Dataset Information

Evidence of Human-Level Bonds Established With a Digital Conversational Agent: Cross-sectional, Retrospective Observational Study.

ABSTRACT:

Background

There are far more patients in mental distress than there is time available for mental health professionals to support them. Although digital tools may help mitigate this issue, critics have suggested that technological solutions that lack human empathy will prevent a bond or therapeutic alliance from being formed, thereby narrowing these solutions' efficacy.

Objective

We aimed to investigate whether users of a cognitive behavioral therapy (CBT)-based conversational agent would report therapeutic bond levels that are similar to those in literature about other CBT modalities, including face-to-face therapy, group CBT, and other digital interventions that do not use a conversational agent.

Methods

A cross-sectional, retrospective study design was used to analyze aggregate, deidentified data from adult users who self-referred to a CBT-based, fully automated conversational agent (Woebot) between November 2019 and August 2020. Working alliance was measured with the Working Alliance Inventory-Short Revised (WAI-SR), and depression symptom status was assessed by using the 2-item Patient Health Questionnaire (PHQ-2). All measures were administered by the conversational agent in the mobile app. WAI-SR scores were compared to those in scientific literature abstracted from recent reviews.

Results

Data from 36,070 Woebot users were included in the analysis. Participants ranged in age from 18 to 78 years, and 57.48% (20,734/36,070) of participants reported that they were female. The mean PHQ-2 score was 3.03 (SD 1.79), and 54.67% (19,719/36,070) of users scored over the cutoff score of 3 for depression screening. Within 5 days of initial app use, the mean WAI-SR score was 3.36 (SD 0.8) and the mean bond subscale score was 3.8 (SD 1.0), which was comparable to those in recent studies from the literature on traditional, outpatient, individual CBT and group CBT (mean bond subscale scores of 4 and 3.8, respectively). PHQ-2 scores at baseline weakly correlated with bond scores (r=-0.04; P<.001); however, users with depression and those without depression had high bond scores of 3.45.

Conclusions

Although bonds are often presumed to be the exclusive domain of human therapeutic relationships, our findings challenge the notion that digital therapeutics are incapable of establishing a therapeutic bond with users. Future research might investigate the role of bonds as mediators of clinical outcomes, since boosting the engagement and efficacy of digital therapeutics could have major public health benefits.

SUBMITTER: Darcy A

PROVIDER: S-EPMC8150389 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundDue to time limitations, the preanesthetic consultation (PAC) is not the best time for patients to integrate information specific to their perioperative care pathway.ObjectiveThe main objectives of this study were to evaluate the effectiveness of a digital companion on patients' knowledge of anesthesia and their satisfaction after real-life implementation.MethodsWe conducted a prospective, monocentric, comparative study using a before-and-after design. In phase 1, a 9-item self-reported anesthesia knowledge test (Delphi method) was administered to patients before and after their PAC (control group: PAC group). In phase 2, the study was repeated immediately after the implementation of a digital conversational agent, MyAnesth (@+PAC group). Patients' satisfaction and their representations for anesthesia were also assessed using a Likert scale and the Abric method of hierarchized evocation.ResultsA total of 600 tests were distributed; 205 patients and 98 patients were included in the PAC group and @+PAC group, respectively. Demographic characteristics and mean scores on the 9-point preinformation test (PAC group: 4.2 points, 95% CI 3.9-4.4; @+PAC: 4.3 points, 95% CI 4-4.7; P=.37) were similar in the two groups. The mean score after receiving information was better in the @+PAC group than in the PAC group (6.1 points, 95% CI 5.8-6.4 points versus 5.2 points, 95% CI 5.0-5.4 points, respectively; P<.001), with an added value of 0.7 points (95% CI 0.3-1.1; P<.001). Among the respondents in the @+PAC group, 82% found the information to be clear and appropriate, and 74% found it easily accessible. Before receiving information, the central core of patients' representations for anesthesia was focused on the fear of being put to sleep and thereafter on caregiver skills and comfort.ConclusionsThe implementation of our digital conversational agent in addition to the PAC improved patients' knowledge about their perioperative care pathway. This innovative audiovisual support seemed clear, adapted, easily accessible, and reassuring. Future studies should focus on adapting both the content and delivery of a digital conversational agent for the PAC in order to maximize its benefit to patients.

