Effects of Data Aggregation on Time Series Analysis of Seasonal Infections

Design and methodology for monitoring the occurrence of foodborne pathogens in meat and poultry in United States: Case study of Campylobacter on chicken

In the United States, the Food Safety and Inspection Service (FSIS), which is part of the U.S. Department of Agriculture, oversees multiple sampling programs that focus on detecting microbial contamination of meat and poultry in slaughter and processing establishments. There have been efforts across scientific disciplines to describe data collection methods and estimation strategies for large scale surveys to allow for the integration of data from multiple sources. The FSIS sample selection methods can be described as a close approximation of either a stratified or two-stage cluster sampling design. These surveys therefore support estimates of pathogen occurrence that are derived under both design- and model-based inferential paradigms. This retrospective study will describe population-level estimation strategies and how changes in pathogen contamination can be monitored across time using a trend analysis. An example based on Campylobacter on broiler chicken carcasses is provided. The example demonstrates that changes in the apparent prevalence during the study period are predominantly the result of changes in laboratory methods.

Seasonality of underweight among infants 1-11 months old in Niger: an exploratory analysis of data from a cluster-randomised trial

INTRODUCTION

Malnutrition is a risk factor for child mortality, with around 45% of deaths in children under 5 globally linked to malnutrition. Seasonality of malnutrition has important implications for the timing of child health programme activities, but evidence is mixed on the nature of such patterns. Moreover, the bulk of the existing evidence is focused on wasting and stunting in children 6-59 months, despite increasing evidence that younger children also face a high risk, and that underweight alone is an important predictor of mortality.

METHODS

This study used data from the cluster-randomised AVENIR trial which compared the effect of biannual distribution of azithromycin vs placebo on mortality in children 1-59 months old in Niger. AVENIR included a biannual census conducted on a rolling basis over 2 years. A subset of 133 781 infants aged 1-11 months from 2904 communities were included in this study, and weight-for-age z-score (WAZ) was calculated at each census. The exposure for this analysis is the day of the year weight was captured. Harmonic regression was used to determine primary and secondary peaks and nadirs of WAZ over time.

RESULTS

Overall, the primary peak of WAZ occurred in late February and the primary nadir occurred in mid-May, aligning with a seasonal temperature increase before the rainy season. A secondary peak in August and a secondary nadir in November were also seen, aligning with the postrainy season.

CONCLUSION

The seasonality of WAZ of infants 1-11 months in Niger may have implications for the timing of programmes aiming to decrease malnutrition.

Seasonality of Acute Malnutrition in African Drylands: Evidence From 15 Years of SMART Surveys

Reduction of wasting, or low weight-for-height, is a critical target for the Zero Hunger Sustainable Development Goal, yet robust evidence establishing continuous seasonal patterns of wasting is presently lacking. The current consensus of greatest hunger during the preharvest period is based on survey designs and analytical methods, which discretize time frame into preharvest/postharvest, dry/wet, or lean/plenty seasons. We present a spatiotemporally nuanced study of acute malnutrition seasonality in African drylands using a 15-year data set of Standardized Monitoring and Assessment of Relief and Transition surveys (n = 412,370). Climatological similarity was ensured by selecting subnational survey regions with 1 rainy season and by spatially matching each survey to aridity and livelihood zones. Harmonic logit regression models indicate 2 peaks of wasting during the calendar year. Greatest wasting prevalence is estimated in April to May, coincident with the primary peak of temperature. A secondary peak of wasting is observed in August to October, coinciding with the primary peak of rainfall and secondary peak of temperature. This pattern is retained across aridity and livelihood zones and is sensitive to temperature, precipitation, and vegetation. Improved subnational estimation of acute malnutrition seasonality can thus assist decision makers and practitioners in data-sparse settings and facilitate global progress toward Zero Hunger.

Forecasting Seasonal Acute Malnutrition: Setting the Framework

BACKGROUND

Malnutrition is an umbrella term that refers to an impairment in nutrition indicative of subsequently compromised human well-being. The term covers the full spectrum of nutritional impairments from a small yet detectable departure from a "norm" to a terminal stage when severe malnutrition could result in death. This broad spectrum of nutritional departures from "the optimum" dictates the need for an ensemble of metrics to capture the complexity of involved mechanisms, risk factors, precipitating events, short-term, and long-term consequences. Ideally, these metrics should be universally applicable to vulnerable populations, settings, ages, and times when people are most susceptible to malnutrition. We should be able to characterize and intervene to minimize the risk of malnutrition, especially child acute malnutrition that could be assessed by anthropometric measurements.

