Piecewise Linear Regression Research Articles

Abstract 11: Geographic Inequalities in Cardiovascular Mortality: Impact of Public Health Initiatives

Introduction: Improvements in therapy and prevention have led to declining cardiovascular mortality in the United States, but it is not clear whether these improvements have narrowed geographic disparities in cardiovascular outcomes. We sought to compare mortality due to cardiovascular disease, heart failure, stroke, and ischemic heart disease in the stroke belt cluster of 11 states versus the rest of the United States. Methods: A retrospective cross-sectional analysis of the CDC WONDER database was done to evaluate the nationwide mortality trends derived from the death certificates of all American residents from 1999 to 2017. Mortality trends for death due to heart failure, stroke, ischemic heart disease or any cardiovascular cause, were identified in the stroke belt and non-stroke belt populations using ICD-10 codes. Piecewise linear regression was used to assess the change in mortality trends. Results: Among 16,111,775 deaths due to cardiovascular causes during the study period, the age-adjusted mortality rates (AAMR) were highest among non-Hispanic Black, males from non-metropolitan areas, living in the stroke belt. In the stroke belt, AAMR due to all cardiovascular causes [Average Annual Percentage Change (AAPC): -2.5 (95% CI:-2.9 to -2.0); p<0.001], stroke [AAPC: -2.9 (95% CI: -3.7 to -2.1); p<0.001] and ischemic heart disease [AAPC: -3.9 (95% CI: -4.3 to -3.5); p<0.001] declined from 1999 to 2017. Similarly, a decrease in cardiovascular [AAPC: -2.6 (95% CI:-3.1 to -2.1); p<0.001], stroke [AAPC:-2.9 (95% CI: -3.2 to -2.2); p<0.001] and ischemic heart disease [AAPC: -4.1 (95% CI: -4.5 to -3.6); p<0.001] mortality was seen in the non-stroke belt region from 1999 to 2017. There was no overall change in heart failure mortality in either regions (p for AAPC >0.05). The gap in age-adjusted mortality estimates for cardiovascular cause of death was 11.8% in 1999 and was 16% in 2017 across the two regions ( Figure 1 ). The mortality gaps were persistent across sub-groups of age, sex, race, and level of urbanization. Conclusions: Despite the overall decline in cardiovascular mortality, significant geographic disparities in cardiovascular mortality persist. Preventive efforts targeting risk factors and improved disease management may attenuate the longstanding geographical heterogeneity in cardiovascular mortality.

Is a switch to a different electronic health record associated with a change in patient satisfaction?

ObjectiveFinancial impacts associated with a switch to a different electronic health record (EHR) have been documented. Less attention has been focused on the patient response to an EHR switch. The Mayo Clinic was involved in an EHR switch that occurred at 6 different locations and with 4 different “go-live” dates. We sought to understand the relationship between patient satisfaction and the transition to a new EHR.Materials and MethodsWe used patient satisfaction data collected by Press Ganey from July 2016 through December 2019. Our patient satisfaction measure was the percent of patients responding “very good” (top box) to survey questions. Twenty-four survey questions were summarized by Press Ganey into 6 patient satisfaction domains. Piecewise linear regression was used to model patient satisfaction before and after the EHR switch dates.ResultsSignificant drops in patient satisfaction were associated with the EHR switch. Patient satisfaction with access (ease of getting clinic on phone, ease of scheduling appointments, etc.) was most affected (range of 6 sites absolute decline: -3.4% to -8.8%; all significant at 99% confidence interval). Satisfaction with providers was least affected (range of 6 sites absolute decline: -0.5% to -2.8%; 4 of 6 sites significant at 99% confidence interval). After 9-15 months, patient satisfaction with access climbed back to pre-EHR switch levels.ConclusionsPatient satisfaction in several patient experience domains dropped significantly and stayed lower than pre–“go-live” for several months after a switch in EHR. Satisfaction with providers declined less than satisfaction with access.

