Rationing Limited Healthcare Resources in the COVID-19 Era and Beyond: Ethical Considerations Regarding Older Adults

Coronavirus disease 2019 (COVID-19) continues to impact older adults disproportionately with respect to serious consequences ranging from severe illness and hospitalization to increased mortality risk. Concurrently, concerns about potential shortages of healthcare professionals and health supplies to address these issues have focused attention on how these resources are ultimately allocated and used. Some strategies, for example, misguidedly use age as an arbitrary criterion that disfavors older adults in resource allocation decisions. This is a companion article to the American Geriatrics Society (AGS) position statement, “Resource Allocation Strategies and Age-Related Considerations in the COVID-19 Era and Beyond.” It is intended to inform stakeholders including hospitals, health systems, and policymakers about ethical considerations that should be considered when developing strategies for allocation of scarce resources during an emergency involving older adults. This review presents the legal and ethical background for the position statement and discusses these issues that informed the development of the AGS positions: (1) age as a determining factor, (2) age as a tiebreaker, (3) criteria with a differential impact on older adults, (4) individual choices and advance directives, (5) racial/ethnic disparities and resource allocation, and (6) scoring systems and their impact on older adults. It also considers the role of advance directives as expressions of individual preferences in pandemics. J Am Geriatr Soc 68:1143–1149, 2020.     

Radiological findings correlate with neurological deficits but not with pain after operatively treated sacral fractures: An 11-year follow-up study of 28 patients

Background and purpose

Neurological deficits and pain are common after displaced sacral fractures. However, little is known about the association between the long-term clinical outcomes and radiological findings. We examined the long-term radiological findings and their correlations with lumbosacral pain and neurological deficits in the lower extremities after surgery for sacral fractures.

Methods

28 consecutive patients with operatively treated displaced sacral fractures were followed for mean 11 (8–13) years. Sensorimotor impairments of the lower extremities were classified according to the American Spinal Injury Association (ASIA). Pain was assessed using a visual analog scale (VAS). All patients underwent conventional radiographic examination and CT, and the images were scrutinized for nonunion, residual displacement, narrowing of the sacral foramina, and post-foraminal encroachment of the L5 and S1 nerves.

Results

There was residual displacement of ≥ 10 mm in 16 of the 28 patients. 26 patients had narrowing of 1 or more neural root foramina in L5-S4. 8 patients reported having no pain, 11 had pain only in the lumbosacral area, and 9 had pain in combination with radiating leg pain. Statistically significant correlations were found between narrowing of the sacral foramina and neurological deficits in the corresponding dermatomes. Significant correlations were also found between post-foraminal encroachment of L5 nerves and both sensory and motor deficits. No correlations were found between pain and radiological findings.

Interpretation

Pathological radiological findings are common 11 years after operatively treated displaced sacral fractures. Sacral foraminal and L5 post-foraminal bony encroachments were common findings and correlated with neurological deficits. However, lumbosacral pain did not correlate with radiological sequelae after fracture healing.High-energy trauma with displaced sacral fracture is frequently associated with concomitant injuries to the intrapelvic soft tissue structures, including the lumbosacral plexus (Huittinen 1972, Denis et al. 1988, Majeed 1992). These injuries may cause considerable morbidity (Pohlemann et al. 1994, Tornetta and Matta 1996, Tötterman et al. 2006). However, little is known about which factors determine long-term clinical outcome in these patients, or what may explain the progression of neurological symptoms observed in a small proportion of patients (Adelved et al. 2012). Pelvic malunions and nonunions have been put forward as prognostic factors for impaired long-term outcome (Matta et al. 1996, Mears and Velyvis 2003, Oransky and Tortora 2007), but long-term structural changes of the sacrum after fracture healing have not been explored.Our primary aim was to assess long-term radiological findings after surgically treated displaced sacral fractures. In addition, we wanted to assess whether pathological radiological findings, including bony structural changes of the sacrum, may contribute to neurological dysfunctions of the lower extremities or to the occurrence of pelvis-related pain.     

Generalizability of machine learning for classification of schizophrenia based on resting‐state functional MRI data

Machine learning has increasingly been applied to classification of schizophrenia in neuroimaging research. However, direct replication studies and studies seeking to investigate generalizability are scarce. To address these issues, we assessed within‐site and between‐site generalizability of a machine learning classification framework which achieved excellent performance in a previous study using two independent resting‐state functional magnetic resonance imaging data sets collected from different sites and scanners. We established within‐site generalizability of the classification framework in the main data set using cross‐validation. Then, we trained a model in the main data set and investigated between‐site generalization in the validated data set using external validation. Finally, recognizing the poor between‐site generalization performance, we updated the unsupervised algorithm to investigate if transfer learning using additional unlabeled data were able to improve between‐site classification performance. Cross‐validation showed that the published classification procedure achieved an accuracy of 0.73 using majority voting across all selected components. External validation found a classification accuracy of 0.55 (not significant) and 0.70 (significant) using the direct and transfer learning procedures, respectively. The failure of direct generalization from one site to another demonstrates the limitation of within‐site cross‐validation and points toward the need to incorporate efforts to facilitate application of machine learning across multiple data sets. The improvement in performance with transfer learning highlights the importance of taking into account the properties of data when constructing predictive models across samples and sites. Our findings suggest that machine learning classification result based on a single study should be interpreted cautiously.