A state-space approach for longitudinal outcomes: An application to neuropsychological outcomes

Longitudinal assessments are crucial in evaluating the disease state and trajectory in patients with neurodegenerative diseases. Neuropsychological outcomes measured over time often have a non-linear trajectory with autocorrelated residuals and a skewed distribution. We propose the adjusted local linear trend model, an extended state-space model in lieu of the commonly used linear mixed-effects model in modeling longitudinal neuropsychological outcomes. Our contributed model has the capability to utilize information from the stochasticity of the data while accounting for subject-specific trajectories with the inclusion of covariates and unequally spaced time intervals. The first step of model fitting involves a likelihood maximization step to estimate the unknown variances in the model before parsing these values into the Kalman filter and Kalman smoother recursive algorithms. Results from simulation studies showed that the adjusted local linear trend model is able to attain lower bias, lower standard errors, and high power, particularly in short longitudinal studies with equally spaced time intervals, as compared to the linear mixed-effects model. The adjusted local linear trend model also outperforms the linear mixed-effects model when data is missing completely at random, missing at random, and, in certain cases, even in data with missing not at random.

Determining What Changed Japanese Suicide Mortality in 2020 Using Governmental Database

The pandemic of 2019 novel coronavirus disease (COVID-19) caused both COVID-19-related health hazards and the deterioration of socioeconomic and sociopsychological status due to governmental restrictions. There were concerns that suicide mortality would increase during the COVID-19 pandemic; however, a recent study reported that suicide mortality did not increase in 21 countries during the early pandemic period. In Japan, suicide mortality was reduced from 2009 to 2019, but both the annual number of suicide victims and the national suicide mortality rates in 2020 increased compared to that in 2019. To clarify the discrepancy of suicide mortality between the first and second half of 2020 in Japan, the present study determines annual and monthly suicide mortality disaggregated by prefectures, gender, age, means, motive, and household factors during the COVID-19 pandemic and pre-pandemic periods using a linear mixed-effects model. Furthermore, the relationship between suicide mortality and COVID-19 data (the infection rate, mortality, and duration of the pandemic) was analysed using hierarchal linear regression with a robust standard error. The average of monthly suicide mortality of both males and females in all 47 prefectures decreased during the first stay-home order (April–May) (females: from 10.1–10.2 to 7.8–7.9; males: from 24.0–24.9 to 21.6 per 100,000 people), but increased after the end of the first stay-home order (July–December) (females: from 7.5–9.5 to 10.3–14.5; males: from 19.9–23.0 to 21.1–26.7 per 100,000 people). Increasing COVID-19-infected patients and victims indicated a tendency of suppression, but the prolongation of the pandemic indicated a tendency of increasing female suicide mortality without affecting that of males. Contrary to the national pattern, in metropolitan regions, decreasing suicide mortality during the first stay-home order was not observed. Decreasing suicide mortality during the first stay-home order was not observed in populations younger than 30 years old, whereas increasing suicide mortality of populations younger than 30 years old after the end of the first stay-home order was predominant. A decrease in suicide mortality of one-person household residents during the first stay-home order was not observed. The hanging suicide mortality of males and females was decreased and increased during and after the end of the first stay-home orders, respectively; however, there was no decrease in metropolitan regions. These results suggest that the suicide mortality in 2020 of females, younger populations, urban residents, and one-person household residents increased compared to those of males, the elderly, rural residents, and multiple-person household residents. Therefore, the unexpected drastic fluctuations of suicide mortality during the COVID-19 pandemic in Japan were probably composed of complicated reasons among various identified factors in this study, and other unknown factors.

Respiratory function declines in children with asthma associated with chemical species of fine particulate matter (PM2.5) in Nagasaki, Japan

Background The differential effects of PM2.5 fractions on children’s lung function remain inconclusive. This study aimed to examine whether lung function in asthmatic children was associated with increased PM2.5 fractions in urban areas in Nagasaki prefecture, Japan, where the air pollution level is relatively low but influenced by transboundary air pollution. Methods We conducted a multiyear panel study of 73 asthmatic children (boys, 60.3%; mean age, 8.2 years) spanning spring 2014–2016 in two cities. We collected self-measured peak expiratory flow (PEF) twice a day and daily time-series data for PM2.5 total mass and its chemical species. We fitted a linear mixed effects model to examine short-term associations between PEF and PM2.5, adjusting for individual and time-varying confounders. A generalized linear mixed effects model was also used to estimate the association for worsening asthma defined by severe PEF decline. Back-trajectory and cluster analyses were used to investigate the long-range transboundary PM2.5 in the study areas. Results We found that morning PEFs were adversely associated with higher levels of sulfate (− 1.61 L/min; 95% CI: − 3.07, − 0.15) in Nagasaki city and organic carbon (OC) (− 1.02 L/min; 95% CI: − 1.94, − 0.09) in Isahaya city, per interquartile range (IQR) increase at lag1. In addition, we observed consistent findings for worsening asthma, with higher odds of severe PEF decline in the morning for sulfate (odds ratio (OR) = 2.31; 95% CI: 1.12, 4.77) and ammonium (OR = 1.73; 95% CI: 1.06, 2.84) in Nagasaki city and OC (OR = 1.51; 95% CI: 1.06, 2.15) in Isahaya city, per IQR increase at lag1. The significant chemical species were higher on days that could be largely attributed to the path of Northeast China origin (for sulfate and ammonium) or both the same path and local sources (for OC) than by other clusters. Conclusions This study provides evidence of the differential effects of PM2.5 fractions on lung function among asthmatic children in urban areas, where the Japanese national standards of air quality have been nearly met. Continuous efforts to promote mitigation actions and public awareness of hazardous transboundary air pollution are needed to protect susceptible children with asthma.