National Trends and Outcomes of Obesity and Morbid Obesity Complicating Benign Hysterectomy [28B

INTRODUCTION: Nearly 40% of women in the United States are obese, presenting unique challenges and considerations for gynecologic surgeons. We sought to examine trends and outcomes of benign hysterectomy in obese women. METHODS: A population-based retrospective study utilizing the National Inpatient Sample was conducted including benign total hysterectomies from 2001–2015 (N=2,680,788). Temporal trends of obesity, perioperative complications, and mortality rates were assessed with piecewise linear regression with log transformation. Pairwise propensity score inverse probability of treatment weighting with a generalized estimating equation was then used to estimate risk of perioperative complications and mortality. RESULTS: Of all women who underwent hysterectomy, 172,517 (6.4%) had Class I-II obesity and 83,513 (3.1%) had Class III (morbid) obesity. During the study period, Class III obesity increased from 0.6% to 6.4% (11.1-fold increase) and Class I-II obesity increased from 1.8% to 11.4% (6.5-fold increase) (both, P<.001). Compared to non-obese women, women with Class I-II (13.2% versus 11.0%, odds ratio [OR] 1.22, 95% confidence interval [CI] 1.18–1.27) and Class III (17.0% versus 11.0%, OR 1.40, 95% CI 1.34–1.47) obesity had significantly increased risk of perioperative complications. Mortality was increased two-fold in women with Class III obesity compared to non-obese women (0.06% versus 0.02%, OR 2.01, 95% CI 1.06–3.81). Over the study period, hysterectomy complication rates increased from 7.4% to 17.0% (2.3-fold increase, P<.001), and there was statistically non-significant increase in surgical mortality (0.01% to 0.03%, 2.4-fold increase, P=.052). CONCLUSION: Obesity, particularly morbid obesity, complicating benign hysterectomy has significantly increased in the United States and is associated with increased perioperative morbidity and mortality.

Assessing cyclone disturbances (1988–2016) in the Sundarbans mangrove forests using Landsat and Google Earth Engine

Cyclone disturbances can cause significant damage to forest vegetation. The Sundarbans spreading across Bangladesh and India, the world’s largest mangrove forest, is frequently exposed to cyclones of various magnitudes. However, the extent and pattern of forest disturbances caused by cyclones in the Sundarbans (both parts) remain poorly understood, and a long-term dataset focused on cyclones and forest disturbances is required. In this study, Google Earth Engine and Landsat images were used to evaluate changes in the normalized difference vegetation index (NDVI) before versus after 21 cyclones that occurred between 1988 and 2016. Supervised classification successfully classified the forest area with an overall accuracy of 86% and Kappa coefficient of 0.80. The percentage of affected forest area (i.e., the area that exhibited negative changes in NDVI values following a cyclone) ranged from 0.5 to 24.1% of the total forest area. Of the 21 focal cyclones, 18 affected less than 10% of the forest area, while two cyclones, Sidr in 2007 (category H5) and a cyclone in 1988 (category H3), affected 24.1% and 20.4%, respectively. Among the cyclone parameters (i.e., maximum wind speed, distance from the Sundarbans, and river water level), wind speed was significantly and positively correlated with affected forest area. Wind speed and affected forest area were nonlinearly related indicated by the piecewise linear regression and cubic regression. The piecewise model estimated a threshold point, suggesting that wind speed had little effects below a breakpoint of 101.9 km h−1. Our analyses, based on a 29-year dataset, suggest that, although the region experienced cyclones almost every year, only the largest cyclones (i.e., in the H3 category or higher) affected 20% or more of the mangrove forest area, and these occurred around once per 7- to 12-year period. Trees with broken stems or uprooted canopies as a result of strong winds are likely to contribute to the reduction in NDVI in the aftermath of a cyclone. From a long-term perspective, such rare yet intense cyclones may have a significant effect on regeneration and species composition in the Sundarbans mangrove forest. Since previous studies only focused on a few cyclones, our results based on 21 cyclones will certainly help better understanding of the effects of cyclones on mangrove forest disturbance.

