Evidence-Based Review of Smartphone Versus Paper Asthma Action Plans on Asthma Control

Abstract

Objective: To summarize and evaluate existing literature regarding the impact of mobile asthma action plans (MAAPs) versus written asthma action plans (WAAPs) on degree of asthma control. Data Sources: PubMed, EMBASE, Web of Science, and ClinicalTrials.gov were searched (2000-January 2019) using the term asthma action plan with each of the following: smartphone, computers, handheld, mobile applications, portable electronic application, portable software application, tablet, or technology. Study Selection and Data Extraction: The search was limited to cohort and randomized controlled trials examining MAAP versus WAAP data. Data extracted included the following: study design, population, intervention, control, outcomes related to asthma control, and potential biases assessed using Cochrane Collaboration’s Risk of Bias Assessment Tool. Data Synthesis: Four of the 41 studies identified were included, each of which were randomized control trials. One study showed significant improvement using a non–asthma-specific assessment tool, 1 study showed improvement only for patients with uncontrolled asthma at baseline, and 2 studies showed no difference in asthma control scores. Overall risk of bias across all studies was low to moderate. Relevance to Patient Care and Clinical Practice: Health care providers should select an asthma action plan (AAP) format based on what the patient is most likely to understand and consistently use. Conclusions: Because of conflicting published data regarding the use of MAAPs versus WAAPs and risk of bias, it is unclear at this time whether one format of AAP is superior to the other for either adolescents or adults.

Background

In 2016, more than 8.3% of children and adults in the United States had an asthma diagnosis.¹ Asthma affects 11.8% of people under the poverty line.¹ Hospitalization for asthma exacerbation occurs in 18.3% of children and 13% of adults with asthma, leading to 2.8 deaths per million children and 13.3 deaths per million adults.¹ To optimize asthma control, and ultimately minimize risk of asthma exacerbation, the Global Initiative for Asthma (GINA) 2018 guidelines recommend all patients with asthma receive training on proper inhaler technique, education on adherence to controller medications, and an asthma action plan (AAP).² An AAP is a step-by-step instructional guide for patients to follow at home to manage their asthma symptoms on a daily basis and assist patients in recognizing and responding to worsening disease.^2,3 The green zone indicates the patient’s asthma is controlled, yellow indicates moderate symptoms, and red indicates severe symptoms.³ These zones are specific to each patient and provide instruction for how to manage asthma on a daily basis, whether it be using medications as directed, changing reliever or controller medications, using oral corticosteroids, or accessing medical care immediately.^2,3 Despite the GINA 2018 guidelines recommending an AAP for every patient with asthma, a majority of patients do not have one. Yawn and colleagues found that 8.9% of children, 2.4% of tweens, and 1% of adults had a copy of a written AAP within their medical record.⁴ Providers may not be aware of the AAP recommendations or lack confidence in patients self-managing their own condition.⁵ Patients may not value AAPs due to the AAP being outdated or simply not reflecting the needs of the patient.⁶

AAPs can be paper-based or technology-based. Technology-based AAPs can be accessed through a smartphone application (app). These apps provide reminders to enter peak flow values, use inhalers, and indicate if additional action is needed for the patient’s asthma care. These mobile AAPs (MAAPs) could benefit both adolescents and adults since 77% of Americans now own a smartphone.⁷ An 8-week single-arm study involving 20 adolescent patients demonstrated high satisfaction with the MAAP and perceived improvement in asthma control.⁸ Investigators in a single-center, nonrandomized study sought to design, develop, and pilot a MAAP for adults. The study found that 86% of patients followed self-management advice provided by the MAAP and 50% of patients acted at least once to reduce their exposure risks when prompted by the MAAP.⁹ Another single-arm study examined the use of a MAAP in 60 adult patients with at least 2 events of urgent health care utilization over the last year. Asthma control improved for the overall study population from inadequate/poor control to well controlled (P < .0001).¹⁰

