A position paper on how cost and quality reforms are changing healthcare in America: focus on nutrition

Assessing calibration and bias of a deployed machine learning malnutrition prediction model within a large healthcare system

Malnutrition is a frequently underdiagnosed condition leading to increased morbidity, mortality, and healthcare costs. The Mount Sinai Health System (MSHS) deployed a machine learning model (MUST-Plus) to detect malnutrition upon hospital admission. However, in diverse patient groups, a poorly calibrated model may lead to misdiagnosis, exacerbating health care disparities. We explored the model's calibration across different variables and methods to improve calibration. Data from adult patients admitted to five MSHS hospitals from January 1, 2021 - December 31, 2022, were analyzed. We compared MUST-Plus prediction to the registered dietitian's formal assessment. Hierarchical calibration was assessed and compared between the recalibration sample (N = 49,562) of patients admitted between January 1, 2021 - December 31, 2022, and the hold-out sample (N = 17,278) of patients admitted between January 1, 2023 - September 30, 2023. Statistical differences in calibration metrics were tested using bootstrapping with replacement. Before recalibration, the overall model calibration intercept was -1.17 (95% CI: -1.20, -1.14), slope was 1.37 (95% CI: 1.34, 1.40), and Brier score was 0.26 (95% CI: 0.25, 0.26). Both weak and moderate measures of calibration were significantly different between White and Black patients and between male and female patients. Logistic recalibration significantly improved calibration of the model across race and gender in the hold-out sample. The original MUST-Plus model showed significant differences in calibration between White vs. Black patients. It also overestimated malnutrition in females compared to males. Logistic recalibration effectively reduced miscalibration across all patient subgroups. Continual monitoring and timely recalibration can improve model accuracy.

Supplemental Nutrition Assistance Program participation and health care expenditures in children

BACKGROUND

The Supplemental Nutrition Assistance Program (SNAP) has well-established positive impacts on child health outcomes, including increased birth weight and decreased likelihood of underweight status. Studies in adult populations suggest that SNAP is associated with lower health care costs, although less is known in children.

METHODS

Retrospective analysis of U.S. children (age <18 years) living in low-income households (< 200% of the federal poverty level) in the 2013-2017 Medical Expenditure Panel Survey. We used multivariable regression, adjusting for sociodemographic and clinical covariates, to model the effect of continuous SNAP enrollment on health expenditures as compared to non-enrollees at 12 and 24 months.

RESULTS

The sample included 5,626 children, of whom 49.2% consistently received SNAP for the entire two-year survey period. Compared with SNAP non-recipients, SNAP-recipient households more often had incomes below 100% FPL (78.3% vs 37.9%), and children in SNAP-recipient households were more often publicly insured (94.9% vs 64.5%). Unadjusted expenditures were lower for children in SNAP-recipient households at 12 ($1222 vs $1603) and 24 months ($2447 vs $3009). However, when adjusting for sociodemographic and clinical differences, no statistically significant differences in health care expenditures, including emergency department, inpatient, outpatient, and prescription costs, were identified.

CONCLUSION

SNAP participant children experience heightened social hardships across multiple domains. There were no differences in short term health care costs based on SNAP enrollment when accounting for differences in sociodemographic and clinical factors. Despite demonstrated child health benefits, we found that sustained enrollment in SNAP over a two-year period did not generate significant short- term health care cost reductions. Our findings suggest that although SNAP is intended to act as a benefit towards the health and well-being of its recipients, unlike among adults, it may not reduce health care costs among children.

Bundled Payment Models in Spine Surgery

STUDY DESIGN

The following is a narrative discussion of bundled payments in spine surgery.

OBJECTIVE

The cost of healthcare in the United States has continued to increase. To lower the cost of healthcare, reimbursement models are being investigated as potential cost saving interventions by driving incentives and quality improvement in fields such a spine surgery.

METHODS

Narrative overview of literature pertaining to bundled payments in spine surgery synthesizing findings from computerized databases and authoritative texts.

RESULTS

Spine surgery is challenging to define payment modes because of high cost variability and surgical decision-making nuances. While implementing bundled care payments in spine surgery, it is important to understand concepts such as value-based purchasing, episodes of care, prospective versus retrospective payment models, one versus two-sided risk, risk adjustment, and outlier protection. Strategies for implementation underscore the importance of risk stratification and modeling, adoption of evidence based clinical pathways, and data collection and dissemination. While bundled care models have been successfully implemented, challenges facing institutions adopting bundled care payment models include financial stressors during adoption of the model, distribution of risks, incentivization of treating only low risk patients, and nuanced variation in procedures leading to variation in costs.

