Integrating post-hospital care by digital counseling tools: A non-randomized proof-of-concept study

Integrating post-hospital care by digital counseling tools: A non-randomized proof-of-concept study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Integrating post-hospital care by digital counseling tools: A non-randomized proof-of-concept study Julia Röglin, Johanna Nitschke, Tobias Kleemann, Steffen Ortmann This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3791558/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background The duration of stays in hospitals have decreased by almost 50% to an average of 7.2 days in 2021 compared to 1992 whilst reliance on internet-based health information has increased. This trend raises concerns about potential misinterpretations and the need for enhanced post-hospital support. Methods This proof-of-concept study established a chat-based recovery counseling service providing nursing expertise and digital counseling options to patients within 7 days after discharge from hospital. Therefore, real nursing professionals where available to respond to patient queries and questions. A chatbot assisted the counselor by suggesting potential responses based on the patient's questions. This chatbot was trained using the expertise of nursing professionals. The study aimed to assess patients' acceptance, nursing professionals' commitment, and patients' willingness to contribute chat interactions and chat content for further research and tool developments. Surveys and interviews were conducted with recovery counselors to explore their attitudes towards digitalization, self-assessed digital competencies, and potential changes to the service structure. Results Within one year, 247 patients across five stations (surgery, oncology, and orthopedics) were introduced to the digital recovery counseling service. Several patients declined to participate in the study, with the main reason for refusal voluntarily given by patients being the lack of a PC in the household (68.86%). Patients in the 51–70 age group showed the highest positive responses. Out of the consenting patients, all but one agreed to donate their chat history, and 21 of the participants registered on the platform. Neurosurgery patients exhibited the highest interest, while oncology patients had limited interest due to pre-existing information. Recovery counselors reported varying degrees of improved digital competencies. The chatbot presented challenges for uniform training across specialties due to its limited dataset, emphasizing the need for a broader question set for comprehensive training. Conclusions The study shows patients acceptance for digital counselling via chat, emphasizing also nursing staff's readiness for digital expansion. Integrating digital training is vital to overcome initial doubts. Patients willingly donate data with clear information, showcasing the chatbot's potential as a nursing expert system. Expanding chat-based nurse counselling may enhance post-inpatient advice, necessitating future considerations for broader deployment. post-inpatient counseling chat-based nursing post-hospitalization Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background In recent years, a notable challenge has arisen concerning the shortening of duration of hospital stays. From 1992 to 2021, there has been an almost 50% reduction of hospitalization time, with the current average stay being 7.2 days ( 1 ).This trend has an impact on patients' recovery and overall well-being. The shortened stays often leave patients feeling inadequately prepared for the post-hospitalization recovery period, potentially resulting in complications and an increased reliance on hospital services ( 2 ). Consequently, this heightened utilization contributes to escalated costs within the healthcare system ( 3 ). The diminished interaction between patients and nursing staff, potentially being caused by the shortened duration of stays, results in a diminished transfer of health and care knowledge. Morris's study underlined this issue, revealing that patients are frequently excluded from the discharge process, hindering their ability to adequately prepare for the recovery phase ( 4 ). Moreover, a study conducted by Cocco et al. in 2018 discovered that 49% of the 400 participants routinely sought health information from the internet. Alarmingly, 34.8% of participants informed themselves about their existing health issues online before seeking emergency room care ( 5 ). The reliance on internet-derived information elevates the risk of misinterpreting symptoms and making inaccurate self-diagnoses. This propensity can exacerbate health issues, subsequently leading to an increased frequency of emergency room visits ( 6 ). To meet these challenges, it is important that hospitals and healthcare systems respond and ensure the best possible care and preparation for patients. The concept of recovery counseling in virtual space supports patients in their post-hospital recovery phase by enabling them to contact a trained nurse for up to seven days after their discharge (specified by Section 115a of the German Social Security Code V) for a predefined period if necessary and for them to answer their nursing questions about recovery. Based on the concept of recovery counseling in virtual space, a digital solution for post-hospital care counseling was established. The sub-goals of the project can be divided into three areas: Patient care, recovery counseling and chat donation. The aim of our project was to empower patients for self-care through targeted support provided by recovery counselors after hospitalization. Success was measured by assessing the acceptance of patients for post-discharge nursing consultations in general, specifically through a corresponding chat-based service. For nursing professionals, the objective was defined as their willingness to engage in counseling activities within virtual spaces. The goal for training was to investigate patients' willingness to contribute their chat interactions as data donations for the further development of the concept. These sub-goals were tested in a non-randomized proof-of-concept study as described in the next sections. Methods The non-randomized proof-of-concept study was approved by the Ethics Committee of the Brandenburg Medical Association, Cottbus office, under the reference “2022-30-ANMF-ff” for implementation at the Carl Thiem Clinic in Cottbus. For the execution of the non-randomized proof-of-concept study, the areas of neurosurgery, trauma surgery, oncology, and orthopedics of the Carl Thiem Clinic were selected. The decision to choose these departments was based on the high patient density and the specific need for post-discharge counseling due to patients' complaints. Before discharge, patients from the mentioned departments were offered the opportunity to participate in the recovery-counseling pilot. They were informed about the project and their participation in the study beforehand and by written consent. Upon agreement to participate, patients were issued a card with their personal access code for the digital counseling service. Patients were free to decide to sign up on the platform and utilize the recovery counseling service themselves. Patients who opted for the service had seven days after hospital leave to engage in online communication with the assigned counseling staff. Patients registered for counseling using a pseudonym. During the counseling sessions, the counselors had no access to the patient's medical records. The goal was to address all questions and provide information through structured communication in the chat. The project involved four internal and one external nursing professionals. Internal nursing professionals came from the fields of neurosurgery, cardiology, and oncology. The external nursing professional has worked in the field of outpatient care and is therefore familiar with the concerns of patients after hospitalization time. For registration, a pseudonym, discharge date, and voluntary information regarding the patient's medical record were required. The patient decided which information to disclose, and which should be included in the profile. All recovery counselors had the opportunity to view this profile retrospectively and then assigned themselves to the patient. A chatbot assisted the counselors in answering patients' questions, providing response options to the queries. Counselors could choose to adopt, modify, or delete the response provided by the chatbot. Crucial to the underlying concept is the constant revision and improvement of the chatbot in its role as an assistant to counselors. During the study briefing, patients were informed about the chatbot. They were informed that the chatbot would not be in direct contact with them but would support the care professional with their advice and knowledge. In addition, it was explained how the chat history is anonymized after a voluntary donation by the patient and that no personal data flows into the chatbot's training. Additionally, patients were given the opportunity to consent to an anonymous donation of the chat history, allowing for the improvement of the chatbot's training and subsequent results. Only the chat content itself was donated, excluding the information that patients had provided in their profiles. This ensured that no personally identifiable data could be incorporated into the training of the chatbot. Seven days after discharge, if consent for data donation has been granted, the chat histories were processed, transferred to a separately protected editorial system, and finally annotated. It was ensured that no personally identifiable information remained in the chat. Therefore, each chat was looked over manually and all personal data was annotated. The chatbot was trained based on the questions and information provided by the patient and the responses given by the counselor. The project aimed to investigate the acceptance for post-discharge nursing consultations among patients, considering the willingness of nursing professionals to engage in counseling activities within digital spaces. Additionally, the study examined whether patients were willing to anonymously contribute their chat interactions for the training of the chatbot. The study's objective was to engage at least 500 patients for their participation. Furthermore, the