Impact of an Interprofessional, Longitudinal, Undergraduate Student-Initiated Home Visit Program Towards Interprofessional Education


 Background

Interprofessional (IP) practice is an essential aspect of healthcare to meet the challenges of an aging population. Tri-Generational HomeCare (TriGen) is an undergraduate student-initiated, longitudinal, IP home visit program focused on older patients with frequent hospital readmissions.
Methods

Teams comprising healthcare undergraduates and secondary school (SS) students performed home visits for frequently admitted patients (three or more visits in past six months) on a fortnightly basis over a period of six months. They were supported and trained by healthcare professionals from Khoo Teck Puat Hospital (KTPH), a regional healthcare system, and North West Community Development Council (NWCDC), a social service organisation. A multimethod approach was used, with the administration of Readiness for Interprofessional Learning Scale (RIPLS) pre- and post-intervention, and gathering post-intervention quantitative and qualitative feedback.
Results

80.1% of 226 undergraduate participants from 2015 to 2018 enrolled in the study. There was a significant decrease in RIPLS total score from a mean of 80.3 pre-intervention to 79.0 post-intervention with a mean difference of 1.3 (-2.5 – -0.089, p = 0.035). Subscale analysis revealed a significant increase in the subscale “Roles and Responsibility” with a mean difference of 0.46 (0.14–0.77, p < 0.01). There was no significant change in the subscale “Teamwork” and a significant decrease in the subscale “Professional Identity” with a mean difference of -0.14 (-2.1– -0.77, p < 0.01). This was discordant to the participants’ feedback. 91.6% of respondents agreed they could “better appreciate the importance of IP collaboration (IPC) in the care of patients”. Qualitative analysis revealed takeaways including learning from and teaching one another, understanding one’s own role and the role of other healthcare professionals, appreciating teamwork and developing friendship with undergraduates from other professions.
Conclusion

We demonstrated the feasibility of an IP, longitudinal, undergraduate-initiated home visit program that may improve attitudes towards one’s role and responsibilities in IP care for future IPC.


Abstract
Background Interprofessional (IP) practice is an essential aspect of healthcare to meet the challenges of an aging population. Tri-Generational HomeCare (TriGen) is an undergraduate student-initiated, longitudinal, IP home visit program focused on older patients with frequent hospital readmissions.

Methods
Teams comprising healthcare undergraduates and secondary school (SS) students performed home visits for frequently admitted patients (three or more visits in past six months) on a fortnightly basis over a period of six months. They were supported and trained by healthcare professionals from Khoo Teck Puat Hospital (KTPH), a regional healthcare system, and North West Community Development Council (NWCDC), a social service organisation. A multimethod approach was used, with the administration of Readiness for Interprofessional Learning Scale (RIPLS) pre-and post-intervention, and gathering postintervention quantitative and qualitative feedback.
Results 80.1% of 226 undergraduate participants from 2015 to 2018 enrolled in the study. There was a signi cant decrease in RIPLS total score from a mean of 80.3 pre-intervention to 79.0 post-intervention with a mean difference of 1.3 (-2.5 --0.089, p = 0.035). Subscale analysis revealed a signi cant increase in the subscale "Roles and Responsibility" with a mean difference of 0.46 (0.14-0.77, p < 0.01). There was no signi cant change in the subscale "Teamwork" and a signi cant decrease in the subscale "Professional Identity" with a mean difference of -0.14 (-2.1--0.77, p < 0.01). This was discordant to the participants' feedback. 91.6% of respondents agreed they could "better appreciate the importance of IP collaboration (IPC) in the care of patients". Qualitative analysis revealed takeaways including learning from and teaching one another, understanding one's own role and the role of other healthcare professionals, appreciating teamwork and developing friendship with undergraduates from other professions.

Conclusion
We demonstrated the feasibility of an IP, longitudinal, undergraduate-initiated home visit program that may improve attitudes towards one's role and responsibilities in IP care for future IPC.

