Centers for Medicaid and Medicare Services

Centers for Medicaid and Medicare Services

A Critique: A Comprehensive Hospital-Based Intervention to Reduce Readmissions For Chronically Ill Patients: A Randomized Controlled Trial by Linden Ariel, and Butterworth Susan White.


Centers for Medicaid and Medicare Services (CMS) is adopting policies aimed at reducing the number of readmissions owing to the fact that they are avoidable yet costly. The approach adopted by CMS involves imposing penalties for hospitals that exceed their anticipated readmission rates within the 30-day period. As a result, hospitals have reacted to this policy by implementing transitional care interventions having the primary objective of lessening the readmissions within the one-month period; nevertheless, little empirical research has been devoted towards investigating the effectiveness of interventions used by hospitals to transition the patient from the hospital to the community. Moreover, the available empirical literature does not offer comprehensive guidance with respect to how effective transitional interventions can be designed in order to cut such hospital readmissions. Thus, Linden and Butterworth (2014) conducted an empirical research to determine whether a comprehensive hospital-based transitional care is effective in attempting to reduce hospital readmissions of patients suffering from congestive heart disease and chronic obstructive pulmonary.

The researchers hypothesized that patients who were subjected to the comprehensive hospital-based intervention would have significantly lower rates of readmission rates when compared to patients placed under the usual transition care. In addition, Linden and Butterworth hypothesized that the comprehensive hospital-based intervention would result in significantly lower rates of mortality as well as other secondary outcomes associated with the emergency department. Linden and Butterworth performed an empirical research study to evaluate the effectiveness of the intervention they had developed – a ninety-day hospital-based transitional care program. The findings of their research did not reveal any statistically significant differences between the rates of readmission for patients placed under the comprehensive hospital-based program and those who were not under the program. Furthermore, the findings of their research did not report ant statistically significant differences in terms of the visits to the emergency department after one month and three months. The researchers also found that that the mortality rates were the same for the patients placed under the program and those under the usual care transition. Basing on the obtained results, the researchers reached a conclusion that standalone community hospitals cannot reduce the rates of readmissions even in case of applying comprehensive, evidence-based intervention components that they can manage to control. As a result, Linden and Butterworth emphasized the need for cooperation between community-based providers and hospitals in order to ensure continued care for patients who have been discharged from the hospital.

Evaluation of the Research Methods

The authors provided a short review of the literature. In their literature review, Linden and Butterworth (2014) argue that, although hospitals have developed interventions aimed at reducing the rate of 30-day readmissions, available empirical evidence does not outline a comprehensive guidance that hospitals can use to design effective transitional care interventions for discharged patients. It is evident that Linden and Butterworth are precise and succinct as regards their argument in the literature review; nevertheless, they did not perform a wide-ranging literature review on the issue under consideration. In the literature review, the authors only mention one meta-analytic research undertaken by Hansen, Hinami, Leung and Williams (2011) that analyzed 43 research studies adopted in hospitals to cut the rates of readmissions. Based on this meta-analytic research, out of the 43 interventions, only 16 were found to be effective. Moreover, the meta-analysis pointed no consistent relationship between these interventions and effectiveness. Another crucial aspect that this meta-analysis revealed is that majority of the studies included in the analysis were performed in academic settings; as a result, there are concerns regarding the practical applications of these interventions in hospital settings. The researchers provided a rationale of their research on grounds that there is a model of transitional care that is effective and generalizable, which highlights the need to perform unending experimentations and continually revise existing interventions to cut hospital readmissions. A stronger argument could have been developed in the literature review if the authors could have performed an extensive review of the literature before reaching the conclusion that existing interventions are not effective.

The problem tackled in the study is both relevant and current in the context of healthcare. There is widespread agreement that hospital readmission is a significant problem of concern for numerous hospitals in the US and abroad (Axon & Williams, 2011; Wong, et al., 2008). In the US, hospitals risk penalties in their Medicare reimbursements when they document higher readmission rates that exceed the expected rates for particular diagnoses (Axon & Williams, 2011). Even though some unplanned readmissions are not related to previous hospital visits, some readmissions are unpreventable. Irrespective of the reason for hospital readmissions, healthcare insurance providers consider unplanned hospital readmissions as a wastage of resources and something that ought to be prevented (Axon & Williams, 2011). The problem of hospital readmissions is not only an issue of healthcare costs, but is also associated with the quality of healthcare services offered at hospitals, which explains the penalties sanctioned by CMS on hospitals exceeding their expected rates of readmissions. Essentially, higher rates of hospital readmissions suggest a poor quality of healthcare services administered at hospitals (Axon & Williams, 2011; Wong, et al., 2008). As a result, hospitals face the challenge of developing an effective transitional care program that seeks to make sure that discharged patients continue getting high-quality healthcare services and reduce the likelihood of their readmissions. Therefore, the issue that Linden and Butterworth (2014) studied is timely as well as relevant in present-day healthcare.

