Decreasing re-hospitalizations and emergency department visits in persons with recent spinal cord injuries using a specialized medical home

Abstract

Objective: To determine whether a specialized medical home can reduce re-hospitalizations and emergency department (ED) visits as compared to reports in the literature for persons with spinal cord injury (SCI) in the first year post-discharge from acute inpatient rehabilitation.

Design: A three-year prospective cohort study.

Setting: An academic free standing inpatient rehabilitation hospital (IRF), participating in the SCI Model Systems network, serving urban, rural and suburban settings in the mid-Atlantic region of the United States of America.

Participants: 176 successive individuals with varying levels of SCI, inclusive of patients requiring mechanical ventilation, discharged from rehabilitation from 2/1/15 to 7/1/17, who met criteria and consented to participate.

Interventions: Pre-discharge communication between the inpatient and outpatient staff was initiated, medication education by a pharmacist was added, 1 month of discharge medications was offered, and proactive phone calls to patients after discharge were instituted. In addition, patients were offered a 24/7 hotline to reach physician and nursing staff, and multidisciplinary outpatient follow-up was provided in conjunction with extensive proactive case management.

Outcome measures: All-cause hospital readmission and ED visits.

Results: Thirty percent of the individuals were readmitted within the first year after discharge from an IRF, and 24% were readmitted within the first year after onset. The incidence of readmission was 0.46 and 0.36 respectively. Forty-one percent had an ED visit.

Conclusion: The results suggest that the medical home interventions decreased the rate and incidence of readmission and the rate of ED usage in the first year.

Introduction

Spinal cord injury (SCI) is a devastating pathology that affects approximately 300,000 persons in the United States, with an estimated 17,700 new SCIs each year.¹ Advancements have been made in pre-hospital treatment, emergency department care, neurologic intensive care and neurological surgical management. The average acute hospital stay has been reduced from 24 days in the 1970s to 11 days currently, and is often followed by inpatient rehabilitation, for which the average length of stay has declined from 98 days in the 1970s to 34 days currently.¹ Immediately after discharge from rehabilitation, patients and their caregivers must cope with the new injury on an emotional and physiologic level while also contending with the logistics of managing the new condition.

Given this often overwhelming situation, it is not surprising that Skelton et al. recently reported a 45% re-hospitalization rate in the first year after inpatient rehabilitation for 168 patients with traumatic SCI (TSCI).² Earlier studies in the US of persons with traumatic SCI report re-hospitalization rates of 27.2–33% in the first year after injury.^3–5

While numerous studies describe the rates and causes of re-hospitalization, and a few studies address the reduction of re-hospitalizations among selected groups with chronic SCI,^6,7 the authors found only one study that successfully reduced the number of days of unplanned re-hospitalization in the first 180 days after discharge from acute rehabilitation, though there was no statistically significant difference in the rate of re-hospitalization.⁸ This pilot study attempts to reduce re-hospitalizations and ED visits in the first year after IRF discharge by applying concepts developed in primary care medical homes (PCMH) to this specialized population.

The PCMH model entails intensive case management and medical services for persons with chronic medical conditions that put them at high risk for re-hospitalization. The goal of a PCMH is to have a designated team of healthcare providers work with the patient to maintain health in a cost effective and clinically efficient manner. Often payment incentives are realigned so that there is a risk-adjusted monthly care coordination payment, a visit-based fee-for-service payment, and a performance-based component for the achievement of quality and efficiency goals.^9,10 The team engages with the patient, understanding and breaking down barriers to health and enabling the patient to manage their condition more independently.

Two studies have demonstrated value in applying medical home principles to specialized populations of persons with physical disabilities. A recent study by Maeng et al. demonstrated the value of a Comprehensive Care Clinic for adolescents and young adults with special care and health needs, transitioning into the adult system of care.¹¹ This study applied a complex care management model with seven principles, similar to those used in a medical home, and demonstrated a 28% reduction in per-member-per-month total cost, driven by decreases in hospitalizations and ED visits. Milligan and Lee also applied medical home principles via an Interprofessional Mobility Clinic for persons with severe physical disabilities, including persons with spinal cord injury.^12,13 Though the evidence was anecdotal, outcomes appear to support the medical home principles for this population. Thus, the authors hypothesized that a specialized medical home, utilizing PCMH principles, might be beneficial to persons with new SCI, as they are discharged from inpatient rehabilitation, to support them in learning to manage their new circumstances.

