Implementation of a telehealth-enhanced home visiting programme for families of young children

Abstract

Introduction

Health systems are applying innovative solutions, such as telehealth and home visiting, to reduce unnecessary healthcare utilization, including emergency department (ED) visits. A large paediatric healthcare system implemented a telehealth-enhanced home visiting programme as an extension of primary care services. The purpose of this paper is three-fold: (1) Examine the process of implementation and the intermediate clinical outcomes; (2) Evaluate patient experiences and acceptability of the programme post-implementation; (3) Identify system, clinic and patient factors influencing implementation of the home visiting programme.

Methods

Implementation of the telehealth-enhanced home visiting programme occurred from July 2018 to March 2019. Longitudinal electronic health records (EHR) and surveys were triangulated with qualitative data to evaluate the preliminary effectiveness, feasibility, and acceptability of the programme.

Results

Of the 948 eligible families, clinical care staff referred 38% of families to the home visiting programme and 49 families (5%) completed the 12-week home visiting programme. Necessary ED utilization significantly increased post-implementation compared with pre-implementation. Families were overall highly satisfied with the programme and its content. Several factors influenced implementation including outer setting (i.e. patient needs and external policy), inner setting (e.g. poor leadership engagement, fully integrated network, and high tension for change), and individual characteristics (e.g. high self-efficacy).

Conclusions

Once families were enrolled, the programme was fairly successful in addressing patient outcomes. The programme and visit process was highly regarded by families and the unlicensed healthcare professionals. Future programme recommendations, such as small programmatic changes and major improvements in the clinic, should be implemented before widespread dissemination.

Keywords

Paediatrics home visits telehealth emergency department utilization preventative care

Introduction

Health systems are applying innovative solutions to reduce primary care-related emergency department (ED) visits¹,² and, thus, healthcare costs.³ One such innovative approach has been telehealth – a method of improving accessibility to health care and education through telecommunication and technology. A systematic review showed that telehealth is not only utilized by medical professionals to treat and communicate with patients but also by the patients themselves.⁴ Telehealth systems have been implemented across an increasing range of healthcare domains such as chronic disease management, diabetes management and heart failure hospitalization^5–7 to reduce costs;⁸ however, there are limited studies focused on how telehealth and home visits can reduce paediatric primary care-related ED visits, and among high-needs paediatric populations in particular.

Health systems are also implementing home visiting programmes as part of the patient-centred medical home concept. The overarching goal of home visiting programmes is to empower families with the tools, education and services that they might not otherwise access in traditional medical settings. Results of 15 randomized control trials have shown improvements to outcomes of young children as a result of home visiting programmes,⁹ including improvement in physical growth,¹⁰,¹¹ adherence to timely immunizations¹² and reduction in hospitalizations,^13–15 developmental delays¹²,^16–19 and child behaviour.¹⁷,¹⁸,²⁰,²¹ In 2017, an extensive evaluation was conducted among 46 home visiting programmes in the United States. Among home visiting programmes that targeted healthy paediatric families, programmes show improvement in adherence to preventative visits.²² However, to our knowledge, little has been explored evaluating the impact of home visiting programmes on ED utilization or outcomes among healthy paediatric populations.²³

In the United States, many families seek services from EDs because of lack of access to primary care clinics in a timely fashion. For example, appointment times may be delayed or interfere with parent work schedules. Being able to provide timely access to primary care via telehealth is one way to re-direct these lower acuity visits to the right level of care and for a lower cost. However, limited literature exists on health system-, clinical care team-, and patient-level implementation barriers and facilitators of combining these two evidence-based strategies (i.e. telehealth and home visiting programme) and the impact on patient outcomes such as healthcare utilization patterns. To fill this gap, a large paediatric healthcare system developed and tested the implementation of a telehealth-enhanced home visiting programme as an extension of primary care services in the Dallas-Fort Worth area. The purpose of this paper is three-fold:

Examine the process of implementation and the intermediate clinical outcomes;

Evaluate patient experiences and acceptability of the programme post-implementation;

Identify system, clinic and patient factors influencing implementation of the home visiting programme.

Methods

Intervention

A large paediatric healthcare system that includes a paediatric hospital and ED, subspecialty outpatient care and primary healthcare medical homes implemented a telehealth-enhanced home visiting programme as an extension of primary care in Dallas, Texas from July 2018 to March 2019. Separate programme content was developed to target three specific patient populations: (1) Newborns and their mothers; (2) ED users for low-acuity or primary care-related illness; and (3) children with an increased risk of becoming overweight or obese, identified by a weight-for-length, or body mass index ≥ 50 percentile adjusted for age and sex.