Project description:We examined mental time travel reflected onto individuals' utterances in real-life conversations using a naturalistic observation method: Electronically Activated Recorder (EAR, a portable audio recorder that periodically and unobtrusively records snippets of ambient sounds and speech). We introduced the term conversational time travel and examined, for the first time, how much individuals talked about their personal past versus personal future in real life. Study 1 included 9,010 sound files collected from 51 American adults who carried the EAR over 1 weekend and were recorded every 9 min for 50 s. Study 2 included 23,103 sound files from 33 young and 48 healthy older adults from Switzerland who carried the EAR for 4 days (2 weekdays and 1 weekend, counterbalanced). 30-s recordings occurred randomly throughout the day. We developed a new coding scheme for conversational time travel: We listened to all sound files and coded each file for whether the participant was talking or not. Those sound files that included participant speech were also coded in terms of their temporal focus (e.g., past, future, present, time-independent) and autobiographical nature (i.e., about the self, about others). We, first, validated our coding scheme using the text analysis tool, Linguistic Inquiry and Word Count. Next, we compared the percentages of past- and future-oriented utterances about the self (to tap onto conversational time travel). Results were consistent across all samples and showed that participants talked about their personal past two to three times as much as their personal future (i.e., retrospective bias). This is in contrast to research showing a prospective bias in thinking behavior, based on self-report and experience-sampling methods. Findings are discussed in relation to the social functions of recalling the personal past (e.g., sharing memories to bond with others, to update each other, to teach, to give advice) and to the directive functions of future-oriented thought (e.g., planning, decision making, goal setting that are more likely to happen privately in the mind). In sum, the retrospective bias in conversational time travel seems to be a functional and universal phenomenon across persons and across real-life situations.

Project description:BackgroundConversational agents (CAs) for chronic disease management are receiving increasing attention in academia and the industry. However, long-term adherence to CAs is still a challenge and needs to be explored. Personalization of CAs has the potential to improve long-term adherence and, with it, user satisfaction, task efficiency, perceived benefits, and intended behavior change. Research on personalized CAs has already addressed different aspects, such as personalized recommendations and anthropomorphic cues. However, detailed information on interaction styles between patients and CAs in the role of medical health care professionals is scant. Such interaction styles play essential roles for patient satisfaction, treatment adherence, and outcome, as has been shown for physician-patient interactions. Currently, it is not clear (1) whether chronically ill patients prefer a CA with a paternalistic, informative, interpretive, or deliberative interaction style, and (2) which factors influence these preferences.ObjectiveWe aimed to investigate the preferences of chronically ill patients for CA-delivered interaction styles.MethodsWe conducted two studies. The first study included a paper-based approach and explored the preferences of chronic obstructive pulmonary disease (COPD) patients for paternalistic, informative, interpretive, and deliberative CA-delivered interaction styles. Based on these results, a second study assessed the effects of the paternalistic and deliberative interaction styles on the relationship quality between the CA and patients via hierarchical multiple linear regression analyses in an online experiment with COPD patients. Patients' sociodemographic and disease-specific characteristics served as moderator variables.ResultsStudy 1 with 117 COPD patients revealed a preference for the deliberative (50/117) and informative (34/117) interaction styles across demographic characteristics. All patients who preferred the paternalistic style over the other interaction styles had more severe COPD (three patients, Global Initiative for Chronic Obstructive Lung Disease class 3 or 4). In Study 2 with 123 newly recruited COPD patients, younger participants and participants with a less recent COPD diagnosis scored higher on interaction-related outcomes when interacting with a CA that delivered the deliberative interaction style (interaction between age and CA type: relationship quality: b=-0.77, 95% CI -1.37 to -0.18; intention to continue interaction: b=-0.49, 95% CI -0.97 to -0.01; working alliance attachment bond: b=-0.65, 95% CI -1.26 to -0.04; working alliance goal agreement: b=-0.59, 95% CI -1.18 to -0.01; interaction between recency of COPD diagnosis and CA type: working alliance goal agreement: b=0.57, 95% CI 0.01 to 1.13).ConclusionsOur results indicate that age and a patient's personal disease experience inform which CA interaction style the patient should be paired with to achieve increased interaction-related outcomes with the CA. These results allow the design of personalized health care CAs with the goal to increase long-term adherence to health-promoting behavior.