Various tools are used in trials to assess a patient’s degree of asthma control. One tool, the validated Asthma Control Test (ACT) that is recognized by the American Thoracic Society (ATS) and the American Academy of Allergy, Asthma, & Immunology (AAAAI) and that is used for patients 12 years of age and older, asks the patient 5 questions about their asthma symptoms and rescue inhaler use.^11,12 ACT scores range from 25 (completely controlled) to 5 (severely uncontrolled). Scores less than or equal to 19 are classified as uncontrolled and a minimum change of 3 points is considered clinically significant. A second validated tool is the Asthma Control Questionnaire (ACQ), also recognized by the ATS and AAAAI and used for patients 12 years of age and older, which is a 7-item questionnaire assessing symptoms and rescue inhaler use.^12,13 Scores range from 0 (completely controlled) to 6 (severely uncontrolled) and a clinically significant difference is a change of 0.5. A third validated tool, the Asthma Quality of Life Questionnaire (AQLQ), recognized by the ATS and used for patients 12 years of age and older, is a 32-item questionnaire addressing symptoms, activity limitations, emotional function, and environmental exposure.¹⁴ Scores range from 1 (low quality of life) to 7 (best quality of life) with a minimum change of 0.5 considered clinically significant. The 12-Item Short-Form Health Survey (SF-12) is a questionnaire, while not specific or validated for patients with asthma, that measures 8 general concepts: physical functioning, role limitations due to physical health problems, bodily pain, general health, vitality (energy/fatigue), social functioning, role limitations due to emotional problems, and mental health (psychological distress and psychological wellbeing).¹⁵ Scores range from 0 to 100, with higher scores indicating better health status. Another validated instrument is the Knowledge, Attitude, and Self-Efficacy Asthma Questionnaire (KASE-AQ), used for patients 18 years of age and older, which is a 60-item questionnaire assessing a patient’s knowledge regarding asthma, attitudes about asthma, and their self-efficacy regarding their perceived ability to control the disorder.¹⁶ Scores range from 20 to 100, with higher scores signifying better knowledge, attitudes, and self-efficacy. A minimal change of 10 is considered significant.¹⁷

One limitation present in the 3 previously discussed trials is the lack of a comparator group. The aim of this review is to compare MAAPs versus written asthma action plans (WAAPs) on the degree of asthma control for both adolescent and adult patients.

Data Sources, Study Selection, and Data Extraction

A search of PubMed, EMBASE, Web of Science, and ClinicalTrials.gov was conducted (2000-Janaury 2019) using the search term asthma action plan along with each of the following: smartphone, computers, handheld, mobile applications, portable electronic application, portable software application, tablet, or technology. The search was initially limited to cohort and randomized controlled trials (RCTs) examining MAAP versus WAAP data. A limit to the English language was set, but if articles were translated to English, they were included in the search. In addition, a manual search of references of pertinent meta-analyses and review articles supplemented the electronic findings. Articles were included if AAPs from apps on smartphones or devices were compared with WAAPs. Review articles, studies only published in abstract form, studies without WAAP comparators, and studies not meeting the inclusion criteria were excluded. Two investigators completed the search and analyzed studies for inclusion. Data extracted from each study included study design, population, intervention, control information, primary end points, outcomes related to the primary analysis, safety assessments, and potential biases and limitations. Three investigators using the Risk of Bias Assessment Tool published by the Cochrane Collaboration assessed potential bias of each study.¹⁸ Bias was assessed across 5 categories: randomization, allocation and concealment, blinding, attrition, and outcome reporting. Discrepancies between investigators were resolved through discussion.

Results

A summary of the search strategy and study selection is presented in Figure 1. Forty-one articles were identified and screened against inclusion and exclusion criteria. Four articles met inclusion criteria, reporting results of comparisons between MAAPs and WAAPs.^19-22 A summary of the features of the MAAPs used in the included trials is provided in Table 1.^19-22 The characteristics and specific data from the included trials are reported in Table 2.^19-22

Figure 1.

Study selection summary.

Table 1.

Summary of Apps Used in the Included Studies.