Background
There is increasing recognition of the need for interprofessional collaboration (IPC), echoed by the World Health Organisation (WHO) in a position paper in 2010.(1) Effective collaborative practices have been shown to yield better health outcomes for patients across different populations and care settings. (2)(3)(4)(5) Interprofessional education (IPE) aims to prepare healthcare professions for effective collaboration, and occurs when two or more professions learn about, from and with each other. (1) Research has shown that prior positive IP experiences allow for future positive IPC. (6) Moreover, it has been recommended for IPE to be introduced early in undergraduate healthcare courses as this may avoid stereotypes forming. (7,8) Singapore faces rapidly rising healthcare costs and an aging population. (9) A small group of patients who are frequent admitters (FA), de ned by Singapore's Ministry of Health (MOH) as having three or more inpatient admissions in a year, account for a disproportionate amount of the healthcare cost. The annual average cost is SGD 30,000 per FA patient compared to SGD 2,400 per non-FA patient with chronic disease. (10) Many of these patients are older, have more physical and psychiatric comorbidities, polypharmacy, and are of a lower socioeconomic strata. (11) To address their complex health and social needs, an interprofessional team approach is needed. In the United States (US), interprofessional home visit teams targeting communities with high utilisation of healthcare resources are called "hotspotters". (12,13) Equivalent programs in Singapore include the Aging-in-Place program (AIP) and Hospital-to-Home program (H2H). (14) "Hotspotting" experience may allow students to understand, learn and contribute to patient's care in the community. (15) Tri-Generational HomeCare (TriGen) was conceptualised in 2014 with the following aims (some similar to the aims described by the Interprofessional Student Hotspotting Learning Collaborative (16)): 1) Reduce ageism; 2) Provide a unique opportunity for students to experience complex needs of these patients in their homes/community; 3) Increase students' awareness of the role of social determinants of health and their ability as future clinicians to impact those determinants; 4) Enhance student's ability to collaborate in interprofessional teams; 5) Develop leadership and mentoring skills.
This study aims to evaluate the effectiveness of TriGen, an IP, longitudinal, undergraduate-initiated, homebased service-learning program focused on older patients with frequent hospital readmissions in improving attitudes towards IPE amongst healthcare undergraduates.

Program Design
TriGen leverages on IP healthcare teams to tackle the challenges of an aging population. Based on the service-learning model of dual objectives of service to the community and learning by the participants (17,18), TriGen aims to provide holistic care to frequently admitted older persons and a structured learning experience for both the healthcare undergraduates and SS students.
TriGen is a collaboration between the National University of Singapore Yong Loo Lin School of Medicine (NUS YLLSoM), Khoo Teck Puat Hospital (KTPH), a regional hospital situated in the Northern part of Singapore, and North West Community Development Council (NWCDC), a grassroots organisation based in the northwest district of Singapore. KTPH developed AIP to reduce readmission rates amongst those Table 1 Sample scenarios for training interprofessional teams

S/No
Learning Objectives. At the end of the exercise, the team will be expected to be able to: Scenario Description 1.
• Identify the normal ranges for vital signs.
• Learn to check the vital signs of a patient using blood pressure machine, glucometer, thermometer.
• Interpret the possible causes for abnormal vital signs.
Mdm Lee is a 53-year old Chinese lady. She has a past medical history of Diabetes Mellitus Type 2 (on insulin therapy twice a day) and recurrent falls. She lives with her husband and domestic helper in a three-room at. She has one daughter who is married and lives apart.
During the simulated home visit, the team is expected to pick up that she has a fever and high capillary glucose reading. On further history taking, she will reveal that she has missed her insulin injections for the last 3 days due to poor appetite. She is likely to be suffering from acute diabetic crisis from both omission of insulin and infection.

2.
• Manage an emergency or accident that may occur during a home visit by assessing the airway, breathing, circulation.
• Communicate with the nurse in-charge of the patient using the Situation-Background-Assessment-Recommendation (SBAR) format.
• Communicate with the patient's next-of-kin regarding the situation. She lives alone in a 1-bedroom rental at.
When you enter the home, you notice that the house is cluttered and dimly lit. The toilet does not have a grab bar and requires the patient to squat. In addition, there is a step the patient must cross to move from the kitchen to the toilet. She has a hemiplegic gait and uses a umbrella to walk around the house.
When you speak to her, you nd out that she has fallen four times in the last month within the house. Midway through the conversation when you mention about her husband, she starts crying.