Association between ambient temperature and COVID-19 infection in 122 cities from China

BackgroundCoronavirus disease 2019 (COVID-19) has become a severe public health problem globally. Both epidemiological and laboratory studies have shown that ambient temperature could affect the transmission and survival of coronaviruses. This study aimed to determine whether the temperature is an essential factor in the infection caused by this novel coronavirus. MethodsDaily confirmed cases and meteorological factors in 122 cities were collected between January 23, 2020, to February 29, 2020. A generalized additive model (GAM) was applied to explore the nonlinear relationship between mean temperature and COVID-19 confirmed cases. We also used a piecewise linear regression to determine the relationship in detail. ResultsThe exposure-response curves suggested that the relationship between mean temperature and COVID-19 confirmed cases was approximately linear in the range of <3 °C and became flat above 3 °C. When mean temperature (lag0–14) was below 3 °C, each 1 °C rise was associated with a 4.861% (95% CI: 3.209–6.513) increase in the daily number of COVID-19 confirmed cases. These findings were robust in our sensitivity analyses. ConclusionsOur results indicate that mean temperature has a positive linear relationship with the number of COVID-19 cases with a threshold of 3 °C. There is no evidence supporting that case counts of COVID-19 could decline when the weather becomes warmer, which provides useful implications for policymakers and the public.

Alterations of hand sensorimotor function and cortical motor representations over the adult lifespan.

Using a cross sectional design, we aimed to identify the effect of aging on sensorimotor function and cortical motor representations of two intrinsic hand muscles, as well as the course and timing of those changes. Furthermore, the link between cortical motor representations, sensorimotor function, and intracortical inhibition and facilitation was investigated. Seventy-seven participants over the full adult lifespan were enrolled. For the first dorsal interosseus (FDI) and abductor digiti minimi (ADM) muscle, cortical motor representations, GABAA-mediated short-interval intracortical inhibition (SICI), and glutamate-mediated intracortical facilitation (ICF) were assessed using transcranial magnetic stimulation over the dominant primary motor cortex. Additionally, participants’ dexterity and force were measured. Linear, polynomial, and piecewise linear regression analyses were conducted to identify the course and timing of age-related differences. Our results demonstrated variation in sensorimotor function over the lifespan, with a marked decline starting around the mid-thirties. Furthermore, an age-related reduction in cortical motor representation volume and maximal MEP of the FDI, but not for ADM, was observed, occurring mainly until the mid-forties. Area of the cortical motor representation did not change with advancing age. Furthermore, cortical motor representations, sensorimotor function, and measures of intracortical inhibition and facilitation were not interrelated.

Profile of regional fat and fat-free soft tissue accumulation in male athletes

BackgroundIt is unclear whether or not the breakpoint (BP), at which the proportion of each of fat mass (FM) and fat-free soft tissue mass (FFSTM) to body mass (BM) alter, exists in male athletes. We examined the hypothesis that in male athletes, the regional FM and FFSTM-BM relationships have a BP, but the body mass at BP (BMBP) differs among the arms, trunk, and legs.MethodsBy using a dual X-ray absorptiometry, whole-body and regional FMs and FFSTMs in the arms, trunk, and legs were estimated in 198 male athletes (20.8 ± 2.1 years; 1.73 ± 0.07 m; 72.7 ± 14.8 kg). To detect the BP in the relationship between each of FM and FFSTM and BM, a piecewise linear regression analysis was used. If a BP was detected in the corresponding relationship, the significant difference between the regression slopes above and below the BP was examined.ResultsThe regression analysis indicated that the BMBP existed in the FM- and FFSTM-BM relationships regardless of region and whole body. For the whole body, BMBP was 81.8 kg for FM and 82.2 kg for FFSTM. In regional FM-BM relationships, BMBP was 80.5 kg for arms, 82.6 kg for trunk, and 63.3 kg for legs, and the regression slopes above the BMBP became higher than those below the BP, and vice versa in regional FFSTM-BM relationships (BMBP 104.6 kg for arms, 80.9 kg for trunk, and 79.0 kg for legs). The relative differences in the slopes between below and above BMBP in the regional FM-BM relationships were higher in the arms and trunk than in the legs, and those in the regional FFSTM-BM relationships in the legs than in the trunk.ConclusionWhole-body and regional FM- and FFSTM-BM relationships for male athletes have breakpoints at which the proportion of the tissue masses to BM alters. The BMBP and differences in the distribution of regional FM and FFSTM around the breakpoint are region specific.

