Cost-effectiveness analysis of digital therapeutics for home-based cardiac rehabilitation for patients with atrial fibrillation after catheter ablation

Abstract

Background

In recent years, numerous guidelines and expert consensuses have recommended that digital technologies and products, such as digital therapeutics (DTx), be incorporated into cardiac rehabilitation (CR).

Objective

The study aims to compare the cost-effectiveness of DTx for home-based CR (HBCR) with that of the conventional HBCR for patients with atrial fibrillation (AF) after catheter ablation in a real-world clinical setting.

Methods

Based on a single-center, prospective, blinded, randomized, parallel controlled trial of DTx-based HBCR for AF patients after catheter ablation, 100 AF patients undergoing catheter ablation were selected and randomized at a 1:1 ratio into a UC group (conventional HBCR) and DT group (DTx-based HBCR). The intervention lasted for 12 weeks. Medical cost data and clinical efficacy data (VO₂max) were acquired from trial data. Effect data (QALYs, anxiety and depression status, health beliefs related to cardiovascular disease, and exercise self-efficacy) were obtained from a patient questionnaire. The cost-effectiveness analysis was conducted to assess the economic benefits of DTx-based HBCR versus conventional HBCR, and the robustness of the results was verified by sensitivity analysis.

Results

In the DT group (n = 49), the mean QALYs increased 0.119 ± 0.09, VO₂max increased 9.3 ± 8.0 mL/(min*kg), anxiety, depression, health belief, and exercise self-efficacy had improved, and the total medical cost was 75,080.55 ± 19,746.62 CNY. In the UC group (n = 48), the mean QALYs increased 0.077 ± 0.06, VO₂max increased 4.9 ± 6.6 mL/(min*kg), anxiety, depression, health belief, and exercise self-efficacy had improved, and the total medical cost was 73,972.66 ± 16,582.04 CNY. The difference in the medical cost was 1107.89 CNY, and the incremental cost-effectiveness ratio was 33,572.42 CNY/QALY, which was lower than the willingness-to-pay threshold of 85,698 CNY (China's per capita GDP in 2022).

Conclusion

DTx-based HBCR is more effective and cost-effective than conventional HBCR.

Introduction

Atrial fibrillation (AF) is a supraventricular tachyarrhythmia with uncoordinated atrial electrical activation and consequently ineffective atrial contraction, which is the most common sustained arrhythmia, with a prevalence of 2–4% in adults worldwide. ¹ AF patients are prone to stroke, thromboembolism, heart failure, and myocardial infarction, seriously endangering their health. ² There are approximately 59.7 million AF patients (including atrial flutter) worldwide and more than 10 million in China.^3–6 AF is a huge burden on patients, doctors, and healthcare systems globally, with annual management costs of 10,100–14,200 dollars/person in the United States and 450–3000 euros/person in Europe. ⁷ In China, the AF treatment cost is 3000 CNY/person, totaling 30 billion CNY.^2,8

As proven by evidence-based medicine, catheter ablation is safe and effective in the treatment of AF, and it is far superior to drug therapy in maintaining sinus rhythm, reducing AF burden, ameliorating AF symptoms, and improving quality of life (QoL), making it one of the main treatments for AF.^9–19 However, the completion of catheter ablation does not mean the end of treatment, and patients often suffer from discomfort in the short term after surgery. Although catheter ablation can significantly improve the symptoms of AF patients, physical problems such as decreased exercise tolerance and low cardiopulmonary fitness still occur after surgery, and patients are often faced with psychological problems, such as poor sleep quality, anxiety, and depression.^20–24 These physical and psychological problems may raise the risk of recrudescence of AF, seriously affecting the patient's QoL and restricting their return to normal work and life. Therefore, professional cardiac rehabilitation (CR) is required for AF patients after catheter ablation.

