Remote monitoring of patients with rheumatoid arthritis using a personal messenger

Full Text

Background

Optimizing medical care in the setting of overwhelming healthcare system burdens as well as staffing and time constraints requires innovative, flexible solutions. Modifications are being made to the medical institutions’ workflow management, patient routing systems, and continuing medical education programs. With a greater number of software being available every year, modern technology breakthroughs offer an array of possibilities.

These devices and software have been employed in diverse medical fields, especially by rheumatologists. Rheumatic and musculoskeletal disorders require long-term (sometimes lifelong) monitoring by specialists. Without appropriate monitoring, these disorders exert permanent effects on patients’ physical and mental health, as well as their social lives [1]. To enhance treatment outcomes, rheumatologists use electronic medical records, artificial intelligence, machine learning, clinical decision support systems, and wearable technology with data transfer capabilities, including mobile devices [2]. This software facilitates patient data classification and rapid, long-distance data transfer. Moreover, it allows delegation of certain routine tasks to digital assistants, streamlines diagnostic search, and minimizes time expenditures for healthcare personnel [3].

The most diverse set of digital tools in rheumatology are available to patients with rheumatoid arthritis (RA), the most prevalent autoimmune inflammatory disease [4, 5]. In recent years, RA incidence in Russia has risen by 17.5%. The prevalence of RA-related disability is rising along with the number of RA patients [6]. A comprehensive understanding of RA mechanisms and treatment approaches, skilled rheumatologists, and advancements in drug therapy and rehabilitation programs enable effective treatment of RA, resulting in remission or minimal disease activity [7–10]. However, maintaining treatment outcomes over the long term remains a challenge in real-world clinical settings. Furthermore, for certain patients, the mitigation in baseline disease activity during treatment was inadequate. This may result from less stringent monitoring of treatment efficacy following therapy initiation at the onset, during relapses, and after treatment [11, 12].

Remote monitoring solutions for patients with RA exhibit substantial clinical promise. Current guidelines indicate that regular monitoring by a rheumatologist during outpatient follow-up enhances the likelihood of achieving and maintaining remission or low-level disease activity, which is the primary objective of RA treatment [4, 13]. Several studies and systematic reviews on remote medical care have been published in the past five years. In 2022, the European League Against Rheumatism (EULAR) published the first guidelines for remote medical care in patients with rheumatic and musculoskeletal disorders [14]. In most publications, patients who use specialized remote monitoring programs typically have better or equivalent treatment outcomes than those who use traditional patient care techniques. However, recent systematic reviews have highlighted several challenges in the development, implementation, funding, and safety and efficacy assessment of remote monitoring software [15].

Aim

To evaluate the safety, efficacy, and technological features of remote monitoring in RA utilizing a personal messenger developed by the Department of Hospital Therapy of the Sechenov University and to examine patient satisfaction parameters.

Materials and methods

Study design

This was a prospective, non-randomized, controlled, open-label, experimental, single-center study (Fig. 1).

 

Fig. 1. Study design.

 

Eligibility criteria

The study included male and female patients over 18 years old with moderate to high disease activity who were discharged for outpatient-based follow-up and had signed a voluntary informed consent form. The Russian and EULAR recommendations were followed in making the diagnosis [3, 8, 12]. The exclusion criteria were as follows:

• Patients who developed RA before the age of 16;
• Patients with malignancies or mental disorders;
• Patients with a history of stroke or a transient ischemic attack during the previous six months;
• Patients with injuries or other conditions that exacerbate pain and restrict joint mobility;
• Patients who were pregnant or lactating; those without smartphones; and those who were not proficient in the joint self-assessment procedure (for the remote monitoring group).

Patients were omitted from the study if they met the exclusion criteria or declined to participate further.

Study setting

Every patient was followed up at the Rheumatology Department of the Sechenov University Clinical Hospital No. 1.

Subgroup analysis

The study included two groups. Group 1 patients received the standard of care with in-person consultations and used remote monitoring software. Group 2 patients only received in-person consultations.

Intervention

Questionnaires were employed to collect data for the remote patient status evaluation. The study used questionnaires validated for clinical studies as well as recommended for treating and monitoring patients with RA and validated for clinical studies. These included the Health Assessment Questionnaire (HAQ) to assess daily life functioning in patients with rheumatoid arthritis; the European Quality of Life Questionnaire (EQ-5D) questions to ascertain patient adherence, duration of morning stiffness, and the number of tender and swollen joints; and a visual analog scale to assess the overall condition [3]. Moreover, the questionnaires were employed to gauge alterations in the condition of the RA patients.