All programmes were designed to: (a) reduce unnecessary healthcare visits, including hospitalizations, readmissions and ED visits, and (b) provide patient education on self-management of common primary care-related paediatric illnesses and illness prevention. A primary care-related illness is a visit of low acuity that can be managed in an office setting or via telehealth consultation. When a child with a primary care-related illness presents to the ED, the cost of care for the same medical complaint increases significantly and adds pressure on tertiary services. In the programmes, experienced unlicensed healthcare professionals (UHPs), including a community health worker (CHW) and a Medical Assistant (MA), received a rigorous training curriculum on home visiting modules. UHPs were hired and trained specifically for this home visiting programme and were not part of the staff at the primary care clinic. Training manual contents were drawn from validated, credible sources of tools and curriculums including American Academy of Pediatrics, HealthyChildren.org, and What to Do When Your Child is Sick.

Families were recruited to the programme during well-child visits. A well-child visit is a routine paediatric visit with a patient’s regular healthcare provider, such as a paediatrician or paediatric nurse practitioner, to ensure proper growth and development and to prevent health problems. If a family was eligible, an alert automatically appeared in the electronic health records (EHR), which notified MAs of patient eligibility. MAs briefly mentioned the programme to the families, and if the family were interested, a referral was sent to the patient’s regular paediatrician or paediatric nurse practitioner for their signature. The nurse contacted the caregiver by telephone and conducted preliminary screening for patient eligibility, and if eligible the family provided verbal consent, scheduled an enrolment visit. Once a family enrolled, an UHP visited the family’s home every week in a step-down frequency (e.g. three home visits during week 1; one home visit per week during weeks 2–12). During each home visit, the UHP would first assess vital signs of the children, including temperature, heart rate and respiratory rate. The UHP would measure weight at some pre-determined interval based on which programme the children were enrolled in. Then, the UHP would connect, via telehealth videoconferencing, to the nurse, who would supervise the visit, make real-time assessments, and answer medical questions. The nurse was physically located in an office building separate from the primary care clinic and hospital. If families reported child illness, the nurse would triage (assess remotely and determine whether child needed additional medical attentions). Visits were conducted Monday through Friday 8 am to 5 pm. Between 10 and 13 home visits were scheduled over the course of a 12-week period for each unique patient family. The average telehealth visit was 10 minutes long, and the average length of the total visit was 45 minutes. In addition, a follow-up visit was conducted at 6-months post-enrolment with each family.

Evaluation design

We used a convergent, iterative, mixed-methods design to accomplish the main aims of this study.

Clinic sample

A primary care clinic, an office where a patient sees their regular physician for routine or sick care visits, was recruited to implement the home visiting programme. See Appendix A for definitions of key aspects of the American healthcare system. All clinic staff were invited to attend a 30-minute home visiting programme presentation that included an explanation of the curriculum and the process to refer a patient. Clinic leadership engagement was gained by speaking with each attending physician on a one-on-one basis by the physician champion. A ‘physician champion’, in this case, was a paediatrician health system executive leader who was committed to the home visiting programme and spearheaded the project from conceptualization, module development to garnering leadership buy-in throughout the health system and primary care clinic to help support and implement the project. In addition, clinic care staff were emailed information about the programme and referral process on multiple occasions. Informational flyers were created and posted throughout the clinic for continued staff engagement and referrals. Informational brochures were created to help provide an overview of the home visiting programme for families and to help decrease the amount of time the MA had to spend on explaining the programme. The clinic was offered no financial incentive.

Patient sample

The study population for this evaluation included families that visited the primary care clinic for well-child care during the implementation period (July 2018–March 2019). There were three populations targeted for the home visiting programme through an automatic alert in the EHR:

Programme #1: Newborns and their mothers. Eligibility criteria included a child aged <3 months;

Programme #2: ED users for low-acuity illness. Eligibility criteria included: (1) child between the age of 3 months and 59 months; and (2) has used the ED unnecessarily for a low-acuity condition (based on the Emergency Severity Index [ESI] level 5 = non-urgent or 4 = less urgent) at least once in the last 12 months prior to enrolment; and

Programme #3: Young children at risk for an unhealthy weight. Eligibility criteria included: (1) children ages 12 months to <49 months); and (2) a weight-for-length, or body mass index ≥50 percentile adjusted for age and sex.

For all three programmes, families were ineligible if patients: (1) had a rating of clinical severity extracted from enrolment, claims, or other encounter-level data diagnosis indicating more than two chronic illnesses; and (2) were currently enrolled in another home visiting programme at the hospital or a home-care programme.