suitability of each of the five pilot stations for this type of digital offering was assessed. A secondary target was to enhance the chatbot based on insights gained from the conducted nursing consultations. The target of engaging 500 patients during the period from April to December 2022 was determined through the distribution of information sheets. Acceptance from patients were derived based on the number of information sheets distributed, commitments to participate in the study, and actual usage on the platform. The number of commitments to the study and the actual platform usage served as an indicator of the acceptance of such an offering, even in chat form. The age of the patients served as an additional validation parameter to draw possible conclusions about usage as well as acceptance patterns. These investigations were conducted both overall and for each individual clinic to determine which stations were best suited for offering such service. The willingness of nursing professionals to engage in a digital offering was qualitatively assessed through surveys and interviews with recovery counselors. Of particular interest was how they perceived their digital competence throughout the project. Patients' consent for data donation was captured through the study documentation. The enhancement of the chatbot was assessed through daily usage by the counselors whether they increasingly accepted, modified, or deleted the chatbot’s responses. The chatbot system is based on AI-driven keyword recognition. Behind this is a neural network that has been trained to use keywords to determine which class the question belongs to. This class was given certain combinations of keywords in advance. There is a predefined answer catalog for each class to ensure that the suggested answers already correspond to the professional standard that a qualified nurse would give. This ensures that, in best-case, no adjustments need to be made by the consultant. During the consultation the chatbot identifies the specific question catalog based on the patient's inquiry and subsequently provides a predefined response from the associated answer catalog. Results Within the project year, a total of 979 patients were introduced to digital recovery counseling by the nursing staff in the five stations (ref. Table 1 ). During these introductions, patients received basic information about the offering. Table 1 Number of patients at different points in time during the study Number of patients Patients being asked during the hospital stay 979 Interest in the offer (agreed to clarification) 329 Written consent to participate in the study 247 Written consent to the anonymous donation of chat history data 246 Use of the platform 15 Following the initial information, patients were asked if they wished to participate in the study. A conspicuous portion of refusals (650 patients) to participate in the study occurred prior to the information session (ref. Figure 1 ). The reasons voluntarily given by the patients were as follows: Lack of interest due to too much study information material (27.19%) Absence of a personal computer or lack of computer skills. (66,86%) Cognitive limitations on the part of the patient. (19,30%) During the patient education process, the study documented the patients' birthdates, revealing that many affirmative responses were within the age group of 51–70 years (ref. Figure 2 ). This result is somewhat surprising, as there was an assumption that patients under 50 years old would be more motivated to use such digital service. However, an examination of the fundamental interest revealed that there were few individuals under the age of 30 admitted to all units during the year, limiting the pool of candidates who could have been informed about the program. In total 247 patients have consented to participate in this study (ref. Figure 3 ). Therefore, there seems to be a general interest in the program. Except for one patient, all patients consented to donate their chat data (ref. Figure 4 ). This can be attributed to the comprehensive clarification provided, including details regarding the chatbot. Patients are well-informed about how their data will be processed, and they have the assurance that no personally identifiable information will be used for the training of the chatbot. Based on Fig. 5 , it is evident that only a fraction of the participating patients (15 patients) registered on the platform. One reason for the limited patient registration might be that within the brief timeframe (seven days post-discharge), patients were unable to recognize their potential needs. This limitation is not solely temporal; rather, patients often lack awareness of the extent of nursing expertise. A user comment illustrates this: " Upon discharge, I was asked to try out this chat, even though I felt I didn't need it. Out of curiosity, I tried it, and I was pleasantly surprised. The personalized interaction in the chat, along with the prompt and competent responses to my questions, I find very beneficial, especially because I could tell that the chat partner was familiar with my condition/operation. I would gladly use this type of recovery counseling again. " (cite of study participant). The next step involves a detailed analysis of each pilot station, with the previously mentioned statistics now specific to each respective unit. In total, 342 patients in Neurosurgery, 251 in Trauma Surgery, 302 in Orthopedics, and 84 patients in Oncology were addressed regarding the post-hospitalization nursing consultation service offering. As depicted in Fig. 6 , patients in Neurosurgery express the highest level of interest in the offered service and utilize it most frequently. Contrary to expectations, in the field of Oncology, despite the high demand in everyday ward activities, there is limited interest in utilizing such a service. The low interest on this specific unit can be attributed to the fact that patients in this group are already informed about many aspects upon admission to the hospital, supported by a substantial amount of informational material. Consequently, although initial interest was shown, it quickly decreased upon encountering the volume of informational materials. A transition into the care routine could address this issue. To assess the willingness of nursing professionals to engage in digital practices, two surveys were conducted with recovery counselors through semi-structured guideline interviews (one at the beginning and one at the end of the study [see supplementary files]). This approach was chosen to allow counselors the flexibility to freely express their opinions while focusing on specific areas. The key themes included the counselors' fundamental attitudes towards "digitalization" and their assessments of: Their own digital competencies, and What changes they deemed necessary in the structure of the service. Regarding their understanding of digitalization and prior exposure to the topic, all counselors consistently stated that they had no previous experiences with such digital tools or services. Nevertheless, they collectively recognized clear advantages, especially for nursing, citing increased speed, modernization, and simplified communication. According to their initial self-assessments of digital competencies, improvements were observed (to varying degrees) over the course of the project year. Counselors reported feeling much more confident in their virtual counseling activities alongside with progress of the study activities. Furthermore, counselors were queried about the challenges they faced in digital counseling. One challenge identified was that patient questions often tended to be highly specialized, and the counselors occasionally struggled to align their respective expertise with the specific focus of the nursing professional. This discrepancy was highlighted as one of the challenges. The recovery counselors continuously reflected on the boundaries between nursing and medical advice, expressing concerns about potentially overstepping these boundaries. Another challenge raised by counselors who were not from the pilot stations was the perceived anonymity of the digital platform, resulting in a lack of familiarity that could impact interactions with patients. Lastly, it was noted that patients sometimes preferred to express their concerns rather than directly answer the counselors' questions. For the chatbot, there were a total of 21 chat logs available. In addition to interpersonal interactions, patients primarily posed nursing-related questions, such as those regarding positioning, wound care, or challenges with daily activities. Other than their medical history, patients did not disclose any additional personally identifiable information, which is noteworthy in a positive manner. Due to the diverse nature of the offering, covering multiple specialties, the chatbot would have needed access to an extensive array of questions. Unfortunately, with only 21 chat logs available, training the chatbot uniformly across different specialties proved to remain challenging. Discussion Based on the concept of recovery counseling in the virtual space, a digital solution for post-hospitalization nursing consultation was established within a hospital setting. By testing the counseling platform prototype, the accompanying study aimed to assess the general acceptance of post-hospitalization nursing consultation and the specific acceptance of a chat-based offering among patients. The statistical analysis unequivocally demonstrates that among patients who volunteered for the study, there is a clear acceptance of post-hospitalization nursing consultation, particularly through a chat-based interface. In the past project year in Neurosurgery, the most common incidents were herniated discs, which typically require 24–36 months for recovery ( 7 ). However, with the current usage period of seven days (currently given by German law), essential recovery-related questions may not be adequately addressed, thereby most probably explaining the observed patterns of non-usage. In the future, it shall be investigated whether a possible extension of the legal time frame for contact with nursing staff could improve the results of a follow-up examination. Even though many patients welcomed our extended nursing and care services, the main share of all patients asked (i.e. 