CR is a medically supervised specialty field that utilizes a multidisciplinary approach to implement comprehensive secondary prevention programs for cardiovascular disease and has been listed as a level I recommendation for the treatment of cardiovascular disease.^1,25 CR is effective at controlling cardiovascular risk factors, reducing cardiovascular risk, improving patients’ compliance with healthy lifestyle behaviors, raising QoL, and lowering the morbidity and mortality of cardiovascular events and the hospitalization rate of patients.^{21,22,26–28} There is a U-shaped correlation between the incidence of AF and exercise, i.e., sedentary behavior or long-term high-intensity exercise could lead to an increase in the incidence of AF, while low- to moderate-intensity exercise can reduce the incidence of AF.^29–34 Therefore, the core of CR is to provide patients with individualized suggestions on exercise training and develop scientific exercise prescriptions so that their daily physical activity can be enhanced to improve cardiopulmonary function, reduce the risk of chronic diseases, and improve the long-term prognosis.^25,35–39

CR can be classified into two modes: center- or hospital-based CR and home-based CR (HBCR), according to the scenarios of nursing care. ⁴⁰ Center- or hospital-based CR is highly beneficial for implementing exercise prescription and monitoring feedback on exercise status, but it is expensive, and extra time and transportation costs are needed for patients to visit hospitals and rehabilitation centers. Due to some limitations in practical use, the popularization and patient participation of center- or hospital-based CR are low.^41–43 In HBCR, the patient exercises at home without a professional present according to the exercise prescription. Since it is not limited by time, traffic, site, and weather, HBCR reduces the difficulty of participating in CR and allows for better compliance, so it has been increasingly valued by medical personnel.^44,45 Many studies have provided evidence for the effectiveness and safety of HBCR interventions,^46–48 and HBCR not only is cheaper but also has similar outcomes to outpatient CR.^49–51 However, conventional HBCR also has some deficiencies: (a) The patient might be afraid of doing the exercises due to the worry that excessive physical activity could lead to the recrudescence of AF, cardiac discomfort, and accidents, so the patient might reduce or even avoid the exercise. (b) Due to lack of effective guidance on the exercise techniques, it is difficult for the patient to master them, resulting in a poor rehabilitation effect. (c) Due to the lack of exercise status monitoring and the inability to adjust the prescription according to patient feedback about the exercise, the patient has low compliance with self-guided exercise and fails to adhere to long-term exercise.

In recent years, numerous guidelines and expert consensuses have recommended that digital technologies and products be incorporated into CR, so that precise management of CR can be provided for patients through remote monitoring, rehabilitation guidance, and information feedback, thereby improving compliance and rehabilitation effect.^2,9,25,35 DTx is an evidence-based medicine modality that collects data and monitors indicators by digital means, helping clinicians’ control and optimize the treatment, management, and prevention of diseases. ⁵² DTx devices are classified as new medical devices in several countries, and DTx and its related products must pass clinical testing before being approved by regulatory authorities. In 2017, the first prescription of DTx was approved by the US Food and Drug Administration. DTx has been applied to disease treatment, rehabilitation and prevention, and patient management in the fields of cardiovascular medicine, endocrinology, orthopedics, ophthalmology, neurology, and psychology.^53–62 As shown in several randomized controlled trials, patients undergoing DTx have better health improvement and symptom relief, and their psychological status and QoL are significantly superior to those of controls. In addition, DTx enhances doctor–patient communication and disease risk factor management, reducing medical costs and enabling precision medicine.^44,45,63

In 2019, the first randomized controlled trial of DTx-based HBCR for AF patients after catheter ablation in China was conducted in the First Affiliated Hospital of Nanjing Medical University.^64–66 Patients in the experimental group received DTx-based HBCR using the Recovery Plus app. The safety and effectiveness of DTx-based HBCR for AF patients after catheter ablation have been verified by a clinical trial. As shown in Figure 1, the procedure of DTx-based HBCR included the formulation of a personalized exercise prescription in the experimental group, remote monitoring by portable and wearable devices, exercise training guidance, and management by a rehabilitation team consisting of electrophysiologists, specialist nurses, and rehabilitative physicians. The Recovery Plus app has been approved for marketing by the National Medical Products Administration. Its user interfaces are shown in Figures 2 to 4, which display the personalized exercise prescription (including target heart rate, exercise action, and exercise duration), monitoring information (including exercise recording and physiological indicators during exercise), and exercise action guidance video for CR, respectively.