Measures and data collection

Process outcomes were extracted from the EHR. Measures included: (1) number of eligible families with visits in the clinic during the recruitment period; (2) proportion of families, through automatic alerts, referred to programme, dismissed by clinical care staff, or declined; (3) proportion of families contacted by programme nurse supervisor; (4) proportion of families enrolled and had at least one home visit; (5) proportion of decliners; (6) reasons for not participating; (7) proportion of patients that completed the programme at 12 weeks and 6 months; and (8) reasons for loss to follow-up.

Primary clinical outcome measures for programme #1 and #2 were proportion of unique families that utilized the ED and mean change in number of ED encounters per child. These measures were extracted from the EHR for eligible patients 6 months pre- and post-enrolment. Further, number of triages made by the nurse supervisor was extracted post-implementation. For programme #3, the main outcome was change in score of Family Nutrition and Physical Activity (FNPA). The FNPA is a 20-item, validated construct that elicits parent report of home environment and practices that may predispose to obesity development (e.g. How often does your child eat breakfast, either at home or at school?). The four response options are: (1) never/almost never; (2) sometimes; (3) often; and (4) very often/always. Higher scores indicated healthier practices. Families self-reported FNPA at the first home visit (i.e. baseline) and 12 weeks post-first home visit. The questionnaire was disseminated via iPad.

Family experience and acceptability of programme was measured using 5-item questions self-reported by families at the end of the programme. Overall satisfaction with programme (i.e. ‘Overall, how would you rate the home visiting programme’, range 1–5), satisfaction with programme content (i.e. ‘I was satisfied with the information provided’, range 1–5), interest in continuing programme (i.e. ‘I would be interested in continuing a programme like this one’, range 1–5), programme met healthcare needs (i.e. ‘The home visiting programme has been helpful to me’, range 1–5) and programme an extension of primary health care (i.e. ‘To what extent do you think the home visiting programme fit as an extension of primary health care at this clinic?’, range (1–7). All responses were rated on Likert scales.

Factors influencing implementation

Post-implementation, semi-structured interviews were conducted by author KKJ with two UHPs and one supervising nurse representing different roles to ascertain level of implementation and changes in clinical practice. Interviews took place in a private conference room within the main hospital. Interviews followed a semi-structured guide that assessed overall satisfaction with programme, integrated care workflows and tasks within the primary care clinic, experience with home visits and culture of clinical organization.²⁴ Each interview lasted 30–45 minutes. Interviews were recorded and professionally transcribed. Throughout the implementation period, clinical care staff at the practice (i.e. MAs and the practice manager) were informally interviewed and observed by the UHPs and the nurse supervisor. The nurse supervisor observed clinic periodically throughout implementation for approximately 10 hours.

Data analysis

Descriptive statistics (tabulations, percentages, means, and standard deviations) were used to describe process outcomes and sociodemographics of families enrolled in the programme at baseline. Paired t-test models were used to evaluate the relationship between pre- and post-implementation in regards to ED utilization and FNPA scores (dependent variables). Patient surveys were descriptively analysed by examining response frequency distributions. When specific survey items were not answered, respondents’ items were excluded from analyses.

Qualitative findings were analysed by grouping emerging findings into categories of themes using an immersion-crystallization approach.²⁵ Then, we connected our findings to the existing literature.²⁶ We conducted a second in-depth comparative analysis using the Consolidated Framework for Implementation Research (CFIR)²⁷ constructs to understand factors that influenced implementation. After considering all 39 CFIR constructs, 10 relevant CFIR constructs that mapped to three domains (i.e. outer setting, inner setting and individual characteristics) were identified and an in-depth analysis was conducted to identify how these factors influenced implementation.

Qualitative and quantitative analyses were first conducted independently to ensure unbiased interpretation. After completing analyses, qualitative and quantitative findings were integrated at the practice level using data-triangulation techniques.²⁵ The Institutional Review Board at the University of Texas Health Science Center at Houston approved this study protocol (IRB# HSC-SPH-17-0790). A Waiver of Informed Consent was granted for de-identified EHR extraction and analyses.

Results

Process outcomes

Figure 1 depicts process outcomes for the pooled analytical population. During the implementation period, a total of 948 patients were eligible for the home visiting programme (i.e. alerts triggered to clinical care teams). Of these eligible families, 34% (n=322) were referred by the clinical care team to the home visiting programme, 50% of eligible family alerts were dismissed, and 16% of families declined to participate while in clinic. Of the referrals from the clinical care team, 74% were contacted by the supervisor nurse and, ultimately, 63 families (7% of all families offered the programme) agreed to participate. Forty-nine families completed the 12-week programme and 29 families completed the 6-month follow-up.

Figure 1.

Process outcomes for pooled sample.