Article	App Name	Description	Share Information With Physician	Graph or Chart	Peak Flow Recorded	Peak Flow Through Bluetooth	Symptoms	Zones (Red, Yellow, and Green)	FEV1	Notifications to Enter Data or Take Medication	Records Inhaler Frequency	Referred to GINA Guidelines	Alerts About Emergencies	Child Friendly
Perry et al19	Unknown	Mobile phone app	*	*	Yes	No	Yes	Yes	*	Yes, weekly	Yes	*	*	Yes
Ryan et al20	t+ asthma app	Mobile phone monitoring system	Yes, with a password on a computer	*	Yes	*	Yes	Yes	*	*	Yes	*	Yes	*
Liu et al21	Unknown	Mobile telephone-based interactive self-care software	Yes, daily	*	Yes	*	Yes	Yes	*	*	Yes	Yes	Yes	*
Kim et al22	snuCare	Smartphone-based asthma care system	Yes	*	Yes	*	Yes	Yes	*	*	*	*	Yes	*

Abbreviations: *, unknown; App, application; FEV₁, forced expiratory volume in 1 second; GINA, Global Initiative for Asthma.

Table 2.

Characteristics of Included Trials.

Study	Design	Population	Intervention	Control	Duration	Efficacy Outcome(s)	Results
Perry et al19	Randomized pilot study	Patients 12-17 years of age with persistent asthma who had been prescribed a controller medication within the previous 6 months and had a smartphone or compatible device	MAAP (n = 17)	WAAP (n = 17)	6 months	Asthma control using ACT scores	ACT score MAAP • Baseline: 21 • 6 months: 22 ACT score WAAP • Baseline: 20 • 6 months: 21 • Between groups (NS) ACT uncontrolled at baseline MAAP (n = 7) • Baseline: 11 • 6 months: 20 (P = .04) ACT uncontrolled at baseline WAAP (n = 8) • Baseline and 6 months: 17 (P = .64)
Ryan et al20	Multicenter RCT	Patients ≥12 years of age with poorly controlled asthma (≥1.5 ACQ) and had access to a smartphone or compatible device	MAAP (n = 145)	WAAP (n = 143)	6 months	• Asthma control using ACQ • Self-efficacy using KASE-AQ • Attitude using KASE-AQ	ACQ mean change • MAAP: 0.75 • WAAP: 0.73 (mean difference −0.02, 95% CI −0.23 to 0.19) KASE-AQ self-efficacy mean change • MAAP: −4.4 • WAAP: −2.4 (mean difference 2.0, 95% CI −0.3 to 4.2) KASE-AQ attitude mean change • MAAP: −1.7 • WAAP: −1.8 (mean difference of mean change −0.2, 95% CI −1.6 to 1.6)
Liu et al21	RCT	Adult patients with moderate-to-severe persistent asthma as defined by the American Thoracic Society	MAAP (n = 46)	WAAP (n = 43)	6 months	• Asthma control using SF-12 physical and mental component scores • ED visits	SF-12 physical at 6 months • MAAP: 45.5 ± 1.4 • WAAP: 40.0 ± 1.5 (P < .01) SF-12 mental at 6 months • MAAP: 50.4 ± 1.1 • WAAP: 44.4 ± 1.4 (P < .01) ED visits • MAAP: 2 • WAAP: 9 (P < .05)
Kim et al22	RCT	Patients >19 years of age with asthma who were considered to have fragile asthma	MAAP (n = 22)	WAAP (n = 22)	8 weeks	• Asthma control using ACT score • Quality of life using QLQAKA	ACT score MAAP • Baseline: 22 • 8 weeks: 21 (P = .920) ACT score WAAP • Baseline: 22 • 8 weeks: 23 (P = .217) QLQAKA MAAP • Baseline: 67 (28-81) • 8 weeks: 70 (26-85) (P = .139) QLQAKA WAAP • Baseline: 69 (29-85) • 8 weeks: 72 (38-84) (P = .027)

Abbreviations: ACQ, Asthma Control Questionnaire; ACT, Asthma Control Test; AQLQ, Asthma Quality of Life Questionnaire; ED, emergency department; KASE-AQ, Knowledge, Attitude, and Self-Efficacy Asthma Questionnaire; MAAP, mobile asthma action plan; N/A, not applicable; NAEPP, National Asthma Education and Prevention Program; NS, not significant; QLQAKA, Quality of Life Questionnaire for Adult Korean Asthmatics; RCT, randomized controlled trial; SF-12, 12-item Short-Form Health Survey; WAAP, written asthma action plan.