For this project, a multi-disciplinary professional team at a NIDILRR-designated regional SCI Center, (National Institute on Disability, Independent Living and Rehabilitation Research), developed a grant-funded specialized medical home for patients with new onset spinal cord injuries. The rehabilitation team, led by physiatrists, included rehabilitation nurses, clinical pharmacists, certified wound ostomy continence nurses (CWOCN), dietitians, seating specialists, physical therapists, peer mentors, and social work trained case managers.

Methods

Participants

All patients with sudden onset of new SCI, (not necessarily traumatic), discharged from this IRF from 2/1/15 to 7/1/17 were screened for possible inclusion. Unlike the model system data set, patients with spinal cord infarcts, bleeds, tumors or other non-traumatic causes of sudden spinal cord injury were included, though the data were separated for comparison purposes. Those who were ambulatory and did not have neurogenic bowel or bladder were excluded. Patients who returned directly to the community after acute rehabilitation as well as those who planned to return to the community after a short course of sub-acute rehabilitation were included. Patients who were anticipated to have long-term institutionalization were not enrolled. Additionally, enrollees needed to be able to return periodically to the IRF’s outpatient center, so patients with distant discharge locations as well as those with insurances that precluded follow-up at the IRF were excluded. Voluntary informed consent was obtained once the discharge disposition, functional status, and follow-up plan were relatively clear, usually in the last two weeks of inpatient rehabilitation. Patients who declined consent were not enrolled.

Interventions

Upon enrolling, patients were introduced to the core medical home team which consisted of a physiatrist, a certified rehabilitation nurse, a clinical pharmacist and a social worker. Prior to discharge from rehabilitation, the inpatient team performed a face-to-face hand-off with the medical home care providers. The physiatrist and pharmacist performed a thorough medication reconciliation, and the pharmacist provided to the patient detailed medication instructions and an easy-to-follow printed medication schedule. Additionally, the pharmacist resolved any formulary and prior authorization issues prior to discharge. Patients typically were discharged with a 30-day supply of medications in hand. Infrequently, due to insurance restrictions, medications needed to be mail-ordered, and a 30-day supply could not be provided prior to discharge. However, patients were always provided sufficient medication to carry them through until the mail-order medications were delivered.

After discharge all participants received proactive support that ranged from phone calls by staff at approximately 1 day, 1 week, 1 month and then every several months after discharge, to complex and on-going case management interventions related to transportation, equipment, supplies, housing, food insecurity, and legal issues. Clinical guidance was provided to help identify potential and emerging medical complications and to ensure that all questions related to care and services were answered. The medical home team was often in contact with home nursing and therapy services, primary care clinicians, and medical and surgical specialists to coordinate and discuss patient-specific care needs. Patients were given access to a 24/7 phone number that they could call for questions or concerns. The hotline was first answered by a senior member of the nursing staff then elevated to an attending physiatrist on-call as needed.

Patients were typically given an outpatient appointment for 2–4 weeks after discharge. Additional visits were scheduled every 2 weeks to 6 months as indicated. Specialized providers, such as wound ostomy continence nurses, clinical dietitians, respiratory services, and urologists were available as needed (Table 1).

Table 1. Template of routine medical home activities.

• Patient with SCI/D admitted to rehab. As discharge approached, medical home inclusion criteria reviewed with case manager. If patient met inclusion criteria then project director met with patient and obtained informed consent. Demographic data obtained from medical record.

• Prior to discharge, face-to-face hand-off between inpatient nurse and case manager and medical home care providers. Team members introduced selves to patient and explained what would happen after discharge. Patient given refrigerator magnet with hotline number. Physiatrist and pharmacist performed thorough medication reconciliation. Pharmacist screened medications for any formulary and prior authorization issues. These were resolved prior to discharge. Pharmacist provided to the patient detailed medication instructions and an easy-to-follow printed medication schedule. Patients discharged with a 30-day supply of medications in hand.

• 24–48 h after discharge, a member of medical home team called the patient at home to insure all was well. Timing and content of subsequent calls based on outcome of this call.

• Member of team called the patient day 5 with scripted questions, if there had been no need for additional contact since first call.

• Some patients and care givers were in regular contact with staff to address questions and needs. Scheduled call to the patient day 25 with scripted questions and reminder about one month follow-up visit.