68%) did not want to participate in our study. The reasons for potential study refusals (such as no PC or PC skills or cognitive limitations) align with findings in other research papers that investigated the utilization of digital offerings among an older demographic. Liu et al. affirmed in their analysis of 48 publications that the utilization of digital offerings by the older population is relatively low ( 8 ). Kalicki et al. cited reasons for the challenges faced by the elderly in adopting digital services, including the absence of a personal computer and the cognitive and sensory limitations of the patients, which restrict their ability to operate technical devices ( 9 ). It is important to note that this study was conducted in a predominantly rural area characterized by an older population. As described by Rios ( 10 ) and Urban ( 11 ) in their respective works, this study also revealed challenges where older individuals encounter difficulties in understanding or operating digital (health) systems. According to those affected, these issues often lead to feelings of stress or anxiety. However, according to Lindberg et al. ( 12 ), Digital Health Services or eHealth applications are seen as an opportunity for primary healthcare and rural communities. Therefore, it is emphasized that digital caregiving should ideally involve a combination of digital and personal aspects. Furthermore, studies by Blusi and colleagues, as well as Paul and colleagues, indicate that older individuals are increasingly reporting positive experiences with eHealth, even in rural areas ( 13 , 14 ). Both works describe positive effects, such as a reduction in feelings of isolation among older individuals. Additionally, an improvement in their independence and positive feedback in palliative care was also mentioned. It is evident that in the future, eHealth offerings, as also applied in this study, should be further expanded, particularly in rural areas. The other focus of this study was on assessing the readiness of nursing staff for digital counseling. Interviews with counselors revealed that many nursing professionals across various fields still have reservations about using digital applications, especially when integrating them into their daily work. It becomes increasingly crucial to introduce nursing professionals to this topic during development and ideally involve them in the creation of digital nursing offers and services. Regular exposure and training in the digital space contribute to nursing professionals feeling more confident over time, as indicated by our survey outcome. The interviewees express openness to our approach as a future consideration, viewing digital offerings as a valuable complement to traditional nursing operations. However, challenges in implementation underscore potential difficulties arising from the lack of direct patient interaction inherent in digital applications. Assigning patients who are already known to the nursing staff due to their inpatient stay could address this, as the patients have subject-specific expertise. To avoid legal issues, clear distinctions between medical and nursing responsibilities should be established. Uncertainties in the digital space led counselors to question routine tasks that would be automatic in a clinical setting, highlighting the need to address these uncertainties for optimal utilization of nursing expertise in digital applications. Support and relief for the nursing profession on hospital wards are urgently needed. Due to the increasing shortage of skilled workers in nursing and the working conditions on the ward, one in five individuals is dissatisfied with their job. This is evidenced by a study conducted in twelve European countries, surveying healthcare professionals from 488 hospitals about their current work situation. Dissatisfaction arises from low wages, limited educational opportunities, and career advancement perspectives. The longstanding nursing shortage has far-reaching consequences, including a decline in the quality of patient care ( 15 ). Additionally, there is a considerable shortage of trained nursing staff, evident through numerous open positions, a high unemployment rate, and a tense situation in the job market ( 16 , 17 , 18 , 19 ). Addressing this can be achieved through digital solutions to support and balance the workload of nursing professionals, ultimately making the profession more attractive. A clear negative outcome of our study is the limited number of data recorded. For machine learning, a large amount of data is typically required. In the project, only 15 chat logs were collected. Despite valuable insights from the chats, adequate training the chatbot and optimizing suggested responses proved to remain challenging. Positive adjustments to the process can be achieved through the following approaches: Optimization of the chatbot structure: Adapting the architecture or data input could enhance the chatbot's performance. Additional chat logs: More chat logs would enable the chatbot to better address specific situations and propose considerably improved responses. Involving nursing professionals in the chatbot process: To familiarize counselors with digitalization, it's essential for them to have a basic understanding of how the chatbot works. This would allow them to be more mindful in daily consultations, knowing (and trusting!) that the chatbot incorporates these responses into its training. Involving nursing professionals in data preparation: To create a comprehensive and robust data foundation, capturing responses from all counselors to the same question or regularly inputting various nursing situations would be beneficial. This approach could integrate diverse expertise into the bot, facilitating its deployment to other units. The topic of artificial intelligence has already made its way into clinical everyday life and nursing management. Studies demonstrate that chatbots can overcome obstacles in conventional care, leading to improved outcomes and strong trust relationships with care managers ( 20 ). However, such approaches still are mainly utilized within nursing and care research only. In medical education and the nursing profession, trust plays a crucial role, fostered through direct contact. This trust should also be maintained in digital applications, such as a chat interface, as it is essential for the practice of nursing and patient care in the future ( 21 ). In their review, Vijayarani and Balamurugan revealed that chatbots in healthcare can be utilized for health information on topics such as breast cancer, obesity, ureteroscopy, as well as for coping with anxiety, depression, and stress, and for psychoeducation. However, they emphasize that chatbots cannot think like humans with wisdom and empathy, necessitating regulatory and evaluation processes. In this study, a nursing professional is involved, benefiting from the knowledge of the expert system while providing personal care advice to the patient ( 22 ). The study by Chang, Kuo and Hwang( 23 ) illustrates the suitability of chatbot applications for knowledge dissemination. They integrated a knowledge-based chatbot into nursing training for university students, resulting in improved student performance. In this way, nursing expert systems can have a positive impact on the nursing profession. Conclusions The study indicates widespread patient acceptance of digital nursing advice through chat interfaces. However, due to brief consultation periods, determining individual patient needs can be challenging, especially considering longer recovery times. Moreover, the research highlights the willingness of nursing staff to empower digital expansion within their profession. Future considerations should prioritize integrating digital applications into training to mitigate initial reservations. Additionally, providing comprehensive information on data usage enhances patients' willingness to contribute data for scientific purposes. The chatbot, functioning as a nursing expert system, demonstrates considerable potential in offering valuable support to nursing professionals in the digital space. Considering the positive outcomes, implementing chat-based nursing aftercare could effectively complement post-hospital nursing advice. Future activities should focus on creating conditions and frameworks for extending such services to other wards, ensuring optimal support for patient recovery. An existing effort, in the form of a primary care concept where each patient is assigned a nursing professional as a point of contact, strengthens patient engagement, and can positively impact recovery. The integration of digital applications such as the chatbot can not only relieve nursing professionals but also enhance their digital competence. This will be further explored in the future. Declarations Ethics approval and consent to participate The non-randomized proof-of-concept study was approved by Ethics Committee of the Brandenburg Medical Association, Cottbus office (https://www.laekb.de/kammer/ethik-kommission), under the following reference “2022-30-ANMF-ff” for implementation at the Carl Thiem Clinic in Cottbus. The consent or refusal to participate in the study was obtained from all participants and is documented as informed consent. All surveyed patients were also informed about the possibility of data donation and its use, and their consent or refusal was obtained in the form of informed consent. Consent for publication Not applicable as no individual person’s data is contained in the manuscript. Availability of data and materials The datasets generated and/or analyzed during the current study are not publicly accessible due to data protection regulations but can be requested from the corresponding author upon reasonable request. The key points from the interviews can be found in the supplementary material. Competing interests The authors declare that they have no competing interests. Funding The study was funded as part of the CHRIS project approved by the Federal Ministry of Health with the funding reference 2521TEL23A. Authors' contributions JR performed the scientific analysis of the study. She was also instrumental in drafting and revising the manuscript. JN conducted the interviews with the nurses and analyzed them. She was also involved in the preparation and revision of the manuscript. JR and JN share authorship. TK was the investigator of this study. SO was responsible for reviewing and editing the manuscript as well as supervising the study. All authors read and approved the final manuscript. Acknowledgements Not applicable. Authors' information B.Sc. Johanna Nitschke: Nursing sciences with integrative training as a nurse; professional experience in neurology; master's degree in vocational education for health professions while working; then a few years as a teacher in medical school and now an expert in nursing development in the competence center of the Carl-Thiem-Clinic’s nursing directorate with a focus on nursing research and academization. M.Sc. Julia Röglin: Employed as a Medical Data Scientist at Thiem Research; focus on applications in the field of machine learning. References Statista. Durchschnittliche Verweildauer in deutschen Krankenhäusern in den Jahren 1992 bis 2022 (in Tagen). [Online].; 2023 [cited 2023 12 13. Available from: https://de.statista.com/statistik/daten/studie/2604/umfrage/durchschnittliche-verweildauer-im-krankenhaus-seit-1992/. Pollack AH, Backonja U, Miller AD, Mishra SR, Khelifi M, Kendall L, et al. Closing the Gap: Supporting Patients' Transition to Self-Management after Hospitalization. CHI '16: Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems. 2016;: 5324-5336. Krook M, Iwarzon M, Siouta E. The Discharge Process - From a Patient’s Perspective. SAGE Open Nursing. 