Table 1 displays process outcomes by programme. The clinical care team was more likely to disregard alerts for eligible patients in programme #2 (i.e. ED users) or programme #3 (i.e. obesity; χ² = 57.68; p-value < 0.001). Programme #1 (i.e. newborn moms) had the highest rate of Enrollment of eligible families (12%; χ² = 26.90; p-value < 0.001). Families in programme #2 were least likely to finish the entire 12-week programme (χ² = 7.33; p-value = 0.02).

Table 1.

Process outcomes across programmes among eligible patients (n = 948).

	Programme #1N = 350 N (%)	Programme #2N = 390 N (%)	Programme #3N = 208 N (%)	X² or Fisher’s Exact
Screened for eligibility				57.68***
Referred to programme	157 (45)	118 (30)	47 (23)
Family declined	73 (21)	50 (13)	30 (14)
Clinical care disregarded alert	120 (34)	222 (57)	131 (63)
Enrolled into programme	43 (12)	11 (3)	11 (5)	26.90***
Completed 12-week programme	29 (8)	10 (3)	10 (5)	12.39**
Completed 6-month follow-up	14 (4)	4 (1)	8 (4)	7.33*

*p-value<0.05; **p-value<0.01; ***p-value<0.001

Participant sample

Enrolled caregiver sociodemographics are presented in Table 2. Briefly, the average age of caregivers was 30 years old (SD = 5.2) and a majority were Hispanic (75%). The majority of caregivers were married (49%), employed (41%) or a homemaker (32%), and were a mother to the eligible child (97%). The majority of eligible children were Hispanic (71%) and 23% of families indicated Spanish as their primary language.

Table 2.

Sample description of enrolled families with at least one home visit (n = 63).

Caregiver	N (%)
Age; years, mean (SD)	29.5 (5.2)
CRG Status
Healthy	37 (55)
History of Acute Disease	19 (28)
Single Minor CD	5 (7)
Language
English	48 (77)
Spanish	14 (23)
Race/Ethnicity
Hispanic	47 (75)
White NH	1 (2)
Black NH	11 (18)
Other	4 (6)
Education
Less than high school graduate	10 (10)
High school graduate	21 (33)
Some college or technical school	23 (37)
College graduate	9 (14)
Annual Household Income
<$8,000	14 (22)
8,000–<25,000	24 (38)
25,000–<50,000	15 (24)
50,000+	10 (6)
Marital status
Married	31 (49)
Divorced/Widowed/Separated	3 (5)
Never married	17 (27)
A member of an unmarried couple	12 (19)
Employment
Employed	26 (41)
Out of work/Unable to Work	14 (22)
Homemaker	20 (32)
A student	3 (5)
Caregiver relationship to enrolled child
Mother	61 (97)
Father	2 (3)
Child
Race/Ethnicity
Hispanic	45 (71)
White NH	0 (0)
Black NH	10 (16)
Other	8 (13)

NH: Non-Hispanic; SD: Standard deviation; CRG: Clinical risk group; CD: Chronic disease.

Intermediate clinical outcomes

Table 3 presents intermediate clinical outcomes post-implementation compared with pre-implementation stratified by programme. Overall, the proportion of unique families that appropriately utilized the ED significantly increased after programme implementation (34% pre vs. 66% post), and, specifically, a significant increase was observed in programme #1 (28% pre vs. 59% post). The mean number of ED visits per child did not significantly change after implementation (1.58 vs. 1.86). Overall, the nurse supervisor completed 121 triage encounters during the implementation period, in which parent reported an acute issue during the UHP visit, which the nurse supervisor assessed remotely. FNPA scores increased post- (64.9) compared with pre-implementation (61.4) and approached significance (p-value = 0.06).

Table 3.

Clinical outcomes pre- and post-implementation and family experience across programmes post-implementation.

	Programme #1N = 29			Programme #2N = 10			Programme #3N = 10			Pooled Sample N = 49
	Pre-	Post-	Test statistic	Pre-	Post-	Test statistic	Pre-	Post-	Test statistic	Pre-	Post-	Test statistic
Intermediate Clinical Outcomes:
Unique children visiting ED (%)	0.28	0.59	3.39^***	0.70	0.60	0.64	0.10	0.30	1.38	0.34	0.66	4.72^***
Number of visits per child (mean, SE)	1.27 (0.14)	1.94 (0.19)	1.89	2.00 (0.58)	1.83 (0.40)	0.23	2.00 (0)	1.00 (0)	–	1.58 (0.23)	1.86 (0.16)
Number of triages during home visit (N)	–	79	–	–	17	–	–	25	–	–	121	–
FNPA (mean, SD)	–	–	–	–	–	–	61.4 (7.09)	64.9 (5.97)	2.14	–	–	–
Family Experiences: (mean, SD)
Satisfaction with programme											4.9 (0.4)
Satisfaction with programme content											4.9 (0.3)
Interest in continuing programme											4.6 (0.8)
Programme met healthcare needs											4.8 (0.6)
Programme extension of primary health care											6.7. (0.7)

FNPA: family nutrition and physical activity; SD: standard deviation; ED: emergency department.