Hotspots of biotic compositional change in lakes along vast latitudinal transects in northern Canada.

Ecotones mark zones of rapid change in ecological structure at various spatial scales. They are believed to be particularly susceptible to shifts caused by environmental transformation, making them key regions for studying the effects of global change. Here, we explored the variation in assemblage structure of aquatic primary producer and consumer communities across latitudinal transects in northeastern North America (Québec-Labrador) to identify spatial patterns in biodiversity that indicated the location of transition zones across the landscape. We analyzed species richness and the cumulative rate of compositional change (expressed as beta-diversity) of diatoms and chironomids to detect any abrupt shifts in the rate of spatial taxonomic turnover. We used principal coordinates analysis to estimate community turnover with latitude, then applied piecewise linear regression to assess the position of ecotones. Statistically significant changes in assemblage composition occurred at 52 and 55°N, corresponding to the transition between closed- and open-crown forest, and to the southern onset of the forest tundra (i.e., the forest limit), respectively. The spatial distribution of ecotones was most strongly related to air temperature for chironomids and to vegetation- and soil-related chemical attributes of lake water for diatoms, including dissolved organic carbon content and water color. Lakes at mid- to high-latitudes currently face pressures from rapidly rising temperatures, accompanied by large increases in organic carbon inputs from their catchments, often leading to browning and its associated effects. The biota at the base of food webs in lakes located in transition zones are disproportionately affected by the cascading effects of these multi-factorial changes, concurrent with pronounced terrestrial greening observed in these regions. Similar patterns of biotic shifts have been observed along alpine aquatic transects, indicating the potential for widespread restructuring of cold, high-altitude and high-latitude freshwater communities due to global change.

The use of internal heart rate loggers in determining cardiac breakpoints of fish.

As marine environments are influenced by global warming there is a need to thoroughly understand the relationship between physiological limits and temperature in fish. One quick screening method of a physiological thermal tipping point is the temperature at which maximum heart rate (ƒHmax) can no longer scale predictably with warming and is referred to as the Arrhenius break temperature (TAB). The use of this method has been successful for freshwater fish by using external electrodes to detect an electrocardiogram (ECG), however, the properties of this equipment pose challenges in salt water when evaluating marine fish. To overcome these challenges, this study aimed to explore the potential use of implantable heart rate loggers to quantify the TAB of Chrysoblephus laticeps, a marine Sparid, following the ECG method protocols where ƒHmax is monitored over an acute warming event and the TAB is subsequently identified using a piece-wise linear regression model. Of the nine experimental fish, only five (56%) returned accurate ƒHmax data. The TAB of successful trials was identified each time and ranged from 18.09 to 20.10 °C. This study therefore provides evidence that implantable heart rate loggers can estimate TAB of fish which can be applied to many marine species.

Survival between synchronous and non-synchronous multiple primary cutaneous melanomas-a SEER database analysis.

BackgroundThere is no criterion to distinguish synchronous and non-synchronous multiple primary cutaneous melanomas (MPMs). This study aimed to distinguish synchronous and non-synchronous MPMs and compare the survivals of them using the Surveillance, Epidemiology, and End Results database.MethodsSynchronous and non-synchronous MPMs were distinguished by fitting the double log transformed distribution of the time interval between the first and second primary cutaneous melanomas (TIFtS) through a piecewise linear regression. The overall and melanoma-specific survivals were compared by the Kaplan–Meier method and Cox proportional hazard model through modeling the occurrence of synchronous MPMs as a time-dependent variable.ResultsThe distribution of TIFtS was composed by three power-law distributions. According to its first inflection point, synchronous MPMs were defined as tumors that occurred within 2 months. The Kaplain–Meier plot revealed a significant inferior survival for synchronous MPMs than non-synchronous MPMs (P < 0.0001), and the occurrence of synchronous MPM was a risk factor for overall survival of cutaneous melanoma (CM) (hazard ratio: 2.213; (95% CI [2.087–2.346]); P < 0.0001).ConclusionsThis study provided data analysis evidences for using 2 months to distinguish synchronous MPMs and non-synchronous MPMs. Furthermore, the occurrence of synchronous MPM was a risk factor for prognosis of patients with CM.

