Angeliki Papana

Global and regional estimates of COPD prevalence: Systematic review and meta-analysis

Background The burden of chronic obstructive pulmonary disease (COPD) across many world regions is high. We aim to estimate COPD prevalence and number of disease cases for the years 1990 and 2010 across world regions based on the best available evidence in publicly accessible scientific databases. Methods We conducted a systematic search of Medline, EMBASE and Global Health for original, population–based studies providing spirometry–based prevalence rates of COPD across the world from January 1990 to December 2014. Random effects meta–analysis was conducted on extracted crude prevalence rates of COPD, with overall summaries of the meta–estimates (and confidence intervals) reported separately for World Health Organization (WHO) regions, the World Banks income categories and settings (urban and rural). We developed a meta–regression epidemiological model that we used to estimate the prevalence of COPD in people aged 30 years or more. Findings Our search returned 37 472 publications. A total of 123 studies based on a spirometry–defined prevalence were retained for the review. From the meta–regression epidemiological model, we estimated about 227.3 million COPD cases in the year 1990 among people aged 30 years or more, corresponding to a global prevalence of 10.7% (95% confidence interval (CI) 7.3%–14.0%) in this age group. The number of COPD cases increased to 384 million in 2010, with a global prevalence of 11.7% (8.4%–15.0%). This increase of 68.9% was mainly driven by global demographic changes. Across WHO regions, the highest prevalence was estimated in the Americas (13.3% in 1990 and 15.2% in 2010), and the lowest in South East Asia (7.9% in 1990 and 9.7% in 2010). The percentage increase in COPD cases between 1990 and 2010 was the highest in the Eastern Mediterranean region (118.7%), followed by the African region (102.1%), while the European region recorded the lowest increase (22.5%). In 1990, we estimated about 120.9 million COPD cases among urban dwellers (prevalence of 13.2%) and 106.3 million cases among rural dwellers (prevalence of 8.8%). In 2010, there were more than 230 million COPD cases among urban dwellers (prevalence of 13.6%) and 153.7 million among rural dwellers (prevalence of 9.7%). The overall prevalence in men aged 30 years or more was 14.3% (95% CI 13.3%–15.3%) compared to 7.6% (95% CI 7.0%–8.2%) in women. Conclusions Our findings suggest a high and growing prevalence of COPD, both globally and regionally. There is a paucity of studies in Africa, South East Asia and the Eastern Mediterranean region. There is a need for governments, policy makers and international organizations to consider strengthening collaborations to address COPD globally.

Simulation study of direct causality measures in multivariate time series

Measures of the direction and strength of the interdependence among time series from multivariate systems are evaluated based on their statistical significance and discrimination ability. The best-known measures estimating direct causal effects, both linear and nonlinear, are considered, i.e., conditional Granger causality index (CGCI), partial Granger causality index (PGCI), partial directed coherence (PDC), partial transfer entropy (PTE), partial symbolic transfer entropy (PSTE) and partial mutual information on mixed embedding (PMIME). The performance of the multivariate coupling measures is assessed on stochastic and chaotic simulated uncoupled and coupled dynamical systems for different settings of embedding dimension and time series length. The CGCI, PGCI and PDC seem to outperform the other causality measures in the case of the linearly coupled systems, while the PGCI is the most effective one when latent and exogenous variables are present. The PMIME outweighs all others in the case of nonlinear simulation systems.

Reducing the bias of causality measures

Measures of the direction and strength of the interdependence between two time series are evaluated and modified to reduce the bias in the estimation of the measures, so that they give zero values when there is no causal effect. For this, point shuffling is employed as used in the frame of surrogate data. This correction is not specific to a particular measure and it is implemented here on measures based on state space reconstruction and information measures. The performance of the causality measures and their modifications is evaluated on simulated uncoupled and coupled dynamical systems and for different settings of embedding dimension, time series length, and noise level. The corrected measures, and particularly the suggested corrected transfer entropy, turn out to stabilize at the zero level in the absence of a causal effect and detect correctly the direction of information flow when it is present. The measures are also evaluated on electroencephalograms (EEG) for the detection of the information flow in the brain of an epileptic patient. The performance of the measures on EEG is interpreted in view of the results from the simulation study.