Remote monitoring

Patients in the remote monitoring group received monthly reminders to complete a questionnaire in the software (mobile application). Patients could request an unannounced consultation and complete an unscheduled survey if their condition deteriorated. The questionnaire responses were immediately reported to the attending physician. The physician contacted the patients by phone in the following scenarios:

• When the responses in the questionnaire indicated adverse developments;
• At the patient’s request and on unscheduled questionnaire completion;
• Insufficient decrease in RA activity.

Where necessary, these patients were referred for a follow-up examination to assess RA activity using DAS28 and CDAI. Additionally, the patients were consulted over the phone or in person.

Personal messenger-based software

The personal messenger-based remote monitoring software Telemedbot consists of interface subsystems, an internal software interface (Application Programming Interface, API), a backup subsystem, and PostgreSQL and Redis database management systems (DBMS) for short-term and long-term data storage.

The interface subsystem is responsible for the application logic, interactions with personal messenger APIs (specifically the use of the Telethon V2 library for the Telegram API), and data presentation in the personal messenger for both patients and physicians. An illustration of how patient data is displayed on the Telegram mobile app is provided in Fig. 2.

 

Fig. 2. Layout of the personal messenger-based software for remote monitoring.

 

The internal API subsystem effectively regulates data management in the DBMS (standard operations of record generation, updating, and removal). Redis is used for caching, and PostgreSQL secures the long-term storage of patient data, questionnaires, and outcomes. Patient data are stored anonymously, with a unique code (nickname) assigned to each patient when creating a new patient account. Consequently, only the physician who created the account can identify the patient.

The backup subsystem ensures that data are regularly uploaded and saved to an external independent object storage service called S3 (Simple Storage Service).

All subsystems run in independent Docker containers and are managed using Docker Compose. All Telemedbot messenger components are located on a virtual server in Russia.

To use Telemedbot, patients and physicians only need an IOS or Android smartphone installed with a personal messenger.

The patient interface communicates with Telemedbot by sending and receiving messages using a dedicated account in a personal messenger. Depending on the physician’s treatment strategy, the patient was reminded to complete a questionnaire on a regular basis (e.g., once a month). Once the patient agreed to complete the questionnaire, the Telemedbot would send successive messages with various questions (single- or multiple-choice, free- or semifree-form responses; in the latter case, the response was checked for conformity with the set regular expression). The questionnaire results, including the partially completed questionnaires, were immediately reported to the physician.

The physician interface also interacts via a personal messenger. Physicians can create and update new accounts as well as review patient data and questionnaire results.

Main study outcomes

The following parameters were assessed during an in-person visit after six months:

• Clinical treatment outcomes;
• Level of patient satisfaction with software-based remote monitoring;
• Time spent by healthcare personnel on remote monitoring.

Additional study outcomes

Assessments were conducted on the self-monitoring skills during treatment, technical difficulties, and willingness to continue monitoring.

Outcomes registration

Treatment efficacy (clinical outcome) was assessed based on RA activity changes from baseline using DAS28.

After six months, to assess overall satisfaction with the messenger-based medical care, patients were asked to rate the technique using the following parameters:

• Convenience and user-friendliness of the software;
• Time needed per month to utilize the software;
• Physician’s time to respond;
• Convenience of format;
• Satisfaction with treatment outcomes over six months.

Each parameter was assessed on a five-point scale, where 1 = very bad, 2 = rather bad than good, 3 = satisfactory, 4 = rather good than bad, and 5 = excellent.

Ethical review

The study was approved by the local ethics committee of the Sechenov First State Medical University (Minutes No. 22-22 of November 3, 2022).

Statistical analysis

Statistical analysis was performed using the StatTech v. 4.2.6 software (StatTech LLC, Russia). Based on the effect size determined in previous studies, an expected minimal significance level of 5%, and a statistical power of 90%, the sample size was estimated to be a minimum of 30 patients in each group. The descriptive statistics for the quantitative parameters are presented as median (Ме) and interquartile range [Q1; Q3]. The Pearson’s chi-square test was used for intergroup comparisons of the categorical variables. Differences were considered significant at a p-value <0.05.

Results

The two study groups were matched based on sex, age, serological parameters (rheumatoid factor [RF] and anti-cyclic citrullinated peptide antibody [anti-CCP] levels), and RA activity parameters at baseline (Table 1).