Figure 1.

Procedure of DTx-based HBCR.

Figure 2.

Exercise prescription.

Figure 4.

Guidance video for rehabilitation.

Figure 3.

Exercise recording.

Building upon the RCT, we measured outcomes of physical health (quality-adjusted life years (QALYs) and maximal oxygen uptake (VO₂max)), mental health (anxiety and depression status, health beliefs related to cardiovascular disease, and exercise self-efficacy), and real-world data on medical costs to analyze the cost-effectiveness of the DTx-based HBCR for AF patients after catheter ablation using the evaluation methods and models of health economics. In this way, the economic and social benefits of DTx for CR were comprehensively assessed from the perspective of healthcare decision-makers (HCDMs), thereby providing decision support and reference information for the application and popularization of DTx.

Methods

Design

A single-center, prospective, blinded, randomized, parallel controlled trial of DTx-based HBCR for AF patients after catheter ablation was conducted from February 2019 to July 2019 in the First Affiliated Hospital of Nanjing Medical University, Jiangsu Province, China. The AF patients undergoing catheter ablation in the Cardiovascular Medicine Department of the First Affiliated Hospital of Nanjing Medical University were enrolled as subjects. The study protocol was approved by the Human Research Ethics Committee of the First Affiliated Hospital of Nanjing Medical University (approval No.: 2020-SR-38) and registered in the US Clinical Trials Registry (Registration No.: NCT 04414007).

The inclusion criteria included (a) patients diagnosed with AF and scheduled to undergo catheter ablation, (b) those aged 18–75 years, (c) those who, or whose primary caregivers, could use smart phones, and (d) those who gave their informed consent and voluntarily participated in the study. The exclusion criteria included (a) history of catheter ablation, (b) inability to complete the questionnaire due to expression and understanding problems, (c) history of mental illness, (d) severe physical disease or severe postoperative complications, and (e) acute heart failure or other contraindications to exercise. Patients who had to stop exercising due to receiving other treatments or being troubled by other diseases were eliminated from the study. All patients enrolled signed the informed consent form. They were randomized at a 1:1 ratio using a computer random number table into a UC group (n = 50) and a DT group (n = 50).

Interventions

In the UC group, routine postoperative nursing care was given, including in-hospital assessment and education and out-of-hospital exercise for 12 weeks. During hospitalization, routine health guidance, including postoperative drugs, diet, activities, and rest, as well as disease-related precautions, was provided by the nurses, and postoperative rehabilitation manuals were distributed. The patients were followed up by telephone weekly after surgery to find out their rehabilitation status. Moreover, re-examination was performed monthly after surgery, and exercise prescriptions were formulated according to the results of cardiopulmonary exercise testing (CPET) following the FITTVP principle (F: exercise frequency; I: exercise intensity; T: exercise time; T: exercise type; V: exercise volume; and P: exercise progression). ⁶⁷ The specifics were as follows: (a) Aerobic exercise predominated. (b) Exercise of low to moderate intensity, reflected by the patient's heart rate during exercise, was prescribed. (c) Exercise was prescribed 3–5 times per week. (d) The cumulative time to reach the target heart rate in each exercise was ≥30 min, and the time of exercising was ≥150 min weekly. The target heart rate (heart rate_target) was defined as (heart rate_max–heart rate_rest)–(40–60)% + heart rate_rest, and the alarm heart rate (heart rate_alarm) was heart rate_target + 20 bpm. ⁶⁸ The exercise rehabilitation status of patients in the UC group was recorded in their exercise diary. In addition, relevant supervision, including medication reminders and diet and lifestyle guidelines, was undertaken.