372 Baseline Gonadotropin Levels do not Predict Testosterone Response in Hypogonadal Men Treated with Clomiphene Citrate

Clomiphene citrate (CC) is commonly used in the treatment of male hypogonadism/infertility and stimulates pituitary gonadotropin release. Patients variably respond to CC treatment, but whether baseline gonadotropin levels can be used to select patients for CC treatment is unknown. To investigate the role of baseline gonadotropins in predicting the biochemical response to CC treatment. Retrospective review of data from hypogonadal men treated with CC in two high-volume fertility centers was performed between 2013 and 2018. Patient age, body mass index (BMI), and baseline hormones (follicle stimulating hormone [FSH], luteinizing hormone [LH], and total testosterone [TT]) were obtained. Men on exogenous testosterone or gonadotropins within 6 months of CC treatment were excluded. CC starting dose varied from 25-50 mg every day to every other day. Response in treatment was measured as changes in continuous TT levels within 6 months of initiating CC treatment. Linear and curvilinear (quadratic fit) correlations were assessed between baseline gonadotropins and changes in TT levels. We defined the optimal slope transition point, fitting a piecewise linear regression at a threshold that maximized the multiple R statistics. Linear regression models adjusted for age, BMI, and time on CC therapy were fitted to assess the associations between baseline LH and FSH levels with treatment response.

Soil Moisture Monitoring of the Plant Root Zone by Using Phenology as Context in Remote Sensing

In this study, the phenological behavior and energy balance of plants are used as a sensory mechanism for root-zone soil moisture monitoring using both in-situ and satellite remote sensing data. The commonly used in-situ measurements are not feasible for mapping soil moisture at large-scale agricultural areas. Local direct root-zone soil moisture measurements cannot be reliably interpolated owing to the high spatial variability of soil structure and the vegetative content. Remote sensing methods are negatively affected by vegetation coverage and density regarding penetration and backscattering characteristics. In order to overcome these limitations, we propose a root-zone soil moisture estimation method utilizing a context-aware data clustering process, which can be applied prior to any statistical analysis, for empirical evaluation of data. In this aspect, the crops’ phenological stages and soil–air temperature differences are defined as the two contexts for data clustering. Parameters such as canopy–air temperature difference, land surface temperature, and solar radiation with respect to plant energy and water processes are used for the analysis. The proposed model is utilized using piecewise linear regression of data obtained from 16 rainfed wheat parcels distributed across Turkey, under different climatic and topographic conditions. It is shown that the proposed context-aware data clustering process enables the nonlinear plant behavior to be analyzed linearly. The correlation value of the whole season increased from 21% to a range between 78% and 95% for different clusters. The outliers became relevant and the parameters became significant after the proposed context-aware data clustering.

Normal QT interval in Ukrainian Riding Horses at rest and during exercise

Measurement of the main parameter of cardiac repolarization, namely QT interval, has a very high diagnostic value in human medicine, since its irregularities may indicate severe life-threatening ventricular tachyarrhythmias. The QT interval may vary not only with heart rate, age, sex, and autonomic tone, but also with horse breeds. Therefore, the description of its reference values for a specific breed is of great importance. The Ukrainian Riding Horse was bred as a show jumping, dressage and three-day eventing breed on the basis of Hanoverian, Thoroughbred and Trakehner stallions and local mares, as well as Hungarian Furioso, Gidran Arab and Nonius mares. Twenty-three horses of the Ukrainian riding breed were included in the study: 8 geldings, 8 mares, and 7 stallions aged 3-11 years. The electrodes for ECG recording were placed according to an adapted base-apex system. The ECG was registered during rest, exercise, and recovery periods. QT intervals were measured from resting to peak exercise levels on the traces of the 2nd lead and plotted against RR intervals. The piecewise regression model was fitted to the data plot. The values of Slope1, Slope2, and RRbend were compared to those of other breeds. The QT/RR relationship was relatively described by the piecewise linear regression model for all sexes (0.95 &lt; r2 &lt; 0.97). The sex of horses of the Ukrainian riding breed had a significant effect on the model. In terms of Slope1, Slope2, and RRbend values, Ukrainian riding horses are closest to Warmbloods, Standardbreds, and Thoroughbreds. The QT interval in horses should be corrected for breed and sex.