CONCLUSION

An alternative for fee for service payments, bundled care payments may lead to higher cost savings and surgeon accountability in a patient's care.

Trends and Novel Research in Hospital Nutrition Care: A Narrative Review of Leading Clinical Nutrition Journals

Hospital malnutrition is a longstanding problem that continues to be underrecognized and undertreated. The aim of this narrative review is to summarize novel, solution-focused, recent research or commentary to update providers on the prevention of iatrogenic malnutrition as well as the detection and treatment of hospital malnutrition. A narrative review was completed using the top 11 clinically relevant nutrition journals. Of the 13,850 articles and editorials published in these journals between 2013 and 2019, 511 were related to hospital malnutrition. A duplicate review was used to select (n = 108) and extract key findings from articles and editorials. Key criteria for selection were population of interest (adult hospital patients, no specific diagnostic group), solution-focused, and novel perspectives. Articles were categorized (6 classified in >1 category) as Screening and Assessment (n = 17), Standard (n = 25), Advanced (n = 12) and Specialized Nutrition Care (n = 8), Transitions (n = 15), Multicomponent (n = 21), Education and Empowerment (n = 9), Economic Impact (n = 3), and Guidelines (n = 4) for summarizing. Research advances in screening implementation, standard nutrition care, transitions, and multicomponent interventions provide new strategies to consider for malnutrition prevention (iatrogenic), detection, and care. However, several areas requiring further research were identified. Specifically, larger and more rigorous studies that examine health outcomes and economic analyses are urgently needed.

MUST-Plus: A Machine Learning Classifier That Improves Malnutrition Screening in Acute Care Facilities

OBJECTIVE

Malnutrition among hospital patients, a frequent, yet under-diagnosed problem is associated with adverse impact on patient outcome and health care costs. Development of highly accurate malnutrition screening tools is, therefore, essential for its timely detection, for providing nutritional care, and for addressing the concerns related to the suboptimal predictive value of the conventional screening tools, such as the Malnutrition Universal Screening Tool (MUST). We aimed to develop a machine learning (ML) based classifier (MUST-Plus) for more accurate prediction of malnutrition.

METHOD

A retrospective cohort with inpatient data consisting of anthropometric, lab biochemistry, clinical data, and demographics from adult (≥ 18 years) admissions at a large tertiary health care system between January 2017 and July 2018 was used. The registered dietitian (RD) nutritional assessments were used as the gold standard outcome label. The cohort was randomly split (70:30) into training and test sets. A random forest model was trained using 10-fold cross-validation on training set, and its predictive performance on test set was compared to MUST.

RESULTS

In all, 13.3% of admissions were associated with malnutrition in the test cohort. MUST-Plus provided 73.07% (95% confidence interval [CI]: 69.61%-76.33%) sensitivity, 76.89% (95% CI: 75.64%-78.11%) specificity, and 83.5% (95% CI: 82.0%-85.0%) area under the receiver operating curve (AUC). Compared to classic MUST, MUST-Plus demonstrated 30% higher sensitivity, 6% higher specificity, and 17% increased AUC.

CONCLUSIONS

ML-based MUST-Plus provided superior performance in identifying malnutrition compared to the classic MUST. The tool can be used for improving the operational efficiency of RDs by timely referrals of high-risk patients.

Immunometabolism: new insights and lessons from antigen-directed cellular immune responses

Modulation of immune responses by nutrients is an important area of study in cellular biology and clinical sciences in the context of cancer therapies and anti-pathogen-directed immune responses in health and disease. We review metabolic pathways that influence immune cell function and cellular persistence in chronic infections. We also highlight the role of nutrients in altering the tissue microenvironment with lessons from the tumor microenvironment that shapes the quality and quantity of cellular immune responses. Multiple layers of biological networks, including the nature of nutritional supplements, the genetic background, previous exposures, and gut microbiota status have impact on cellular performance and immune competence against molecularly defined targets. We discuss how immune metabolism determines the differentiation pathway of antigen-specific immune cells and how these insights can be explored to devise better strategies to strengthen anti-pathogen-directed immune responses, while curbing unwanted, non-productive inflammation.