In the DT group, DTx-based HBCR management was offered for 12 weeks based on standard rehabilitation nursing care. The specifics were as follows: (a) In the outpatient re-examination 1 month after surgery, CPET was conducted and an exercise prescription was formulated, with the same content as that in the UC group. The exercise prescription was sent to the patient's mobile phone through the Shukang (Recovery Plus Inc., China) app, and both heart rate_target and heart rate_alarm were set. The heart rate of the patient during exercise was monitored in real time through the portable and wearable smart heart rate belt, and an alarm was raised in case of abnormality. (b) Through the portable and wearable smart heart rate belt, the heart rhythm and heart rate during daily life activities and during exercise were monitored, and abnormal electrocardiogram signals were captured in real time and transmitted to the monitoring center through the patient's mobile phone so that the rehabilitation team could make a judgment and promptly give its opinions on diagnosis and treatment. (c) The patient's total time of exercise, effective time (heart rate_target maintained) of exercise, time to recover to resting heart rate, and subjective feeling about the exercise were recorded by the Recovery Plus app, based on which the professional rehabilitation team adjusted the exercise prescription and answered questions one to one online.

Outcome measurement

Primary outcomes were QALYs and VO₂max. Secondary outcomes were exercise self-efficacy, health beliefs related to cardiovascular disease, and anxiety and depression status. These outcomes reflected the patient's physical and psychological health status and QoL.

QALYs are a standardized universal indicator of health outcomes. This measure has been widely used in clinical trials and population studies, and its good reliability, validity, and sensitivity have been verified in many disease fields and populations. After a 12-week trial, the patient completed the Chinese version of EQ-5D-5L (developed by the EuroQol Group). Following the literature, ⁶⁹ QALYs were calculated using the health utility value in line with China's real-world situation. Higher QALY values indicate better health status.

VO₂max refers to the maximum oxygen content taken in by the body during exercise with a respiratory exchange rate ≥1.10. Measured by CPET (Jaeger MS-CPX) and breath-by-breath analysis, ²¹ VO₂max is the gold standard for assessing cardiopulmonary function and exercise tolerance. ⁷⁰ In this study, VO₂max was measured at baseline and 12 weeks later by the Bruce method using the professional exercise treadmill (h/p/cosmos, Germany) and MasterScreen CPX system (Jaeger, Germany).

The psychological condition of the patients was detected through their anxiety and depression statuses. Anxiety and depression statuses were assessed using the seven-item Generalized Anxiety Disorder Scale (GAD-7) and the Patient Health Questionnaire-9 (PHQ-9), respectively, after 12 weeks. The lower the GAD-7 and PHQ-9 scores, the milder the anxiety and depression.^71,72

Health beliefs were assessed using the Health Beliefs Related to Cardiovascular Disease Scale after 12 weeks. The scale consists of 25 items in such dimensions as cognitive sensitivity and severity of cardiovascular disease and the benefits and barriers of diet and exercise. ⁷³ In this study, the Chinese version of the scale and the 4-point Likert scale (Cronbach's α: 0.932) were used.

Exercise self-efficacy was assessed at baseline and after 12 weeks through the International Physical Activity Questionnaire. The questionnaire consists of 10 items and reflects the confidence of the subject in undertaking regular physical activity and exercise.^74,75 In this study, the differences in exercise performance were compared by measuring the percentage of patients with moderate and high levels of activity. The International Physical Activity Questionnaire was translated into Chinese by the two-step method (bilingual translation and reverse translation), and it was then compared and validated by bilingual experts specialized in questionnaire design and cross-cultural validation (Cronbach's α: 0.84).

Cost-effectiveness analysis

The difference in cost-effectiveness was compared by the incremental cost-effectiveness ratio (ICER) between the DT group and UC group. ICER was calculated by the formula below:

$$\mathrm{ICER}={\displaystyle \frac{\mathrm{COS}{\mathrm{T}}_{\mathrm{DT}\mathrm{Group}}-\mathrm{COS}{\mathrm{T}}_{\mathrm{UC}\mathrm{Group}}}{\mathrm{EFFEC}{\mathrm{T}}_{\mathrm{DT}\mathrm{Group}}-\mathrm{EFFEC}{\mathrm{T}}_{\mathrm{UC}\mathrm{Group}}}}$$

where EFFECT involved QALYs, VO₂max, exercise self-efficacy, health beliefs related to cardiovascular disease, and anxiety and depression status and COST represented medical costs during the trial, including examination fees, treatment fees, medicine fees, hospitalization fees, nursing care fees, and DTx fees. Cost data were acquired from the patient's electronic medical record. According to the recommendations of the World Health Organization and China Guidelines for Pharmacoeconomic Evaluations, the willingness-to-pay (WTP) threshold was set as 85,698 CNY/QALY (China's per capita GDP in 2022).^76,77