 

Table 1. Clinical characteristics of the patients
Parameter	Remote monitoring	In-person monitoring
Number of patients, n	30	30
Male, n (%)	6 (20.0)	4 (13.3)
Female, n (%)	24 (80.0)	26 (86.7)
Age, years, М ± SD	52.20 ± 15.23	54.10 ± 12.62
DAS28, Me [Q1–Q3]	4.46 [3.76–5.62]	4.70 [4.12–5.59]
Moderate RA activity* (%)	20 (66.7)	17 (56.7)
High RA activity (%)	10 (33.3)	13 (43.3)
RF*+, n (%)	25 (83.3)	22 (73.3)
anti-CCP*+, n (%)	10 (33.3)	10 (33.3)
Note. RA, rheumatoid arthritis; RF, rheumatoid factor; anti-CCP, anticyclic citrullinated peptide antibody.

 

After six months, RA activity was assessed in both groups using DAS28 (Table 2). To achieve optimal RA control, disease activity must be minimal, or the patient should be in remission. By the end of the follow-up, Group 1 patients who used Telemedbot attained a state of low disease activity or remission more frequently than those in Group 2 (see Table 2, Fig. 3). In the remote monitoring group, 9 (30.0%) and 11 (36.7%) patients achieved remission and low disease activity, respectively, compared to 3 (10.0%) and 8 (26.7%) patients in the in-person monitoring group (see Table 2). Thus, the disease was effectively controlled in 20 (66.7%) patients in Group 1, compared to 11 (36.7%) patients in Group 2 (see Fig. 3).

 

Fig. 3. RA control in the groups after six months.

 

Observed intergroup differences were likely due to earlier detection of worsening and absence of on-treatment improvements in the remote monitoring group, which enabled timely treatment modifications. During the follow-up period, 11 (36.6%) patients in the remote monitoring group had unfavorable changes, such as increasing pain and tender/swollen joint counts, which required an unscheduled consultation. One (3.3%) patient required previous prescriptions to be explained once more. Four patients (13.3%) received remote treatment adjustments, whereas six (20%) patients were recommended an unscheduled in-person appointment, follow-up examination, and inpatient treatment adjustment.

The analysis of patient satisfaction with medical care utilizing remote monitoring software indicated that most Group 1 (20 patients, 66.7%) patients were completely satisfied with treatment outcomes, comparable to the proportion of patients who achieved RA activity control. Most patients (27 patients, 90.0%) reported that the physician responded immediately. The convenience of the chatbot was rated excellent by 24 (80.0%) patients and good by five (16.7%) patients. One (3.3%) patient encountered difficulties in using the software and considered it inconvenient. One patient (3.3%) considered the time required to complete the questionnaires to be excessive, while another patient (3.3%) deemed it satisfactory. The remaining patients rated the application as good (3 patients, 10.0%) or excellent (25 patients, 83.3%) (Fig. 4). Three (10.0%) patients encountered technical issues (temporary switch-off during the software update process, patient mobile device issues) (Fig. 5).

 

Fig. 4. Subjective patient assessment of the remote monitoring software: 1 = very bad, 2 = rather bad than good, 3 = satisfactory, 4 = rather good than bad, and 5 = excellent.

 

Fig. 5. Subjective patient assessment of the remote monitoring software (2).

 

A greater understanding of self-monitoring and self-assessment while undergoing RA treatment was reported by patients (25 patients, 83.3%) who utilized the remote monitoring program. In total, 24 (80.0%) patients consented to use the chatbot for continued monitoring (see Fig. 5).

Discussion

In recent years, remote monitoring in RA patients has become a convenient and accessible tool for enhancing treatment outcomes. Although the number of available monitoring applications and software is rapidly expanding, only a few of them have been scientifically proven to be effective and safe. A systematic review of mobile applications for RA patients by Luo et al. revealed that only seven of the 20 assessed applications were designed in consultation with healthcare professionals [16]. Very few applications are assessed in clinical studies prior to release, and software proven to be effective in clinical studies is not widely available. The lack of information on data transfer and storage makes it challenging to evaluate the confidentiality of the available mobile applications. There have been few studies on technical solutions for the remote diagnosis of relapses [17]. According to most studies in a systematic review by Marques et al., the management of RA patients employing dedicated applications provides comparable or superior outcomes compared to conventional in-person appointments in terms of efficacy, safety, compliance, and user experience. Publication bias cannot be excluded in more than half of the analyzed randomized clinical studies, as positive outcomes are more likely to be published than negative ones [15]. Remote monitoring applications can boost patient engagement in therapy. Greater awareness of the disease and treatment modalities, confidence in the outcome from following physician advice, and improved self-assessment skills contribute to favorable treatment outcomes [18]. Notably, all remote monitoring studies in RA patients focused on clinical safety. However, the lack of knowledge on the storage and transfer of patient data makes it nearly impractical to evaluate cybersecurity, especially personal data security.

