Designing a Fall Prediction Model for Inpatient Rehabilitation Facilities Using Readily Available Data.

OBJECTIVE: To create a fall risk assessment tool for inpatient rehabilitation facilities (IRFs) using available data and compare its predictive accuracy with that of the Morse Fall Scale (MFS). DESIGN: We conducted a secondary analysis from a retrospective cohort study. Using a nomogram that displayed the contributions of QI codes associated with falls in a multivariable logistic regression model, we created a novel fall risk assessment, the Inpatient Rehabilitation Fall Scale (IRF Scale). To compare the predictive accuracy of the IRF Scale and MFS, we used receiver operator characteristic (ROC) curve analysis. SETTING: We included data from 4 IRFs owned and operated by Intermountain Health. PARTICIPANTS: Data came from the medical records of 1699 patients. All participants were over the age of 14 and were admitted and discharged from 1 of the 4 sites between January 1 and December 31, 2020. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURE(S): We assigned point values on the IRF Scale based on the adjusted associations of QI codes with falls. Using ROC curve analysis, we discovered an optimal cutoff score, sensitivity, specificity, and overall AUC of the IRF Scale and MFS. RESULTS: ROC curve analysis revealed the optimal IRF Scale cutoff score of 22.4 yielded a sensitivity of 0.74 and a specificity of 0.63. With an AUC of 0.72, the IRF Scale demonstrated acceptable accuracy at identifying patients who fell in our retrospective cohort. CONCLUSIONS: Because the IRF Scale uses readily available data, it minimizes staff assessment and outperforms the MFS at identifying patients who previously fell. Prospective research is needed to investigate if it can adequately identify in advance which patients will fall during their IRF stay.

Using Quality Indicator Codes to Identify Patients' Fall Risk in Inpatient Rehabilitation Facilities.

OBJECTIVE: To discover if quality indicator (QI) codes are associated with patient falls in inpatient rehabilitation facilities (IRFs). DESIGN: This retrospective cohort study explored differences between patients who fell and those who did not fall. We analyzed potential associations between QI codes and falls using univariable and multivariable logistic regression models. SETTING: We collected data from electronic medical records at 4 IRFs. PARTICIPANTS: In 2020, our 4 data collection sites admitted and discharged a total of 1742 patients older than 14 years . We only excluded patients (N=43) from statistical analysis if they were discharged before admission data had been assigned. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Using a data extraction report, we collected age, sex, race and ethnicity, diagnosis, falls, and QI codes for communication, self-care, and mobility performance. Staff documented communication codes on a 1-4 scale and self-care and mobility codes on a 1-6 scale, with higher codes representing greater independence. RESULTS: Ninety-seven patients (5.71%) fell in the 4 IRFs over a 12-month period. The group who fell had lower QI codes for communication, self-care, and mobility. When adjusting for bed mobility, transfer, and stair-climbing ability, low performance with understanding, walking 10 feet, and toileting were significantly associated with falls. Patients with admission QI codes below 4 for understanding had 78% higher odds of falling. If they were assigned admission QI codes below 3 for walking 10 feet or toileting, they had 2 times greater odds of falling. We did not find a significant association between falls and patients' diagnosis, age, sex, or race and ethnicity in our sample. CONCLUSIONS: Communication, self-care, and mobility QI codes appear to be significantly associated with falls. Future research should explore how to use these required codes to better identify patients likely to fall in IRFs.

Variation in Outcomes and Number of Visits Following Care Guideline Implementation: Part 2 of an Analysis of 12 355 Patients After Total Knee Arthroplasty.

OBJECTIVE: To describe the variation in outcomes and number of visits before and after implementing a care guideline for total knee arthroplasty (TKA) rehabilitation. DESIGN: Nonrandomized intervention study. METHODS: We compared 2558 patients with TKA who received care that was not standardized (non-care guideline [NCG] group) to 9797 patients with TKA who received care according to the care guideline (CG). We fit 2 Bayesian hierarchical linear regression models using the Knee Outcome Survey - Activities of Daily Living (KOS-ADL) change score and number of physical therapy (PT) visits as the response variables, controlling for relevant predictor variables. We also compared the ratio of the standard deviations of the KOS-ADL change scores and the number of PT visits within and between clinics. RESULTS: The overall estimated mean improvement in KOS-ADL change score was 23.0 points (95% confidence interval [CI]: 20.3, 25.7) in the NCG group and 28.7 points (95% CI: 27.5, 29.7) in the CG group; the mean difference was 5.6 (2.7-8.6). Mean KOS-ADL change scores were higher in the CG group than the NCG group in every clinic, although only 8 clinics improved significantly. The number of PT visits did not change meaningfully (NCG: mean, 10.7 [95% CI: 9.9, 11.5]; CG: mean, 10.5 [95% CI: 9.9, 10.9]). Variation in KOS-ADL change score decreased by 4% within clinics (CG-NCG ratio: 0.96 [95% CI: 0.93, 0.99]) and 63% between clinics (CG-NCG ratio: 0.37 [95% CI: 0.21, 0.62]). Variation in number of visits decreased by 7% within clinics (CG-NCG ratio: 0.93 [95% CI: 0.90, 0.96]) and 19% between clinics (CG-NCG ratio: 0.81 [95% CI: 0.39, 1.49]). CONCLUSION: Implementing a care guideline for TKA rehabilitation may improve outcomes and reduce unwarranted variation in practice within clinics and especially between clinics within a large health care system. J Orthop Sports Phys Ther 2023;53(3):151-158. Epub: 12 December 2022. doi:10.2519/jospt.2022.11370.

Improved Outcomes Following a Care Guideline Implementation: Part 1 of an Analysis of 12 355 Patients After Total Knee Arthroplasty.

OBJECTIVE: To describe the application and examine the influence of a continuous quality improvement intervention, which had a goal of standardizing care to reduce the proportion of patients who do not have a meaningful improvement in patient-reported outcomes following total knee arthroplasty (TKA). DESIGN: Continuous quality improvement. METHODS: A physical therapy (PT) care guideline was initiated in 2013 for patients following TKA. The Knee Outcome Survey - Activities of Daily Living (KOS-ADL) was measured at every visit, and scores were extracted from a clinical outcomes database to calculate the proportion of patients who did not achieve a minimal clinically important difference. Based on logistic regression analysis, we compared the proportion of patients who did not progress on the KOS-ADL in a non-care guideline group (2008-2012) to a care guideline (CG) group (2014-2019). RESULTS: This study included 12 355 patients (aged 18-92 years) following TKA incurring at least 3 PT visits from 2008 to 2019. The percentage of patients who did not progress in the non-care guideline group was 25.8% and in the care guideline group 14.3% (P<0.001). The relationship between care guideline adherence and lack of progression on the KOS-ADL was statistically significant, X2 (df = 1) = 148.7, P<.001. CONCLUSION: The percentage of patients who did not achieve meaningful progress on the KOS-ADL declined significantly in the 6 years after implementing a TKA care guideline without an increase in the number of clinical visits. The standardized care guideline was associated with meaningful improvements for patients following TKA when applied in conjunction with PT access to outcome data, feedback through audits, performance goals, and financial incentives. J Orthop Sports Phys Ther 2023;53(3):143-150. Epub: 12 December 2022. doi:10.2519/jospt.2022.11369.