Background
Individual patient data (IPD) meta‐analysis allows for the exploration of heterogeneity and can identify subgroups that most benefit from an intervention (or exposure), much more ...successfully than meta‐analysis of aggregate data. One‐stage or two‐stage IPD meta‐analysis is possible, with the former using mixed‐effects regression models and the latter obtaining study estimates through simpler regression models before aggregating using standard meta‐analysis methodology. However, a comprehensive comparison of the two methods, in practice, is lacking.
Methods
We generated 1000 datasets for each of many simulation scenarios covering different IPD sizes and different between‐study variance (heterogeneity) assumptions at various levels (intercept and exposure). Numerous simulation settings of different assumptions were also used, while we evaluated performance both on main effects and interaction effects. Performance was assessed on mean bias, mean error, coverage, and power.
Results
Fully specified one‐stage models (random study intercept or fixed study‐specific intercept; random exposure effect; and fixed study‐specific effects for covariate) were the best performers overall, especially when investigating interactions. For main effects, performance was almost identical across models unless intercept heterogeneity was present, in which case the fully specified one‐stage and the two‐stage models performed better. For interaction effects, differences across models were greater with the two‐stage model consistently outperformed by the two fully specified one‐stage models.
Conclusions
A fully specified one‐stage model should be preferred (accounting for potential exposure, intercept, and, possibly, interaction heterogeneity), especially when investigating interactions. If non‐convergence is encountered with a random study intercept, the fixed study‐specific intercept one‐stage model should be used instead.
Severe coronavirus disease 2019 (COVID-19) is associated with increased risk of venous thromboembolism events (VTE). This study performed a systematic review in PubMed/EMBASE of studies reporting the ...prevalence of VTE in patients with COVID-19 who were totally screened/assessed for deep vein thrombosis (DVT) and/or for pulmonary embolism (PE). Among 47 candidate studies (n = 6459; 33 in Europe), 17 studies (n = 3973; weighted age 63.0 years, males 60%, intensive care unit (ICU) 16%) reported the prevalence of PE with a pooled estimate of 32% (95% CI: 25, 40%), and 32 studies (n = 2552; weighted age 62.6 years, males 57%, ICU 49%) reported the prevalence of DVT with a pooled estimate of 27% (95% CI: 21, 34%). A total of 36 studies reported the use of at least prophylactic antithrombotic treatment in the majority of their patients. Meta-regression analysis showed that the prevalence of VTE was higher across studies with a higher percentage of ICU patients and higher study population mean D-dimer values, and lower in studies with mixed dosing of anticoagulation in ⩾ 50% of the population compared to studies with standard prophylactic dosing of anticoagulation in < 50% of the population. The pooled odds ratio for death in patients with COVID-19 and VTE versus those without VTE (17 studies, n = 2882) was 2.1 (95% CI: 1.2, 3.6). Hospitalized patients with severe COVID-19 are at high VTE risk despite prophylactic anticoagulation. Further research should investigate the individualized VTE risk of patients with COVID-19 and the optimal preventive antithrombotic therapy. PROSPERO Registration No.: CRD42020185543.
Heterogeneity has a key role in meta-analysis methods and can greatly affect conclusions. However, true levels of heterogeneity are unknown and often researchers assume homogeneity. We aim to: a) ...investigate the prevalence of unobserved heterogeneity and the validity of the assumption of homogeneity; b) assess the performance of various meta-analysis methods; c) apply the findings to published meta-analyses.
We accessed 57,397 meta-analyses, available in the Cochrane Library in August 2012. Using simulated data we assessed the performance of various meta-analysis methods in different scenarios. The prevalence of a zero heterogeneity estimate in the simulated scenarios was compared with that in the Cochrane data, to estimate the degree of unobserved heterogeneity in the latter. We re-analysed all meta-analyses using all methods and assessed the sensitivity of the statistical conclusions. Levels of unobserved heterogeneity in the Cochrane data appeared to be high, especially for small meta-analyses. A bootstrapped version of the DerSimonian-Laird approach performed best in both detecting heterogeneity and in returning more accurate overall effect estimates. Re-analysing all meta-analyses with this new method we found that in cases where heterogeneity had originally been detected but ignored, 17-20% of the statistical conclusions changed. Rates were much lower where the original analysis did not detect heterogeneity or took it into account, between 1% and 3%.