• Patient seen in clinic by multidisciplinary team at one month. If active issues, such as a pressure injury requiring management, were identified before discharge or on any of the calls, patient was seen in clinic earlier.

• Patient contacted at 5 months, if no recent contact, and scheduled for 6 month visit. Typically, supplies and medications were reordered every 3 months. Sometimes insurers mandated in-person visits prior to re-order.

• Patient seen in clinic six months post discharge.

• Patient contacted at 11 months and reminded about annual visit.

• Patient seen in clinic for first annual visit post discharge.

Data collection

Demographic data were collected from the medical record after informed consent was obtained. Data related to re-hospitalizations and emergency department visits were gathered in real-time, when the team was involved in the process, and during follow-up clinic visits and phone calls otherwise (Table 2).

Table 2. Demographic characteristics of individuals with SCI/D.

	Percent (Number)
Age (mean ± SD)	44.2 ± 19.9 years
Sex	Male: 134; Female: 42
Race	Asian (1%), Black (38%), White (61%)
Ethnicity	Hispanic or Latino (6%), Not Hispanic or Latino (94%)
Marital status	Single (56%), Married (36%), Divorced (3%), Separated (0%), Widowed (5%)
Payer category	Commercial 82 (47%) Medical Assistance 49 (28%), Medicare 33 (19%), Workers Comp 7 (4%), self-pay 2 (1%), military 3 (2%)
Level of injury	Paraplegia (46%), Tetraplegia (54%)
Traumatic AIS A-C C1–C4	(20%) 36
Traumatic AIS A-C C5–C8	(11%) 20
Traumatic AIS A-C para	(34%) 60
Traumatic AIS D any	(15%) 27
Nontraumatic incomplete tetraplegia	(9%) 16
Nontraumatic incomplete paraplegia	(7%) 13
Nontraumatic complete tetraplegia	(1%) 1
Nontraumatic complete paraplegia	(2%) 3

Statistical analysis

Categorical variables were summarized using frequencies and percentages. Continuous variables were summarized using means and standard deviations. Groups were compared using ANOVA for continuous variables and Pearson’s chi-square test of Fisher’s exact test for categorical variables. The number of readmissions and the number of ED visits were summarized using incidence rates with study comparisons using Poisson regression. Readmissions and ED visits were summarized for all patients and for the subset with traumatic injuries. All analyses were performed using SAS 9.4 and SAS/STAT 14.3 (SAS Institute, Cary, NC, USA).

Results

One hundred and seventy six individuals were enrolled in the program while 126 people were excluded. Of those excluded, 27 had high levels of function and were not thought to be at high risk of re-hospitalization, 53 required long-term institutional placement, 16 were discharged to receive care at another facility, (acute care, out of area rehabilitation, VA facility, hospice, prison), 14 had insurance that precluded follow-up, 11 lived too far away or were otherwise unable to return for follow-up care and 5 declined to participate.

Of the 176 participants, 5 died within the first year; 3 with non-traumatic SCIs (NTSCI) and 2 with TSCI. Three patients with TSCI had insurance changes mid-year and could not continue the program. The exclusion of these participants skews the results in a favorable direction. To minimize this artifact and to avoid biasing the outcomes, particularly as those who died may have been the most fragile, their data were not excluded.

Thirty percent of the 176 participants were readmitted to hospital within the first year after discharge from the IRF, and 24% were readmitted within the first year after SCI onset. The incidence of readmission was 0.46 within one year of IRF discharge and 0.36 within one year of onset. Eighty-nine percent of the readmissions were related to the SCI. The four most common causes for all readmissions were urologic pathologies (39%), respiratory complications (19%), infections (unspecified) (15%) and cardiovascular problems (11%).

Forty-one percent of the 176 participants visited the ED within the first year after discharge from an IRF, and 36% visited the ED within the first year after onset. The incidence of ED visits was 0.74 within one year of discharge from rehab and 0.60 within one year of onset. Fifty-five percent of the ED visits resulted in hospital admissions. Of those visits that did not result in admission, 85% were related to the SCI (Tables 3 and 4)

Table 3. Readmission/ED Visits within the 1st year post rehab discharge.

