Timing and pattern of readmission in individuals with spinal cord injury in the setting of a specialized medical home

Abstract

Background

The rate of readmission in individuals with spinal cord injury (SCI) is known to be high (28% to 45%) during the first year post-injury and post-rehabilitation. However, there are several critical gaps in our knowledge including the timing pattern of medical complications and the pattern of health complications associated with readmissions.

Objective

To identify the timing and pattern of complications associated with hospital readmissions in individuals with traumatic SCI and non-traumatic spinal cord disorders (SCI) post-discharge from an inpatient rehabilitation facility (IRF).

Design

Secondary analysis of a three year prospective cohort study.

Setting

An SCI medical home.

Participants

Individuals who were readmitted (n = 53) within the first year (N = 176) post-discharge from an IRF.

Interventions

N/A.

Outcome Measures

Timing and pattern of all-cause hospital readmissions.

Results

Eighty one percent of the readmitted patients experienced readmission within the first six months after discharge, and 36% of the initial readmissions occurred within 30 days of discharge from an IRF. The trend line for the timing of the first readmission post-discharge from an IRF was curvilinear, with a sharp decrease in the number of new patients readmitted for months 1–7 and then a slight increase between 9 and 12 months. Urological and respiratory complications were related to repeat readmissions.

Conclusion

The patient is at the greatest risk of readmission in the first 6 months, with a secondary increase in risk at 9 months. Possible reasons may include reduction in in-home and outpatient therapy and skilled nursing over the first year post-SCI.

Background

The rate of readmission in individuals with spinal cord injury (SCI) is known to be high (28% to 45%) during the first year post-injury^1–3 and first year post-inpatient rehabilitation.⁴

Health conditions associated with these high rates of readmission include urinary tract infection, respiratory infections, and skin and subcutaneous tissue injury (pressure injuries).^1,2 Even though these conditions are considered preventable, DeJong et al. have shown that 36% of individuals with SCI were readmitted at least once in the 12-month period following injury.³ These results were obtained from studying health services utilization from six geographically dispersed rehabilitation centers in the USA. Research has indicated that the current high rate of readmission may only be reduced by the innovative care and monitoring of the individuals with SCI in the community.^1–3,5,6

There are several critical gaps in our knowledge of the factors associated with unplanned readmission. First, is there a timing pattern of medical complications, post-discharge from an inpatient rehabilitation facility (IRF), which result in readmission? Knowing the timing pattern may enable healthcare providers to target outreach proactively, in order to intercede and prevent readmissions. Prior research from 16 SCI Model System centers (1995–2002) in the USA has indicated the medical complications that are associated with readmission over a period of time (1–20 years).² However, the timing-pattern of medical complications that result in readmission within the first year or the following years has not been described. Second, is there a pattern of health complications associated with repeat readmissions for individuals with SCI? While prior research has indicated the major health conditions associated with readmission,^1–4 there is a lack of information about the pattern of health complications associated with readmission.

To address these knowledge gaps this study aimed to identify the timing of hospital readmissions in individuals with traumatic SCI and non-traumatic spinal cord disorders of acute onset (here onward SCI) after discharge from an IRF. The data used in this study was collected from an IRF that developed a specialized SCI Medical Home to ensure delivery of integrated, coordinated, and cost-effective healthcare services for persons with SCI discharged to the community or short-term subacute care.⁷ The IRF is located in a large city in the USA and provides rehabilitation services to surrounding urban, suburban and rural areas. Adopting principles from primary care medical home research,⁸ the SCI Medical home aimed to reduce the readmission rate by providing proactive phone calls to patients after discharge, a 24/7 hotline for patients, and a multidisciplinary follow-up clinic with extensive case management. A secondary aim of this study was to assess the pattern of health conditions associated with repeat readmissions for individuals with SCI. By elucidating the timing and pattern of medical complications, resources can be better directed to yield maximal results in terms of reducing patient morbidity and readmission.

Methods

Participants

Potential candidates for the IRF's Medical Home were identified as they progressed through their rehabilitation and were screened by a multidisciplinary team. Although the team tracked comorbidities and adverse socio-economic factors that were expected to impact readmissions, based on the literature, it became clear early in the project that the diagnosis of SCI was sufficient to put someone at relatively high risk for readmission, unless the person was ambulatory and did not have neurogenic bowel or bladder. Thus all individuals with SCI were included (N = 176) unless they met one or more of the exclusion criteria (n = 126). Patients enrolling in the medical home included those who returned directly to the community as well as those who returned to the community after a planned short stay in a sub-acute rehabilitation facility. Fifty three of the 126 required long-term institutional placement and were excluded, and 16 were discharged to receive care at another facility, (acute care, out of area rehabilitation, Veterans Affairs facility, hospice, prison). Participants also needed to be able to return periodically to the IRF's outpatient center, so distant discharge locations or inability to return for follow-up care (n = 11) as well as insurance issues that precluded follow-up at the IRF (n = 14) were exclusion criteria. Patients who were ambulatory and who did not have neurogenic bowel or bladder were excluded (n = 27) since they were not considered to be at high risk for readmission. Five declined to participate.⁷ Informed consent was obtained during the inpatient rehabilitation stay to enable the exchange of health outcome data.