Body mass index (BMI) is associated with serum thyroid-stimulating hormone (TSH) level in infertile women: a cross-sectional study.

Studies suggest that there is a relationship between body mass index (BMI) and thyroid-stimulating hormone (TSH) levels. But conflicting evidence exists regarding the relationship between the two variables. Moreover, thyroid function is closely related to female fertility and has certain effects on infertility. Therefore, the present study will explore the relationship between BMI and TSH levels in patients with infertility in our center. We retrospectively analyzed relevant indicators of 2,789 in Tubal Factor Infertility patients undergoing assisted reproduction technology from January 2016 to December 2018 in our center in order to analyze the relationship between BMI and serum TSH level. The medical histories of patients were reviewed. The relationship between BMI and TSH was assessed using smooth curve fitting and multivariate regression model. The smoothing curve fitting between BMI and TSH exhibited a non-linear relationship, and the resulting curve exhibited a two-stage change and a breakpoint. By multivariate piecewise linear regression, we found that the TSH level was increased with the increase of BMI when the BMI was greater than 25.3 kg/m2 (β 0.06, 95% CI 0.02, 0.01; p = 0.0028). In contrast, the TSH level was decreased with the increase of BMI when the BMI was less than 25.3 kg/m2 (β -0.02, 95% CI -0.05, 0.00; p = 0.0573). Collectively, our study described a non-linear relationship between BMI and TSH level in infertile patients after adjustment of potential confounders. However, such causal relationship between BMI and TSH in infertile women still needs to be further clarified in future investigations.

Pioneering Methods for Enhancing PPI and Phenotype Networks for Candidate Disease Prioritization

The physical contacts of high-specificity between two or more protein molecules constitute Protein-Protein Interactions (PPIs). PPI networks are modeled through graphs where node denotes proteins and edges denote interaction between proteins. The PPI network plays an important role to identify the interesting disease gene candidates. But, the PPI network usually contains false interactions. Many techniques have been proposed to reconstruct PPI network to remove false interactions and improve ranking of candidate disease. Random Walk with Restart on Diffusion profile (RWRDP) and Random Walk on a Reliable Heterogeneous Network (RWRHN) was two among them. In these methods, Gene topological similarity was incorporated with original PPI network to reconstruct new PPI network. Phenotype network was constructed by calculating similarity between gene phenotypes. The reconstructed network and phenotype networks were combined to rank candidate disease genes. However, the PPI reconstruction was fully related with the quality of protein interaction data. In order to enhance the reconstruction of PPI, a Piecewise Linear Regression (PLR) based protein sequence similarity measure and Bat Algorithm based gene expression similarity were proposed with RHN. In this paper, additional measure called Interaction Level Sub cellular Localization Score (ILSLS) is proposed to further reduce the false interaction in the reconstruction of PPI network. ILSLS is the combination of Normalized Sub cellular Localization score (NSL) and Protein Multiple Location Prediction score (PMLP). The proposed work is named as Random Walker on Optimized Trustworthy Heterogeneous Sub Cellular localization aware Network (RW-OTHSN). In order to enhance the ranking of RW-OTHSN, phenotype structure is considered while construction phenotype network to rank the candidate disease genes. The phenotype structure is characterized based on h*-sequence model which identify highly discriminative signatures with only a small number of genes. This proposed work is named as Random Walker on Optimized Trustworthy Heterogeneous Sub Cellular localization and Phenotype structure aware Network (RW-OTHSPN). The efficiency of the proposed methods are evaluated on PPI network database in terms of Average degree, Relative Frequency for PPI reconstruction, Number of successful predictions, precision and recall for candidate disease gene ranking.