2020. Morris J. Registered Nurses’ Perceptions of the Discharge Planning Process for Adult Patients in an Acute Hospital. Journal of Nursing Education and Practice. 2012. Cocco AM, Zordan R, Taylor DM, Weiland TJ, Dilley SJ, Kant J, et al. Dr Google in the ED: searching for online health information by adult emergency department patients. Medical Journal of Australia. 2018; 209(8): 342-347. Ekström A, Kurland L, Farrokhnia N, Castrén M, Nordberg M. Forecasting Emergency Department Visits Using Internet Data. Annals of Emergency Medicine. 2015;: 436-442. Wong JJ, Côté P, Quesnele JJ, Stern PJ, Mior SA. The course and prognostic factors of symptomatic cervical disc herniation with radiculopathy: a systematic review of the literature. Spine J. 2014;: 1781-9. Liu L, Stroulia E, Nikolaidis I, Miguel-Cruz A, Rincon AR. Smart homes and home health monitoring technologies for older adults: A systematic review. International Journal of Medical Informatics. 2016. Kalicki AV, Moody KA, Franzosa E, Gliatto PM, Ornstein KA. Barriers to telehealth access among homebound older adults. Journal of the American Geriatrics Society. 2021;: 2404-2411. Rios GR. eHealth Literacy and Older Adults: A Review of Literature. Topics in Geriatric Rehabilitation. 2013;: 116-125. Urban M. ‘This really takes it out of you!’ The senses and emotions in digital health practices of the elderly. Digital Health. 2017. Lindberg J, Bhatt R, Ferm A. Older people and rural eHealth: perceptions of caring relations and their effects on engagement in digital primary health care. Scand J Caring Sci. 2021; 35(4): 1322-1331. Blusi M, Kristiansen L, Jong M. Exploring the influence of Internet-based caregiver support on experiences of isolation for older spouse caregivers in rural areas: a qualitative interview study. International Journal of Older People Nursing. 2014; 10(3): 211-220. Paul LR, Salmon C, Sinnarajah A, Spice R. Web-based videoconferencing for rural palliative care consultation with elderly patients at home. Supportive Care in Cancer. 2019; 27: 3321-3330. Aiken LH, Sloane DM, Bruyneel L, Van den Heede K, Sermeus W. Nurses' reports of working conditions and hospital quality of care in 12 countries in Europe. Int J Nurs Stud. 2013;: 143-53. Ariste R, Béjaoui A, Dauphin A. Critical analysis of nurses' labour market effectiveness in Canada: The hidden aspects of the shortage. Int J Health Plann Manage. 2019;: 1144-1154. Drennan VM, Ross F. Global nurse shortages-the facts, the impact and action for change. Br Med Bull. 2019;: 25-37. Winter V, Schreyögg J, Thiel A. Hospital staff shortages: Environmental and organizational determinants and implications for patient satisfaction. Health Policy. 2020; 124(4): 380-388. Xu H, Intrator O, Bowblis JR. Shortages of Staff in Nursing Homes During the COVID-19 Pandemic: What are the Driving Factors? Journal of the American Medical Directors Association. 2020; 21(10): 1371-1377. Schario ME, Bahner CA, Widenhofer TV, Rajaballey JI, Thatcher EJ. Chatbot-Assisted Care Management. Professional Case Management. 2022; 27(1): 19-25. Toader DC, Boca G, Toader R, Măcelaru M, Toader C, Ighian D, et al. The Effect of Social Presence and Chatbot Errors on Trust. Sustainability. 2020; 12(1). Vijayarani M, Balamurugan G. Chatbot in Health Care – a review. RGUHS Journal of Nursing Sciences. 2019; 9(1): 5-10. Chang CY, Kuo SY, Hwang GH. Chatbot-facilitated Nursing Education. Educational Technology & Society. 2022; 25(1): 15-27. Additional Declarations No competing interests reported. Supplementary Files FirstInterview.docx SecondInterview.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3791558","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":264617152,"identity":"bb51a4cb-c726-4146-8a48-04d1f15ef001","order_by":0,"name":"Julia Röglin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxUlEQVRIiWNgGAWjYBACxgYwZcHDwN4A4jATrUWCh4HnAGMjUVqgQAKIEojUwtzA+/DBh18SMuaSj58/nMFgLUeEw9iNDWf2SfBYzk4zbNzAkG5MhBY2NmneHgkeg9s5jI0PGA4nNhCl5S9Iy80zYC31xGlh+AHUcoOHEeiwwwmEHdbMxmzY2wDUcibNcOYMg3RDgrYYtrcxPvjxx8be4PjhBx97KqzlCdpi2Ayyqg3GNSCogYEBYugfIlSOglEwCkbByAUAKs83+TWlusQAAAAASUVORK5CYII=","orcid":"","institution":"Thiem-Research GmbH","correspondingAuthor":true,"prefix":"","firstName":"Julia","middleName":"","lastName":"Röglin","suffix":""},{"id":264617153,"identity":"617ca124-7d03-4566-a7c4-b39128bb32f7","order_by":1,"name":"Johanna Nitschke","email":"","orcid":"","institution":"Carl-Thiem-Klinikum Cottbus","correspondingAuthor":false,"prefix":"","firstName":"Johanna","middleName":"","lastName":"Nitschke","suffix":""},{"id":264617154,"identity":"73cbdc6a-0adb-4365-963e-77b7a89b83af","order_by":2,"name":"Tobias Kleemann","email":"","orcid":"","institution":"Carl-Thiem-Klinikum Cottbus","correspondingAuthor":false,"prefix":"","firstName":"Tobias","middleName":"","lastName":"Kleemann","suffix":""},{"id":264617155,"identity":"a7d2a984-56d3-4dcf-83f9-e6657fcd8bd8","order_by":3,"name":"Steffen Ortmann","email":"","orcid":"","institution":"Thiem-Research GmbH","correspondingAuthor":false,"prefix":"","firstName":"Steffen","middleName":"","lastName":"Ortmann","suffix":""}],"badges":[],"createdAt":"2023-12-22 10:59:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3791558/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3791558/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":49140123,"identity":"8bc326cd-6cf0-44f0-a5ee-777b9c3a7e18","added_by":"auto","created_at":"2024-01-03 18:16:23","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":200119,"visible":true,"origin":"","legend":"\u003cp\u003eInterest in the offer\u003c/p\u003e","description":"","filename":"Figure1interestintheoffer.png","url":"https://assets-eu.researchsquare.com/files/rs-3791558/v1/1da0d1dcbc364998c4b04c4c.png"},{"id":49140949,"identity":"b59d6df4-6c08-4192-8e3d-6cb47c9626fb","added_by":"auto","created_at":"2024-01-03 18:24:23","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":123041,"visible":true,"origin":"","legend":"\u003cp\u003eAge of commitments\u003c/p\u003e","description":"","filename":"Figure2ageofcommitments.png","url":"https://assets-eu.researchsquare.com/files/rs-3791558/v1/1bf8e5b0933eee20bcfa8235.png"},{"id":49140119,"identity":"a0a152ce-a80c-4392-96be-4dfdc047cc71","added_by":"auto","created_at":"2024-01-03 18:16:23","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":160574,"visible":true,"origin":"","legend":"\u003cp\u003eDemand for the offer seen after the clarification.\u003c/p\u003e","description":"","filename":"Figure3commitmenttothestudy.png","url":"https://assets-eu.researchsquare.com/files/rs-3791558/v1/e4206e506c631d39ebf9d743.png"},{"id":49140126,"identity":"4e966328-ab8c-4711-b29c-e2c3f3b93dbd","added_by":"auto","created_at":"2024-01-03 18:16:23","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":179609,"visible":true,"origin":"","legend":"\u003cp\u003eAgreement to donate chat data after agreeing to the study.\u003c/p\u003e","description":"","filename":"Figure4agreementtodonatechatdata.png","url":"https://assets-eu.researchsquare.com/files/rs-3791558/v1/2f91ec734b442d12e391f9e3.png"},{"id":49140124,"identity":"dc281209-ad94-44d4-a1cb-5e74dbd1fbd0","added_by":"auto","created_at":"2024-01-03 18:16:23","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":174132,"visible":true,"origin":"","legend":"\u003cp\u003eRegistration on the platform after acceptance of the study\u003c/p\u003e","description":"","filename":"Figure5useoftheoffer.png","url":"https://assets-eu.researchsquare.com/files/rs-3791558/v1/8fc78159585138551ebecc1f.png"},{"id":49140950,"identity":"3e9d19e5-39e5-44b1-95b8-aac75c148ce7","added_by":"auto","created_at":"2024-01-03 18:24:23","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":217958,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of the individual stations for the evaluation statistics\u003c/p\u003e","description":"","filename":"Figure6statisticaloverviewofthepilotstations.png","url":"https://assets-eu.researchsquare.com/files/rs-3791558/v1/d5b4636b011894865519153b.png"},{"id":51357767,"identity":"9a82d465-1439-42e7-a7ca-081bd342d1a1","added_by":"auto","created_at":"2024-02-20 07:31:12","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":785087,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3791558/v1/bdf125e3-6b3b-4bdc-bcce-d71424d66f9c.pdf"},{"id":49140121,"identity":"d0757014-2371-4a45-a7f0-4a1acd422bc4","added_by":"auto","created_at":"2024-01-03 18:16:23","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":37641,"visible":true,"origin":"","legend":"","description":"","filename":"FirstInterview.docx","url":"https://assets-eu.researchsquare.com/files/rs-3791558/v1/c77053d30a443c3de177b24a.docx"},{"id":49140122,"identity":"b4f5c991-d100-432f-98c5-f656edd0cd83","added_by":"auto","created_at":"2024-01-03 18:16:23","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":32020,"visible":true,"origin":"","legend":"","description":"","filename":"SecondInterview.docx","url":"https://assets-eu.researchsquare.com/files/rs-3791558/v1/6c2bb6592fe9c151884602af.