The robustness of the model was tested by sensitivity analysis. In the one-way sensitivity analysis, two-way sensitivity analysis was performed on the input cost data to test the degree of influence of each parameter on the results, with the parameter change range of 80–120% of the baseline. In the probabilistic sensitivity analysis, assuming that the medical cost data were gamma-distributed, the QALY data were beta-distributed, and the state transition probability was lognormal-distributed. Simulations were performed 10,000 times using the Monte Carlo method to test the probability of cost-effectiveness in the DT group and UC group under different WTP thresholds.

Data collection and analysis

The general demographic data and medical data were collected from all enrolled patients during hospitalization. CPET was performed in the outpatient re-examination 1 month after surgery and after the 12-week intervention. After the 12-week intervention, the exercise diary of patients in the UC group was recycled, and the implementation situation of exercise prescription in the DT group was directly found through the background of Recovery Plus app. The data were statistically analyzed by SPSS 22.0 and compared by the two-independent-samples-t-test between two groups and by paired t-test within each group before vs. after intervention. The rank sum test and the χ² test were used for ordered categorical variables. The test level was α = 0.05 (two-sided), and P < 0.05 was considered statistically significant.

Results

In total, 138 AF patients undergoing catheter ablation were preliminarily screened, and a total of 100 patients were enrolled. The test procedure is shown in Figure 5. During the 12-week intervention period, three patients withdrew from the study (one in the DT group and two in the UC group), so 49 patients in the DT group and 48 patients in the UC group were finally included in the analysis. ⁶⁴ There were no cardiovascular-related adverse events caused by exercise. The demographic data of patients and trial-related clinical indicators were investigated at baseline. The data were acquired from the electronic medical record database of the First Affiliated Hospital of Nanjing Medical University (Table 1).

Figure 5.

Study procedure.

Table 1.

Baseline data of patients.

	DT group (n = 49)	UC group (n = 48)	P value
Sex			0.726
Male	31 (63.3%)	32 (66.7%)
Female	18 (36.7%)	16 (33.3%)
Age	57 ± 9	57 ± 11	0.744
Educational level			0.953
Primary school or below	7 (14.3%)	7 (14.5%)
Junior high school	11 (22.3%)	9 (18.8%)
Senior high school	22 (44.9%)	24 (50.0%)
Bachelor degree or above	9 (18.4%)	8 (16.7%)
Preoperative New York Heart Association (NYHA) grade			0.964
I	17 (34.7%)	17 (35.4%)
II	23 (46.9%)	21 (43.8%)
III	7 (14.3%)	7 (14.6%)
IV	2 (4.1%)	3 (6.2%)
Type of AF			0.181
Paroxysmal	34 (69.4%)	22 (56.3%)
Persistent	15 (30.6%)	26 (43.7%)
Course of AF			0.708
≤1 year	15 (30.6%)	11 (22.9%)
1–5 years	18 (36.7%)	22 (45.8%)
5–10 years	7 (14.3%)	5 (10.4%)
≥10 years	9 (18.4%)	10 (20.9%)
Times of ablation			0.509
First	36 (73.5%)	38 (79.2%)
Repeated	13 (26.5%)	10 (20.8%)
Complications
Hypertension	22 (44.9%)	26 (54.2%)	0.361
Diabetes mellitus	5 (10.2%)	3 (6.2%)	0.497
Coronary heart disease	7 (14.3%)	7 (8.3%)	0.355
Health status within 1 year of the end of the intervention
Stable	44 (89.80%)	41 (85.42%)	0.453
Recrudesce	5 (10.20%)	7 (14.58%)	0.513
Death	0 (0%)	0 (0%)	0.621

Table 2 shows the outcome measures of patients in the DT group and UC group pre- and postintervention. At baseline, there were no significant differences between the two groups in QALYs (P = 0.538), VO₂max (P = 0.462), anxiety status (P = 0.752), depression status (P = 0.957), health beliefs (P = 0.047), and exercise self-efficacy (P = 0.646).

