Although several systematic reviews regarding transitional care for older people have been already conducted [14, 20], to our knowledge, this is the first systematic review on the use of TCM components addressing the geriatric patients (> 65 years old and with multi-morbidity), and all-cause readmission rate reduction. The present systematic review had as an objective to identify and summarize the different TCM components implemented in the included studies to guarantee safe transitions from hospital to community dwelling in order to reduce hospital readmissions in geriatric patients. Another objective was to recognize the Transitional Care Model components’ role and impact on readmission rates reduction. Addressing this very specific population – which will be an important issue due to the demographic change in the upcoming future – the findings of this systematic review provide valuable information that can provide guidance to health care professionals or to the development of evidence-based transitional care interventions. The increasing number of geriatric patients implies the utmost need to adapt the structures and methodologies of the current public health care systems. The diversity of professionals taking the lead in geriatric care is supported by the included three intervene studies: one being lead by occupational therapy (Australia), one conducted by pharmacy (Spain) and the last one in cardiology setting (US). This also demonstrate the diversity of the three international health care systems.
Although we realize that due to our established inclusion criteria, we only came up with three included studies – demonstrating the need for future studies with high quality, and larger sample sizes – the extracted information is very solid and of high research standards. Within the excluded studies are the basic studies conducted on the TCM assessment by Naylor and colleagues, since they did not meet the inclusion criteria related to participants age (> 65 years old), nor the intervention period [17, 24, 25]. Similarly, other studies were excluded for not meeting the eligibility requirements, e.g., Cao et al., 2017 and Bekelman et al., 2018, which did not meet the age criteria [47, 48]. With regard to “healthy aging” as proposed by the World Health Organization in 2015, all efforts have to be made for geriatric patients with multi-morbidity and chronic conditions to help them stay independent even in the context of care transitions [49, 50].
Primary outcome: reduction of the readmission rate
This systematic review found meaningful differences between the three included trials, which are important for designing future trials and for the identification of relevant aspects in the improvement of transitional care in geriatric patients.
Aspects such as the length of the intervention were different between the three trials. The trial of Rich and colleagues [30] evidenced a successful reduction of the readmission rate at three months of intervention and follow up. For its part, Lopez Cabezas and colleagues [29] found this success in the reduction of the readmission rate at two and six months of intervention and follow up, but not at 12 months of intervention and follow up. A non-significant readmission rate was observed by Clemson and colleagues [28] after one month intervention, at the three months of follow up (Table 2). These different findings suggest that the length of the intervention as one aspect seems to influence the readmission rate. Future research is needed, to evidence the optimal length of transitional care for geriatric patients. Furthermore, the needed staff is being an economical issue for the public health care systems.
The included trials were rated in this systematic review as moderate- to high-intensity interventions of transitional care (Table 3). Based on these results, it is possible to hypothesize that the intensity level may have a relevant effect in reducing the readmission rate. Verhaegh and colleagues evidenced as a result of their meta-analysis that high-intensity interventions were associated with reduced short-term, intermediate-term and long-term readmissions [38]. They found a significant association between the first home visit within the first three days after hospital discharge and the reduction of short-term readmission rates. The results are related with the inverse relationship between early follow-up and risk of readmission already established by Hernandez and colleagues [51]. An explanation for these associations could be the characteristics of geriatric patients, who need a complex care, considering their high vulnerability and fragility. Particularly, for those patients who live alone, intensive supportive interventions after hospital discharge may play an important role in the prevention of hospital readmissions.
In relation to the multicomponent intervention approach, recent literature reviews evidenced a significant association between the number of transition components included in an intervention and the probability of success in the reduction of readmissions [20, 52,53,54]. Additionally, the research group of Burke inquired about the specific role of each component in reducing readmission rates. The component Cp5: Assessing / Managing Risk and Symptoms was the component most likely to reduce readmissions. An application of the components Cp9: Fostering Coordination and Cp6: Education/ Promoting Self-Management exhibited also a significant effect in reducing readmission rates [53]. Furthermore, these findings are in line with the existing research of Koelling and colleagues, who evaluated an educational component individually, evidencing less risk of rehospitalization in patients receiving the education intervention compared to patients receiving usual care [55].