OBJECTIVES

The main challenge in reaching such an ambitious goal is the complexity of measuring, characterizing, explaining, predicting, and preventing malnutrition at any dimension: temporal or spatial and at any scale: a person or a group. The expansive body of literature has been accumulated on many temporal aspects of malnutrition and seasonal changes in nutritional (anthropometric) status. The research community is now shifting their attention to predictive modeling of child malnutrition and its importance for clinical and public health interventions. This communication aims to provide an overview of challenges for understanding child malnutrition from a perspective of predictive modeling focusing on well-documented seasonal variations in nutritional outcomes and exploring "the systems approach" to tackle underlining conceptual and practical complexities to forecast seasonal malnutrition in an accurate and timely manner. This generalized approach to forecasting seasonal malnutrition is then applied specifically to child acute malnutrition.

Global seasonal and pandemic patterns in influenza: An application of longitudinal study designs

The confluence of growing analytic capacities and global surveillance systems for seasonal infections has created new opportunities to further develop statistical methodology and advance the understanding of the global disease dynamics. We developed a framework to characterise the seasonality of infectious diseases for publicly available global health surveillance data. Specifically, we aimed to estimate the seasonal characteristics and their uncertainty using mixed effects models with harmonic components and the δ-method and develop multi-panel visualisations to present complex interplay of seasonal peaks across geographic locations. We compiled a set of 2 422 weekly time series of 14 reported outcomes for 173 Member States from the World Health Organization's (WHO) international influenza virological surveillance system, FluNet, from 02 January 1995 through 20 June 2021. We produced an analecta of data visualisations to describe global travelling waves of influenza while addressing issues of data completeness and credibility. Our results offer directions for further improvements in data collection, reporting, analysis and development of statistical methodology and predictive approaches.

Associations between extreme precipitation, drinking water, and protozoan acute gastrointestinal illnesses in four North American Great Lakes cities (2009-2014)

Climate change is already impacting the North American Great Lakes ecosystem and understanding the relationship between climate events and public health, such as waterborne acute gastrointestinal illnesses (AGIs), can help inform needed adaptive capacity for drinking water systems (DWSs). In this study, we assessed a harmonized binational dataset for the effects of extreme precipitation events (≥90th percentile) and preceding dry periods, source water turbidity, total coliforms, and protozoan AGIs - cryptosporidiosis and giardiasis - in the populations served by four DWSs that source surface water from Lake Ontario (Hamilton and Toronto, Ontario, Canada) and Lake Michigan (Green Bay and Milwaukee, Wisconsin, USA) from January 2009 through August 2014. We used distributed lag non-linear Poisson regression models adjusted for seasonality and found extreme precipitation weeks preceded by dry periods increased the relative risk of protozoan AGI after 1 and 3-5 weeks in three of the four cities, although only statistically significant in two. Our results suggest that the risk of protozoan AGI increases with extreme precipitation preceded by a dry period. As extreme precipitation patterns become more frequent with climate change, the ability to detect changes in water quality and effectively treat source water of varying quality is increasingly important for adaptive capacity and protection of public health.

Evaluating Completeness of Foodborne Outbreak Reporting in the United States, 1998-2019