When evidence for heterogeneity is lacking, standard practice is to assume homogeneity and apply a simpler fixed-effect meta-analysis. We find that assuming homogeneity often results in a misleading analysis, since heterogeneity is very likely present but undetected. Our new method represents a small improvement but the problem largely remains, especially for very small meta-analyses. One solution is to test the sensitivity of the meta-analysis conclusions to assumed moderate and large degrees of heterogeneity. Equally, whenever heterogeneity is detected, it should not be ignored.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Meta-analysis of clinical trials targeting rare events face particular challenges when the data lack adequate number of events and are susceptible to high levels of heterogeneity. The standard ...meta-analysis methods (DerSimonian Laird (DL) and Mantel–Haenszel (MH)) often lead to serious distortions because of such data sparsity. Applications of the methods suited to specific incidence and heterogeneity characteristics are lacking, thus we compared nine available methods in a simulation study. We generated 360 meta-analysis scenarios where each considered different incidences, sample sizes, between-study variance (heterogeneity) and treatment allocation. We include globally recommended methods such as inverse-variance fixed/random-effect (IV-FE/RE), classical-MH, MH-FE, MH-DL, Peto, Peto-DL and the two extensions for MH bootstrapped-DL (bDL) and Peto-bDL. Performance was assessed on mean bias, mean error, coverage and power. In the absence of heterogeneity, the coverage and power when combined revealed small differences in meta-analysis involving rare and very rare events. The Peto-bDL method performed best, but only in smaller sample sizes involving rare events. For medium-to-larger sample sizes, MH-bDL was preferred. For meta-analysis involving very rare events, Peto-bDL was the best performing method which was sustained across all sample sizes. However, in meta-analysis with 20% or more heterogeneity, the coverage and power were insufficient. Performance based on mean bias and mean error was almost identical across methods. To conclude, in meta-analysis of rare binary outcomes, our results suggest that Peto-bDL is better in both rare and very rare event settings in meta-analysis with limited sample sizes. However, when heterogeneity is large, the coverage and power to detect rare events are insufficient. Whilst this study shows that some of the less studied methods appear to have good properties under sparse data scenarios, further work is needed to assess them against the more complex distributional-based methods to understand their overall performances.
Meta-analysis (MA) is a statistical methodology that combines the results of several independent studies considered by the analyst to be ‘combinable’. The simplest approach, the fixed-effects (FE) ...model, assumes the true effect to be the same in all studies, while the random-effects (RE) family of models allows the true effect to vary across studies. However, all methods are only correct asymptotically, while some RE models assume that the true effects are normally distributed. In practice, MA methods are frequently applied when study numbers are small and the normality of the effect distribution unknown or unlikely. In this article, we discuss the performance of the FE approach and seven frequentist RE MA methods: DerSimonian–Laird, Q-based, maximum likelihood, profile likelihood, Biggerstaff–Tweedie, Sidik–Jonkman and Follmann–Proschan. We covered numerous scenarios by varying the MA sizes (small to moderate), the degree of heterogeneity (zero to very large) and the distribution of the effect sizes (normal, skew-normal and ‘extremely’ non-normal). Performance was evaluated in terms of coverage (Type I error), power (Type II error) and overall effect estimation (accuracy of point estimates and error intervals).
Summary Background Introduced in 2004, the UK's Quality and Outcomes Framework (QOF) is the world's largest primary care pay-for-performance programme. We tested whether the QOF was associated with ...reduced population mortality. Methods We used population-level mortality statistics between 1994 and 2010 for the UK and other high-income countries that were not exposed to pay-for-performance. The primary outcome was age-adjusted and sex-adjusted mortality per 100 000 people for a composite outcome of chronic disorders that were targeted by the QOF. Secondary outcomes were age-adjusted and sex-adjusted mortality for ischaemic heart disease, cancer, and a composite of all non-targeted conditions. For each study outcome, we created a so-called synthetic UK as a weighted combination of comparison countries. We then estimated difference-in-differences models to test whether mortality fell more in the UK than in the synthetic UK after the QOF. Findings Introduction of the QOF was not significantly associated with changes in population mortality for the composite outcome (−3·68 per 100 000 population 95% CI −8·16 to 0·80; p=0·107), ischaemic heart disease (−2·21 per 100 000 –6·86 to 2·44; p=0·357), cancer (0·28 per 100 000 –0·99 to 1·55; p=0·679), or all non-targeted conditions (11·60 per 100 000 –3·91 to 27·11; p=0·143). Interpretation Although we noted small mortality reductions for a composite outcome of targeted disorders, the QOF was not associated with significant changes in mortality. Our findings have implications for the probable effects of similar programmes on population health outcomes. The relation between incentives and mortality needs to be assessed in specific disease domains. Funding None.
The potential impact of the COVID-19 pandemic on population mental health is of increasing global concern. We examine changes in adult mental health in the UK population before and during the ...lockdown.
