De laatste jaren groeit het bewustzijn onder wetenschappers en het brede publiek over hoe innig financiën, gezondheid en welzijn met elkaar verstrengeld zijn. Je kunt problemen van mensen vaak niet reduceren tot een aantal sub-problemen die je onafhankelijk van elkaar kunt oplossen, vanuit de gedachte dat alle oplossingen bij elkaar opgeteld het totaalprobleem verhelpen.4 Uit onderzoek weten we inmiddels steeds meer over hoe problemen met elkaar in verband staan. Beleidsmakers en uitvoerende beroepskrachten en vrijwilligers die de complexe relatie tussen financiële, fysieke en mentale factoren begrijpen, bevinden zich in de unieke positie om deze inzichten te vertalen naar hun initiatieven. Dit rapport biedt een overzicht van academische en grijze literatuur, en de lessen die we hieruit kunnen trekken.
ObjectiveTo compare estimates of effect and variability resulting from standard linear regression analysis and hierarchical multilevel analysis with cross-classified multilevel analysis under various scenarios.Study design and settingWe performed a simulation study based on a data structure from an observational study in clinical mental health care. We used a Markov chain Monte Carlo approach to simulate 18 scenarios, varying sample sizes, cluster sizes, effect sizes and between group variances. For each scenario, we performed standard linear regression, multilevel regression with random intercept on patient level, multilevel regression with random intercept on nursing team level and cross-classified multilevel analysis.ResultsApplying cross-classified multilevel analyses had negligible influence on the effect estimates. However, ignoring cross-classification led to underestimation of the standard errors of the covariates at the two cross-classified levels and to invalidly narrow confidence intervals. This may lead to incorrect statistical inference. Varying sample size, cluster size, effect size and variance had no meaningful influence on these findings.ConclusionIn case of cross-classified data structures, the use of a cross-classified multilevel model helps estimating valid precision of effects, and thereby, support correct inferences.
MULTIFILE
Objective: We determined the prevalences of hyperoxemia and excessive oxygen use, and the epidemiology, ventilation characteristics and outcomes associated with hyperoxemia in invasively ventilated patients with coronavirus disease 2019 (COVID–19). Methods: Post hoc analysis of a national, multicentre, observational study in 22 ICUs. Patients were classified in the first two days of invasive ventilation as ‘hyperoxemic’ or ‘normoxemic’. The co–primary endpoints were prevalence of hyperoxemia (PaO2 > 90 mmHg) and prevalence of excessive oxygen use (FiO2 ≥ 60% while PaO2 > 90 mmHg or SpO2 > 92%). Secondary endpoints included ventilator settings and ventilation parameters, duration of ventilation, length of stay (LOS) in ICU and hospital, and mortality in ICU, hospital, and at day 28 and 90. We used propensity matching to control for observed confounding factors that may influence endpoints. Results: Of 851 COVID–19 patients, 225 (26.4%) were classified as hyperoxemic. Excessive oxygen use occurred in 385 (45.2%) patients. Acute respiratory distress syndrome (ARDS) severity was lowest in hyperoxemic patients. Hyperoxemic patients were ventilated with higher positive end–expiratory pressure (PEEP), while rescue therapies for hypoxemia were applied more often in normoxemic patients. Neither in the unmatched nor in the matched analysis were there differences between hyperoxemic and normoxemic patients with regard to any of the clinical outcomes. Conclusion: In this cohort of invasively ventilated COVID–19 patients, hyperoxemia occurred often and so did excessive oxygen use. The main differences between hyperoxemic and normoxemic patients were ARDS severity and use of PEEP. Clinical outcomes were not different between hyperoxemic and normoxemic patients.