Perry et al conducted a 6-month randomized pilot study to compare a MAAP to a WAAP for asthma control in adolescents.¹⁹ Patients received education on both symptom-based and peak flow–based asthma monitoring and proper use of AAPs. Included patients were 12 to 17 years of age with persistent asthma, prescribed a controller medication in the past 6 months, and had a smartphone. The primary outcome was asthma control after 6 months using the ACT score. Seventeen patients in each group was needed for 80% power to detect a mean difference of 4 from baseline ACT scores. Thirty-four patients were enrolled, with a median age of 15.4 years, 38% were male, 62% African American, and 26.5% had private insurance plans. After 6 months, there was not a significant difference between ACT scores between the 2 groups. Between both groups, 15 patients had ACT scores less than 19 at baseline, indicating uncontrolled asthma. For these patients, a significant improvement in ACT score was observed for the MAAP group but not for the WAAP group. With a lower ACT at baseline for the MAAP group, one could speculate there was more room for improvement than for the WAAP group. After study completion, 92% of participants indicated they would continue using the MAAP and 100% would recommend it to a friend; the WAAP group was not surveyed. The per-protocol study design reduces generalizability, and self-efficacy data were not reported. While satisfaction with MAAPs was high, both WAAPs and MAAPs were equally effective for adolescents with controlled asthma. MAAPs may be more beneficial for adolescents with uncontrolled asthma, but with such a small sample size, this cannot be confirmed.

Ryan et al conducted a 6-month RCT across 32 centers to determine whether MAAPs improved asthma control compared with WAAPs.²⁰ Patients received education on both symptom-based and peak flow–based asthma monitoring and proper use of AAPs. Patients ≥12 years old with poorly controlled asthma (an ACQ score ≥ 1.5) and with access to a smartphone were included in the study. Two hundred eighty-eight patients were randomized to use MAAP (n = 145) or WAAP (n = 143). Baseline characteristics were similar between the MAAP and WAAP groups (66% and 59% females, 97% and 99% whites, mean age of 46.6 ± 18.0 years and 51.5 ± 17.7 years, 13% and 11% smokers, 29% and 24% on Step 1 therapy [short-acting β-agonist], 33% and 30% on Step 2 therapy [inhaled corticosteroids]), and ACQ score of 2.32 ± 0.73 and 2.29 ± 0.77, respectively. There was no significant difference between groups for the mean change in ACQ, KASE-AQ self-efficacy scale, or the KASE-AQ attitude scale. However, more patients in the MAAP group showed an improvement in the mini-AQLQ, by more than minimum important difference, at 6 months (29.7% in the MAAP group and 27.3% in the WAAP group, P = .03). While the investigators did not meet power for the per-protocol analysis, it is important to note that it is still the largest study comparing MAAPs versus WAAPs. Selection bias is a possibility with patients volunteering for enrollment. The study was 6 months in duration and specifically evaluated patients with uncontrolled asthma. The authors concluded that MAAPs did not improve asthma control or increase self-efficacy compared with WAAPs.