Study related procedures

Upon enrolling, patients were introduced to the medical home care team which consisted of physiatrists, certified rehabilitation nurses, clinical pharmacists, and case managers with backgrounds in social work. Prior to discharge, 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 patient was provided with detailed medication instructions and an easy-to-follow medication schedule. Additional key aspects included the provision of a 30-day supply of medication, a series of proactive calls to the patient from the team, provision of a hotline number with which patients and care providers could reach the team 24/7, scheduled follow-up visits with the team, and the availability of specialized providers such as wound ostomy continence nurses, clinical dietitians, respiratory therapists, and urologists. Whenever possible, issues were handled by phone, since transportation was often challenging for patients. Detailed information about the study related procedures is presented elsewhere.⁷

Data and statistical analysis

Data was validated by the SCI Medical Home team. Descriptive statistics were used to present demographics, readmissions, and medical complications associated with readmissions. Mann–Whitney U tests were performed to compare: 1) the number of repeat readmissions for individuals with SCI who were readmitted within the first 30-days of discharge from an IRF and after the first 30-days, and 2) age of individuals who were readmitted once and multiple times. Chi-square tests were performed to assess demographic (sex, race and ethnicity, and level of injury) differences between individuals who were readmitted once and multiple times. A trend line was developed between the number of individuals who were first readmitted and the timing of their readmission using Microsoft Excel 2016. All statistical analyses were performed using IBM SPSS Statistics software (version 24) with a statistical significance at an alpha level of 0.05.

Results

Table 1 presents the demographic characteristics of the participants who took part in the study.

Table 1.

Demographic characteristics of individuals with Spinal Cord Injury (SCI).

Age (mean ± SD)	44.2 ± 19.9 years
Sex	Male: 134 Female: 42
Race in % (Number)	Asian: 1% (2) Black: 38% (67) White: 61% (107)
Ethnicity	Hispanic or Latino: 6% (11) Not Hispanic or Latino: 94% (165)
Marital status	Single: 56% (99) Married: 36% (63) Divorced: 3% (5) Separated: 0% (0) Widowed: 5% (9)
Level of injury	Paraplegia: 46% (81) Tetraplegia: 54% (95)
Traumatic	ASIA (American Spinal Injury Association) Impairment Scale (AIS) AIS A-C C1- C4: 20% (36) AIS A-C C5- C8: 11% (20) AIS A-C paraplegia: 34% (60) AIS D any: 15% (27)
Nontraumatic	Incomplete tetraplegia: 9% (16) Incomplete paraplegia: 7% (13) Complete tetraplegia: 1% (1) Complete paraplegia: 2% (3)

The demographic composition of the presented cohort is representative of the SCI population within model system statistics for average age (35.6 ± 17.2 years), sex (80.4% male), and neurologic category (52.4% tetraplegia, 43.5% paraplegia, 0.6% normal, and 3.5% unknown).⁹ The average age is 44.2 ± 19.9 years old with a 76% male predominance and a roughly even split between paraplegic (46%) and tetraplegic (54%) injuries. Table 2 highlights the incidence of readmission, readmissions related to SCI, medical conditions associated with readmissions, and FIM motor scores.⁷

Table 2.

Readmissions in individuals with Spinal Cord Injury (SCI) within the first year post-discharge from Inpatient Rehabilitation Facility (IRF).

	Percent (Number)
Total Number of patients at least one year post-discharge from an IRF	100% (176)
Incidence of readmissions within first year of discharge from an IRF	0.46 (81)
Unique patients readmitted (number of individuals readmitted one or more times)	30% (53)
Readmissions related to SCI	89% (72)
Readmissions not related to SCI	11% (9)
Top conditions for all readmissions
Urologic	39% (28)
Pneumonia/Respiratory	19% (14)
Infection (unspecified)	15% (11)
Cardiovascular	11% (8)
Pressure Ulcer	6% (4)
Bowel	4% (3)
Other	6% (4)
Functional Independence Measure (FIM) Motor scores (mean ± SD)
Readmitted	34.9 ± 17.8
Not readmitted	48.6 ± 17.4

Fig. 1 shows the timing pattern of first, second, and third readmissions post-discharge from IRF. In addition, the trend line indicates a sharp decrease in the number of new patients readmitted for months 1–7 and then a slight increase between 9–12 months. Six of the 19 participants (31.6%) who were readmitted within the first 30 days post-discharge from an IRF had repeat (two or more) readmissions. Eleven of the 34 participants (32.4%) who were readmitted after the first 30 days of post-discharge from an IRF had repeat readmissions. Median number of repeat readmissions for individuals who were readmitted within the first 30 days post-discharge from an IRF and after 30 days was 2.5 (n = 6) and 2.0 (n = 11), respectively. The median number of repeat readmissions for those with a first admission within the first 30 days compared to those with repeat readmissions after 30 days was not significantly different (P = 0.591).