*p-value<0.05; **p-value<0.01; ***p-value<0.001.

Family experiences

Table 3 also presents family experiences with the programme 12 weeks post-enrolment. Overall, families were highly satisfied with the programme and the programme content, were interested in continuing the programme, and thought the programme met their healthcare needs.

Factors influencing implementation

Factors, and subsequent evidence, influencing implementation, organized by CFIR domains and constructs, are presented in Table 4. Each team within the programme excelled in communication with each other, as evidenced by the highly integrated EHR system. The post-programme perception survey found that the majority of families, UHPs and the nurse supervisor all found utility in the programme. UHP and caregiver self-efficacy increased by the end of the programme, and families felt more health literate and empowered to self-manage common paediatric illnesses when the UHP was not present. There were several barriers to programme implementation such as the clinical care team were either sceptical of, or did not believe in the value of the programme, which led to low levels of overall buy-in.

Table 4.

Factors influencing implementation.

Construct	Description	Facilitator	Barrier
OUTER SETTING
Patient Needs & Resources	Patient needs are accurately known by organization.		-Home visiting programme targeting preventative care not ideal for familiesEvidence: “Having it be a preventative program, and it being aimed at healthy individuals, if it was shortened or aimed at first time mothers in particular, or mothers who had older children and now with newer children it would have been more helpful. (…) A lot of participants, because they had healthy children, didn’t have such a strong attachment of keeping appointments. They didn’t find in any harm in canceling, or not being home. No urgency in the program itself.”- Nurse supervisor
External Policy & Incentives	External strategies to spread interventions.	-Reduction in ED utilization from a health system perspectiveEvidence: -Patient-centred medical homes national policy
INNER SETTING
Networks & Communications	The nature and quality of formal and informal communications within an organization.	-Outstanding communication within organization-Integrated EHR systemEvidence: -Several teams worked together on this within the health system, Data intelligence and IT, CHPG, Project Management, Virtual Health, Care management, Behavioural health, Injury prevention, Nurse Advice line, Marketing, Security , Home Health, Accreditation/Regulatory affairs, Legal, Privacy. -Because of the integrated paediatric system, health IT was able to add workflow sheets and referrals within the EHR	Restricted capacity to orient and motivate practice team
Tension for Change	Stakeholders perceive the current situation as intolerable or needing change.		-Clinical care team didn’t perceive current system needing change-Stress on remaining clinic because of system closureEvidence: -MAs (4 out of 10) didn’t believe in programme at all. The rest were sceptical. They thought it was a burden for families and families would not allow people to come in their homes; very invasive. Didn’t think things needed to be changed. (Observation) -“There was a lot of frustration that we couldn’t become a cohesive team. I’m talking about with the clinical care team and the management piece.” UHP 1
Relative Priority	Individuals’ shared perception of the importance of the implementation within the organization.		-Low priority for clinical care team-Clinical care team perceived programme as an unwanted disruption in their normal daily workflow process. Evidence: -MAs indiscriminately decline the BPAs, mainly to ensure that the BPAs do not fire again and interrupt their regular work flow in Epic. (Observation) -“MAs are hesitant to change their workflow because they will get in trouble if they aren’t rooming enough patients quick enough or missing something on an order because they had to talk about the program to patients. They didn’t think this was going to be a long term project so they didn’t care” –Nurse supervisor-“The MA’s saw it as extra work, if they got to it. And providers too. Families would ask about the program and providers would say I have no idea what this is.” – Nurse supervisor-“The program was not a natural extension of primary care like it was intended. It felt very disjointed. A lot of that comes from the provider. So the provider not explaining the program and not explaining why it’s such a great value and saying the nurse is going to call her this is her name and this is her number and this is the material. I think that’s because there was a push to see more patients and there was provider turnover.” –Nurse supervisor-“From the very beginning, there was a clash between the care team and having them talk to the parent. These ladies are going on routine instinct. It interrupted their service and their routine and the way they normally handle patients. So I just felt like we were in the way.” –UHP 1
Learning Climate	A climate in which: (a) leaders express their own fallibility and need for team members’ assistance and input; (b) team members feel that they are essential, valued, and knowledgeable partners in the change process; and (c) there is sufficient time and space for reflective thinking and evaluation.		-No reflective thinking for clinical care team-UHPs did not feel valued or heard Evidence: -UHPs don’t feel heard or truly valued as an important part of the home visiting program. Their input was never taken into account by leaders on best workflow changes for within the practice, which stemmed from a lack of insight and understanding into the challenges experienced both from a funding/FTE perspective (Observation) -“I know that 2–3 months into it, I was ready to tweak the program. And I said to myself, as soon as we have a big meeting, I wanted this to be added to this, and this to be changed. But we never got that chance.” UHP 1-“With the management piece I felt that we weren’t being listened to.” –UHP 1