Table 2.

Outcome measures pre- and postintervention (Data are shown as mean ± SD).

Outcome	DT group (n = 49)			UC group (n = 48)
	Pre	Post	P value	Pre	Post	P value
Primary outcomes
QALYs	0.789 ± 0.095	0.909 ± 0.073	0.007	0.799 ± 0.091	0.876 ± 0.092	0.03
VO2max (mL/(min*kg))	19.1 ± 4.7	27.3 ± 5.6	<0.001	18.7 ± 4.9	22.9 ± 6.3	<0.001
Secondary outcomes
Anxiety	7.2 ± 3.6	3.8 ± 2.8	0.037	7.1 ± 3.5	5.1 ± 2.4	0.018
Depression	10.2 ± 3.1	5.1 ± 3.5	0.192	10.1 ± 3.3	7.0 ± 3.5	0.549
Health belief	70.4 ± 2.5	79.7 ± 8.4	<0.001	69.9 ± 2.1	74.7 ± 6.6	<0.001
Exercise self-efficacy	50.6 ± 15.2	61.7 ± 15.6	<0.001	50.3 ± 14.5	52.8 ± 17.4	0.417

Table 3 presents the changes in outcome measures after the intervention. After 12 weeks of interventions, both groups had significant improvements in primary and secondary outcomes compared to baseline. In the DT group, the mean QALYs were 0.909 ± 0.073, which increased by 0.119 ± 0.09; VO₂max was 27.3 ± 5.6 mL/(min*kg), which increased by 9.3 ± 8.0 mL/(min*kg); the GAD-7 score was 3.8 ± 2.8 points, which decreased by 3.3 ± 3.1 points; the PHQ-9 score was 5.1 ± 3.5 points, which decreased by 5.1 ± 3.6 points; the health belief score was 79.7 ± 8.4 points, which increased by 11.1 ± 10.5 points; and the exercise self-efficacy score was 61.7 ± 15.6 points, which increased by 8.3 ± 4.8 points. In the UC group, the mean QALYs were 0.876 ± 0.092, which increased by 0.077 ± 0.058; VO₂max was 22.9 ± 6.3 mL/(min*kg), which increased by 4.9 ± 6.6 mL/(min*kg); the GAD-7 score was 5.1 ± 2.4 points, which decreased by 2.7 ± 3.2 points; the PHQ-9 score was 7.0 ± 3.5 points, which decreased by 3.1 ± 2.9 points; the health belief score was 74.7 ± 6.6 points, which increased by 2.5 ± 15.2 points; and the exercise self-efficacy score was 52.8 ± 17.4 points, which increased by 4.2 ± 5.3 points. This indicates that HBCR can improve patients’ cardiopulmonary function, while the effectiveness of DTx-based HBCR was also verified. This is consistent with the findings of previous studies.^21,22,26,27 In addition, patients in the DT group had a lower GAD-7 score and PHQ-9 score than the UC group indicating lower levels of anxiety and depression in the DT group, which provides the evidence that the patients in the DT group had better psychological health than the UC group.

Table 3.

Changes in outcome measures. (Data are shown as mean ± SD).