Public health agencies routinely collect time-referenced records to describe and compare foodborne outbreak characteristics. Few studies provide comprehensive metadata to inform researchers of data limitations prior to conducting statistical modeling. We described the completeness of 103 variables for 22,792 outbreaks publicly reported by the United States Centers for Disease Control and Prevention’s (US CDC’s) electronic Foodborne Outbreak Reporting System (eFORS) and National Outbreak Reporting System (NORS). We compared monthly trends of completeness during eFORS (1998−2008) and NORS (2009−2019) reporting periods using segmented time series analyses adjusted for seasonality. We quantified the overall, annual, and monthly completeness as the percentage of outbreaks with blank records per our study period, calendar year, and study month, respectively. We found that outbreaks of unknown genus (n = 7401), Norovirus (n = 6414), Salmonella (n = 2872), Clostridium (n = 944), and multiple genera (n = 779) accounted for 80.77% of all outbreaks. However, crude completeness ranged from 46.06% to 60.19% across the 103 variables assessed. Variables with the lowest crude completeness (ranging 3.32−6.98%) included pathogen, specimen etiological testing, and secondary transmission traceback information. Variables with low (<35%) average monthly completeness during eFORS increased by 0.33−0.40%/month after transitioning to NORS, most likely due to the expansion of surveillance capacity and coverage within the new reporting system. Examining completeness metrics in outbreak surveillance systems provides essential information on the availability of data for public reuse. These metadata offer important insights for public health statisticians and modelers to precisely monitor and track the geographic spread, event duration, and illness intensity of foodborne outbreaks.

Spatiotemporal patterns of the COVID-19 epidemic in Mexico at the municipality level

In recent history, Coronavirus Disease 2019 (COVID-19) is one of the worst infectious disease outbreaks affecting humanity. The World Health Organization has defined the outbreak of COVID-19 as a pandemic, and the massive growth of the number of infected cases in a short time has caused enormous pressure on medical systems. Mexico surpassed 3.7 million confirmed infections and 285,000 deaths on October 23, 2021. We analysed the spatio-temporal patterns of the COVID-19 epidemic in Mexico using the georeferenced confirmed cases aggregated at the municipality level. We computed weekly Moran’s I index to assess spatial autocorrelation over time and identify clusters of the disease using the “flexibly shaped spatial scan” approach. Finally, we compared Euclidean, cost, resistance distances and gravitational model to select the best-suited approach to predict inter-municipality contagion. We found that COVID-19 pandemic in Mexico is characterised by clusters evolving in space and time as parallel epidemics. The gravitational distance was the best model to predict newly infected municipalities though the predictive power was relatively low and varied over time. This study helps us understand the spread of the epidemic over the Mexican territory and gives insights to model and predict the epidemic behaviour.

Modeling the Global Dynamic Contagion of COVID-19

The COVID-19 infections have profoundly and negatively impacted the whole world. Hence, we have modeled the dynamic spread of global COVID-19 infections with the connectedness approach based on the TVP-VAR model, using the data of confirmed COVID-19 cases during the period of March 23rd, 2020 to September 10th, 2021 in 18 countries. The results imply that, (i) the United States, the United Kingdom and Indonesia are global epidemic centers, among which the United States has the highest degree of the contagion of the COVID-19 infections, which is stable. South Korea, France and Italy are the main receiver of the contagion of the COVID-19 infections, and South Korea has been the most severely affected by the overseas epidemic; (ii) there is a negative correlation between the timeliness, effectiveness and mandatory nature of government policies and the risk of the associated countries COVID-19 epidemic affecting, as well as the magnitude of the net contagion of domestic COVID-19; (iii) the severity of domestic COVID-19 epidemics in the United States and Canada, Canada and Mexico, Indonesia and Canada is almost equivalent, especially for the United States, Canada and Mexico, whose domestic epidemics are with the same tendency; (iv) the COVID-19 epidemic has spread though not only the central divergence manner and chain mode of transmission, but also the way of feedback loop. Thus, more efforts should be made by the governments to enhance the pertinence and compulsion of their epidemic prevention policies and establish a systematic and efficient risk assessment mechanism for public health emergencies.

The Effect of Interdependences of Referral Behaviors on the Quality of Ambulatory Care: Evidence from Taiwan

Purpose

The purpose of this study is to investigate the effect of interdependences of healthcare providers’ referral behaviors on the quality of ambulatory care. The significance of this study is to address the concern regarding the low quality of ambulatory care due to the lack of a compulsory referral system under Taiwan’s National Health Insurance system.

Methods

We applied the dynamic connectedness network analysis to estimate the total connectedness index of the referral behavior network, which was separated into the horizontal and vertical referral behavior components in order to measure the interdependences of horizontal and vertical referral behaviors across hospitals and local clinics, respectively.