S/No
Learning Objectives. At the end of the exercise, the team will be expected to be able to: When you speak to him, he revealed the following: • He was made to join TriGen by his teacher. Nevertheless, he was willing to join the programme as his best friend is in it too.
• He has been experiencing an increase in workload and stress from school (Examinations are approaching and he has been scolded by his teachers and parents for not faring well in recent class tests) • He nds it challenging to interact with the older person as his grandparents passed away when he was very young.
This IPE program was designed based on educational principles for adult learners outlined by Knowles et al.(19) First, the program provided the healthcare undergraduates with opportunities for experiential learning anchored in the service-learning approach. Second, the program was largely problem-based group learning with most of the training sessions being team-based and scenario-based. MDMs were also problem-based; these sessions encouraged the undergraduates to grapple with issues of their patients and brainstorm ideas to address them. Third, the service they provided in this program modeled the work they may engage in after graduation. What they learned in this program was of immediate relevance to their current study and future practice. Lastly, the program provided autonomy to healthcare undergraduates to direct their own learning. This program is voluntary, and most participated because they were drawn by the elements of IPE and IPC. In addition, the program allowed exibility for further self-study of topics of interest.
TriGen was built upon the contact hypothesis, which proposes that contact between different groups can change attitudes for the better. (20) Several factors are needed to ensure the success of these contacts: equal status of participants; positive expectations; a co-operative atmosphere; successful joint work; a concern for and understanding of differences as well as similarities between different professions; the experience of working together as equals. (21) Evaluation Approach This study used the Kirkpatrick's framework expanded by Barr

Quantitative Measures
The RIPLS was administered pre-and post-intervention. The RIPLS was formulated as a 19-item questionnaire comprising 3 subscales (Teamwork and Collaboration; Professional Identity; and Roles and Responsibilities), wherein some items are negatively coded.(23) Higher RIPLS scores imply greater readiness for IP learning. The RIPLS was chosen as it has been shown to be a valid and useful tool for measuring attitudes towards IPE especially in the context of healthcare students, (24,25) and has been validated for use in the local context. (26) Statistical Analysis The Shapiro-Wilk test was used to assess if the data followed a normal distribution. (27) A paired t-test comparing baseline and post-intervention responses was computed for each survey item to determine signi cant attitude differences (p ≤ 0.05). One-way analysis of variance (ANOVA) was performed to assess for demographic factors that correlated with pre-intervention and magnitude of change in RIPLS scores. If one-way ANOVA demonstrated an overall difference between groups, post-hoc Tukey's honestly signi cant difference (Tukey's HSD) test was performed. Pearson correlation was used to identify correlations in baseline and post-intervention scores. For all statistical analyses, the Statistical Package for Social Sciences (SPSS, Version 23.0, Chicago, Illinois) was used.

Internal Consistency
Cronbach's alpha was used to assess internal consistency of RIPLS, α = 0.848 for RIPLS total score.

Qualitative Measures
Post-intervention qualitative feedback regarding their learning experiences was collected through online surveys. Questions include: What did you learn about interprofessional collaboration? What are your learning points after completing the project? Would you recommend this project to your peers, what are your reasons? Thematic analysis was performed on the students' qualitative descriptions of their learning experiences.

Thematic Analysis
A grounded theory approach was used.(28) Constant comparison analysis was used to identify patterns in participants' responses and develop a coding schema. Two coders (KYYN and GYCL) independently identi ed major themes from text within all transcripts, with reference to the research questions. Both coders discussed and resolved any disagreements. A common coding schema was generated and applied to all the transcripts.

