A comparison of survival models for prediction of eight-year revision risk following total knee and hip arthroplasty.

BACKGROUND: There is increasing interest in the development and use of clinical prediction models, but a lack of evidence-supported guidance on the merits of different modelling approaches. This is especially true for time-to-event outcomes, where limited studies have compared the vast number of modelling approaches available. This study compares prediction accuracy and variable importance measures for four modelling approaches in prediction of time-to-revision surgery following total knee arthroplasty (TKA) and total hip arthroplasty (THA). METHODS: The study included 321,945 TKA and 151,113 THA procedures performed between 1 January 2003 and 31 December 2017. Accuracy of the Cox model, Weibull parametric model, flexible parametric model, and random survival forest were compared, with patient age, sex, comorbidities, and prosthesis characteristics considered as predictors. Prediction accuracy was assessed using the Index of Prediction Accuracy (IPA), c-index, and smoothed calibration curves. Variable importance rankings from the Cox model and random survival forest were also compared. RESULTS: Overall, the Cox and flexible parametric survival models performed best for prediction of both TKA (integrated IPA 0.056 (95% CI [0.054, 0.057]) compared to 0.054 (95% CI [0.053, 0.056]) for the Weibull parametric model), and THA revision. (0.029 95% CI [0.027, 0.030] compared to 0.027 (95% CI [0.025, 0.028]) for the random survival forest). The c-index showed broadly similar discrimination between all modelling approaches. Models were generally well calibrated, but random survival forest underfitted the predicted risk of TKA revision compared to regression approaches. The most important predictors of revision were similar in the Cox model and random survival forest for TKA (age, opioid use, and patella resurfacing) and THA (femoral cement, depression, and opioid use). CONCLUSION: The Cox and flexible parametric models had superior overall performance, although all approaches performed similarly. Notably, this study showed no benefit of a tuned random survival forest over regression models in this setting.

Higher frequency of vertebrate-infecting viruses in the gut of infants born to mothers with type 1 diabetes.

BACKGROUND: Microbial exposures in utero and early life shape the infant microbiome, which can profoundly impact on health. Compared to the bacterial microbiome, very little is known about the virome. We set out to characterize longitudinal changes in the gut virome of healthy infants born to mothers with or without type 1 diabetes using comprehensive virome capture sequencing. METHODS: Healthy infants were selected from Environmental Determinants of Islet Autoimmunity (ENDIA), a prospective cohort of Australian children with a first-degree relative with type 1 diabetes, followed from pregnancy. Fecal specimens were collected three-monthly in the first year of life. RESULTS: Among 25 infants (44% born to mothers with type 1 diabetes) at least one virus was detected in 65% (65/100) of samples and 96% (24/25) of infants during the first year of life. In total, 26 genera of viruses were identified and >150 viruses were differentially abundant between the gut of infants with a mother with type 1 diabetes vs without. Positivity for any virus was associated with maternal type 1 diabetes and older infant age. Enterovirus was associated with older infant age and maternal smoking. CONCLUSIONS: We demonstrate a distinct gut virome profile in infants of mothers with type 1 diabetes, which may influence health outcomes later in life. Higher prevalence and greater number of viruses observed compared to previous studies suggests significant underrepresentation in existing virome datasets, arising most likely from less sensitive techniques used in data acquisition.

Effect of placental restriction and neonatal exendin-4 treatment on postnatal growth, adult body composition, and in vivo glucose metabolism in the sheep.

Intrauterine growth restriction (IUGR) increases the risk of adult type 2 diabetes (T2D) and obesity. Neonatal exendin-4 treatment can prevent diabetes in the IUGR rat, but whether this will be effective in a species where the pancreas is more mature at birth is unknown. Therefore, we evaluated the effects of neonatal exendin-4 administration after experimental restriction of placental and fetal growth on growth and adult metabolic outcomes in sheep. Body composition, glucose tolerance, and insulin secretion and sensitivity were assessed in singleton-born adult sheep from control (CON; n = 6 females and 4 males) and placentally restricted pregnancies (PR; n = 13 females and 7 males) and in sheep from PR pregnancies that were treated with exendin-4 as neonates (daily sc injections of 1 nmol/kg exendin-4; PR + exendin-4; n = 11 females and 7 males). Placental restriction reduced birth weight (by 29%) and impaired glucose tolerance in the adult but did not affect adult adiposity, insulin secretion, or insulin sensitivity. Neonatal exendin-4 suppressed growth during treatment, followed by delayed catchup growth and unchanged adult adiposity. Neonatal exendin-4 partially restored glucose tolerance in PR progeny but did not affect insulin secretion or sensitivity. Although the effects on glucose tolerance are promising, the lack of effects on adult body composition, insulin secretion, and insulin sensitivity suggest that the neonatal period may be too late to fully reprogram the metabolic consequences of IUGR in species that are more mature at birth than rodents.

Bioelectrical phase angle values in a clinical sample of ambulatory rehabilitation patients.