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Integrating post-hospital care by digital counseling tools: A non-randomized proof-of-concept study","fulltext":[{"header":"Background","content":"\u003cp\u003eIn recent years, a notable challenge has arisen concerning the shortening of duration of hospital stays. From 1992 to 2021, there has been an almost 50% reduction of hospitalization time, with the current average stay being 7.2 days (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e).This trend has an impact on patients' recovery and overall well-being. The shortened stays often leave patients feeling inadequately prepared for the post-hospitalization recovery period, potentially resulting in complications and an increased reliance on hospital services (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Consequently, this heightened utilization contributes to escalated costs within the healthcare system (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). The diminished interaction between patients and nursing staff, potentially being caused by the shortened duration of stays, results in a diminished transfer of health and care knowledge. Morris's study underlined this issue, revealing that patients are frequently excluded from the discharge process, hindering their ability to adequately prepare for the recovery phase (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). Moreover, a study conducted by Cocco et al. in 2018 discovered that 49% of the 400 participants routinely sought health information from the internet. Alarmingly, 34.8% of participants informed themselves about their existing health issues online before seeking emergency room care (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). The reliance on internet-derived information elevates the risk of misinterpreting symptoms and making inaccurate self-diagnoses. This propensity can exacerbate health issues, subsequently leading to an increased frequency of emergency room visits (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo meet these challenges, it is important that hospitals and healthcare systems respond and ensure the best possible care and preparation for patients. The concept of recovery counseling in virtual space supports patients in their post-hospital recovery phase by enabling them to contact a trained nurse for up to seven days after their discharge (specified by Section 115a of the German Social Security Code V) for a predefined period if necessary and for them to answer their nursing questions about recovery. Based on the concept of recovery counseling in virtual space, a digital solution for post-hospital care counseling was established. The sub-goals of the project can be divided into three areas: Patient care, recovery counseling and chat donation.\u003c/p\u003e \u003cp\u003eThe aim of our project was to empower patients for self-care through targeted support provided by recovery counselors after hospitalization. Success was measured by assessing the acceptance of patients for post-discharge nursing consultations in general, specifically through a corresponding chat-based service. For nursing professionals, the objective was defined as their willingness to engage in counseling activities within virtual spaces. The goal for training was to investigate patients' willingness to contribute their chat interactions as data donations for the further development of the concept. These sub-goals were tested in a non-randomized proof-of-concept study as described in the next sections.\u003c/p\u003e "},{"header":"Methods","content":"\u003cp\u003e The non-randomized proof-of-concept study was approved by the Ethics Committee of the Brandenburg Medical Association, Cottbus office, under the reference “2022-30-ANMF-ff” for implementation at the Carl Thiem Clinic in Cottbus. For the execution of the non-randomized proof-of-concept study, the areas of neurosurgery, trauma surgery, oncology, and orthopedics of the Carl Thiem Clinic were selected. The decision to choose these departments was based on the high patient density and the specific need for post-discharge counseling due to patients' complaints. Before discharge, patients from the mentioned departments were offered the opportunity to participate in the recovery-counseling pilot. They were informed about the project and their participation in the study beforehand and by written consent. Upon agreement to participate, patients were issued a card with their personal access code for the digital counseling service. Patients were free to decide to sign up on the platform and utilize the recovery counseling service themselves. Patients who opted for the service had seven days after hospital leave to engage in online communication with the assigned counseling staff. Patients registered for counseling using a pseudonym. During the counseling sessions, the counselors had no access to the patient's medical records. The goal was to address all questions and provide information through structured communication in the chat. The project involved four internal and one external nursing professionals. Internal nursing professionals came from the fields of neurosurgery, cardiology, and oncology. The external nursing professional has worked in the field of outpatient care and is therefore familiar with the concerns of patients after hospitalization time. For registration, a pseudonym, discharge date, and voluntary information regarding the patient's medical record were required. The patient decided which information to disclose, and which should be included in the profile. All recovery counselors had the opportunity to view this profile retrospectively and then assigned themselves to the patient. A chatbot assisted the counselors in answering patients' questions, providing response options to the queries. Counselors could choose to adopt, modify, or delete the response provided by the chatbot. Crucial to the underlying concept is the constant revision and improvement of the chatbot in its role as an assistant to counselors. During the study briefing, patients were informed about the chatbot. They were informed that the chatbot would not be in direct contact with them but would support the care professional with their advice and knowledge. In addition, it was explained how the chat history is anonymized after a voluntary donation by the patient and that no personal data flows into the chatbot's training. Additionally, patients were given the opportunity to consent to an anonymous donation of the chat history, allowing for the improvement of the chatbot's training and subsequent results. Only the chat content itself was donated, excluding the information that patients had provided in their profiles. This ensured that no personally identifiable data could be incorporated into the training of the chatbot. Seven days after discharge, if consent for data donation has been granted, the chat histories were processed, transferred to a separately protected editorial system, and finally annotated. It was ensured that no personally identifiable information remained in the chat. Therefore, each chat was looked over manually and all personal data was annotated. The chatbot was trained based on the questions and information provided by the patient and the responses given by the counselor.\u003c/p\u003e\u003cp\u003eThe project aimed to investigate the acceptance for post-discharge nursing consultations among patients, considering the willingness of nursing professionals to engage in counseling activities within digital spaces. Additionally, the study examined whether patients were willing to anonymously contribute their chat interactions for the training of the chatbot. The study's objective was to engage at least 500 patients for their participation. Furthermore, the suitability of each of the five pilot stations for this type of digital offering was assessed. A secondary target was to enhance the chatbot based on insights gained from the conducted nursing consultations.\u003c/p\u003e\u003cp\u003eThe target of engaging 500 patients during the period from April to December 2022 was determined through the distribution of information sheets. Acceptance from patients were derived based on the number of information sheets distributed, commitments to participate in the study, and actual usage on the platform. The number of commitments to the study and the actual platform usage served as an indicator of the acceptance of such an offering, even in chat form. The age of the patients served as an additional validation parameter to draw possible conclusions about usage as well as acceptance patterns. These investigations were conducted both overall and for each individual clinic to determine which stations were best suited for offering such service. The willingness of nursing professionals to engage in a digital offering was qualitatively assessed through surveys and interviews with recovery counselors. Of particular interest was how they perceived their digital competence throughout the project. Patients' consent for data donation was captured through the study documentation. The enhancement of the chatbot was assessed through daily usage by the counselors whether they increasingly accepted, modified, or deleted the chatbot’s responses.\u003c/p\u003e\u003cp\u003eThe chatbot system is based on AI-driven keyword recognition. Behind this is a neural network that has been trained to use keywords to determine which class the question belongs to. This class was given certain combinations of keywords in advance. There is a predefined answer catalog for each class to ensure that the suggested answers already correspond to the professional standard that a qualified nurse would give. This ensures that, in best-case, no adjustments need to be made by the consultant. During the consultation the chatbot identifies the specific question catalog based on the patient's inquiry and subsequently provides a predefined response from the associated answer catalog.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eWithin the project year, a total of 979 patients were introduced to digital recovery counseling by the nursing staff in the five stations (ref. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). During these introductions, patients received basic information about the offering.