Outcome	DT group (n = 49)	UC group (n = 48)	P value
Primary outcomes
QALYs	0.119 ± 0.09	0.077 ± 0.058	<0.001
VO2max (mL/(min*kg))	9.3 ± 8.0	4.9 ± 6.6	0.003
Secondary outcomes
Anxiety	−3.3 ± 3.1	−2.7 ± 3.2	0.762
Depression	−5.1 ± 3.6	−3.1 ± 2.9	0.207
Health belief	11.1 ± 10.5	2.5 ± 15.2	0.002
Exercise self-efficacy	8.3 ± 4.8	4.2 ± 5.3	<0.001

The medical cost data in the two groups were acquired from the electronic medical record database of the First Affiliated Hospital of Nanjing Medical University, as shown in Table 4. The total medical cost was 75,080.55 ± 19746.62 CNY in the DT group and 73,972.66 ± 16582.04 CNY in the UC group, a difference of 1107.89 CNY.

Table 4.

Medical costs in the DT group and UC group (data are shown as mean ± SD).

Medical costs (CNY)	DT group (n = 49)	UC group (n = 48)	P value
Examination	4944.06 ± 2879.13	4649.13 ± 1090.36	0.133
Treatment	65,424.39 ± 17014.3	67,153.91 ± 16244.74	0.447
Medicine	3000.89 ± 6193.89	1365.17 ± 485.36	0.001
Hospitalization	557.29 ± 336.66	480.41 ± 161.13	0.002
Nursing care	354.92 ± 259.61	324.04 ± 293.86	0.251
DTx	799	0	–
Total	75,080.55 ± 19,746.62	73,972.66 ± 16,582.04	0.343

Cost-effectiveness analysis

Table 5 shows the cost-effectiveness ratio in the DT group and UC group. Compared with those in the UC group, the mean annual QALYs increased by 0.0324, VO₂max increased by 4.5 mL/(min*kg), the GAD-7 score decreased by 0.6 points, the PHQ-9 score decreased by 1.9 points, the health belief score increased by 5.0 points, and the exercise self-efficacy score increased by 8.8 points in the DT group. The results showed that patients in the DT group had significantly higher VO₂max improvement than the UC group, indicating that patients had improved cardiopulmonary capacity.

Table 5.

ICER in the DT group and UC group.

Outcome	Difference	ICER
Primary indicators
QALYs	0.033	33,572.42
VO2max (mL/(min*kg))	4.5	246.20
Secondary indicators
Anxiety	−1.3	852.22
Depression	−1.9	583.10
Health belief	5.0	221.58
Exercise self-efficacy	8.8	125.90

The ICERs were 33,572.42 CNY/QALY, 246.20 CNY/(mL/(min*kg)) of VO₂max, 852.22 CNY/GAD-7 score, 583.10 CNY/PHQ-9 score, 221.58 CNY/health belief score, and 125.90 CNY/exercise self-efficacy score. The ICER of 33,572.42 CNY/QALY was lower than the WTP threshold of 85,698 CNY (China's per capita GDP in 2022). The above results demonstrate that DTx-based HBCR is highly cost-effective for AF patients after catheter ablation.

Sensitivity analysis

The tornado diagram of the one-way sensitivity analysis is shown in Figure 6. The results showed that the treatment fees had the greatest impact on ICER, followed by examination fees, medicine fees, DTx fees, hospitalization fees, and nursing care fees. Regardless of the changes in parameters within their ranges, ICER remained below the WTP threshold (China's per capita GDP in 2022).

Figure 6.

Results of one-way sensitivity analysis.

Probabilistic sensitivity analysis was performed through Monte Carlo simulation according to the parameter distribution of relevant items input into the model. The ICER scatter plot and cost-effectiveness acceptability curve are shown in Figures 7 and 8, respectively. In the DT group, 63.7% of the scatter points fell below the WTP threshold of 85,698 CNY (China's per capita GDP in 2022), while 92.4% of the scatter points fell below the WTP threshold of 257,094 CNY (three times China's per capita GDP in 2022). Therefore, the probability of cost-effectiveness in the DT group was higher than that in the UC group. In the DT group, the probability of cost-effectiveness was 85.2% under the WTP threshold of 85,698 CNY (China's per capita GDP in 2022), and it was 98.7% under the WTP threshold of 257,094 CNY (three times China's per capita GDP in 2022). Moreover, the probability of cost-effectiveness increased with the increase in the WTP threshold in the DT group.