EBP 2.0: Implementing and Sustaining Change: The Malnutrition Readmission Prevention Protocol

This is the fifth article in a new series about evidence-based practice (EBP) that builds on AJN's award-winning previous series-Evidence-Based Practice, Step by Step-published between 2009 and 2011 (to access the series, go to http://links.lww.com/AJN/A133). This follow-up series features exemplars illustrating the various strategies that can be used to implement EBP changes-one of the most challenging steps in the EBP process.

Identifying malnutrition: From acute care to discharge and beyond

Nutrition assessment and intervention significantly contribute to the well-being of patients. NPs should advocate that patients be appropriately evaluated and implement recommendations as part of a comprehensive care plan to avoid malnutrition in patients while they are in the hospital and when they return home.

Combating hospital malnutrition: Dietitian-led quality improvement initiative

BACKGROUND & AIM

Malnutrition is serious but under-diagnosed problem among hospitalized patients as approximately one-third patients may become malnourished during their stay. We audited our clinical nutrition practice (CNP) to understand the gaps and planned quality improvement initiatives to strengthen CNP through early delivery of oral nutrition supplement (ONS) prescribed.

METHOD

A prospective study was conducted in a tertiary care hospital in three phases between Dec'15 and Feb'17. Phase-1(P1): To assess the prevalence of malnutrition on admission and CNP; Phase-2(P2): Sensitization of Clinical Dietitians' to improve the nutrition care process; Phase-3(P3): Coalition of Multidisciplinary Team (MDT) to improve the timely delivery of ONS. Patients' demographics, baseline nutritional status (Subjective Global Assessment (SGA)), details of nutrition intervention and outcomes were recorded. Data were analyzed using SPSS version 20.0.

RESULTS

Out of 1443 patients (P1-500; P2-428; P3-515) enrolled, 63% were men and 37% women with mean age of 53.6 ± 15.49y. A significant improvement in the nutritional diagnosis as per SGA (0.031) and increasing trend in ONS prescription (P1:10(2%); P2:56(13%); P3:83(16%) (p = 0.000) was evident, which was attributed to the sensitization of Dietitians. Early delivery of ONS within 6 h from the time of prescription (EDONS) was progressively higher in P3 (P1:10%; P2:16.1%; P3:44.6%) and notable reversal of delayed delivery (>24 h) (P1:40%, P2:16%, P3:8.4%) (p = 0.000). Remarkable decline in the Length of Stay (LOS) was shown between the phases (P1:6.82 ± 10.3d; P2:5.68 ± 4.07d; P3:5.60 ± 4.74d) (p = 0.010). Vegetarians (6.54d) had a longer stay compared with non-vegetarians (5.88d) (p = 0.036). Lower BMI correlated negatively with LOS (p = 0.026). Univariant analysis showed a significant increase in LOS with admission history on weight loss (No weight loss = 5.52d, <10% loss = 7.74d, 10-15% = 7.51d, >15% = 8.32 d) (p = 0.001), severe malnutrition (12.67d) and EDONS in shortening the LOS (<6 h = 6.11, 6-12 h = 7.97, 12-24 h = 6.83, >24 h = 17.3 days) (p = 0.000). Multiple Linear regression analysis using ANOVA indicated that EDONS was the only major contributing factor in lowering the LOS (p = 0.002). Re-admission was higher in non ONS group (72%: non ONS; 26%: ONS group) (p = 0.000).

CONCLUSION

Dietitian-Led multi-professional inter-disciplinary Coalition, bridging the knowledge gap among MDT with regular sensitization about the importance of hospital malnutrition and EDONS, has led to the early identification and intervention of nutrition risk which enhanced progressive decrease in LOS and healthcare cost.

Addressing Disease-Related Malnutrition in Healthcare: A Latin American Perspective

Alarmingly high rates of disease-related malnutrition have persisted in hospitals of both emerging and industrialized nations over the past 2 decades, despite marked advances in medical care over this same interval. In Latin American hospitals, the numbers are particularly striking; disease-related malnutrition has been reported in nearly 50% of adult patients in Argentina, Brazil, Chile, Costa Rica, Cuba, Dominican Republic, Ecuador, Mexico, Panama, Paraguay, Peru, Puerto Rico, Venezuela, and Uruguay. The tolls of disease-related malnutrition are high in both human and financial terms-increased infectious complications, higher incidence of pressure ulcers, longer hospital stays, more frequent readmissions, greater costs of care, and increased risk of death. In an effort to draw attention to malnutrition in Latin American healthcare, a feedM.E. Latin American Study Group was formed to extend the reach and support the educational efforts of the feedM.E. Global Study Group. In this article, the feedM.E. Latin American Study Group shows that malnutrition incurs excessive costs to the healthcare systems, and the study group also presents evidence of how appropriate nutrition care can improve patients' clinical outcomes and lower healthcare costs. To achieve the benefits of nutrition for health throughout Latin America, the article presents feedM.E.'s simple and effective Nutrition Care Pathway in English and Spanish as a way to facilitate its use.

