Search

Abstract Many countries use national nutrition surveys (NNSs) to assess their population’s health and nutrition needs. However, NNS aims, approaches, tools, and measurements vary among countries. To date, there has been no review evaluating the NNSs and their practices worldwide to help conduct future NSSs. Therefore, this narrative review was conducted to 1) explore and tabulate current NNSs in five continents (Asia, Europe, Africa, North America, and Australia) and 2) help lay the foundation for establishing clear guidelines for future NNSs. The NNSs were identified using two approaches. First, an electronic database search was conducted with key terms in PubMed database. Second, a general web-based search on the survey webpages of governmental organizations was conducted using the same key terms to identify eligible surveys. The review included general adult population (≥ 18 years) with a cross-sectional design, excluding NNSs related to household-only surveys, specific age groups, or insufficient sample sizes. A total of 41 NNSs were identified in 37 countries across four continents: Asia (n = 15), Europe (n = 21), North America (n = 3), and Australia (n = 2). Broad differences between the surveys were identified, including survey purposes and designs, definitions of geographic areas and target groups, and dietary assessments. Currently, there are 26 ongoing NNSs, while 15 have ended. Among the ongoing NNSs, the cycles of the surveys were either at regular intervals (n = 8) or irregular intervals (n = 8). Of the 41 surveys, 24-h dietary recalls were used in 27 surveys, while only 6 surveys used diet diaries and 8 surveys relied on FFQs. Some surveys (n = 17) utilized multiple tools to assess dietary intake. Most of the surveys that assessed biochemical status (n = 12) focused on blood glucose, haemoglobin A1c (HbA1c), and lipid status, whereas some surveys (n = 6) tested for vitamin and mineral status in blood and/or urine samples. The review identified key characteristics, time frames, sampling methods, and dietary and physical assessment methods obtained from different surveys worldwide. The information organized in this review will be important for researchers, policymakers, and public health programme developers in developing and improving NNS.

Background: Recognizing the scarcity of data on caffeine consumption among university students in Malaysia, this study was designed to first characterize the habitual caffeine intake within this group. This study further investigated the potential association between these caffeine consumption patterns and sleep quality and mental health status. Methodology: The study employed an online self-administered questionnaire to assess habitual caffeine intake from various products. Sleep quality and mental health were evaluated using the Pittsburgh Sleep Quality Index (PSQI) and the Depression, Anxiety, and Stress Scale-21 (DASS-21), respectively. Results: 78% of the respondents consumed caffeinated products daily within the safe level recommended by the European Food Safety Authority (£400mg/day). Only 2.6% of the students consumed caffeine exceeding the recommended limit. The median total daily caffeine intake was 100mg/day. Most students had poor sleep quality (77.3%), with 43.5%, 49.5%, and 33.0% reporting varying levels of depression, anxiety, and stress, respectively. Odds ratio calculations revealed respondents consuming less than or more than 400 mg/day of caffeine did not show significantly different odds of poor sleep quality, depression, anxiety, or stress compared to non-consumers. Conclusions: This study highlights the need for further localized research on caffeine's effects among Malaysian university students regarding caffeine consumption, sleep, and mental health.

Accurate detection of anomalies in multivariate time series data has attracted much attention due to its importance in a wide range of applications. Since it is difficult to obtain accurately labeled data, many unsupervised anomaly detection algorithms for multivariate time series data have been developed. However, building such a system is challenging since it requires capturing temporal dependencies in each time series and must also encode the inter-correlations between different pairs of time series. To meet this challenge, we propose a Multi Scale Convolutional Variational Autoencoder (MSCVAE) to detect anomalies in multivariate time series data. Firstly, multi scale attribute matrices are constructed from multivariate time series to characterize multiple levels of the system states at different time steps. Then, given the attribute matrices, a convolutional variational autoencoder is employed to generate reconstructed attribute matrices, and also an attention-based ConvLSTM network is used to capture the temporal patterns. In addition, a new ERR-based threshold setting strategy is developed to optimize anomaly detection performance instead of relying on the traditional ROC-based threshold setting strategy with an imbalanced dataset. Finally, the proposed framework is assessed by means of experiments on four datasets. The experimental results show that our proposed framework is superior to competing algorithms in terms of model performance and robustness, demonstrating that our model is effective in detecting anomalies in multivariate time series.