Ethics
Ethical approval was obtained from the NUS institutional review board (B-15-272) to evaluate the learning outcomes of the program. Study participation was entirely voluntary and anonymous. Informed consent was taken from participants before the commencement of data collection. There were no incentives provided to study participants.
Results 226 healthcare undergraduates participated in TriGen from 2015 to 2018, with a response rate of 80.1%. school students also participated in TriGen.  . Therapy students had a signi cantly higher baseline score compared to pharmacy students (p = 0.043). There was no difference in "teamwork" subscale score between genders, previous exposure to IPE projects and current or past participation in activities outside of the faculty. (Table 5)  For the "roles and responsibility" subscale, there was a signi cant difference between undergraduates of different faulty (p < 0.001). Social work undergraduates had a lower score 8.1 (7.3-8.8) compared to pharmacy undergraduates 9.6 (9.2-10.0) medical undergraduates 10.0 (9.5-10.5), nursing undergraduates 10.4 (9.7-11.2). Therapists had the highest score of 12.2 (9.9-14.5). (Table 7) (Table 4) For the "teamwork" subscale, there was no signi cant difference between the pre-and post-RIPLS score. (Table 5) For the subscale "professional identity", there was a signi cant decrease in the post-intervention score (mean difference − 1.4, -2.1 --0.77, p < 0.001). This was seen in the following subgroups: both genders, early years, both previous involvement and non-involvement in IPE projects, and current or past participation in activities outside of the faculty. (Table 6) For the subscale "roles and responsibility", there was a signi cant increase in the post-intervention score (mean difference 0.46, 0.14-0.77, p = 0.005). This was seen in the following subgroups: female, later years, social work, no previous involvement in IPE projects and current or past participation in activities outside of the faculty. (Table 7) Pearson correlation performed on baseline versus post-intervention scores showed a signi cant positive We also analysed the individual items of the RIPLS. (Table 9) Negatively coded statements like "the function of nurses and therapists is mainly to provide support for doctors" (Item 17) and "I am not sure what my professional role will be" (Item 18) showed signi cant increases in scores post-intervention (0.23, p = 0.005 and 0.17, p = 0.016 respectively). Other signi cant ndings include a decrease in scores for the statements "shared learning with other health and social care professionals will help me to communicate better with patients and other healthcare professionals" (Item 13) (-0.14, p = 0.013), and "shared learning will help to clarify the nature of patient problems (Item 15) (-0.10, p = 0.034).     In the subscale "teamwork", we found that undergraduates in their later years had a lower baseline score as compared to undergraduates in their early years. We postulate that this could be due to undergraduates with more clinical experiences appreciating the challenges of IPE in actual clinical practice. In the subscale "roles and responsibility", we found that social work undergraduates had the lowest baseline score. This is likely because these undergraduates have minimal exposure to social work in the healthcare setting unless they choose such elective modules in their senior years of study.
For the total RIPLS and subscale scores, there is a consistently signi cant positive correlation between baseline and post-intervention, yet a consistently signi cant negative correlation between baseline and change in scores. These ndings suggest that undergraduates who had a higher baseline scores may be more idealistic about IPE and have these expectations tempered after experiencing the di culties surrounding IPE. In addition, there could potentially be a ceiling effect wherein participants who are at baseline more ready for IPE stand to bene t less from the program in this area.