Cp 1: screening
All of the three trials performed this component due to the study design. However, geriatric patients should also be identified in the daily routine of the hospital. According to Greysen and colleagues [56] participants who are more fragile (poorer physical function, older age, suffering from multi-morbidity, impairment in activities of daily living, etc.) are the ones who tend to present higher readmission rates. There are evaluated geriatric assessments, e.g. ISAR score to identify these patients [57]. Beyond that, their special needs should be considered individually. De Wit and Schuurmans [58] suggested that this approach could lead to a slower deterioration in the condition of the patients and limited unplanned (re-)admissions.
Cp 2: staffing
Although the traditional TCM uses advanced practice registered nurses to provide the hospital to community dwelling intervention, one positive aspect of our finding is that different professions seem to successfully implement the TCM model, which broadens the possible implementation process of this model. All three studies used different professions to implement their transitional care intervention. Trained occupational therapists, who aimed to provide patients with self-care skills that allow them to cope with daily living issues and return to their daily life activities, conducted the intervention in the trial of Clemson and colleagues [28]. Pharmacists carried out the intervention in the trial of López Cabezas and colleagues [29], who were experts on medicines. Thus, they were supportive in improving medication adherence as they provided education regarding medical doses, frequencies and number of dose intervals for the medical treatment of patients [59]. Rich and colleagues [30] used a multidisciplinary team and reported a significantly reduced readmission rate. These findings are in line to a recent study where particularly multidisciplinary network in heart failure management has been related to a reduction of rehospitalization, prolonged survival, and improved quality of life [60]. De Wit and Schuurmans [58] also call for a multidisciplinary collaboration across the different sectors.
It has to be taken into account that internationally there are different standards in the education of the healthcare team. For example, nurses or therapists are educated academically or non-academically, depending on the country of their education.
Finally, it could be possible that the impact of the contact given by the different professionals of the studies evaluated, could vary mainly because of the nature of the professions included (Occupational Therapists, Pharmacists and a multidisciplinary team), as well as by the uni- or multidisciplinary aspect of the work team. This difference may have had an impact not only on the desired outcomes but operationally on the transition components that could be applied according to the professions carrying out the interventions and follow-up. Therefore, it could be suggested that the joint work of a multidisciplinary team would have a greater impact on patient contact.
Cp 3: maintaining relationships
This component was included in all three trials by caring for the patients and their family caregivers, both in the hospital and in the community dwelling. The healthcare professionals keep a relationship with patients and their caregivers through visits and telephone calls to prepare and accompany the patients during the implementation of the care plan and to meet their current and future needs. This approach is fairly consistent with the description of the TCM [31], in which maintaining relationships is a key feature. In addition, Le Berre and colleagues confirmed that this component leads to better adherence and disease control in geriatric patient, when the same person accompanies the patient in the transition from hospital to community dwelling [14]. Our findings support the importance of this component.
Cp 4: engaging patients and caregivers
This component was applied only in two of the three trials [28, 30]. These trials engaged the people of their intervention groups in different ways.
Rich and colleagues [30] used a limited approach of patient engagement. Medication reconciliation by a geriatric cardiologist and modifications to the medications were made when necessary. These activities were carried out in cooperation with the patient, who additionally was required to keep a weight chart. In contrast, the patients in the trial of Clemson and colleagues [28] were asked to set client-centered goals. Additionally, only the research group around Clemson [28] mentioned the aspect of caregivers’ engagement, which was carried out depending on the availability of the patient’s family member but the authors did not report on the impact of their intervention on caregiver engagement.
It should be noted that a limited approach to caregiver engagement may reduce the impact of this component in the transition process, considering that care after hospital discharge generates a difficult burden on families [24]. If caregivers support and engagement can be included significantly in this component, however, this could relieve care giver burden substantially.