BACKGROUND: Phase angle (PhA) is derived from the resistance and reactance measurements obtained from bioelectric impedance analysis (BIA) and is considered indicative of cellular health and membrane integrity. This study measured PhA values of rehabilitation patients and compared them to reference values, measures of functional ability and serum C-reactive protein (CRP) levels to explore their utility as a clinical tool to monitor disease progression and treatment efficacy. METHODS: This cross-sectional observational study was conducted on 215 ambulatory rehabilitation patients aged 20 - 94 years. All participants had been hospitalised for a stroke, orthopaedic or other condition resulting in a functional limitation. PhA was derived from BIA analysis and functional ability characterised using the Functional Independence Measure (FIM), timed up and go (TUG) and maximal quadriceps strength (MQS). Serum levels of CRP were also collected. RESULTS: Stroke patients had the highest PhA (5.3 degrees) followed by elective orthopaedic surgery (5.0 degrees) with the other group (4.3 degrees) significantly lower than both previous categories (p < 0.001). Ambulatory rehabilitation patients' PhA values were dependent on age and sex (p < 0.001), lower than published age matched healthy reference values (p < or = 0.05) and similar to other hospitalised or sick groups, but also higher than values reported in critically ill patients. Patients with CRP values less than 10 mg.L-1 had significantly (p = 0.005) higher mean PhA values. Furthermore, the highest functional status quartiles had significantly higher PhAs (p < or = 0.04) for the FIM, MQS and TUG measures. CONCLUSION: The results suggest that the phase angles of rehabilitation patients are between those of healthy individuals and seriously ill patients, thereby supporting claims that PhA is indicative of general health status. Phase angles are a potentially useful indicator of functional status in patients commencing an ambulatory rehabilitation program with a normal hydration status.

Nutritional status of adults participating in ambulatory rehabilitation.

AIMS: To assess the overall nutritional status of older adults participating in ambulatory rehabilitation and determine its association with relevant outcomes including physical function and quality of life. DESIGN: Cross-sectional. SETTING: Ambulatory rehabilitation service in the Southern region of Adelaide, Australia. SUBJECTS: A total of 229 participants recruited as part of a RCT between June 2005 and June 2006, stroke (n=83), elective orthopedic procedure (n=44) and other medical condition (n=102). METHODS: Nutritional status was measured using Mini Nutritional Assessment (MNA), Simplified Nutrition Appetite Questionnaire (SNAQ) and Body Mass Index. Functional performance was assessed using the Modified Barthel Index (MBI) and quality of life was measured using the Short Form-36 (SF-36). RESULTS: Sixty-three percent of participants were malnourished or at risk of malnutrition according to the MNA and a third had a risk of >or= 5% weight loss in the subsequent six months, according to the SNAQ. Participants with a diagnosis other than stroke or elective orthopedic procedure were the most vulnerable, with 53% (n=74/140) classified as at risk of malnutrition or malnourished and a longer length of stay in hospital. Functional performance was no different for participants assessed as at risk of malnutrition or malnourished compared to the well nourished, but the SF-36 mental component score was significantly higher for those who were well nourished (p=0.003). CONCLUSION: Findings emphasise the magnitude of the malnutrition problem in ambulatory rehabilitation settings. Further research is required to evaluate the resource implications against expected benefits of providing nutrition interventions at this point.

Falls in hospitalized patients: can nursing information systems data predict falls?

Falls among inpatients are the most frequently reported critical incidents in hospitals and can have tragic consequences that affect morbidity and mortality. The present study aimed to determine whether certain nursing units of care identified on patient care plans can be used to predict falls among hospitalized inpatients. A retrospective analysis of 7167 inpatient admissions in the 2002 calendar year was conducted. Faller status was ascertained from the hospital's accident and incident monitoring system, and nursing units of care activated in the hospital's nursing information system were identified. Twelve nursing units of care predicted falls. Logistic regression analyses showed that nursing units of care related to patient safety, confusion, incontinence, medication, mobility, and sleep were significant risk factors for falls among inpatients. The number of nursing units of care activated also predicted falls. Data collected from nursing information systems can be used to identify patients at high risk of falls.

Corrected arm muscle area: an independent predictor of long-term mortality in community-dwelling older adults?

OBJECTIVES: Older people are at risk of undernutrition because of a number of physiological conditions and lifestyle factors. The purpose of this study was to explore the predictive relationship of corrected arm muscle area (CAMA) with 8-year mortality in a representative sample of older Australians. DESIGN: Prospective cohort study: The Australian Longitudinal Study of Ageing. SETTING: Community. PARTICIPANTS: One thousand three hundred ninety-six participants aged 70 and older. MEASUREMENTS: Trained observers measured baseline weight, height, mid upper arm circumference, and triceps skinfold thickness using standard techniques. Body mass index (BMI) and CAMA were calculated. Baseline BMI and CAMA measurements were categorized according to cutoff values proposed by Garrow et al. and Friedman et al., respectively. Subsequent analyses were undertaken using Cox proportional hazards regression. RESULTS: After adjustment for potential confounders (baseline age, gender, marital status, smoking, self-rated health, ability to conduct activities of daily living, comorbidity, cognition performance, and presence of depression), those older Australians with a low CAMA (

Social networks among older Australians: a validation of Glass's model.

BACKGROUND: The purpose of the study was to validate a set of measures of social networks that will be useful in epidemiological studies of the health of older persons. METHODS: Confirmatory factor analysis was used to test a multi-dimensional model of social networks. Data were drawn from The Australian Longitudinal Study of Ageing, conducted in the greater metropolitan area of Adelaide, South Australia. Participants were 1477 persons aged > or = 70 years who lived in the community or an institution and took part in the Australian Longitudinal Study of Ageing. Social networks with children, other relatives, friends and confidants were considered. Demographic indicators of gender, age group, income, education, marital status and place of residence were also measured, and social networks were compared for categories of these variables. RESULTS: Networks with children, relatives, friends and confidants were substantiated in these analyses. Differences in networks between categories of the demographic variables were evident. DISCUSSION: Adoption of the proposed measures will allow greater comparability between studies of older persons. This could lead to a better understanding of the effect of specific social networks upon health.