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eNumber of patients at different points in time during the study\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber of patients\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePatients being asked during the hospital stay\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e979\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInterest in the offer (agreed to clarification)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e329\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWritten consent to participate in the study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e247\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWritten consent to the anonymous donation of chat history data\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e246\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUse of the platform\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFollowing the initial information, patients were asked if they wished to participate in the study. A conspicuous portion of refusals (650 patients) to participate in the study occurred prior to the information session (ref. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The reasons voluntarily given by the patients were as follows:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eLack of interest due to too much study information material (27.19%)\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eAbsence of a personal computer or lack of computer skills. (66,86%)\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eCognitive limitations on the part of the patient. (19,30%)\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eDuring the patient education process, the study documented the patients' birthdates, revealing that many affirmative responses were within the age group of 51\u0026ndash;70 years (ref. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThis result is somewhat surprising, as there was an assumption that patients under 50 years old would be more motivated to use such digital service. However, an examination of the fundamental interest revealed that there were few individuals under the age of 30 admitted to all units during the year, limiting the pool of candidates who could have been informed about the program.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn total 247 patients have consented to participate in this study (ref. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Therefore, there seems to be a general interest in the program.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eExcept for one patient, all patients consented to donate their chat data (ref. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This can be attributed to the comprehensive clarification provided, including details regarding the chatbot. Patients are well-informed about how their data will be processed, and they have the assurance that no personally identifiable information will be used for the training of the chatbot.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBased on Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, it is evident that only a fraction of the participating patients (15 patients) registered on the platform. One reason for the limited patient registration might be that within the brief timeframe (seven days post-discharge), patients were unable to recognize their potential needs. This limitation is not solely temporal; rather, patients often lack awareness of the extent of nursing expertise. A user comment illustrates this: \"\u003cem\u003eUpon discharge, I was asked to try out this chat, even though I felt I didn't need it. Out of curiosity, I tried it, and I was pleasantly surprised. The personalized interaction in the chat, along with the prompt and competent responses to my questions, I find very beneficial, especially because I could tell that the chat partner was familiar with my condition/operation. I would gladly use this type of recovery counseling again.\u003c/em\u003e\" (cite of study participant).\u003c/p\u003e \u003cp\u003eThe next step involves a detailed analysis of each pilot station, with the previously mentioned statistics now specific to each respective unit. In total, 342 patients in Neurosurgery, 251 in Trauma Surgery, 302 in Orthopedics, and 84 patients in Oncology were addressed regarding the post-hospitalization nursing consultation service offering.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAs depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, patients in Neurosurgery express the highest level of interest in the offered service and utilize it most frequently. Contrary to expectations, in the field of Oncology, despite the high demand in everyday ward activities, there is limited interest in utilizing such a service. The low interest on this specific unit can be attributed to the fact that patients in this group are already informed about many aspects upon admission to the hospital, supported by a substantial amount of informational material. Consequently, although initial interest was shown, it quickly decreased upon encountering the volume of informational materials. A transition into the care routine could address this issue.\u003c/p\u003e \u003cp\u003e To assess the willingness of nursing professionals to engage in digital practices, two surveys were conducted with recovery counselors through semi-structured guideline interviews (one at the beginning and one at the end of the study [see supplementary files]). This approach was chosen to allow counselors the flexibility to freely express their opinions while focusing on specific areas. The key themes included the counselors' fundamental attitudes towards \"digitalization\" and their assessments of:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eTheir own digital competencies, and\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eWhat changes they deemed necessary in the structure of the service.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eRegarding their understanding of digitalization and prior exposure to the topic, all counselors consistently stated that they had no previous experiences with such digital tools or services. Nevertheless, they collectively recognized clear advantages, especially for nursing, citing increased speed, modernization, and simplified communication. According to their initial self-assessments of digital competencies, improvements were observed (to varying degrees) over the course of the project year. Counselors reported feeling much more confident in their virtual counseling activities alongside with progress of the study activities.\u003c/p\u003e \u003cp\u003eFurthermore, counselors were queried about the challenges they faced in digital counseling. One challenge identified was that patient questions often tended to be highly specialized, and the counselors occasionally struggled to align their respective expertise with the specific focus of the nursing professional. This discrepancy was highlighted as one of the challenges. The recovery counselors continuously reflected on the boundaries between nursing and medical advice, expressing concerns about potentially overstepping these boundaries. Another challenge raised by counselors who were not from the pilot stations was the perceived anonymity of the digital platform, resulting in a lack of familiarity that could impact interactions with patients. Lastly, it was noted that patients sometimes preferred to express their concerns rather than directly answer the counselors' questions.\u003c/p\u003e \u003cp\u003eFor the chatbot, there were a total of 21 chat logs available. In addition to interpersonal interactions, patients primarily posed nursing-related questions, such as those regarding positioning, wound care, or challenges with daily activities. Other than their medical history, patients did not disclose any additional personally identifiable information, which is noteworthy in a positive manner. Due to the diverse nature of the offering, covering multiple specialties, the chatbot would have needed access to an extensive array of questions. Unfortunately, with only 21 chat logs available, training the chatbot uniformly across different specialties proved to remain challenging.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eBased on the concept of recovery counseling in the virtual space, a digital solution for post-hospitalization nursing consultation was established within a hospital setting. By testing the counseling platform prototype, the accompanying study aimed to assess the general acceptance of post-hospitalization nursing consultation and the specific acceptance of a chat-based offering among patients.\u003c/p\u003e \u003cp\u003eThe statistical analysis unequivocally demonstrates that among patients who volunteered for the study, there is a clear acceptance of post-hospitalization nursing consultation, particularly through a chat-based interface. In the past project year in Neurosurgery, the most common incidents were herniated discs, which typically require 24\u0026ndash;36 months for recovery (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). However, with the current usage period of seven days (currently given by German law), essential recovery-related questions may not be adequately addressed, thereby most probably explaining the observed patterns of non-usage. In the future, it shall be investigated whether a possible extension of the legal time frame for contact with nursing staff could improve the results of a follow-up examination.