	All Subjects (N = 176)		Traumatic Injuries (N = 143)
	Readmissions	ED Visits	Readmissions	ED Visits
Number of patients (%)	53 (30%)	73 (41%)	46 (32%)	62 (43%)
Number of readmissions/ED Visits (incidence)	81 (0.46)	131 (0.74)	72 (0.50)	114 (0.80)
Readmissions related to SCI	72 (89%)		64 (89%)
ED Visits resulting in readmission		72 (55%)		65 (57%)
ED Visits not resulting in readmission, not related to SCI		9 (7%)		7 (6%)
ED Visits not resulting in readmission, related to SCI		50 (38%)		42 (37%)
Top conditions related to SCI	(out of 72)	(out of 50)	(out of 64)	Out of 42
Urologic	28 (39%)	22 (44%)	27 (42%)	21 (50%)
Respiratory	14 (19%)	6 (12%)	14 (22%)	6 (14%)
Pain		8 (16%)	1 (2%)	5 (12%)
Infection (unspecified)	11 (15%)		8 (12%)
Cardiovascular	8 (11%)	3 (6%)	6 (9%)	2 (5%)
Pressure injury	4 (6%)	4 (8%)	3 (5%)	4 (10%)
Bowel	3 (4%)	3 (6%)	3 (5%)	3 (7%)
Other	4 (6%)		2 (3%)	1 (2%)

Table 4. Readmissions/ED Visits for individuals with SCI/D during the 1st year post injury.

	All Subjects (N = 176)		Traumatic Injuries (N = 143)
	Readmissions	ED Visits	Readmissions	ED Visits
Number of patients (%)	42 (24%)	64 (36%)	37 (26%)	53 (37%)
Number of Readmissions/ED Visits (incidence)	63 (0.36)	106 (0.60)	57 (0.40)	90 (0.63)
Readmissions related to SCI	57 (90%)		51
ED Visits resulting in readmission		57 (54%)		52 (58%)
ED Visits not resulting in readmission, not related to SCI		9 (8%)		7 (8%)
ED Visits not resulting in readmission, related to SCI		40 (38%)		31 (34%)

Discussion

Re-hospitalization and emergency department visits

Throughout the SCI literature, the rates of ED utilization and re-hospitalization are noted to be much higher than the general population, particularly in the first year. The literature is not consistent as how to define the first year. Most studies report one year from the date of injury,^1,4 but others utilize 1 year from discharge from IRF, which is a more logical comparator to use when reviewing the effects of actions post discharge.^2,5

By the time of discharge from rehabilitation, caregivers and patients managing new SCI must be able to learn and apply new skill sets that may include bladder management with catheterization, bowel management and functional activities. Often multiple medications must be procured and organized and new equipment mastered. Patients must follow up with multiple specialists, procure specialized transportation and navigate insurance issues. Failure to master these skills may lead to medical complications and re-hospitalizations.

Scant information is available about ED visitation by persons with spinal cord injuries. Skelton et al. reported a rate of 57% in the first year from IRF discharge but incidence was not reported.² The authors are not aware of any literature citing the total incidence of ED visits in the first year for persons with SCI.

More data are available about hospital readmission rates. According to DeJong, during a 5-year period (2007–2012), about 36.2% of individuals with traumatic SCI, discharged from six geographically dispersed rehabilitation hospitals in the US, were re-hospitalized at least once in the 12-month period following injury.⁴ There was marked variability between centers – from 27.8% to 50%, with some of the difference accounted for by case mix index. These data were collected by annual survey, which has the risk of underreporting as well as the risk of confusing ED visits or observational status with formal hospital readmission. During this pilot project, in the 12-month period following injury, the re-hospitalization rate was 24%. Although there was close contact with most participants, the risk of underreporting still exists.

Using survey data collected by phone every 3 months after discharge from inpatient rehabilitation, Skelton et al. found a 45% re-hospitalization rate for those discharged from one center from 2009 to 2013.² Cardenas found a re-hospitalization rate of 28–37% using data from 1995 to 2002.⁵ These data may be less relevant currently, as the demographics of persons with SCI and the healthcare industry have experienced significant change over the past 2 decades. In fact, Cardenas et al. found a trend suggestive of increasing re-hospitalization rates in the years observed, though the trend was observational and not statistically significant. The re-hospitalization rate utilizing our specialized SCI medical home was 30% with an incidence of 0.46, in the year following discharge from rehabilitation.