Leadership Engagement	Commitment, involvement, and accountability of leaders and managers with the implementation.		-Low commitment from clinical leadership-No accountability of clinical managers with implementationEvidence: -“The practice manager said to me ‘Your program sounds awesome but we have done tons of these pilot programs before. Your program isn’t going to stick around.’ Right off the bat. When you have that from leadership, it trickles down no doubt. He was well-respected and very vocal and I have no doubt this trickled down from the very beginning.” –Nurse Supervisor
CHARACTIERISTICS OF INDIVIDUALS
Knowledge & Beliefs about the Intervention	Individuals’ attitudes toward and value placed on the intervention.	-Majority of families found utility in programme-UHP and Nurse supervisor found utility in programmeEvidence: -“I loved the program. From the very beginning I knew it was going to be beneficial not just for me as a CHW, but for the families. There’s something lost between the actual clinic visit with the doctor to the actual understanding of what a child needs to do. And home visiting gave us the opportunity to sit with the parent and really explain things in their own level of understanding. I have some moms crying when I told them this was the 12^th visit and we would be back in a couple of months for the 6 month follow-up visit. And others that were in shock, like really it’s over?” – UHP 1-“Families were very open to learn. Especially during the flu season. It was a lot of help that they didn’t have to go to the ED. Something that could be resolved without 3-4 hours waiting at the ED. They were very grateful.” UHP 2-“The families were amazed that someone could come to the house, take vitals, answer questions, virtual health it was great for first time moms. Everyone was amazed we could provide this service. It breaks a lot of taboos.” UHP 2-“One of the 3 week old triplets were coughing. Mom said she thinks they are having trouble breathing. I took vitals and Amy was able to triage right away. They were admitted to the hospital in one hour. Just with that one visit, the whole program was worth it. They stayed in the hospital for 3 weeks until they were stable. Who knew what would happen if we weren’t there.” UHP 2	-Some families did not find utility in programme-Hours of home visiting not ideal for familiesEvidence: “Some families said this is my 6th kid. I have had 5 before. I already have a routine. We still gave them the information and they respectfully stopped the program.” UHP 2-“A lot of people during the enrollment attempts, would have been interested in evening and weekend hours. And moms who went back to work for [program #1]. Between 5-8 at night and weekends. No stipulations on the weekends. Evening hours would have been good.”- Nurse supervisor-“Families want home visits on weekends. A lot of parents were asking if we could see them Sat mornings. The families I talked to were asking for after working hours. Because a lot of them need to work. Or even Saturday mornings. We lost a lot of people because they couldn’t do M-F 8-5pm”- UHP 1
Self-efficacy	Individual belief in their own capabilities to execute courses of action to achieve implementation goals.	-High parent self-efficacy at end of the programme-CHW believed in their own capabilities to achieve home visiting programme-Families knew how to leverage educational materials to help themselves when UHP wasn’t thereEvidence: - “A couple of parents got really into it. One parent bought their own MyPlate on Amazon. One started a rewards program with her daughter because she wasn’t eating with the family”-Nurse supervisor-“I really think that I have the knack to speak to families. Most families felt at ease with me. I was always asked to have coffee and lunch. And of course I didn’t want to cross that line.” UHP 1-“We would teach something, give them information on fevers and headaches and sore throats, and the following week we would reiterate. Families had an idea of what to do and where to find it in their parent handbook. They knew where to go for content.” – UHP 2

Discussion

The goal of this study was to evaluate multi-level implementation factors and the subsequent effectiveness of combining two evidence-based strategies (telehealth and enhanced home visiting programme) at an integrated, paediatric health system. Overall, there were a variety of system- and patient-level factors that significantly facilitated this paediatric telehealth home visiting programme. First, the health system was integrated – a hospital organization that provides both acute patient care and multispecialty care – which allowed for closed-loop communication between a variety of multidisciplinary teams, continuity of care between the clinic and home visiting programme, and tools necessary to implement and evaluate a complex programme. The programme was able to leverage the integrated electronic health system for data extraction, a pre-existing telehealth programme and programmatic use for developing this telehealth enhancement, as well as a strong delivery team to help facilitate the aims of this project. The programme was also carefully developed, using validated sources for each target population. Families expressed overall satisfaction with the programme: the content was well received and families recognized the utility of the important programme. Moreover, the UHPs also greatly benefited from the programme. They reported that the training expanded their capacity as healthcare professionals and thought this position was valuable in their own career development and trajectory.