\u003c/p\u003e \u003cp\u003eEven though many patients welcomed our extended nursing and care services, the main share of all patients asked (i.e. 68%) did not want to participate in our study. The reasons for potential study refusals (such as no PC or PC skills or cognitive limitations) align with findings in other research papers that investigated the utilization of digital offerings among an older demographic. Liu et al. affirmed in their analysis of 48 publications that the utilization of digital offerings by the older population is relatively low (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Kalicki et al. cited reasons for the challenges faced by the elderly in adopting digital services, including the absence of a personal computer and the cognitive and sensory limitations of the patients, which restrict their ability to operate technical devices (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). It is important to note that this study was conducted in a predominantly rural area characterized by an older population. As described by Rios (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e) and Urban (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e) in their respective works, this study also revealed challenges where older individuals encounter difficulties in understanding or operating digital (health) systems. According to those affected, these issues often lead to feelings of stress or anxiety. However, according to Lindberg et al. (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e), Digital Health Services or eHealth applications are seen as an opportunity for primary healthcare and rural communities. Therefore, it is emphasized that digital caregiving should ideally involve a combination of digital and personal aspects. Furthermore, studies by Blusi and colleagues, as well as Paul and colleagues, indicate that older individuals are increasingly reporting positive experiences with eHealth, even in rural areas (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Both works describe positive effects, such as a reduction in feelings of isolation among older individuals. Additionally, an improvement in their independence and positive feedback in palliative care was also mentioned. It is evident that in the future, eHealth offerings, as also applied in this study, should be further expanded, particularly in rural areas.\u003c/p\u003e \u003cp\u003eThe other focus of this study was on assessing the readiness of nursing staff for digital counseling. Interviews with counselors revealed that many nursing professionals across various fields still have reservations about using digital applications, especially when integrating them into their daily work. It becomes increasingly crucial to introduce nursing professionals to this topic during development and ideally involve them in the creation of digital nursing offers and services. Regular exposure and training in the digital space contribute to nursing professionals feeling more confident over time, as indicated by our survey outcome. The interviewees express openness to our approach as a future consideration, viewing digital offerings as a valuable complement to traditional nursing operations. However, challenges in implementation underscore potential difficulties arising from the lack of direct patient interaction inherent in digital applications. Assigning patients who are already known to the nursing staff due to their inpatient stay could address this, as the patients have subject-specific expertise. To avoid legal issues, clear distinctions between medical and nursing responsibilities should be established. Uncertainties in the digital space led counselors to question routine tasks that would be automatic in a clinical setting, highlighting the need to address these uncertainties for optimal utilization of nursing expertise in digital applications.\u003c/p\u003e \u003cp\u003eSupport and relief for the nursing profession on hospital wards are urgently needed. Due to the increasing shortage of skilled workers in nursing and the working conditions on the ward, one in five individuals is dissatisfied with their job. This is evidenced by a study conducted in twelve European countries, surveying healthcare professionals from 488 hospitals about their current work situation. Dissatisfaction arises from low wages, limited educational opportunities, and career advancement perspectives. The longstanding nursing shortage has far-reaching consequences, including a decline in the quality of patient care (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Additionally, there is a considerable shortage of trained nursing staff, evident through numerous open positions, a high unemployment rate, and a tense situation in the job market (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). Addressing this can be achieved through digital solutions to support and balance the workload of nursing professionals, ultimately making the profession more attractive.\u003c/p\u003e \u003cp\u003eA clear negative outcome of our study is the limited number of data recorded. For machine learning, a large amount of data is typically required. In the project, only 15 chat logs were collected. Despite valuable insights from the chats, adequate training the chatbot and optimizing suggested responses proved to remain challenging. Positive adjustments to the process can be achieved through the following approaches:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eOptimization of the chatbot structure: Adapting the architecture or data input could enhance the chatbot's performance.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eAdditional chat logs: More chat logs would enable the chatbot to better address specific situations and propose considerably improved responses.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eInvolving nursing professionals in the chatbot process: To familiarize counselors with digitalization, it's essential for them to have a basic understanding of how the chatbot works. This would allow them to be more mindful in daily consultations, knowing (and trusting!) that the chatbot incorporates these responses into its training.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eInvolving nursing professionals in data preparation: To create a comprehensive and robust data foundation, capturing responses from all counselors to the same question or regularly inputting various nursing situations would be beneficial. This approach could integrate diverse expertise into the bot, facilitating its deployment to other units.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eThe topic of artificial intelligence has already made its way into clinical everyday life and nursing management. Studies demonstrate that chatbots can overcome obstacles in conventional care, leading to improved outcomes and strong trust relationships with care managers (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). However, such approaches still are mainly utilized within nursing and care research only. In medical education and the nursing profession, trust plays a crucial role, fostered through direct contact. This trust should also be maintained in digital applications, such as a chat interface, as it is essential for the practice of nursing and patient care in the future (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). In their review, Vijayarani and Balamurugan revealed that chatbots in healthcare can be utilized for health information on topics such as breast cancer, obesity, ureteroscopy, as well as for coping with anxiety, depression, and stress, and for psychoeducation. However, they emphasize that chatbots cannot think like humans with wisdom and empathy, necessitating regulatory and evaluation processes. In this study, a nursing professional is involved, benefiting from the knowledge of the expert system while providing personal care advice to the patient (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). The study by Chang, Kuo and Hwang(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e) illustrates the suitability of chatbot applications for knowledge dissemination. They integrated a knowledge-based chatbot into nursing training for university students, resulting in improved student performance. In this way, nursing expert systems can have a positive impact on the nursing profession.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThe study indicates widespread patient acceptance of digital nursing advice through chat interfaces. However, due to brief consultation periods, determining individual patient needs can be challenging, especially considering longer recovery times. Moreover, the research highlights the willingness of nursing staff to empower digital expansion within their profession. Future considerations should prioritize integrating digital applications into training to mitigate initial reservations. Additionally, providing comprehensive information on data usage enhances patients' willingness to contribute data for scientific purposes. The chatbot, functioning as a nursing expert system, demonstrates considerable potential in offering valuable support to nursing professionals in the digital space. Considering the positive outcomes, implementing chat-based nursing aftercare could effectively complement post-hospital nursing advice. Future activities should focus on creating conditions and frameworks for extending such services to other wards, ensuring optimal support for patient recovery. An existing effort, in the form of a primary care concept where each patient is assigned a nursing professional as a point of contact, strengthens patient engagement, and can positively impact recovery. The integration of digital applications such as the chatbot can not only relieve nursing professionals but also enhance their digital competence. This will be further explored in the future.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthics approval and consent to participate\u003c/h2\u003e\n\u003cp\u003eThe non-randomized proof-of-concept study was approved by Ethics Committee of the Brandenburg Medical Association, Cottbus office (https://www.laekb.de/kammer/ethik-kommission), under the following reference \u0026ldquo;2022-30-ANMF-ff\u0026rdquo; for implementation at the Carl Thiem Clinic in Cottbus.\u003c/p\u003e\n\u003cp\u003eThe consent or refusal to participate in the study was obtained from all participants and is documented as informed consent. All surveyed patients were also informed about the possibility of data donation and its use, and their consent or refusal was obtained in the form of informed consent.\u003c/p\u003e\n\u003ch2\u003eConsent for publication\u003c/h2\u003e\n\u003cp\u003eNot applicable as no individual person\u0026rsquo;s data is contained in the manuscript.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eAvailability of data and materials\u003c/h2\u003e\n\u003cp\u003eThe datasets generated and/or analyzed during the current study are not publicly accessible due to data protection regulations but can be requested from the corresponding author upon reasonable request. The key points from the interviews can be found in the supplementary material.