Project description:BackgroundA World Health Organization 2017 report stated that major depression affects almost 5% of the human population. Major depression is associated with impaired psychosocial functioning and reduced quality of life. Challenges such as shortage of mental health personnel, long waiting times, perceived stigma, and lower government spends pose barriers to the alleviation of mental health problems. Face-to-face psychotherapy alone provides only point-in-time support and cannot scale quickly enough to address this growing global public health challenge. Artificial intelligence (AI)-enabled, empathetic, and evidence-driven conversational mobile app technologies could play an active role in filling this gap by increasing adoption and enabling reach. Although such a technology can help manage these barriers, they should never replace time with a health care professional for more severe mental health problems. However, app technologies could act as a supplementary or intermediate support system. Mobile mental well-being apps need to uphold privacy and foster both short- and long-term positive outcomes.ObjectiveThis study aimed to present a preliminary real-world data evaluation of the effectiveness and engagement levels of an AI-enabled, empathetic, text-based conversational mobile mental well-being app, Wysa, on users with self-reported symptoms of depression.MethodsIn the study, a group of anonymous global users were observed who voluntarily installed the Wysa app, engaged in text-based messaging, and self-reported symptoms of depression using the Patient Health Questionnaire-9. On the basis of the extent of app usage on and between 2 consecutive screening time points, 2 distinct groups of users (high users and low users) emerged. The study used mixed-methods approach to evaluate the impact and engagement levels among these users. The quantitative analysis measured the app impact by comparing the average improvement in symptoms of depression between high and low users. The qualitative analysis measured the app engagement and experience by analyzing in-app user feedback and evaluated the performance of a machine learning classifier to detect user objections during conversations.ResultsThe average mood improvement (ie, difference in pre- and post-self-reported depression scores) between the groups (ie, high vs low users; n=108 and n=21, respectively) revealed that the high users group had significantly higher average improvement (mean 5.84 [SD 6.66]) compared with the low users group (mean 3.52 [SD 6.15]); Mann-Whitney P=.03 and with a moderate effect size of 0.63. Moreover, 67.7% of user-provided feedback responses found the app experience helpful and encouraging.ConclusionsThe real-world data evaluation findings on the effectiveness and engagement levels of Wysa app on users with self-reported symptoms of depression show promise. However, further work is required to validate these initial findings in much larger samples and across longer periods.

Project description:Data regarding the prevalence of hypertriglyceridemia in the Russian population are lacking, despite triglyceride (TG)-mediated pathways being causal in cardiovascular disease. The prevalence of mixed dyslipidemia and severe hypertriglyceridemia in the Russian population (PROMETHEUS) was undertaken to address this gap.This was an observational, cross-sectional retrospective study. Data from adults with a full/partial lipoprotein record who had blood analyses done at an INVITRO laboratory in Russia between January 1, 2011 and December 31, 2013 were analyzed. The primary endpoint was the prevalence of hypertriglyceridemia (TG ? 1.7 mmol/L); secondary endpoints included prevalence of borderline high, high, and very high TG and severe hypertriglyceridemia, defined as a TG level of 1.7 to <2.3, 2.3 to <5.6, ?5.6, and ?10.0 mmol/L, respectively. Statistical analyses involved the Wilcoxon and the Chi square tests. Correlations between log-transformed TG and low- and high-density lipoprotein cholesterol (LDL-C and HDL-C) and total cholesterol (TC) were assessed. The correlation between glycated hemoglobin (HbA1c) and TG levels in a nested sample of subjects with HbA1c and TG data was also assessed using a log-linear model.The full dataset and nested sample comprised 357,072 and 54,602 individuals, respectively. Prevalence of hypertriglyceridemia, borderline high TG, high TG, very high TG, and severe hypertriglyceridemia in the full dataset was 29.2, 16.2, 12.9, 0.11, and 0.011%, respectively; corresponding rates in the nested sample were 19.0, 17.2, 0.25, and 0.016%, respectively. TG levels were 16.4% higher in males versus females; males had a greater risk of hypertriglyceridemia (risk ratio 1.25; 95% CI 1.24, 1.26; P < 0.0001). Prevalence of hypertriglyceridemia increased with age, peaking at 40-49 years in males (42.8%) and 60-69 years in females (34.4%); a 0.61% increase in TG levels for each year of life was predicted. Hypertriglyceridemia prevalence increased over time. Correlations between TG and LDL-C, HDL-C, TC, and HbA1c (nested sample only) were observed.Almost one-third of Russians have hypertriglyceridemia, but severe disease (TG ? 10.0 mmol/L) is rare. Although the risk of hypertriglyceridemia was greater in males versus females, its prevalence increased with age, regardless of sex. TG was associated with HbA1c, LDL-C, HDL-C, and TC.

