (apps) have the potential to be used as a supportive health tool, given the increasingly common use of smartphones, the powerful onboard computing capacity of the latest devices, and the unique relationship individuals have with their mobile phones [17]. However, the number of mHealth interventions including apps to track adherence are not as large as in SMS. In a recent review, including articles aimed to track medication adherence in patients with chronic diseases, only 23.4 % (25/107) of studies used an app or a more specialized software [8]. On the contrary, the use of SMS exclusively was the most commonly tool in 40.2 % (43/107) of studies. This is a striking result considering that in April 2012, 680 apps to enhance adherence were commercially available [18]. Still, mixed results are also found. Different medication reminder apps improved adherence to medication in hypertensive [19] or diabetic [20] patients, while no significant differences were shown in another study conducted with asthma patients [21]. Nevertheless, not all chronic patients will be willing to use a smartphone app to track adherence [22]. In an exploratory study (35 patients, mean age 67 years) assessing the usability and usefulness of existing medication management apps for older adults, participants stated that they would consider using an app to monitor adherence only if their health or memory was declining or they needed to manage medications for a relative [23]. Moreover, in order to assess the usability of the mHealth app, screen size [24], font size, comprehensive texts or the vocabulary used in the app, should correspond to the target users [25]. Currently, adherence can be monitored and medication reminders sent using several commercially available apps, although few are available in Spanish or Catalan and few incorporate a bidirectional communication tool. The objective of the present study was to evaluate our web and smartphone–based medication self-management platform, Medplan. Medplan incorporates medication reminders and healthy lifestyle advice and facilitates bidirectional communication between health care professionals and patients. In
J Med Syst (2016)4:9
the first phase of our study, we designed and developed Medplan. Here, we report the results of the second phase, in which a pilot study was conducted to assess the feasibility and effectiveness of the platform in terms of improvement in adherence, patient–professional communication, and health outcomes.
Methods Study design We performed a 6-month single-arm, prospective, pre-post intervention study. During the pre-intervention phase, patients were followed according to their usual care schedule; during the intervention phase, participants used Medplan. The study protocol was approved by the Clinical Research Ethics Committee of Hospital Clínic, Barcelona, Spain. Study participants Study participants were recruited from Casanova Primary Care Clinic and Hospital Clinic, which are located in the center of Barcelona, and were cared for by 4 physicians at the primary care clinic. The inclusion criteria were treatment for heart failure and/or hypertension and/or dyslipidemia for more than 1 month, age ≥18 years, ownership of an iOS (iPhone) or Android smartphone device, and registration in the Catalan Electronic Prescription System (CEPS). The study population also included HIV-infected patients participating in an integrated care program. These patients are well controlled HIV patients with undetectable viral loads and its follow-up is coordinated between primary and secondary care, as these patient’s profiles are similar to other chronic diseases. In fact, the aim of this program is that HIV infections may be considered like any other infection or chronic disease by patients, primary care physicians, as well as the rest of the population. The exclusion criteria were not having a smartphone, using a device that was not Android or iOS (e.g. Blackberry or Windows Phone), residence in a nursing home, cognitive disabilities or language barriers preventing use of electronic devices, or refusal to participate in the study. Study procedures Once the patients provided their consent at the initial visit, demographic and other baseline data were collected. Patients were asked to fill out the quality of life questionnaire EQ-5D [26] at baseline and at the third and sixth months. They also completed the Simplified Medication Adherence Questionnaire (SMAQ) [27] at baseline and every 6 weeks. The SMAQ is composed of 6 questions: 1) Do you ever forget
Page 3 of 14 9
to take your medicines? Yes/No; 2) Do you always take your medication at the indicated time? Yes/No; 3) Sometimes, if you feel worse, do you stop taking your medicines? Yes/No; 4) Did you not take any of your medicines over the past weekend? Yes/No 5) Last week, how often have you not taken your medicines? A: none; B: 1-2; C:3-5; D:6-10; E: more than 10; 6) Since the last visit, how many days have you not taken any medicine at all? Number of days. A patient is considered to be non adherent if he responded to question 1=yes and/or 2=no and/or 3=yes and/or 4= yes and/or 5= C, D or E and/or 6= > 2 days. Question 5 can be used as a semiquantitative measure: A: 95–100 % adherence; B: 85–94 %; C: 65–84 %; D: 30–64 %; E