E cacy of the Program
There was a signi cant decrease in the total RIPLS score post-intervention. This was mainly driven by the decrease in the subscale "professional identity" (possible reasons are explored in the next section). There was a signi cant increase in the subscale "roles and responsibility", spearheaded by signi cant improvements in the scoring for two negatively coded statements "the function of nurses and therapists is mainly to provide support for doctors" and "I am not sure what my professional role will be". The subscale "teamwork" saw no signi cant changes.
This was discordant with our quantitative feedback and qualitative feedback which, in contrast, suggested that the undergraduates had improved attitudes towards IPE. In our quantitative self-reported feedback, majority of the undergraduates expressed that they better appreciated the importance of IPC for patient care and many felt that that the MDMs were useful for their learning.
In our qualitative analysis, undergraduates expressed ve major themes in their learning pertaining to IPE.
Firstly, they shared that they have learned from one another, and are equipped to teach individuals from other professions. This was a major focus on the program as undergraduates were encouraged to share their knowledge and skills with team members. Being able to freely learn from and teach one another requires mutual trust and respect which are key elements of collaborative practices. (32) In addition, the undergraduates reported better understanding of their own professional roles and the roles of other healthcare professionals. These two areas are recognised as crucial components of collaborative practice as de ned by the Canadian Interprofessional Health Collaborative framework. (32) In this aspect, there is agreement between RIPLS and the qualitative feedback provided as the subscale "roles and responsibility" had a signi cant increase. Undergraduates also shared that they learned about teamwork, speci cally, con ict resolution and compromise. Finally, undergraduates appreciated the opportunities to meet fellow undergraduates from different faculties. It has been observed in many successful IPE programs that informal social interactions are potentially as important as the actual IPE activities. (33) We observed that the relationships between the undergraduates built over through the program often persisted beyond the end of the program. These relationships can bene t the institution and healthcare system. (6) Discordance Between RIPLS and Quantitative/qualitative Feedback We found the decrease in RIPLS score post-intervention unexpected due to the positive feedback received in the quantitative and qualitative feedback. We postulate a few reasons. Firstly, the RIPLS has been described to have psychometrics issues. Most studies publishing data on exploratory or con rmatory analysis of the RIPLS show much variation in factor structure. As a result, multiple researchers have added or rearranged items and relabel subscales.(34) Secondly, RIPLS is insensitive to course improvements and to pre-post change as suggested by Schmitz et al (2015). (35) We chose the RIPLS at the start of 2014 as it had been widely used and validated and simple to administer. However, during this study, we discovered the potential issues as described above. Thirdly, having gone through a six-month period of fortnightly interaction, undergraduates may have had a better understanding of the challenges of IPE and realities of collaborating within IP teams. Their idealism may have been tampered with a dose of realism. Lestari et al. described how nursing and midwifery undergraduates have lower RIPLS scores as compared to medical and dentistry undergraduates as they had prior clinical experience and likely observed less than ideal interactions amongst members of healthcare teams.(29) Similarly, Makino et al.
found that alumni who have studied in an IPE program had a lower mean score on the Modi ed Attitudes Toward Health Care Teams Scale (ATHCTS) as compared to undergraduates. (36) The authors suggested that the negative attitude of alumni may be due to their professional experience in the context of realworld medicine. Several structural issues in clinical practice have been identi ed to contribute to this trend, e.g. competition between professionals(37) and power struggles. (38) This could be the reason why a short intervention such as that presented by Chua et al., where undergraduates attend a one-day conference together, could lead to an increase in RIPLS. (26) Undergraduates rarely have the opportunity to work on real-world problems together and experience the reality and challenges of IPE. Our data seems to support this hypothesis as undergraduates in their later years (with presumptively more real-life experience) had a lower baseline score as compared to undergraduates in their early years. In addition, there was a consistently signi cant negative correlation between baseline and change in scores in all subscales. Moreover, the qualitative data elicited several practical barriers to IPE, which could have partly accounted for the decrease in RIPLS scores.

Barriers to IPE
Undergraduates reported four main barriers: (1) time constraints, (2) unmotivated team members, (3) burden of administrative tasks, (4) lack of suitable patients. Lack of time as a barrier has been brought up by various authors. (39,40) As this program is voluntary, undergraduates are taking time off their already packed curriculum to participate. We are working with faculties to allow accumulation of academic credits from participation in the program. Participants also cited unmotivated team members as reason for their negative experience. This challenge is mitigated by the voluntary nature of participation. Further measures include more stringent selection of participants and faculty intervention for unmotivated team members. As participants were contributing to clinical care, they were required to document their visits. Multiple studies showed that physicians deemed documentation using electronic health records and administrative work burdensome and excessive time spent on these may be associated with physicians' burnout. (41,42) With feedback from participants regarding the burden of documentation, we are working to streamline and reduce the administrative requirements. Lastly, participants cited the lack of suitable patients to care for and to learn about IPC. This re ected the importance of choice and t of the patients for the program as well.

Strengths and Limitations
The strength of our study lies in the use of established evaluation framework (Kirkpatrick). In addition, the sample size is large. Also, this a multimethod study approach where both quantitative and qualitative methods contribute to an integrated inference for complex program evaluation. (22) The limitations of our study include it being single-institution. The participants are volunteers and a self-selected group. Hence, results may not be generalisable. In addition, there was no control arm for the intervention.