Project description:BackgroundHIV testing rates in sub-Saharan Africa remain below the targeted threshold, and primary care facilities struggle to provide adequate services. Innovative approaches that leverage digital technologies could improve HIV testing and access to treatment.ObjectiveThis study aimed to examine the feasibility and acceptability of Nolwazi_bot. It is an isiZulu-speaking conversational agent designed to support HIV self-testing (HIVST) in KwaZulu-Natal, South Africa.MethodsNolwazi_bot was designed with 4 different personalities that users could choose when selecting a counselor for their HIVST session. We recruited a convenience sample of 120 consenting adults and invited them to undertake an HIV self-test facilitated by the Nolwazi_bot. After testing, participants completed an interviewer-led posttest structured survey to assess their experience with the chatbot-supported HIVST.ResultsParticipants (N=120) ranged in age from 18 to 47 years, with half of them being men (61/120, 50.8%). Of the 120 participants, 111 (92.5%) had tested with a human counselor more than once. Of the 120 participants, 45 (37.5%) chose to be counseled by the female Nolwazi_bot personality aged between 18 and 25 years. Approximately one-fifth (21/120, 17.5%) of the participants who underwent an HIV self-test guided by the chatbot tested positive. Most participants (95/120, 79.2%) indicated that their HIV testing experience with a chatbot was much better than that with a human counselor. Many participants (93/120, 77.5%) reported that they felt as if they were talking to a real person, stating that the response tone and word choice of Nolwazi_bot reminded them of how they speak in daily conversations.ConclusionsThe study provides insights into the potential of digital technology interventions to support HIVST in low-income and middle-income countries. Although we wait to see the full benefits of mobile health, technological interventions including conversational agents or chatbots provide us with an excellent opportunity to improve HIVST by addressing the barriers associated with clinic-based HIV testing.

Project description:BackgroundThe rising incidence of chronic diseases is a growing concern, especially in Singapore, which is one of the high-income countries with the highest prevalence of diabetes. Interventions that promote healthy lifestyle behavior changes have been proven to be effective in reducing the progression of prediabetes to diabetes, but their in-person delivery may not be feasible on a large scale. Novel technologies such as conversational agents are a potential alternative for delivering behavioral interventions that promote healthy lifestyle behavior changes to the public.ObjectiveThe aim of this study is to assess the feasibility and acceptability of using a conversational agent promoting healthy lifestyle behavior changes in the general population in Singapore.MethodsWe performed a web-based, single-arm feasibility study. The participants were recruited through Facebook over 4 weeks. The Facebook Messenger conversational agent was used to deliver the intervention. The conversations focused on diet, exercise, sleep, and stress and aimed to promote healthy lifestyle behavior changes and improve the participants' knowledge of diabetes. Messages were sent to the participants four times a week (once for each of the 4 topics of focus) for 4 weeks. We assessed the feasibility of recruitment, defined as at least 75% (150/200) of our target sample of 200 participants in 4 weeks, as well as retention, defined as 33% (66/200) of the recruited sample completing the study. We also assessed the participants' satisfaction with, and usability of, the conversational agent. In addition, we performed baseline and follow-up assessments of quality of life, diabetes knowledge and risk perception, diet, exercise, sleep, and stress.ResultsWe recruited 37.5% (75/200) of the target sample size in 1 month. Of the 75 eligible participants, 60 (80%) provided digital informed consent and completed baseline assessments. Of these 60 participants, 56 (93%) followed the study through till completion. Retention was high at 93% (56/60), along with engagement, denoted by 50% (30/60) of the participants communicating with the conversational agent at each interaction. Acceptability, usability, and satisfaction were generally high. Preliminary efficacy of the intervention showed no definitive improvements in health-related behavior.ConclusionsThe delivery of a conversational agent for healthy lifestyle behavior change through Facebook Messenger was feasible and acceptable. We were unable to recruit our planned sample solely using the free options in Facebook. However, participant retention and conversational agent engagement rates were high. Our findings provide important insights to inform the design of a future randomized controlled trial.