A similar 6-month study by Liu et al compared the degree of asthma control using a MAAP versus a WAAP in patients with moderate-to-severe asthma.²¹ Patients were instructed to measure and record daily symptoms and peak expiratory flow rate. A total of 120 patients were enrolled in the study with 60 patients randomized to the MAAP group and 60 to the WAAP group. Eighty-nine patients were analyzed at the end of the study. Baseline characteristics were similar between the MAAP and WAAP groups (mean age 50.0 ± 1.9 and 54.0 ± 2.4 years, 51.2% and 46.8% male, SF-12 physical component score 41.6 ± 1.5 and 43.2 ± 1.4, and SF-12 mental component score 48.6 ± 1.1 and 48.6 ± 1.2), respectively. After 6 months, MAAP patients had a significantly higher SF-12 physical component score both compared with baseline and WAAP patients. The SF-12 mental component did not have a significant change from baseline in MAAP patients at 6 months. However, WAAP patients had a significant decrease in the SF-12 mental component at 6 months both compared with baseline and the MAAP group. WAAP patients had significantly more visits to the emergency department. The authors concluded that a MAAP could be a promising tool for improving asthma control. The external validity of the study is limited by an unclear risk of bias. In addition, the authors used a general health survey, SF-12, rather than an asthma-specific questionnaire to assess the patient’s degree of disease control. Other limitations include the lack of a power calculation, a broad description of inclusion criteria, no description of the randomization process, and a high loss to follow-up. A 2013 Cochrane Review, which included the results of Ryan et al²⁰ and Liu et al²¹, concluded that there was insufficient evidence to advise the use of a MAAP over a WAAP²³; the authors of this review agree with the Cochrane Review assessment.

Kim et al conducted an 8-week RCT to evaluate the feasibility and efficacy of a MAAP compared with a WAAP in adult patients with asthma.²² Patients were instructed to measure and record daily symptoms and peak expiratory flow. The study included patients >19 years old who were considered to have fragile asthma from one clinic in South Korea. Patients who were not familiar with or not willing to use a MAAP were excluded. Patients were randomized to be MAAP “users” or “nonusers.” Forty-four patients were enrolled, 22 in the user group and 22 in the nonuser group. Baseline characteristics were similar between the user and nonuser groups (median age 49 [19-72] and 51 [34-67] years, and male–female ratio 4:18 and 8:14), respectively. The number of patients treated with short-term or an increased dose of systemic corticosteroids in the previous year was higher in the user group versus the nonuser group (13 versus 6, P = .035) and in the past 8 weeks (7 versus 1, P = .020). The study did not show a significant difference in ACT scores in either the user group or the nonuser group. The nonuser group had a significant increase in the Quality of Life Questionnaire for Adult Korean Asthmatics (QLQAKA) compared with the user group. While the median QLQAKA change for both groups was 3 from baseline to 8 weeks, it is speculated that the narrower range for the nonuser group at 8 weeks led to the statistically significant difference. While the authors concluded that a MAAP could be helpful in the monitoring and management of asthma, there are many study limitations to consider. Limitations of the study include a short study duration (8 weeks), absence of a power calculation, many controlled patients at baseline, a slight difference in the number of patients controlled at baseline for the 2 groups, and comparing the user and nonuser group 8-week data to baseline instead of to the other group’s 8-week data. In addition, external validity of the study is reduced due to a South Korean patient population consisting mostly of women with a median age around 50 years.

Discussion

Risk-of-Bias Assessment Summary

Overall risk of bias across all studies was low to moderate based on at least one category being ranked unclear or not applicable. Unclear was used when studies failed to report the mechanisms with which randomization, concealment, attrition, or reporting was accomplished. Blinding was not possible by the nature of the intervention, as it would be obvious to the patients whether they were using the MAAP or WAAP. Attrition was addressed in the majority of studies, providing insight into how patients were lost to follow-up and reasons for study discontinuation. There was no unresolved concern for influence over reporting or bias that would affect the validity of reported results and conclusions. Although for one study²¹ no study protocol was available. A summary of the risk of bias assessment is presented in Table 3.

Table 3.

Risk-of-Bias Assessment^a.

Study	Random Sequence Generation	Allocation Concealment	Blinding	Attrition	Selective Outcome Reporting	Comments
Perry et al19	Yes	Unclear	No	Yes	No	Did not report self-efficacy data
Ryan et al20	Yes	Yes	No	Unclear	Yes	Reasons for missing follow-up data not provided
Liu et al21	Unclear	Unclear	No	Unclear	Unclear	High loss to follow-up
Kim et al22	Unclear	Unclear	No	Yes	Yes	Median ACT scores show patients controlled at baseline

Abbreviation: ACT, Asthma Control Test.