\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThe study was funded as part of the CHRIS project approved by the Federal Ministry of Health with the funding reference 2521TEL23A.\u003c/p\u003e\n\u003ch2\u003eAuthors\u0026apos; contributions\u003c/h2\u003e\n\u003cp\u003eJR performed the scientific analysis of the study. She was also instrumental in drafting and revising the manuscript. JN conducted the interviews with the nurses and analyzed them. She was also involved in the preparation and revision of the manuscript. JR and JN share authorship. TK was the investigator of this study. SO was responsible for reviewing and editing the manuscript as well as supervising the study.\u003c/p\u003e\n\u003cp\u003eAll authors read and approved the final manuscript.\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eAuthors\u0026apos; information\u003c/h2\u003e\n\u003cp\u003eB.Sc. Johanna Nitschke: Nursing sciences with integrative training as a nurse; professional experience in neurology; master\u0026apos;s degree in vocational education for health professions while working; then a few years as a teacher in medical school and now an expert in nursing development in the competence center of the Carl-Thiem-Clinic\u0026rsquo;s nursing directorate with a focus on nursing research and academization.\u003c/p\u003e\n\u003cp\u003eM.Sc. Julia R\u0026ouml;glin: Employed as a Medical Data Scientist at Thiem Research; focus on applications in the field of machine learning.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eStatista. Durchschnittliche Verweildauer in deutschen Krankenh\u0026auml;usern in den Jahren 1992 bis 2022 (in Tagen). [Online].; 2023 [cited 2023 12 13. Available from: https://de.statista.com/statistik/daten/studie/2604/umfrage/durchschnittliche-verweildauer-im-krankenhaus-seit-1992/.\u003c/li\u003e\n \u003cli\u003ePollack AH, Backonja U, Miller AD, Mishra SR, Khelifi M, Kendall L, et al. Closing the Gap: Supporting Patients\u0026apos; Transition to Self-Management after Hospitalization. CHI \u0026apos;16: Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems.\u0026nbsp;2016;: 5324-5336.\u003c/li\u003e\n \u003cli\u003eKrook M, Iwarzon M, Siouta E. The Discharge Process - From a Patient\u0026rsquo;s Perspective.\u0026nbsp;SAGE Open Nursing. 2020.\u003c/li\u003e\n \u003cli\u003eMorris J. Registered Nurses\u0026rsquo; Perceptions of the Discharge Planning Process for Adult Patients in an Acute Hospital.\u0026nbsp;Journal of Nursing Education and Practice. 2012.\u003c/li\u003e\n \u003cli\u003eCocco AM, Zordan R, Taylor DM, Weiland TJ, Dilley SJ, Kant J, et al.\u0026nbsp;Dr Google in the ED: searching for online health information by adult emergency department patients.\u0026nbsp;Medical Journal of Australia. 2018; 209(8): 342-347.\u003c/li\u003e\n \u003cli\u003eEkstr\u0026ouml;m A, Kurland L, Farrokhnia N, Castr\u0026eacute;n M, Nordberg M. Forecasting Emergency Department Visits Using Internet Data.\u0026nbsp;Annals of Emergency Medicine. 2015;: 436-442.\u003c/li\u003e\n \u003cli\u003eWong JJ, C\u0026ocirc;t\u0026eacute; P, Quesnele JJ, Stern PJ, Mior SA.\u0026nbsp;The course and prognostic factors of symptomatic cervical disc herniation with radiculopathy: a systematic review of the literature.\u0026nbsp;Spine J. 2014;: 1781-9.\u003c/li\u003e\n \u003cli\u003eLiu L, Stroulia E, Nikolaidis I, Miguel-Cruz A, Rincon AR. Smart homes and home health monitoring technologies for older adults: A systematic review.\u0026nbsp;International Journal of Medical Informatics. 2016.\u003c/li\u003e\n \u003cli\u003eKalicki AV, Moody KA, Franzosa E, Gliatto PM, Ornstein KA.\u0026nbsp;Barriers to telehealth access among homebound older adults.\u0026nbsp;Journal of the American Geriatrics Society. 2021;: 2404-2411.\u003c/li\u003e\n \u003cli\u003eRios GR. eHealth Literacy and Older Adults: A Review of Literature.\u0026nbsp;Topics in Geriatric Rehabilitation. 2013;: 116-125.\u003c/li\u003e\n \u003cli\u003eUrban M. \u0026lsquo;This really takes it out of you!\u0026rsquo; The senses and emotions in digital health practices of the elderly.\u0026nbsp;Digital Health. 2017.\u003c/li\u003e\n \u003cli\u003eLindberg J, Bhatt R, Ferm A. Older people and rural eHealth: perceptions of caring relations and their effects on engagement in digital primary health care.\u0026nbsp;Scand J Caring Sci. 2021; 35(4): 1322-1331.\u003c/li\u003e\n \u003cli\u003eBlusi M, Kristiansen L, Jong M. Exploring the influence of Internet-based caregiver support on experiences of isolation for older spouse caregivers in rural areas: a qualitative interview study.\u0026nbsp;International Journal of Older People Nursing. 2014; 10(3): 211-220.\u003c/li\u003e\n \u003cli\u003ePaul LR, Salmon C, Sinnarajah A, Spice R. Web-based videoconferencing for rural palliative care consultation with elderly patients at home.\u0026nbsp;Supportive Care in Cancer. 2019; 27: 3321-3330.\u003c/li\u003e\n \u003cli\u003eAiken LH, Sloane DM, Bruyneel L, Van den Heede K, Sermeus W. Nurses\u0026apos; reports of working conditions and hospital quality of care in 12 countries in Europe.\u0026nbsp;Int J Nurs Stud. 2013;: 143-53.\u003c/li\u003e\n \u003cli\u003eAriste R, B\u0026eacute;jaoui A, Dauphin A. Critical analysis of nurses\u0026apos; labour market effectiveness in Canada: The hidden aspects of the shortage.\u0026nbsp;Int J Health Plann Manage. 2019;: 1144-1154.\u003c/li\u003e\n \u003cli\u003eDrennan VM, Ross F. Global nurse shortages-the facts, the impact and action for change.\u0026nbsp;Br Med Bull. 2019;: 25-37.\u003c/li\u003e\n \u003cli\u003eWinter V, Schrey\u0026ouml;gg J, Thiel A. Hospital staff shortages: Environmental and organizational determinants and implications for patient satisfaction.\u0026nbsp;Health Policy. 2020; 124(4): 380-388.\u003c/li\u003e\n \u003cli\u003eXu H, Intrator O, Bowblis JR. Shortages of Staff in Nursing Homes During the COVID-19 Pandemic: What are the Driving Factors?\u0026nbsp;Journal of the American Medical Directors Association. 2020; 21(10): 1371-1377.\u003c/li\u003e\n \u003cli\u003eSchario ME, Bahner CA, Widenhofer TV, Rajaballey JI, Thatcher EJ.\u0026nbsp;Chatbot-Assisted Care Management. Professional Case Management. 2022; 27(1): 19-25.\u003c/li\u003e\n \u003cli\u003eToader DC, Boca G, Toader R, Măcelaru M, Toader C, Ighian D, et al. The Effect of Social Presence and Chatbot Errors on Trust.\u0026nbsp;Sustainability. 2020; 12(1).\u003c/li\u003e\n \u003cli\u003eVijayarani M, Balamurugan G. Chatbot in Health Care \u0026ndash; a review. RGUHS Journal of Nursing Sciences.\u0026nbsp;2019; 9(1): 5-10.\u003c/li\u003e\n \u003cli\u003eChang CY, Kuo SY, Hwang GH. Chatbot-facilitated Nursing Education. Educational Technology \u0026amp; Society. 2022; 25(1): 15-27.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"post-inpatient counseling, chat-based nursing, post-hospitalization","lastPublishedDoi":"10.21203/rs.3.rs-3791558/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3791558/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThe duration of stays in hospitals have decreased by almost 50% to an average of 7.2 days in 2021 compared to 1992 whilst reliance on internet-based health information has increased. This trend raises concerns about potential misinterpretations and the need for enhanced post-hospital support.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThis proof-of-concept study established a chat-based recovery counseling service providing nursing expertise and digital counseling options to patients within 7 days after discharge from hospital. Therefore, real nursing professionals where available to respond to patient queries and questions. A chatbot assisted the counselor by suggesting potential responses based on the patient's questions. This chatbot was trained using the expertise of nursing professionals. The study aimed to assess patients' acceptance, nursing professionals' commitment, and patients' willingness to contribute chat interactions and chat content for further research and tool developments. Surveys and interviews were conducted with recovery counselors to explore their attitudes towards digitalization, self-assessed digital competencies, and potential changes to the service structure.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eWithin one year, 247 patients across five stations (surgery, oncology, and orthopedics) were introduced to the digital recovery counseling service. Several patients declined to participate in the study, with the main reason for refusal voluntarily given by patients being the lack of a PC in the household (68.86%). Patients in the 51\u0026ndash;70 age group showed the highest positive responses. Out of the consenting patients, all but one agreed to donate their chat history, and 21 of the participants registered on the platform. Neurosurgery patients exhibited the highest interest, while oncology patients had limited interest due to pre-existing information. Recovery counselors reported varying degrees of improved digital competencies. The chatbot presented challenges for uniform training across specialties due to its limited dataset, emphasizing the need for a broader question set for comprehensive training.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThe study shows patients acceptance for digital counselling via chat, emphasizing also nursing staff's readiness for digital expansion. Integrating digital training is vital to overcome initial doubts. Patients willingly donate data with clear information, showcasing the chatbot's potential as a nursing expert system. Expanding chat-based nurse counselling may enhance post-inpatient advice, necessitating future considerations for broader deployment.\u003c/p\u003e","manuscriptTitle":"Integrating post-hospital care by digital counseling tools: A non-randomized proof-of-concept study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-03 18:16:18","doi":"10.21203/rs.3.rs-3791558/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"95284dd6-b7f1-477d-b3ee-05a7a2f8d55b","owner":[],"postedDate":"January 3rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-06-17T08:59:41+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-03 18:16:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3791558","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3791558","identity":"rs-3791558","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}