Calculations from the NIDILRR SCI Model System database reflect an incidence of re-hospitalization of approximately 0.50 within one year of onset.^1,14 These data include only traumatic etiologies of SCI. For the study population, (sudden onset of traumatic and non-traumatic etiologies), the incidence of hospital readmission was 0.36 within one year of onset. When traumatic etiologies were analyzed in isolation, the incidence was 0.40. While the difference in re-hospitalization incidence between the model systems dataset and the study population did not reach significance when NTSCI’s were excluded, the effect may have been dampened by the study exclusion criteria, which eliminated higher functioning patients. Future studies should include follow-up of these patients.

The primary objective of this project was to learn if a specialized medical home could reduce the rate of re-hospitalizations and ED visits for persons with recent onset of SCI. The rate of re-hospitalization appeared to be significantly less for the project participants, measured from date of onset or date of discharge from rehabilitation, when compared to recent reports in the literature, suggesting that the specialized medical home may have reduced rates of re-hospitalizations. Though the incidence of re-hospitalization within one year of onset was reduced, it was not significantly different from reported rates. The rate of ED visits within one year of discharge from rehabilitation was also significantly different than that reported in the literature (Table 5).

Table 5. Comparison to previous studies.

Time-frame	Study	Sample Size	Patients with Readmission	p-value (compared to this report)	Readmission Incidence	p-value (compared to this report)	Patients with ED Visit	p-value (compared to this report)
One Year Post Injury	Dejong 2013	951	344 (36%)	0.010
	Model System 2015	22652	7733 (34%)	0.024	11336 (0.50)	0.067
	This Report – Traumatic Only	143	37 (26%)		57 (0.40)
One Year Post Discharge	Skelton	168	76 (45%)	0.019			96 (57%)	0.015
	This Report – Traumatic Only	143	46 (32%)				62 (43%)

Proportions of patients with readmission/ED Visits compared using Pearson’s chi-square test. Incidence rates compared using Poisson regression.

Reasons for re-hospitalization

The two top reasons for re-hospitalization in the medical home population were the same as reported earlier by Cardenas, DeJong, and Skelton, including diseases of the genitourinary system and of the respiratory system.^2,4,5 Re-hospitalizations related to pressure injuries were not as prevalent as previously reported, perhaps due to the aggressive outpatient care that was provided. Even though the conditions noted above are considered preventable, research has indicated that the current high rate of re-hospitalization may only be reduced by a continuum of care and monitoring of individuals with SCI in the community.^4,5,15,16

Application of medical home concepts

Research on medical homes shows that they are successful when they include person and family-centered care that is team-based, collaborative, coordinated, integrated, accessible, comprehensive, equitable, continuous, and of high value.¹⁷ For the Comprehensive Care Clinic, Maeng also noted the importance of the multidisciplinary care team, comprehensive assessment, case management, emergent care, caregiver support, self-management education/training and medication reconciliation.¹¹ All of these principles were integrated into the SCI Medical Home.

Milligan noted anecdotal evidence that an Interprofessional Mobility Clinic for persons with severe physical disabilities was able to reduce reliance on more costly acute care and specialist resources by managing and preventing secondary complications, such as bladder infections.^12,13 This model incorporated the same principles as the SCI medical home but delivered care in a different manner. Though the evidence was anecdotal, outcomes appear to support the principles.

In contrast, Mackelprang et al. used telephone counseling to try to reduce medical complications and health care utilization and to improve psychosocial outcomes during the first year post-SCI rehabilitation. In this study, all of the principles of the medical home, such as 24/7 access and emergent care were not applied, and no significant differences were observed between telephone counseling and usual care.¹⁸

Lessons learned

Over the course of this project, many lessons were learned and processes were continually refined to improve care, efficiency, and patient outcomes. The importance of the social worker in managing the social determinants of health cannot be over-stated. Typical actions included navigating insurance hurdles related to medication, supplies, equipment, and transportation, scheduling transportation to medical appointments, reaching out to discharge planners at skilled nursing facilities to coordinate smooth transitions home, and providing support and assurance to care providers and patients. Less typical actions included active interventions to prevent homelessness, advocating on behalf of patients with governmental agencies, and connecting patients with necessary legal services.