This component was included by the two trials, only in the setting of the hospital. However, in the course of the transition from one setting to the other, adjustments to the care plan may become necessary. Therefore, it is likely to be important to adjust the care plan also in the home environment. Thus, at home the patients and their family caregivers should be engaged again. In addition, it is worth mentioning to include the individual values and preferences in the care plan [31].
De Wit and Schuurmans [58] strongly encourage engaging geriatric patients to look after their own health. Likewise, a recent systematic review calls for integration of patients as full members of the care team; i.e. the patients should not only be informed, but also be empowered to participate [61]. However, Shearer and colleagues [62] stated that the well-defined empowerment intervention strategies were limited as well as not linked to theoretical frameworks. Therefore, it is recommended that future studies - designed to improve patient empowerment - should be better linked with established theoretical frameworks. In addition, these interventions should also take into account factors influencing hospital readmission, such as the discharge from hospital to patients own home when the patients depend on the help of someone else at home [63, 64]. It could be interesting to inquire about the correlation of such factors to the patients’ empowerment aspect, for instance, if engagement and empowerment of a patient for his own care could be less, when he has a greater expectation of family care, in order to identify the best way to involve patients and caregivers in the care plan. Another aspect could also be the impact on the health of the informal care, issues related to the older caregivers, such as physical and cognitive health problems. In this regard, a recent review have shown that the health of the older informal caregivers is at risk [65].
Cp 5: assessing / managing risk and symptoms
This component was applied in all three trials. However, the implementation of this component was carried out differently in each study, based on the type of assessments and its goal. Clemson and colleagues [28] conducted measurements focused on a person’s functional ability to perform ADLs as well as on a person’s participation in life tasks and roles. They did not assess the symptoms of the disease nor did they evaluate other non-functional risks that may lead to the development of adverse events. Nonetheless, the assessment of the instrumental ADLs made it possible to draw conclusions indirectly about risk factors. López Cabezas and colleagues [29] and Rich and colleagues [30] applied this component more comprehensively, with the assessment of symptoms of the disease as well as risk factors for adverse events such as the side effects of the medication and the quality of life (QoL). It is noteworthy here that patients with a specific diagnosis (heart failure) were included in those two studies.
Especially in the geriatric population with the burden of multi-morbidity, it may be necessary to assess and manage the risk and symptoms individually. By looking at the domains of activity and participation, further undiscovered needs or dangers could be revealed. In general, Burke and colleagues [53] showed that this has been one of the components most associated with the reduction of readmission after discharge.
Cp 6: education/promoting self-management
It became evident, that only the two trials [29, 30] that showed a reduction of readmission rate, applied an educational component. Both trials applied this educational component at pre- as well as at post-discharge. Jones and colleagues also pointed out the importance of this component and its implementation in both settings [66]. The educational component in the included trials was characterized by information and guidance related to illness, diet and medication. Lopéz Cabezas and co-workers [29] oriented their program to the social and cultural level of each patient. Furthermore, Jones and colleagues suggested an individualized educational approach [66].
Rich and colleagues affirmed that their educational component focused on intensive teaching and contributed to achieve significant readmission reductions. Particularly, since the educational component allowed to reinforce patient’s knowledge in the follow up, to guarantee adherence with medications and to provide information to recognize and manage persistent symptoms. These findings are in line with the research of Koelling and colleagues [55] and Burke and colleagues [53] who observed a lower risk of hospital readmission in patients receiving an education intervention. Furthermore, these results are in agreement with Hirschman and collaborators who mentioned that the educational component is important to reduce readmission rates, since education and self-management promotion allows to monitor, identify, understand, and answer to symptoms avoiding their exacerbation and worsening of the chronic condition [31].
Cp 7: collaborating
This component was only included by Rich and colleagues [30]. It was applied at the hospital as well as in the community setting. The transition from hospital to community was developed collaboratively with a social worker and a member of the care team, facilitating a consensus on a plan of care. Thus, the collaborative work between multiple healthcare professionals who are not linked in the same network can provide a more complete approach of care [24]. As the WHO has called for person-centered and integrated care, integrating initiatives on service and organizational level seems mandatory to install the TCM component “collaboration”. This Integrated Care for Older People (ICOPE) approach of the WHO supports the collaboration components by integrating health and social care to improve the management of the geriatric persons [67].