Results

Our results suggest that the interdependences of referral behaviors increase the quality of ambulatory care. The harmful effect on the quality of ambulatory care from the interdependences of horizontal referral behaviors within the local clinics sector is more significant than that from the interdependences of horizontal referral behaviors within the hospital sector, and the negative effect on the overall and chronic composite measures of avoidable hospital admissions from the interdependences of vertical behaviors associated with local clinics is more substantial than that from the interdependences of vertical behaviors within the hospital sector.

Conclusion

These results not only highlight the significance of care collaboration between local clinics and hospitals to restrain avoidable hospital admissions of chronic diseases for a better overall quality of ambulatory care, but they also suggest that the surveillance system established for the quality of ambulatory care under the global budget payment scheme for the local clinics sector should target ambulatory care for patients with acute conditions.

How Seasonality of Malnutrition Is Measured and Analyzed

Seasonality is a critical source of vulnerability across most human activities and natural processes, including the underlying and immediate drivers of acute malnutrition. However, while there is general agreement that acute malnutrition is highly variable within and across years, the evidence base is limited, resulting in an overreliance on assumptions of seasonal peaks. We review the design and analysis of 24 studies exploring the seasonality of nutrition outcomes in Africa's drylands, providing a summary of approaches and their advantages and disadvantages. Over half of the studies rely on two to four time points within the year and/or the inclusion of time as a categorical variable in the analysis. While such approaches simplify interpretation, they do not correspond to the climatic variability characteristic of drylands or the relationship between climatic variability and human activities. To better ground our understanding of the seasonality of acute malnutrition in a robust evidence base, we offer recommendations for study design and analysis, including drawing on participatory methods to identify community perceptions of seasonality, use of longitudinal data and panel analysis with approaches borrowed from the field of infectious diseases, and linking oscillations in nutrition data with climatic data.

Seasonal synchronization of foodborne outbreaks in the United States, 1996-2017

Modern food systems represent complex dynamic networks vulnerable to foodborne infectious outbreaks difficult to track and control. Seasonal co-occurrences (alignment of seasonal peaks) and synchronization (similarity of seasonal patterns) of infections are noted, yet rarely explored due to their complexity and methodological limitations. We proposed a systematic approach to evaluate the co-occurrence of seasonal peaks using a combination of L-moments, seasonality characteristics such as the timing (phase) and intensity (amplitude) of peaks, and three metrics of serial, phase-phase, and phase-amplitude synchronization. We used public records on counts of nine foodborne infections abstracted from CDC's FoodNet Fast online platform for the US and ten representative states from 1996 to 2017 (264 months). Based on annualized and trend-adjusted Negative Binomial Harmonic Regression (NBHR) models augmented with the δ-method, we determined that seasonal peaks of Campylobacter, Salmonella, and Shiga toxin-producing Escherichia Coli (STEC) were tightly clustered in late-July at the national and state levels. Phase-phase synchronization was observed between Cryptosporidium and Shigella, Listeria, and Salmonella (ρ = 0.51, 0.51, 0.46; p < 0.04). Later peak timing of STEC was associated with greater amplitude nationally (ρ = 0.50, p = 0.02) indicating phase-amplitude synchronization. Understanding of disease seasonal synchronization is essential for developing reliable outbreak forecasts and informing stakeholders on mitigation and preventive measures.

An analecta of visualizations for foodborne illness trends and seasonality

Disease surveillance systems worldwide face increasing pressure to maintain and distribute data in usable formats supplemented with effective visualizations to enable actionable policy and programming responses. Annual reports and interactive portals provide access to surveillance data and visualizations depicting temporal trends and seasonal patterns of diseases. Analyses and visuals are typically limited to reporting the annual time series and the month with the highest number of cases per year. Yet, detecting potential disease outbreaks and supporting public health interventions requires detailed spatiotemporal comparisons to characterize spatiotemporal patterns of illness across diseases and locations. The Centers for Disease Control and Prevention's (CDC) FoodNet Fast provides population-based foodborne-disease surveillance records and visualizations for select counties across the US. We offer suggestions on how current FoodNet Fast data organization and visual analytics can be improved to facilitate data interpretation, decision-making, and communication of features related to trend and seasonality. The resulting compilation, or analecta, of 436 visualizations of records and codes are openly available online.