We hypothesized the home visiting programme would significantly decrease ED utilization by equipping families with the needed tools to empower themselves to know when and when not to go to the hospital. The use of telemedicine has shown to decrease the need of hospital visits, and particularly emergency visits, in past studies.²⁸ However, in this study, ED utilization significantly increased, particularly among programme #1 (i.e. mothers of newborns). This is likely due to the unexpected number of triages the nurse supervisor made during the home visiting tele-videoconferences. Overall, the nurse supervisor made 121 triages, with the majority made for programme #1 (79 triages for 29 families). This finding suggests that having direct contact with a nurse increased necessary ED utilization, instead of decreasing unnecessary ED utilization rates. In other words, the telehealth visit prompted the nurse to triage many different symptoms. If she had not been available, the parents perhaps would have observed the patient at home for a longer period of time or attempted to seek care in the primary care clinic. This study found both benefits and drawbacks to telemedicine for programme #1. Future programmes should keep in mind this caveat when implementing telemedicine programmes for new mothers.

There were several clinic- and patient-level barriers that interfered with the home visiting programme, and participant recruitment in particular. First, the practice site lacked stakeholder engagement and leadership buy-in, which led to lower than expected referrals and enrolment. In the future, leaders who intend to implement this type of programme need to engage clinic-level stakeholders (i.e. physicians/providers, practice administrators, programme managers, physicians, nurses, MAs, and other care team members) from project initiation. Top-down engagement, when implementing a home visiting programme in primary care, is not effective and reflects very low recruitment numbers in this study. Past literature has consistently shown that lack of stakeholder buy-in impedes on implementation of public health programmes.²⁹ Finally, at the patient level, many families declined to participate or dropped out of the programme over time because the programme failed to match logistical needs of families. For example, this programme only offered home visiting during the week from 8 am to 5 pm. A large proportion of families that dropped out were mothers returning to work after completing maternity leave and the home visiting times no longer worked with their schedule. Families suggested future programmes should have nights and weekend hours, particularly Saturdays, to complete the home visiting programme successfully. Past home visiting programmes have found the greatest success once they addressed logistic and cultural challenges, such as scheduling visits.³⁰

Limitations

Results should be considered in light of several considerations. First, the programme was implemented in one primary care clinic, which limits generalizability. However, our programme seemed to target a Hispanic population, which may generate hypotheses for other primary care network clinics across the nation. Second, there was a small sample size of patients given the unforeseen system-level change prior to implementation. Although this did reduce power for statistical analyses, significant changes were still detected. Future research should aim to implement child health programmes across larger populations. Third, we had limited information on ED visits and reasons for declining to participate. Future research should extract more nuanced EHR data, such as whether ED visits led to admissions or the frequency of ED visits per family.

Conclusion

In conclusion, the telehealth-enhanced home visiting programme increased necessary ED utilization among certain programmes. The programme and visit process was highly regarded by families and the UHPs. Small programmatic changes should be made in the future, such as changing hours of home visiting to meet family needs. However, major improvements need to be made in the clinic process. For example, families could be recruited in the waiting area, referrals could be made in person between the MA and home visiting programme staff (opposed to follow-up phone calls), and physicians/providers could help to pitch the programme and recommend that families enrol. Future research should also include a consumer input strategy to gather more information on participation and drop-out rates, and particularly, the impact of this programme on cost-savings for health systems.

Footnotes

Acknowledgments

We would like to sincerely thank Dr. Michiko Clutter for her assistance in the design and collection of data.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Katelyn K Jetelina

References

Garcia

Bernstein

Bush

MA.

Emergency department visitors and visits: Who used the emergency room in 2007?

National Center for Health Statistics 2010.

Hoot

Aronsky

Systematic review of emergency department crowding: Causes, effects, and solutions. Ann Emerg Med 2008; 52: 126–136.

Weinick

Burns

Mehrotra

Many emergency department visits could be managed at urgent care centers and retail clinics. Health Aff 2010; 29: 1630–1636.

Koch

Home telehealth—Current state and future trends. Int J Med Inform 2006; 75: 565–576.

Polisena

Coyle

, et al. Home telehealth for chronic disease management: A systematic review and an analysis of economic evaluations. Int J Technol Assess Health Care 2009; 25: 339–349.