Analysis of Nutrition Education in Osteopathic Medical Schools

Purpose. Describe nutrition education at US colleges of osteopathic medicine; determine if it meets recommended levels. Method. We surveyed 30 US colleges of osteopathic medicine (US COM) with a four-year curriculum about the amount and form of required nutrition education during the 2012/13 academic year. The online survey asked about hours of required nutrition across all 4 years and also in what types of courses this instruction occurred. We performed descriptive statistics to analyze the data. Results. Twenty-six institutions (87% response rate) completed the survey. Most responding US COM (22/26, 85%) do not meet the recommended minimum 25 hours of nutrition education; 8 (31%) provide less than half as much. Required nutrition instruction is largely confined to preclinical courses, with an average of 15.7 hours. Only 7 of the 26 responding schools report teaching clinical nutrition practice, providing on average 4.1 hours. Conclusions. Most US COM are inadequately preparing osteopathic physicians for the challenges they will face in practice addressing the nutritional concerns of their patients. Doctors of osteopathy cannot be expected to properly treat patients or guide the prevention of cardiovascular disease, obesity, cancer, diabetes, and metabolic syndrome if they are not trained to identify and modify the contributing lifestyle factors.

Individualised dietary counselling for nutritionally at-risk older patients following discharge from acute hospital to home: a systematic review and meta-analysis

BACKGROUND

Many older patients are undernourished after hospitalisation. Undernutrition impacts negatively on physical function and the ability of older patients to perform activities of daily living at home after discharge from acute hospital. The present study aimed to evaluate the evidence for an effect of individualised dietary counselling following discharge from acute hospital to home on physical function, and, second, on readmissions, mortality, nutritional status, nutritional intake and quality of life (QoL), in nutritionally at-risk older patients.

METHODS

A systematic review of randomised controlled trials was conducted. The overall quality of the evidence was assessed according to Grading of Recommendations Assessment, Development and Evaluation system (GRADE) criteria.

RESULTS

Four randomised controlled trials (n = 729) were included. Overall, the evidence was of moderate quality. Dietitians provided counselling in all studies. Meta-analyses showed a significant increase in energy intake [mean difference (MD) = 1.10 MJ day(-1), 95% confidence interval (CI) = 0.66-1.54, P < 0.001], protein intake (MD = 10.13 g day(-1), 95% CI = 5.14-15.13, P < 0.001) and body weight (BW) (MD = 1.01 kg, 95% CI = 0.08-1.95, P = 0.03). Meta-analyses revealed no significant effect on physical function assessed using hand grip strength, and similarly on mortality. Narrative summation of effects on physical function using other instruments revealed inconsistent effects. Meta-analyses were not conducted on QoL and readmissions as a result of a lack of data.

CONCLUSIONS

Individualised dietary counselling by dietitians following discharge from acute hospital to home improved BW, as well as energy and protein intake, in older nutritionally at-risk patients, although without clearly improving physical function. The effect of this strategy on physical function and other relevant clinical outcomes warrants further investigation.

The need to advance nutrition education in the training of health care professionals and recommended research to evaluate implementation and effectiveness

Nutrition is a recognized determinant in 3 (ie, diseases of the heart, malignant neoplasms, cerebrovascular diseases) of the top 4 leading causes of death in the United States. However, many health care providers are not adequately trained to address lifestyle recommendations that include nutrition and physical activity behaviors in a manner that could mitigate disease development or progression. This contributes to a compelling need to markedly improve nutrition education for health care professionals and to establish curricular standards and requisite nutrition and physical activity competencies in the education, training, and continuing education for health care professionals. This article reports the present status of nutrition and physical activity education for health care professionals, evaluates the current pedagogic models, and underscores the urgent need to realign and synergize these models to reflect evidence-based and outcomes-focused education.