Figure 7.

Scatter plot of ICER for the DT group versus the UC group.

Figure 8.

Cost-effectiveness acceptability curve.

Discussion

This study is the first randomized controlled trial of DTx-based HBCR for AF patients after catheter ablation in China. The results showed that DT was superior to UC in physical outcomes (cardiopulmonary endurance and health status) and psychological outcomes (anxiety and depression status, health beliefs related to cardiovascular disease, and exercise self-efficacy), suggesting that AF patients receiving DTx-based HBCR after catheter ablation would have better rehabilitation outcomes than those receiving conventional HBCR. Multiple randomized controlled trials have shown that DTx-based HBCR achieves better health improvement and symptom relief, and the patients’ psychological status and QoL are significantly better than those of controls.^53–62 The above findings are similar to the conclusion in this study, further demonstrating the safety and effectiveness of DTx in CR.

Furthermore, cost-effectiveness analysis on DTx-based HBCR for AF patients after catheter ablation was conducted with real-world data using the evaluation methods of health economics. The results showed that the ICER was 33,572.42 CNY/QALY between the DT group and the UC group, far below the WTP threshold of 85,698 CNY (China's per capita GDP in 2022), suggesting that DTx-based HBCR is highly cost-effective. When we tested the robustness of the model by sensitivity analysis, the model passed both one-way sensitivity analysis and probabilistic sensitivity analysis, and the results were valid and reliable. The cost of DTx refers to the management fees of DTx, which had a small impact on the results, indicating that most patients will accept the cost of DTx within a reasonable range. The ICER of DTx-based HBCR is lower than the WTP threshold, and DTx-based HBCR can reduce the fatigue of patients from visiting the hospital and traveling due to their difficulty moving, so DTx-based HBCR is more cost-effective.^78–80 Telehealth CR is cost-effective, and third-party payer funding for telehealth CR promotes patient engagement, which will increase overall CR utilization. ⁸¹ The above findings are consistent with the conclusions of this study, strongly suggesting that DTx can not only provide more precise disease management but also help save on medical costs.

In addition to the improvement of physiological outcomes in AF patients, the outcomes of psychological health (anxiety and depression status, health beliefs related to cardiovascular disease, and exercise self-efficacy) were also measured in this study to perform cost-effectiveness analysis on DTx-based HBCR from multiple perspectives. The results showed that the DT group had better psychological health status than the UC group, both their anxiety and their depression were relieved, and the health beliefs related to cardiovascular disease and exercise self-efficacy were noticeably better. CR is the comprehensive management of patients with cardiovascular diseases that aims to reduce physiological and psychological risk factors for cardiovascular diseases, improve QoL, and help patients return to normal work and life as soon as possible. The results of this study and the conclusions of relevant guidelines, consensuses, and studies corroborate each other, further confirming the value of CR. As real-world evidence, the research data in this paper supplement the theoretical foundation of CR.

For HCDMs, DTx achieves the migration of treatment scenarios and the transformation of intervention subjects, expands the ability of patients to obtain active clinical nursing within and outside the traditional scenarios (e.g., home-based nursing, remote nursing, and digital healthcare), provides an optional innovative therapeutic regimen for special patient groups (e.g., low-income, rural, and urban groups), and effectively improves medical accessibility. DTx provides technical support services for patients, nursing staff, and other end-users, mitigating the differences in economic and social factors, such as regional and medical resource allocation.

Conclusion

In conclusion, DTx-based HBCR for AF patients after catheter ablation is more effective and achieves better health status than conventional HBCR. DTx-based HBCR has a lower ICER than the WTP threshold, so it is highly cost-effective.

Limitation

First, the sample was small in this single-center trial. Therefore, the findings may vary in other areas. Second, only patients who had smart phones and could operate relevant software were enrolled, but these are younger and more educated than the average population with AF, resulting in an analysis bias. Finally, the follow-up period was short, so a longer-term follow-up study is required to determine all the effects of DTx-based HBCR.