Cp 8: promoting continuity
Promoting continuity could help to prevent breakdowns in care across settings [31]. Several approaches to promoting continuity were used in the three trials [28,29,30]. The same people as in the hospital were also responsible for the patients in the community dwelling. A person of the care team could be reached by phone in case of problems or doubts at designated times. The first contact (telephone or home visit), which was made by the responsible healthcare professional was in a time interval known to the patient [61].
Cp 9: fostering coordination
With respect to this component, Rich and colleagues [30] provided assistance in the case of emotional, social, but also economic or transportation problems during the discharge process. This points out, that not only an explicit standard for multidisciplinary communication is important, but also explicit standards for processes and systems are needed to ensure provider accountability, which would contribute to a successful transition [68]. Moreover, this component has been previously identified as being used frequently in interventions with an effect on the reduction of short term, intermediate term and long term readmissions [38]. In the present systematic review, the only trial that performed a long term intervention did not include this component and did not obtain a successful reduction of long term readmission [29]. Therefore, it is suggested that future studies should explore the effect of this component to the success of reducing long term readmissions.
In conclusion of the applied TCM components in our systematic review experiencing the “real” environment of geriatric patients will reveal barriers as well as supportive factors that can often not be detected from an inside-hospital view. This demonstrates the need of the community dwelling components of the TCM. Although not all components have been used by the three included trials, they addressed both care settings (hospital AND community) demonstrating the need of such an approach to reduce successfully the re-admission rate in a geriatric population.
Another interesting difference between the three trials is the nature of the TCM team. Looking at the history of the TCM, the focus is on a nurse-led intervention. However, in the included trials the configuration of the care team was not limited to this profession.
Regarding the multidisciplinary care team, the need for a team approach to improve the care of patients with chronic conditions has previously been emphasized [60]. In this case, only the trial carried out by Rich and colleagues [30] applied a multidisciplinary approach, the results of which were positive in reducing the readmission rates at the end of follow-up. These authors implemented more multidisciplinary care activities related to the management of heart failure, such as monitoring of symptoms and assessment of cardiovascular risk. The other two included trials [28, 29] performed the intervention using one health professional, and one of them [29] obtained successful results in the reduction of the readmission rate at two and six months of follow-up.
In conclusion, our systematic review demonstrated that the sum of the integrated components of the TCM is also responsible for a successful transition from the hospital to the community especially in geriatric patients. According to the results of this review, these factors are the intensity level and length of intervention, a multicomponent intervention approach and the specific role of each component, and the multidisciplinary nature of the care team. Future studies should focus on the optimal combination of these factors. In general, all of the nine defined TCM components by Hirschman and colleagues [31] were included across the three trials. Clemson and colleagues [28] and López Cabezas and colleagues [29] both included six of the nine components in their interventions, whereas Rich and colleagues [30] implemented all TCM components. In particular, in these studies more components were applied in the hospital settings than in the community dwelling, suggesting a potential imbalance in the inpatient setting support versus the in-home-follow-up support. However, it was not possible to establish from the three trials, whether this difference between how many components were used in hospital and at home may have had an effect on the successful transition to the community. Two trials affirmed that it was difficult to recognize which components were the most effective, since they administered a multifactorial intervention [29, 30]. According to the previous information, there is no clear evidence regarding which components were the most effective decreasing readmission rates. However, in light of the different results, the present systematic review attempts to evidence which components may have played a key role decreasing readmission rates.
Secondary outcomes
There were also different results in relation to the secondary outcomes of interest.