Our study demonstrates that monitoring RA patients with moderate or high disease activity using the Telemedbot personal messenger facilitates the timely accomplishment of treatment goals: remission or minimal disease activity. The treatment efficacy after six months, as determined by evaluating RA activity using DAS28, was significantly greater in the remote monitoring group than in the conventional in-person monitoring group. One significant advantage of telemonitoring is the possibility to maintain the obtained results via regular monitoring of patients for deterioration and inadequate improvements while on treatment. The method has demonstrated high patient satisfaction with treatment outcomes, increased patient engagement in therapy, and boosted the user-friendliness of the messenger.

Study limitations

Despite the reliability of the findings, the study exhibits several limitations. Although the study sample can be considered representative, the sample size precludes a multivariate analysis of the impact of individual patient characteristics, such as drug therapy variations, on treatment efficacy. The study findings can serve as a basis for future, more extensive randomized controlled studies of remote digital monitoring in RA patients.

Conclusion

This study confirms the efficacy of the RA treatment monitoring program developed at Sechenov University. Telemedbot exhibits the potential to enhance access to medical care by facilitating direct communication with physicians and providing patients with information support. The software promotes more frequent monitoring of changes in the condition, early detection of elevated RA activity, and timely treatment. Moreover, remote monitoring mitigates the need for in-person appointments, which is especially crucial for patients with mobility issues and those residing in remote areas.

Additional information

Funding source. This study was not supported by any external sources of funding.

Competing interests. The authors declare that they have no competing interests.

Authors’ contribution. All authors made a substantial contribution to the conception of the work, acquisition, analysis, interpretation of data for the work, drafting and revising the work, final approval of the version to be published and agree to be accountable for all aspects of the work. Yu.A. Prokofieva — collection and analysis of the material, text writing, literature review, editing; I.V. Menshikova — concept and design of the study, analysis of the material, approval of the final version of the article, editing; Yu.N. Belenkov, M.V. Kozhevnikova — conclusion, editing; E.A. Zheleznykh — concept and design of the study, approval of the final version of the article, editing; Z.V. Alborova — literature review, collection and analysis of the material, article editing.

About the authors

Yuliya A. Prokofeva

Sechenov First Moscow State Medical University

Author for correspondence.
Email: ulyaprokofeva@gmail.com
ORCID iD: 0000-0001-8658-3435
SPIN-code: 3545-2640
Russian Federation, Moscow

Yuri N. Belenkov

Sechenov First Moscow State Medical University

Email: belenkov_yu_n@staff.sechenov.ru
ORCID iD: 0000-0002-3014-6129
SPIN-code: 5661-4691

MD, Dr. Sci. (Medicine), academician member of the Russian Academy of Sciences

Russian Federation, Moscow

Maria V. Kozhevnikova

Sechenov First Moscow State Medical University

Email: kozhevnikova_m_v@staff.sechenov.ru
ORCID iD: 0000-0003-4778-7755
SPIN-code: 8501-9812
Russian Federation, Moscow

Elena A. Zheleznykh

Sechenov First Moscow State Medical University

Email: zheleznykh_e_a@staff.sechenov.ru
ORCID iD: 0000-0002-2596-192X
SPIN-code: 2941-4875

MD, Cand. Sci. (Medicine)

Russian Federation, Moscow

Zarina V. Alborova

Sechenov First Moscow State Medical University

Email: Zari.Alborova2002@yandex.ru
ORCID iD: 0009-0004-6090-4922
Russian Federation, Moscow

Irina V. Menshikova

Sechenov First Moscow State Medical University

Email: menshikova_i_v@staff.sechenov.ru
ORCID iD: 0000-0003-3181-5272
SPIN-code: 5373-7486

MD, Dr. Sci. (Medicine), Professor

Russian Federation, Moscow

References

Supplementary files