Polisena

Tran

Cimon

, et al. Home telehealth for diabetes management: A systematic review and meta-analysis. Diabetes Obes Metab 2009; 11.

Wakefield

Ward

Holman

, et al. Evaluation of home telehealth following hospitalization for heart failure: A randomized trial. Telemed J E Health 2008; 14: 753–761.

Finkelstein

Speedie

, and Potthoff

Home telehealth improves clinical outcomes at lower cost for home healthcare. Telem J E Health 2006; 12: 128–136.

Peacock

Konrad

Watson

, et al. Effectiveness of home visiting programs on child outcomes: A systematic review. BMC Public Health 2013; 13.

10.

Lee

Mitchell-Herzfeld

Lowenfels

, et al. Reducing low birth weight through home visitation. Am J Prev Med 2009; 36: 154–160.

11.

le Roux

Comulada

, et al. Home visits by neighborhood Mentor Mothers provide timely recovery from childhood malnutrition in South Africa: Results from a randomized controlled trial. Nutr J 2010; 9: 56.

12.

Scheiwe

Hardy

Watt

RG.

Four-year follow-up of a randomized controlled trial of a social support intervention on infant feeding practices. Matern Child Nutr 2010; 6: 328–337.

13.

Caldera

Burrell

Rodriguez

, et al. Impact of a statewide home visiting program on parenting and on child health and development. Child Abuse Negl 2007; 31: 829–852.

14.

Hamadani

Huda

Khatun

, et al. Psychosocial stimulation improves the development of undernourished children in rural Bangladesh. J Nutr 2006; 136: 2645–2652.

15.

Calvo

Gómez-Suárez

Soriano

, et al. A home telehealth program for patients with severe COPD: The PROMETE study. Respir Med 2014; 108: 453–462.

16.

Grantham-McGregor

Powell

Walker

, et al. Nutritional supplementation, psychosocial stimulation, and mental development of stunted children: The Jamaican Study. Lancet 1991; 338: 1–5.

17.

Black

Dubowitz

Hutcheson

, et al. A randomized clinical trial of home intervention for children with failure to thrive. Pediatrics 1995; 95: 807–814.

18.

Nair

Schuler

Black

, et al. Cumulative environmental risk in substance abusing women: Early intervention, parenting stress, child abuse potential and child development. Child Abuse Negl 2003; 27: 997–1017.

19.

Moss

Dubois-Comtois

Cyr

, et al. Efficacy of a home-visiting intervention aimed at improving maternal sensitivity, child attachment, and behavioral outcomes for maltreated children: A randomized control trial. Dev Psychopathol 2011; 23: 195–210.

20.

Mejdoubi

van den Heijkant

SCCM

van Leerdam

FJM

, et al. The effect of VoorZorg, the Dutch nurse-family partnership, on child maltreatment and development: A randomized controlled trial. PLoS ONE 2015; 10: e0120182.

21.

Ordway

Sadler

Holland

, et al. A home visiting parenting program and child obesity: A randomized trial. Pediatrics 2018; 141: e20171076.

22.

Goyal

Folger

Sucharew

, et al. Primary care and home visiting utilization patterns among at-risk infants. J Pediatr 2018; 189: 240–246.

23.

Brickhouse

Haldiman

Evani

The impact of a home visiting program on children’s utilization of dental services. Pediatrics 2013; 132(Suppl2): S147–S152.

24.

Woodson

Gunn

Clark

, et al. Designing health information technology tools for behavioral health clinicians integrated within a primary care team. J Innov Health Inform 2018; 25: 153.

25.

Borkan

Immersion/crystallisation. In: Crabtree

Miller

(eds) Doing Qualitative Research. Thousand Oaks, CA: Sage Publications, 1999, pp. 179–194.

26.

Crabtree

Miller

WL.

Doing Qualitative Research, Second edition. Thousand Oaks, CA: Sage Publications, 1999.

27.

Damschroder

Aron

Keith

, et al. Fostering implementation of health services research findings into practice: A consolidated framework for advancing implementation science. Implement Sci 2009; 4: 1–15.

28.

Robinson

Gund

Sjöqvist

, et al. Using telemedicine in the care of newborn infants after discharge from a neonatal intensive care unit reduced the need of hospital visits. Acta Paediatr 2016; 105: 902–909.

29.

Jetelina

Woodson

Gunn

, et al. Evaluation of an electronic health record (EHR) tool for integrated behavioral health in primary care. J Am Board Fam Med 2018; 31: 712–723.

30.

Knierim

Moore

Raghunath

, et al. Home visitations for delivering an early childhood obesity intervention in Denver: Parent and patient navigator perspectives. Matern Child Health J 2018; 22: 1589–1597.