Evaluation of Mutual Information Estimators for Time Series

We study some of the most commonly used mutual information estimators, based on histograms of fixed or adaptive bin size, k-nearest neighbors and kernels and focus on optimal selection of their free parameters. We examine the consistency of the estimators (convergence to a stable value with the increase of time series length) and the degree of deviation among the estimators. The optimization of parameters is assessed by quantifying the deviation of the estimated mutual information from its true or asymptotic value as a function of the free parameter. Moreover, some commonly used criteria for parameter selection are evaluated for each estimator. The comparative study is based on Monte Carlo simulations on time series from several linear and nonlinear systems of different lengths and noise levels. The results show that the k-nearest neighbor is the most stable and less affected by the method-specific parameter. A data adaptive criterion for optimal binning is suggested for linear systems but it is found to be rather conservative for nonlinear systems. It turns out that the binning and kernel estimators give the least deviation in identifying the lag of the first minimum of mutual information from nonlinear systems, and are stable in the presence of noise.

Prevalence of rheumatoid arthritis in low– and middle–income countries: A systematic review and analysis

Background Rheumatoid arthritis (RA) is an autoimmune disorder that affects the small joints of the body. It is one of the leading causes of chronic morbidity in high–income countries, but little is known about the burden of this disease in low– and middle–income countries (LMIC). Methods The aim of this study was to estimate the prevalence of RA in six of the World Health Organizations (WHO) regions that harbour LMIC by identifying all relevant studies in those regions. To accomplish this aim various bibliographic databases were searched: PubMed, EMBASE, Global Health, LILACS and the Chinese databases CNKI and WanFang. Studies were selected based on pre–defined inclusion criteria, including a definition of RA based on the 1987 revision of the American College of Rheumatology (ACR) definition. Results Meta–estimates of regional RA prevalence rates for countries of low or middle income were 0.40% (95% CI: 0.23–0.57%) for Southeast Asian, 0.37% (95% CI: 0.23–0.51%) for Eastern Mediterranean, 0.62% (95% CI: 0.47–0.77%) for European, 1.25% (95% CI: 0.64–1.86%) for American and 0.42% (95% CI: 0.30–0.53%) for Western Pacific regions. A formal meta–analysis could not be performed for the sub–Saharan African region due to limited data. Male prevalence of RA in LMIC was 0.16% (95% CI: 0.11–0.20%) while the prevalence in women reached 0.75% (95% CI: 0.60–0.90%). This difference between males and females was statistically significant (P < 0.0001). The prevalence of RA did not differ significantly between urban and rural settings (P = 0.353). These prevalence estimates represent 2.60 (95% CI: 1.85–3.34%) million male sufferers and 12.21 (95% CI: 9.78–14.67%) million female sufferers in LMIC in the year 2000, and 3.16 (95% CI: 2.25–4.05%) million affected males and 14.87 (95% CI: 11.91–17.86%) million affected females in LMIC in the year 2010. Conclusion Given that majority of the world’s population resides in LMIC, the number of affected people is substantial, with a projection to increase in the coming years. Therefore, policy makers and health–care providers need to plan to address a significant disease burden both socially and economically.

An Estimate of the Prevalence of COPD in Africa: A Systematic Analysis

Abstract Background: Chronic obstructive pulmonary disease (COPD) is among the leading causes of death globally, accounting for about 3 million deaths worldwide in 2011. We aimed to estimate the prevalence of COPD in Africa in the year 2010 to provide the information that could assist health policy in the region. Methods: We conducted a systematic review of Medline, EMBASE and Global Health for studies on COPD published between 1990 and 2012. We included original population based studies providing estimates of the prevalence of COPD. We considered the reported estimates in terms of the mean age of the sample, sex ratio, the year of study and the country of the study as possible covariates. Results from two different types of studies, i.e., based on spirometric and non-spirometric diagnosis of COPD, were further compared. The United Nation Population Divisions population figures were used to estimate the number of COPD cases in the year 2010. Results: Our search returned 243 studies, from which only 13 met our selection criteria and only five were based on spirometry. The difference in the median prevalence of COPD in persons aged 40 years or older based on spirometry data (13.4%; IQR: 9.4%–22.1%) and non-spirometry data (4.0%; IQR: 2.1%–8.9%) was statistically significant (p = 0.001). There was no significant effect of the gender or the year of the study on the reported prevalence of COPD in either set of studies. The prevalence of COPD increased with age in spirometry-based studies (p = 0.017), which is a plausible finding suggesting internal consistency of spirometry-based estimates, while this trend was not observed in studies using other case definitions. When applied to the appropriate age group (40 years or more), which accounted for 196.4 million people in Africa in 2010, the estimated prevalence translates into 26.3 million (18.5–43.4 million) cases of COPD. Comparable figures for the year 2000 based on the same prevalence rates would amount to 20.0 million (14.1–33.1), suggesting an increase of 31.5% over a decade that is attributable to ageing of the African population alone. Conclusion: Our findings suggest that COPD is likely to already represent a very large public health problem in Africa. Moreover, rapidly ageing African population should expect a steady increase in the number of COPD cases in the next decade and beyond. The quantity and quality of available evidence does not match the size of the problem. There is a need for more research on COPD prevalence, but also incidence, mortality and risk factors in Africa. We hope this study will raise awareness of COPD in Africa and encourage further research.