^aYes, low risk of bias; no, high risk of bias; and unclear, unclear risk of bias.

Implications for Practice

Since the 2013 Cochrane Review, results from other studies comparing the efficacy of MAAPs and WAAPs have been published.^19,22 With these new data, the results of the current review may provide additional guidance for clinicians looking to use an AAP to optimize a patient’s asthma control. Each study included patients with different degrees of asthma control at baseline. In the Perry et al study, only a small subset of patients with uncontrolled asthma demonstrated an improvement in asthma control.¹⁹ The improvement in asthma control scores observed at the end of the Ryan et al study for the MAAP group was not statistically significant.²⁰ There is a potential that patients with uncontrolled asthma may not be using a WAAP, or not using one successfully, and a MAAP helps increase successful use of an AAP. In studies evaluating patient perceptions of a MAAP, user satisfaction was generally high and patients demonstrated willingness to use a MAAP in the future.^22,24-27 As electronic information is often more commonly available than paper information, patients with a MAAP can easily access a user-friendly and personalized plan. This is particularly important when a patient’s symptoms are worsening or in the event of an exacerbation, when a patient’s ability to use an AAP is vital. Ultimately, health care providers should consider which AAP format is best for each patient and is most likely to assist with improved asthma control. However, providers should be aware that many MAAPs available through app stores often provide recommendations not based on current evidence.²⁸ None of the MAAPs used in the studies evaluated above are available through an app store; many were created by the investigative teams and were only supplied by the provider offices. If a patient was interested in trying a MAAP, then the AsthmaMD app may be a viable customizable option since its features include peak expiratory flow recording and data sharing with the provider, although at the time of publication no studies were identified examining the benefit of this specific app.²⁹ However, this app requires a high level of understanding of the patient’s AAP. Thus, it might be beneficial for the provider to assist in downloading the MAAP, customizing it with the patient’s medications, and instructing the patient on its use. If a patient has had good success with a WAAP, and perhaps is uncomfortable with technology, there is no need to switch to a MAAP.

Limitations

There were limitations throughout these studies. First, when comparing MAAPs to WAAPs blinding to the intervention is not possible; therefore, there is always a potential level of bias. Second, almost all the studies used different apps. Each app had its own strengths and limitations, making it difficult to make a conclusion because outcomes could have depended on the app used in a certain study. For example, in a few studies, the apps provided information sharing between patients and providers and were considered very user friendly.^20-22 In another study, patients felt the app provided too many reminders.¹⁹ Third, varieties of different scales were used to evaluate asthma control and quality of life, making it difficult to compare results between studies. Use of a validated tool that focuses on asthma control rather than quality of life, such as ACT, would allow for more appropriate study-to-study comparison. Last, the studies were limited due to small sample size, inclusion of few patients with uncontrolled asthma, and short duration of treatment.

Conclusions

Because asthma is a chronic disease, future studies are needed to examine the long-term efficacy comparing WAAPs and MAAPs. The longest study to date examines the use of WAAPs and MAAPs for 6 months. Future studies of at least 1 year including patients with uncontrolled asthma at baseline will better determine the impact of MAAPs versus WAAPs on asthma control as measured by the occurrence of asthma exacerbations. Future studies should use well-designed apps that are easy for patients to use and allow for data sharing with providers.

This review critically evaluated the published literature describing the efficacy of MAAPs versus WAAPs in adolescents and adults. Risk-of-bias assessment demonstrated a low to moderate risk of bias. There were conflicting data between the trials. Most trials demonstrated no significant difference between MAAPs and WAAPs. In the Perry et al study, a small subset of adolescent patients with uncontrolled asthma had improved control after 6 months; however, the small sample size and nonsignificant difference in baseline degree of asthma control were major limitations.¹⁹ Because of the conflicting data within studies and risk of bias, it is unclear at this time whether one format of AAP is superior to the other. Health care providers should select an AAP format based on which one the patient is most likely to understand and consistently use. In adult patients, Ryan et al and Kim et al showed that both the MAAP and the WAAP could improve asthma control, suggesting that either format can be beneficial for patients.^20,22