In case of the QoL, only one of the evaluated trials [30] found a significant increase in the used QoL score. Rich and colleagues [30, 46] used a more specific instrument for their study population, the Chronic Heart Failure Questionnaire, which has been widely validated in older people with heart failure, and has shown adequate sensitivity in detecting clinically important changes over time as well as adequate scores for interpretability [30]. On the other hand, although López Cabezas and colleagues used a validated Spanish version of the EuroQol questionnaire [29], this questionnaire is not a specific instrument to assess QoL in people with heart failure, nor is it a questionnaire with items easy to interpret by the older population, especially with low cultural level, as discussed by López Cabezas and colleagues in their study. Apparently, these findings may be influenced by the instruments applied to measure this construct as well as to the characteristics of the study population. Regarding cost savings, both trials, Rich and colleagues [30] and López Cabezas and colleagues [29], showed lower costs for intervention vs usual care, where specifically Rich and colleagues suggest a long-term cost saving due to the multicomponent approach used. This is in line with other authors, who have also identified an overall reduction in the healthcare system costs due to transitional care interventions [69].
Clemson and colleagues [28] did not find improvements in their primary outcomes such as ADLs and participation in life roles and activities. Two other studies in this field, not included in this review, however found statistically significant improvements in the used measurement regarding ADLs [70, 71]. Apart from the fact that Clemson and colleagues [28] could not find any significant results in their trial, the importance of this domain seems to be proven. Future studies, which will be engaged with the optimal implementation of transitional care, should consider the domain of ADLs and participation.
Finally, it is important to highlight the need to conduct trials focused in the geriatric population over 65 years, which allow a better identification of the TCM role according to the care needs of this population. The results previously reported in the literature on the effects of the TCM, vary greatly, especially due to the variety of populations evaluated. For instance, the systematic review carried out by Coffey and colleagues evaluated studies that applied transition care, each with a specific study population such as new mothers, infants and children, adolescents, older people, among others. They observed mixed findings, in which the results of some studies varied in relation to the cost effectiveness, outcomes as the number of hospitalizations as well as the quality of life [72]. Added to this, in the present systematic review we observed a low number of trials conducted in the geriatric population over 65 years, evidencing a limited evaluation of the TCM in this population.
Limitations and strength of this systematic review
In the present review, only three trials were included that met the precise inclusion criteria on which the present review was based. The still existing fragmentation in most public health care systems especially in geriatric patients’ needs growing realization to overcome this barrier. Our strict inclusion criteria with the special focus on geriatric patients could have excluded other studies with valuable information.
In addition, it was not possible to obtain a clear description of the control group conditions of each trial. As different care and discharge routine could have effects on the acceptance and implementation on the results of TCM implementation it was interesting to see that in all three studies the components of maintaining relationship and continuity was applied. One could hypothesize that in daily discharge and transitional routine being applied for the control group, especially these components of the TCM are not applied. It should be considered that these were carried out in countries with different health systems, where the standard hospital discharge procedure may vary.
Moreover, a gap related to the evaluation of the fidelity of interventions was evidenced, which did not allow this aspect to be addressed in the present review and could pose a bias on the results. Nevertheless, as we followed strictly the protocol with obtaining risk of bias we think that no fidelity information is needed with regard to our main objective. Even so, we consider the inclusion of fidelity assessment criteria in further trials to be relevant, especially in studies that evaluate transitional care interventions in geriatric populations (over 65 years old). Given that characteristics such as multi-morbidity, the application of interventions in multiple sites (Hospital and Home) and the complexity of these interventions (several components of the intervention, multidisciplinary team, among others) could limit the maintenance of the trials fidelity.
On the other hand, the strength of our review is the strict focus on the geriatric population, providing concrete information on the effects of multi-component interventions in reducing readmission in the geriatric population – individuals over 65 years of age with multi-morbidity. In view of the significant increase in this population in the upcoming years, effective and realistic approaches are needed to reduce the readmission rate of these highly vulnerable people. We therefore think that this systematic review will add valuable information not disease oriented but addressing a growing percentage of population putting the health care systems on the edge in the future. Furthermore, the health of the informal carer in this population needs to be taken into account as well. An additional strength of our study is – although we only included three trials – all trials had more than 50 participants in each trial arm, and were of good quality, strengthening our findings and providing a solid base for future research, and designing new transitional care intervention in the geriatric population.