In this secondary analysis of a national, longitudinal cohort study, households that took part in Waves 8 or 9 of the UK Household Longitudinal Study (UKHLS) panel, including all members aged 16 or older in April, 2020, were invited to complete the COVID-19 web survey on April 23-30, 2020. Participants who were unable to make an informed decision as a result of incapacity, or who had unknown postal addresses or addresses abroad were excluded. Mental health was assessed using the 12-item General Health Questionnaire (GHQ-12). Repeated cross-sectional analyses were done to examine temporal trends. Fixed-effects regression models were fitted to identify within-person change compared with preceding trends.
Waves 6-9 of the UKHLS had 53 351 participants. Eligible participants for the COVID-19 web survey were from households that took part in Waves 8 or 9, and 17 452 (41·2%) of 42 330 eligible people participated in the web survey. Population prevalence of clinically significant levels of mental distress rose from 18·9% (95% CI 17·8-20·0) in 2018-19 to 27·3% (26·3-28·2) in April, 2020, one month into UK lockdown. Mean GHQ-12 score also increased over this time, from 11·5 (95% CI 11·3-11·6) in 2018-19, to 12·6 (12·5-12·8) in April, 2020. This was 0·48 (95% CI 0·07-0·90) points higher than expected when accounting for previous upward trends between 2014 and 2018. Comparing GHQ-12 scores within individuals, adjusting for time trends and significant predictors of change, increases were greatest in 18-24-year-olds (2·69 points, 95% CI 1·89-3·48), 25-34-year-olds (1·57, 0·96-2·18), women (0·92, 0·50-1·35), and people living with young children (1·45, 0·79-2·12). People employed before the pandemic also averaged a notable increase in GHQ-12 score (0·63, 95% CI 0·20-1·06).
By late April, 2020, mental health in the UK had deteriorated compared with pre-COVID-19 trends. Policies emphasising the needs of women, young people, and those with preschool aged children are likely to play an important part in preventing future mental illness.
None.
Purpose
The adverse impact of hearing loss (HL) extends beyond auditory impairment and may affect the individuals' psychosocial wellbeing. We aimed to examine whether there exists a causal ...psychosocial pathway between HL and depression in later life, via socioeconomic factors and quality of life, and whether hearing aids usage alleviates depressive symptoms over time.
Methods
We examined the longitudinal relationship between HL and depressive symptoms (CES-D) applying dynamic cross-lagged mediation path models. We used the full dataset of participants aged 50–89 years (74,908 person-years), from all eight Waves of the English Longitudinal Study of Ageing (ELSA). Their quality of life (CASP-19) and their wealth were examined as the mediator and moderator of this relationship, respectively. Subgroup analyses investigated differences among those with hearing aids within different models of subjectively and objectively identified HL. All models were adjusted for age, sex, retirement status and social engagement.
Results
Socioeconomic position (SEP) influenced the strength of the relationship between HL and depression, which was stronger in the lowest versus the highest wealth quintiles. The use of hearing aids was beneficial for alleviating depressive symptoms. Those in the lowest wealth quintiles experienced a lower risk for depression after the use of hearing aids compared to those in the highest wealth quintiles.
Conclusion
HL poses a substantial risk for depressive symptoms in older adults, especially those who experience socioeconomic inequalities. The early detection of HL and provision of hearing aids may not only promote better-hearing health but could also enhance the psychosocial wellbeing of older adults, particularly those in a lower SEP.
Nurses as substitutes for doctors in primary care Laurant, Miranda; van der Biezen, Mieke; Wijers, Nancy ...
Cochrane database of systematic reviews,
07/2018, Letnik:
2019, Številka:
2
Journal Article
Recenzirano
Odprti dostop
Background
Current and expected problems such as ageing, increased prevalence of chronic conditions and multi‐morbidity, increased emphasis on healthy lifestyle and prevention, and substitution for ...care from hospitals by care provided in the community encourage countries worldwide to develop new models of primary care delivery. Owing to the fact that many tasks do not necessarily require the knowledge and skills of a doctor, interest in using nurses to expand the capacity of the primary care workforce is increasing. Substitution of nurses for doctors is one strategy used to improve access, efficiency, and quality of care. This is the first update of the Cochrane review published in 2005.
Objectives
Our aim was to investigate the impact of nurses working as substitutes for primary care doctors on:
• patient outcomes;
• processes of care; and
• utilisation, including volume and cost.
Search methods
We searched the Cochrane Central Register of Controlled Trials (CENTRAL), part of the Cochrane Library (www.cochranelibrary.com), as well as MEDLINE, Ovid, and the Cumulative Index to Nursing and Allied Health Literature (CINAHL) and EbscoHost (searched 20.01.2015). We searched for grey literature in the Grey Literature Report and OpenGrey (21.02.2017), and we searched the International Clinical Trials Registry Platform (ICTRP) and ClinicalTrials.gov trial registries (21.02.2017). We did a cited reference search for relevant studies (searched 27.01 2015) and checked reference lists of all included studies. We reran slightly revised strategies, limited to publication years between 2015 and 2017, for CENTRAL, MEDLINE, and CINAHL, in March 2017, and we have added one trial to ‘Studies awaiting classification’.
