Summary
The probability of readmission of patients with hospital-acquired pressure injuries (PI) is believed to be high. Since specific features of patients readmitted with PI are known, the predictive model can be developed to determine the probability of these individuals returning to a healthcare organization within one month after discharge. This model will be developed using retrospective data from hospitals and binomial distribution to predict if specific patients with PI have a chance of readmission to prevent such situations in the future. The incoming patients will be classified as having a new condition or being admitted for a PI. These events will be used in calculating the binomial distribution model. The parameters that will be used to screen individuals who developed PI during a hospital stay are age and diabetes. Knowing the statistics and potential dangers for a specific patient population can help prevent readmissions through proper wound care and individual home visits by clinicians. Moreover, Lean and Six Sigma principles will be implemented to remove unnecessary data. Lastly, the predictive model will be tested in hospitals where data for the project is obtained.
Introduction
Pressure injury (PI) is one of the most prevalent complications of extended hospitalization. PI is localized skin and tissue damage caused by prolonged immobility, which causes shear stress on a particular body area, reducing blood flow to muscles and skin (McGee et al., 2019; Stewart et al., 2022). Since specific features of patients readmitted with PI are known, the predictive model can be developed to determine the probability of these individuals returning to a healthcare organization within one month after discharge. This model will be developed using retrospective data from hospitals and binomial distribution to predict if specific patients with PI have a chance of readmission to prevent such situations in the future.
Background of the Issue
Patients are often readmitted to healthcare facilities because of the worsening of PIs. Indeed, the readmission rate to hospitals is 30% higher in those with pressure wounds than without PI (Park et al., 2019). PI complications are often caused by such bacterial infections as Staphylococcus aureus and Escherichia coli (Singh et al., 2015). One of the essential predictors of 30-day readmission in patients with pressure wounds is frequent hospital stays and comorbidities like diabetes, AIDS, and cardiovascular diseases (Table A1). More than 2.5 million hospitalized individuals in the United States develop PIs, and 60,000 of them die of complications (Padula et al., 2019). Thus, there is a high demand for preventive strategies to reduce the incidence of PI and readmission rates.
Preventive Strategies
Since the problem of PIs has existed for a long time, various preventive strategies have been developed. For example, when patients stay longer than three days in healthcare facilities, clinicians try to monitor them closely, cleaning their skin, repositioning, and balancing their nutrition (Citty et al., 2019). Indeed, malnutrition due to metabolic diseases like diabetes is the most significant contributor to the development of PIs and the poor healing of these wounds. Therefore, prophylaxis among patients at risk can improve health outcomes.
Methodology
The incoming patients can be classified as having a new condition or being admitted for a PI. These events will be used in calculating binomial distribution, which allows to “give the probability for the occurrence of any event that can have two outcomes” (Kros & Rosenthal, 2016, p. 168). The data for statistical analysis will be obtained from the electronic health records of two local hospitals. The two criteria will be used to screen individuals who developed PI during a hospital stay: age and diabetes status. Moreover, Lean and Six Sigma principles will be implemented to remove unnecessary data to improve the model.
Implementation of the Model
The predictive model developed during this project can be applied to in hospital settings to estimate if a patient is at risk of PI and start preventive strategies. The local hospitals, where the patient data from electronic health records are obtained, will receive an offer to implement this tool to the admitted patients. The model’s performance will be evaluated monthly until the end of the year to estimate its usefulness in identifying patients at risk.
Recommendations
In this project, three recommendations for improving patient outcomes were developed. Firstly, it is essential to examine the skin of hospitalized diabetic patients during the first 24 hours of admission (Stewart et al., 2022). Secondly, elderly and malnourished patients should receive additional nutritional support and frequent repositioning (Citty et al., 2019). Thirdly, Lean and Six Sigma principles suggest that, in this case, waiting for physicians’ orders should be minimized, and nurses need to implement the predictive model to identify patients at risk and take preventive measures.
Conclusion
Healthcare-associated pressure injuries frequently occur in immobilized patients with comorbidities who have a high probability of returning to a healthcare institution. This project will develop the predictive readmission model for these individuals using retrospective hospital data and binomial distribution to build strategies to prevent PI and readmission in the population at risk. The model may be implemented in hospitals to estimate hospitalized individuals at risk of PI and introduce preventive strategies.
References
Chandra, A., Rahman, P. A., Sneve, A., McCoy, R. G., Thorsteinsdottir, B., Chaudhry, R., … & Takahashi, P. Y. (2019). Risk of 30-day Hospital Readmission among Patients Discharged to Skilled Nursing Facilities: Development and Validation of a Risk-Prediction Model. Journal of the American Medical Directors Association, 20(4), 444-450. Web.
Citty, S. W., Cowan, L. J., Wingfield, Z., & Stechmiller, J. (2019). Optimizing Nutrition Care for Pressure Injuries in Hospitalized Patients. Advances in Wound Care, 8(7), 309-322. Web.
Kros, J. F., & Rosenthal, D. A. (2016). Statistics for health care management and administration: Working with Excel (3rd ed.). John Wiley & Sons, Inc.
McGee, W. T., Nathanson, B. H., Lederman, E., & Higgins, T. L. (2019). Pressure Injuries at Intensive Care Unit Admission as a Prognostic Indicator of Patient Outcomes. Critical Care Nurse, 39(3), 44–50. Web.
Padula, W. V., Chen, Y. H., & Santamaria, N. (2019). Five‐Layer Border Dressings as Part of a Quality Improvement Bundle to Prevent Pressure Injuries in US Skilled Nursing Facilities and Australian Nursing Homes: A Cost‐Effectiveness Analysis. International Wound Journal, 16(6), 1263-1272. Web.
Park, S. K., Park, H. A., & Hwang, H. (2019). Development and Comparison of Predictive Models for Pressure Injuries in Surgical Patients: A Retrospective Case-Control Study. Journal of Wound Ostomy & Continence Nursing, 46(4), 291-297. Web.
Singh, R., Dhayal, R. K., Sehgal, P. K., & Rohilla, R. K. (2015). To Evaluate Antimicrobial Properties of Platelet-Rich Plasma and Source of Colonization in Pressure Ulcers in Spinal Injury Patients. Ulcers, 2015, 1-7. Web.
Stewart, T. P., Black, J. M., & Alderden, J. (2022). The Past, Present, and Future of Deep-Tissue (Pressure) Injury. Advances in Skin and Wound Care, 78–80.
Appendix A
Table A1: Hospital Readmissions for PIs in Patients with Comorbidities (Chandra et al., 2019)