DETECTION OF DIRECT CAUSAL EFFECTS AND APPLICATION TO EPILEPTIC ELECTROENCEPHALOGRAM ANALYSIS

An extension of transfer entropy, called partial transfer entropy (PTE), is proposed to detect causal effects among observed interacting systems, and particularly to distinguish direct from indirect causal effects. PTE is compared to a linear direct causality measure, the Partial Directed Coherence (PDC), on known linear stochastic systems and nonlinear deterministic systems. PTE performs equally well as PDC on the linear systems and better than PDC on the nonlinear systems, both being dependent on the selection of the measure specific parameters. PTE and PDC are applied to electroencephalograms of epileptic patients during the preictal, ictal and postictal states, and PTE turns out to detect better changes of the strength of the direct causality at specific pairs of electrodes and for the different states.

Time series feature evaluation in discriminating preictal EEG states

Statistical discrimination of states in the preictal EEG is attempted using a large number of measures from linear and nonlinear time series analysis. The measures are organized in two categories: correlation measures, such as autocorrelation and mutual information at specific lags and new measures derived from oscillations of the EEG time series, such as mean oscillation peak and mean oscillation period. All measures are computed on successive segments of multichannel EEG windows selected from early, intermediate and late preictal states from four epochs. Hypothesis tests applied for each channel and epoch showed good discrimination of the preictal states and allowed for the selection of optimal measures. These optimal measures, together with other standard measures (skewness, kurtosis, largest Lyapunov exponent) formed the feature set for feature-based clustering and the feature-subset selection procedure showed that the best preictal state classification was obtained with the same optimal features.

Assessment of resampling methods for causality testing: A note on the US inflation behavior

Different resampling methods for the null hypothesis of no Granger causality are assessed in the setting of multivariate time series, taking into account that the driving-response coupling is conditioned on the other observed variables. As appropriate test statistic for this setting, the partial transfer entropy (PTE), an information and model-free measure, is used. Two resampling techniques, time-shifted surrogates and the stationary bootstrap, are combined with three independence settings (giving a total of six resampling methods), all approximating the null hypothesis of no Granger causality. In these three settings, the level of dependence is changed, while the conditioning variables remain intact. The empirical null distribution of the PTE, as the surrogate and bootstrapped time series become more independent, is examined along with the size and power of the respective tests. Additionally, we consider a seventh resampling method by contemporaneously resampling the driving and the response time series using the stationary bootstrap. Although this case does not comply with the no causality hypothesis, one can obtain an accurate sampling distribution for the mean of the test statistic since its value is zero under H0. Results indicate that as the resampling setting gets more independent, the test becomes more conservative. Finally, we conclude with a real application. More specifically, we investigate the causal links among the growth rates for the US CPI, money supply and crude oil. Based on the PTE and the seven resampling methods, we consistently find that changes in crude oil cause inflation conditioning on money supply in the post-1986 period. However this relationship cannot be explained on the basis of traditional cost-push mechanisms.

A Nonparametric Causality Test: Detection of Direct Causal Effects in Multivariate Systems Using Corrected Partial Transfer Entropy

In a recent work we proposed the corrected transfer entropy (CTE), which reduces the bias in the estimation of transfer entropy (TE), a measure of Granger causality for bivariate time series making use of the conditional mutual information. An extension of TE to account for the presence of other time series is the partial TE (PTE). Here, we propose the correction of PTE, termed Corrected PTE (CPTE), in a similar way to CTE: time shifted surrogates are used in order to quantify and correct the bias, and the estimation of the involved entropies of high-dimensional variables is made with the method of k-nearest neighbors. CPTE is evaluated on coupled stochastic systems with both linear and nonlinear interactions. Finally, we apply CPTE to economic data and investigate whether we can detect the direct causal effects among economic variables.