Despite the best intentions of staff, it was difficult to keep track of the necessary proactive calls to patients. Proactive calls tended to be deferred in favor of more urgent reactive calls. The addition of tracking software in the third year of the project added an important layer of accountability that appeared to contribute to further decreases in re-hospitalizations and ED visits. It is important to note that it was rarely a particular proactive call that prevented an adverse event. Rather, it is thought that the repeated calls from the team helped to forge relationships between the patients and the outpatient care team, and that those relationships increased the likelihood that a patient or caregiver would call at a later date with a question or concern. It was these later calls from patients and caregivers that enabled the team to intervene when a problem was small, and thus prevented a more significant issue. Historically, in the absence of a strong relationship between the care team and the patient, the patients rarely called with questions. The criticality of such relationships is noted throughout the PCMH literature.¹⁷

The 24/7 hotline was also vital to the success of the medical home. Though the clinic averaged only 2–3 off-hours calls per week, these calls often resulted in physician action that prevented ED visits. Typically, calls involved suspected urinary tract infections (UTIs) or other urologic issues. Occasionally they involved acute respiratory infections or autonomic dysreflexia.

One early concern was that the hotline would be used for routine calls that could have been made during business hours. In fact, this was extremely rare. During off-hours, the answering service contacted a nurse supervisor who was on-duty in the hospital. That person answered questions and provided triage, contacting the physician on-call as needed. For routine needs, the supervisor directed patients to call back during regular business hours.

One unexpected benefit of the hotline was that it was used by visiting home nurses and therapists to clarify care. This was extremely beneficial to the patients and helped the home care professionals feel more comfortable caring for this population. On a few occasions, physicians fielded calls from ED physicians related to autonomic dysreflexia and urologic issues.

At several points in the project, the outpatient team worked with the inpatient team to develop targeted efforts to reduce UTI’s and pneumonia. The inpatient team began to discharge each patient with a neurogenic bladder with a specimen cup and a catheter kit, even if the person was able to void volitionally. The specimen cups made it easy for a patient calling the hotline with symptoms of a UTI to collect a urine sample prior to starting an antibiotic. The catheter kit was invaluable if a person developed urinary retention or was suspected of having autonomic dysreflexia. Staff held pointed conversations with patients prior to discharge about the likelihood of urinary issues and the importance of calling the hotline at the earliest signs of a problem. Patients were encouraged to purchase thermometers and automated blood pressure cuffs. These basic efforts paid frequent dividends.

The extensive medication reconciliation process prior to discharge and the provision of a 30-day supply of all medications were also critical.¹⁹ These processes eliminated many urgent situations in the first few days after discharge due to patients being unable to obtain medications as a result of unexpected copays, extensive prior authorization processes and initial denials. Additionally, it insured that patients had all of the bowel medications needed for their routines, including the over-the-counter medications. Prior to the institution of this process, it was relatively common for patients to skip pieces of their bowel programs in order to save money, and this soon became problematic.

Limitations of study

Data from this study were collected from patients discharged from a single rehabilitation hospital and may not be generalizable. This hospital had key infrastructure, such as fulltime rehabilitation nurses and a social worker with extensive experience with this population in place in its Lifetime Follow-up Clinic, prior to the initiation of this study. It is possible that the data are not fully accurate due to reliance on patient report for some outcomes. The lack of baseline data make definitive conclusions about the effectiveness of the intervention impossible. Claims data, which would have improved the accuracy and strength of the results, were not available.

It is also important to note that the care provided within the medical home evolved and improved as lessons were learned throughout the project. The team typically met twice monthly and utilized a continuous quality improvement approach that enabled increasing success over time.

Conclusion

The rate of readmission and ED usage in the first year post-SCI were decreased, suggesting that the actions of a medical home team made a positive impact. This project demonstrates a model of care with potential to reduce costly re-hospitalizations and ED visits. However, these benefits can only be realized in the context of a payment model that recognizes and supports proactive care that addresses both the physical and social determinants of health, such as that provided in this SCI specialized medical home. Next steps should include partnering with payors to track associated costs and savings for this program and attempting to scale out the program to additional sites. Additionally, future research should explore which sub-populations of communities of persons with disabilities would most benefit from this model.

Disclaimer statements

Contributors None.

Funding The authors would like to acknowledge The Craig H. Neilsen Foundation (Grant#320077) and the members of the Medical Home team.

Conflicts of interest The authors have no conflicts of interest to disclose.