Separating Paternal and Maternal Contributions to Thermal Transgenerational Plasticity

Climate change is rapidly altering the thermal environment in terrestrial and aquatic systems. Transgenerational thermal plasticity (TGP) – which occurs when the temperatures experienced by the parental generation prior to the fertilization of gametes results in a change in offspring reaction norms – may mitigate the effects of climate change. Although “maternal effects” have been widely studied, relatively little is known about TGP effects in vertebrates, particularly paternal contributions. We used artificial fertilization to cross sheepshead minnow (Cyprinodon variegatus) parents exposed to either low (26°C) or high (32°C) temperatures and measured growth rates of the offspring over the first 8 weeks of life at both low and high temperatures. A linear mixed effects model was employed to quantify the effects of maternal, paternal, and offspring temperatures on offspring growth and fecundity. We found that the offspring growth rate up to 63 days post-hatch was affected by both the temperature they experienced directly and parental temperatures prior to fertilization. Growth was lowest when neither parents’ temperature matched the offspring temperature, indicating a strong transgenerational effect. Notably, offspring growth was highest when all three (offspring, sire, and dam) temperatures matched [although the three-way interaction was found to be marginally non-significant (P = 0.155)], suggesting that TGP effects were additive across significant sire-offspring (P < 0.001) and dam-offspring interactions (P < 0.001). Transgenerational effects on fecundity (GSI) were suggestive for both maternal and paternal effects, but not significant. The finding that thermal TGP is contributed by both parents strongly suggests that it has an epigenetic basis.

PSII-7 Farrowing age is a predictor of sow lifetime performance

Sow lifetime performance can be predicted by sow performance at an early age; however, few studies have examined the relationship between farrowing age (FA) after parity 1 and sow lifetime performance. Therefore, our objective was to examine the association between FA in different parities and sow lifetime performance. Data were extracted from 198,043 sows entered into 155 Spanish herds between 2011 - 2013 and removed between 2011 - 2016. The 1st to 99th percentiles of FA records were analyzed, with FA defined as the number of days from birth to the farrowing date in each parity. A two-level linear mixed-effects model was applied to examine the relationship between FA and lifetime performance of sows (SAS University Edition). Means of FA + SD (range) in parities 1 and 3 were 368.4 + 34.8 (238 - 460) days and 663.5 + 41.4 (501 - 773) days, respectively. The ranges of FA in parity 1 and 3 were 238 - 460 days and 501 - 773 days, respectively. Lifetime performance decreased with higher FA in both parities. For example, in parity 1, as FA increased from 290 to 410 days, lifetime piglets born alive (PBA) decreased by 4.4 piglets, annualized lifetime PBA decreased by 3.5 piglets and nonproductive days increased by 16.8 days. Similarly, in parity 3, as FA increased from 590 to 710 days, lifetime PBA decreased by 5.3 piglets, annualized lifetime PBA decreased by 3.0 piglets and nonproductive days increased by 39.5 days. Also, sows with a lower FA in parity 1 had fewer PBA in that parity than sow with a higher FA (P < 0.05). However, they also had fewer nonproductive days and a shorter weaning-to-first-mating interval and more annualized lifetime PBA. Therefore, these data indicate that FA can be used to predict lifetime performance of the sows.

PSII-11 Herd size associated with sow lifetime performance in Spanish breeding herds

It is critical for producers to maximize sows’ reproductive potential in commercial herds in order to improve economic efficiency. While it is generally known that large herds have better reproductive performance than small herds, few studies have assessed how much the lifetime reproductive performance of sows is associated with herd size. Therefore, our objective was to examine the relationship between six herd size groups and sow lifetime performance. Data were extracted from 166,335 sows which were entered into 155 Spanish breeding herds from 2011 - 2013 and removed by 2017. Herds were categorized into six groups based on the 10th, 25th, 50th, 75th and 90th percentiles of average sow inventory in 2016: 87 - 195, 196 - 342, 343 - 596, 597 - 1,025, 1,026 - 2,152 and 2,153 - 3,669 sows. A two-level linear mixed-effects model was applied to examine the relationship between herd size and reproductive performance of sows (SAS University Edition). No differences were found between herd size groups and either lifetime total born or lifetime piglets born alive. However, the largest herd group had 23.0 fewer lifetime nonproductive days (NPD), and 2.1 more annualized lifetime piglets weaned than the mid-size (343 - 596 sows) herds (P < 0.05). Also, parity at removal was 0.3 lower in the largest herds than the mid-size herds (P < 0.05). Furthermore, the largest herds also had 0.8 - 1.0 days shorter weaning-to-first-mating interval (WMI), 2.8 - 4.9% higher farrowing rates and 10.2 - 11.9 days shorter repeat intervals than the mid-size herds (P < 0.05). In conclusion, the shorter WMI, fewer NPD and improved reproductive productivity in the largest herds suggest that these herds have better lactational management and quicker culling decision making than mid-size herds.