The research design used by Linden and Butterworth (2014) is a parallel group, stratified randomized control study. This research design fits the nature of the research that the researchers undertook. RCTs serve the main purpose of evaluating the effectiveness of an intervention while comparing its effect on participants who received the intervention against those placed under usual care or received an alternative intervention (Wong, et al., 2008). The randomization process used by the authors encompassed random stratified sampling that had four strata comprising of two disease conditions and two hospitals. There is no doubt that the RCT design adopted by Linden and Butterworth is suitable considering the objectives of the research, which involved determining the effectiveness of comprehensive hospital-based interventions in significantly reducing the rate of hospital readmissions.

The population in Linden and Butterworth’s study comprises of people suffering from obstructive pulmonary disease and congestive heart failure in nine counties found in northern part of California and the southern part of Oregon. The setting of the research was two non-profit hospitals situated in Grand Pass and Medford, Oregon. The study raises issues associated with the representatives of the sample owing to the fact that the study was performed only in two hospitals. Even though the setting of the research was appropriate when considering the nature and the purpose of the research, Linden and Butterworth could have looked into the possibility of performing the research in numerous hospitals scattered in various locations across the various countries found in northern California and southern Oregon. The sample size in the study was 512 patients who were divided into two groups randomly with the usual care having 259 patients whereas the comprehensive hospital-based intervention (treatment group) had 253 patients. When using previous studies as benchmark for assessing the appropriateness of the sample size (Axon & Williams, 2011; Daly, Douglas, Kelley, O’Toole, & Montenegro, 2005; Wong, et al., 2008), it is evident that the size of the sample used by Linden and Butterworth (2014) is sufficiently large and produced robust findings. Essentially, the authors used a relatively large sample size that was utilized in past studies. Moreover, the authors provided a rationale for their sample size by computing the needed sample size and utilizing the over-dispersed observed Poisson observations for determining the sample size required for each disease condition in order to produce the sufficient sample power needed in detecting a true difference between the two groups. Overall, the sample used in the study by Linden and Butterworth was satisfactory with respect to size; however, there still exist issues that are associated with the representativeness of the sample. It can be explained by the fact that participants were drawn from two hospitals only. The researchers tried to compensate for this problem by making use of random sampling in order to address problems associated with bias in the sample, which also contributed to the validity of the results presented.

The study by Linden and Butterworth seems practical, which can be attributed to the fact that the researchers outlined a detailed explanation of the research methods they used as well as the intervention adopted in the research. As a result, it is possible to replicate the research in other settings. In addition, the methods used in collecting data are described in detail. However, the results of the study cannot be applied since the researchers did not affirm the effectiveness of the comprehensive hospital-based intervention. In other words, the researchers did not report statistically significant differences between patients placed under the comprehensive hospital-based intervention and those placed under usual transition program. However, some aspects of the study can be applied. The one aspect discovered by the study that can be applied in healthcare relates to the finding that devising interventions that place emphasis on patient self-management are capable of lessening the mortality rates of patients having chronic obstructive pulmonary disease. Linden and Butterworth also commended the need for enhanced collaboration between hospitals and community-based care providers in order to guarantee continued care even after hospitals discharge patients.

One potential improvement that could have been made to the research study conducted by Linden and Butterworth (2014) involves the use of at least two hospital locations in order to have a more representative sample. It is evident that the article is written in a straightforward and clear manner since the findings of the research are presented in accordance with the outcome variables. Linden and Butterworth also presented their findings with respect to the purpose of their research. Moreover, they positioned their findings in the context of past research studies. Although the authors justified their study based on the fact that there is no effective transitional care intervention, the intervention that they developed and tested was not effective either. The only research that can be made better is through revising their intervention and subsequently experimenting the intervention and iteratively revising it to a point where an effective intervention can be discovered. Another important area emphasized by the authors that require additional research relates is the collaboration between community-based care providers and hospitals. Further research is needed to determine the specific aspects that require the collaboration such as referrals as well as feedback and update between these providers.


The authors of the research investigated a relevant and current issue in healthcare. Some strengths of the study are evident, including presenting a precise and succinct argument in the review of literature, using an appropriate study design with respect to the purpose of the research and a sufficient size of the sample that yielded robust findings. The study also draws its strengths from clarity and practicality. However, some weaknesses are also evident including conducting an incomprehensive literature review and problems with sample representativeness. An overall assessment of the study is that, regardless these limitations, the results of the study contribute to the literature on nursing.