Selection criteria
Randomised trials evaluating the outcomes of nurses working as substitutes for doctors. The review is limited to primary healthcare services that provide first contact and ongoing care for patients with all types of health problems, excluding mental health problems. Studies which evaluated nurses supplementing the work of primary care doctors were excluded.
Data collection and analysis
Two review authors independently carried out data extraction and assessment of risk of bias of included studies. When feasible, we combined study results and determined an overall estimate of the effect. We evaluated other outcomes by completing a structured synthesis.
Main results
For this review, we identified 18 randomised trials evaluating the impact of nurses working as substitutes for doctors. One study was conducted in a middle‐income country, and all other studies in high‐income countries. The nursing level was often unclear or varied between and even within studies. The studies looked at nurses involved in first contact care (including urgent care), ongoing care for physical complaints, and follow‐up of patients with a particular chronic conditions such as diabetes. In many of the studies, nurses could get additional support or advice from a doctor. Nurse‐doctor substitution for preventive services and health education in primary care has been less well studied.
Study findings suggest that care delivered by nurses, compared to care delivered by doctors, probably generates similar or better health outcomes for a broad range of patient conditions (low‐ or moderate‐certainty evidence):
• Nurse‐led primary care may lead to slightly fewer deaths among certain groups of patients, compared to doctor‐led care. However, the results vary and it is possible that nurse‐led primary care makes little or no difference to the number of deaths (low‐certainty evidence).
• Blood pressure outcomes are probably slightly improved in nurse‐led primary care. Other clinical or health status outcomes are probably similar (moderate‐certainty evidence).
• Patient satisfaction is probably slightly higher in nurse‐led primary care (moderate‐certainty evidence). Quality of life may be slightly higher (low‐certainty evidence).
We are uncertain of the effects of nurse‐led care on process of care because the certainty of this evidence was assessed as very low.
The effect of nurse‐led care on utilisation of care is mixed and depends on the type of outcome. Consultations are probably longer in nurse‐led primary care (moderate‐certainty evidence), and numbers of attended return visits are slightly higher for nurses than for doctors (high‐certainty evidence). We found little or no difference between nurses and doctors in the number of prescriptions and attendance at accident and emergency units (high‐certainty evidence). There may be little or no difference in the number of tests and investigations, hospital referrals and hospital admissions between nurses and doctors (low‐certainty evidence).
We are uncertain of the effects of nurse‐led care on the costs of care because the certainty of this evidence was assessed as very low.
Authors' conclusions
This review shows that for some ongoing and urgent physical complaints and for chronic conditions, trained nurses, such as nurse practitioners, practice nurses, and registered nurses, probably provide equal or possibly even better quality of care compared to primary care doctors, and probably achieve equal or better health outcomes for patients. Nurses probably achieve higher levels of patient satisfaction, compared to primary care doctors. Furthermore, consultation length is probably longer when nurses deliver care and the frequency of attended return visits is probably slightly higher for nurses, compared to doctors. Other utilisation outcomes are probably the same. The effects of nurse‐led care on process of care and the costs of care are uncertain, and we also cannot ascertain what level of nursing education leads to the best outcomes when nurses are substituted for doctors.
Glycemic variability is emerging as a measure of glycemic control, which may be a reliable predictor of complications. This systematic review and meta-analysis evaluates the association between HbA1c ...variability and micro- and macrovascular complications and mortality in type 1 and type 2 diabetes.
Medline and Embase were searched (2004-2015) for studies describing associations between HbA1c variability and adverse outcomes in patients with type 1 and type 2 diabetes. Data extraction was performed independently by two reviewers. Random-effects meta-analysis was performed with stratification according to the measure of HbA1c variability, method of analysis, and diabetes type.
Seven studies evaluated HbA1c variability among patients with type 1 diabetes and showed an association of HbA1c variability with renal disease (risk ratio 1.56 95% CI 1.08-2.25, two studies), cardiovascular events (1.98 1.39-2.82), and retinopathy (2.11 1.54-2.89). Thirteen studies evaluated HbA1c variability among patients with type 2 diabetes. Higher HbA1c variability was associated with higher risk of renal disease (1.34 1.15-1.57, two studies), macrovascular events (1.21 1.06-1.38), ulceration/gangrene (1.50 1.06-2.12), cardiovascular disease (1.27 1.15-1.40), and mortality (1.34 1.18-1.53). Most studies were retrospective with lack of adjustment for potential confounders, and inconsistency existed in the definition of HbA1c variability.
HbA1c variability was positively associated with micro- and macrovascular complications and mortality independently of the HbA1c level and might play a future role in clinical risk assessment.