Figure 1.

Pattern of readmissions (Top) and polynomial trend line for first readmission (Bottom) per month post-discharge from an Inpatient Rehabilitation Facility (IRF).

Seventeen individuals with SCI had multiple readmissions. The results indicated that urinary tract infection (UTI), pneumonia, and both UTI and pneumonia, were present in 88.2%, 29.4%, and 23.5% of repeat readmissions, respectively. A demographic comparison between individuals who were readmitted once (n = 36) vs. multiple times (n = 17) indicated no significant differences for age (P = 0.116), sex (P = 0.067), race (P = 0.697), ethnicity (P = 0.962), level of injury (P = 0.199), or marital status (P = 0.127).

Discussion

Timing of readmission

The results of the original study on which this secondary analysis was based showed that 30% of the individuals were readmitted within the first year after discharge from an IRF, and 24% were readmitted within the first year after onset.⁷ The rate of readmission to acute care in our study is on the lower end of other studies,^1–4 especially in light of the fact that a full year post-discharge from an IRF was included. Several studies^1–3 list rates of readmission from date of injury which presumably entails less than a full year in the community. The highest rate of readmission in our study was in the first month, which was expected as the patient, family and community care providers learn to manage this complex and typically unfamiliar condition without the continuous support of an on-site rehabilitation team. The rate of hospital readmission reaches a nadir at the 6th month. Unexpectedly, the rate of readmission again increased somewhat at the ninth month. The reasons are unclear, but it is possible that in-home and outpatient therapy and skilled nursing supports may have receded by that time.

Causes of readmission

Cardenas et al. and DeJong et al. identified three etiologies that were associated with readmission including genitourinary system (e.g. urinary tract infection), respiratory system (e.g. pneumonia), and skin-related diseases (e.g. pressure injuries).^2,3 In addition to respiratory and urologic complications, Jaglal et al. also identified musculoskeletal, gastrointestinal, and cardiovascular-related conditions as major reasons for readmission.¹ Research by Stillman et al., of the Skelton et al. dataset from above, reported that UTI, autonomic dysreflexia, pressure injuries, bone and soft tissue injuries, and bowel problems were the top five complications in their study.¹⁰

The findings from this study are consistent with previous studies^1–3,10 in that urological and respiratory complications were the first and second causes of readmission, particularly infection of those systems. Unlike previous reports,^2,3,10 the current study did not find complications of the integumentary system, such as decubitus ulceration, to be a leading cause of hospitalization. This was likely due to the aggressive and on-going outpatient management of pressure injuries by certified wound nurses in the follow-up clinic, patient education regarding avoiding emergency departments when wounds were found, and a trend for office procedures and outpatient based surgery for pressure injuries.

Recommendations for future studies

Further study into the most efficient means and frequency of patient follow up to reduce readmission is needed. As telehealth emerges as a practical tool with which to treat patients, the best way to integrate its use into the healthcare continuum needs to be determined. In addition, our study found no significant demographic differences between individuals with single and multiple readmissions, which may be due to small sample size and needs to be further investigated in a larger sample size.

Limitations

Data from this study were collected from patients discharged from a single IRF and may not be generalizable. In addition, the IRF had fulltime rehabilitation nurses and a social worker/case manager with extensive experience with this population in its Lifetime Follow-up Clinic, prior to the initiation of this study. While the SCI Medical Home project seemed to have lowered the absolute number of admissions,⁷ it is less likely but possible that the timing of readmissions was affected. Due to the unavailability of administrative claims data, the study relied on patient self-report which may have introduced errors in the timing of readmissions. However, extensive efforts were made to collect pertinent and accurate information. Regional practice trends and medical resources such as the availability of 24/7 retail pharmacy may have also influenced the current outcomes. Furthermore, the care provided within the medical home evolved and improved as lessons were learned and experience gained throughout the project.

Conclusion

This study identifies urological and respiratory infections as being the most common causes of repeat readmissions in the first year after discharge from an IRF. The patient is at the greatest risk of readmission in the first 6 months with a secondary increase of risk at 9 months. Possible reasons may include reduction in in-home and outpatient therapy and skilled nursing over the first year post-SCI. Based on the findings of this study, the medical home team began tracking the bladder management systems in use by patients at discharge to see what could be learned. The team might consider an additional touch-point with patients around 8–9 months after discharge, to try to prevent the uptick in readmissions, late in the first year. Future clinical trials should assess the readmission and timing of readmission across a control group (usual care) and medical-home group.

Disclaimer statements

Contributors None.

Funding The work is supported by the Craig H. Neilsen Foundation (Grant#320077).

Conflicts of interest The authors report no conflicts of interest.