Combining regression and mixed-integer programming to model counterinsurgency

Counterinsurgencies are a type of violent struggle between state and non-state actors in which one group attempts to gain or maintain influence over a certain portion of the population. When an insurgency (i.e., non-state actor) challenges a host nation (i.e., state actor), often an external counterinsurgent force intervenes. While researchers have categorized insurgencies with social science techniques and United States Army doctrine has established possible counterinsurgency strategies, little research prescribes host nation and counterinsurgent force strength. To this end, we develop a mixed-integer program to provide an estimate of the number of forces required to maximize the probability of a favorable resolution to the counterinsurgent and host nation countries, while minimizing unfavorable resolutions and the number of counterinsurgent deaths. This program integrates: (i) a multivariate piecewise-linear regression model to estimate the number of counterinsurgent deaths each year and (ii) a logistic regression model to estimate the probability of four types of conflict resolution over a 15-year time horizon. Constraints in the model characterize: (i) upper and lower limits on the number of counterinsurgent and host nation forces and their annual rates of increase and decrease, (ii) the characteristics of the type of counterinsurgency, (iii) an estimation of the number of counterinsurgent deaths, and (iv) an estimation of the probability of one of four resolutions. We use Somalia as a case study to estimate how counterinsurgent strategies affect the probability of obtaining each conflict resolution. We conclude that a strategy focusing on building and empowering a stable host nation force provides the highest probability of achieving a positive resolution to the counterinsurgency. Senior leaders can use this information to guide strategic decisions within a counterinsurgency.

Continuous Wetting on the Tibetan Plateau during 1970–2017

Based on daily observation records at 277 meteorological stations on the Tibetan Plateau (TP) and its surrounding areas during 1970–2017, drought evolution was investigated using the Standardized Precipitation Evapotranspiration Index (SPEI). First, the spatiotemporal changes in the growing season of SPEI (SPEIgs) were re-examined using the Mann–Kendall and Sen’s slope approach—the piecewise linear regression and intensity analysis approach. Then, the persistence of the SPEIgs trend was predicted by the Hurst exponent. The results showed that the SPEIgs on the TP exhibited a significant increasing trend at the rate of 0.10 decade−1 (p &lt; 0.05) and that there is no significant trend shift in SPEIgs (p = 0.37), indicating that the TP tended to undergo continuous wetting during 1970–2017. In contrast, the areas surrounding the TP underwent a significant trend shift from an increase to a decrease in SPEIgs around 1984 (p &lt; 0.05), resulting in a weak decreasing trend overall. Spatially, most of the stations on the TP were characterized by an increasing trend in SPEIgs, except those on the Eastern fringe of TP. The rate of drought/wet changes was relatively fast during the 1970s and 1980s, and gradually slowed afterward on the TP. Finally, the consistent increasing trend and decreasing trend of SPEIgs on the TP and the area East of the TP were predicted to continue in the future, respectively. Our results highlight that the TP experienced a significant continuous wetting trend in the growing season during 1970–2017, and this trend is likely to continue.

The Association between the 2009 WIC Food Package Change and Early Childhood Obesity Risk Varies by Type of Infant Package Received

BackgroundIn 2009, the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) food packages were updated. WIC-participating children in Los Angeles County receiving the new food package, compared with the old, had lower obesity risk at age 4. ObjectiveTo determine if the association between the new WIC food package and obesity varies by type of infant WIC food package received. DesignLongitudinal study using administrative data on children participating in WIC in Los Angeles County continuously from birth until age 4. Children were compared across periods: Before (period 1: born 2003-2004), straddling (period 2: born 2005-2008), and after (period 3: born 2009-2011) the food package change. Children were further grouped as Fully Breastfed, Mostly Breastfed, Mostly Formula Fed, and Fully Formula Fed based on the type of food packages received during the first year of life. Participants/settingWIC-participating children in Los Angeles County between 2003 and 2016 (N=116,991). Main outcome measuresWeight-for-height z-score growth trajectories from 0 to 4 years and obesity (body mass index-for-age≥95th percentile) at age 4. Statistical analyses performedChildren were matched across periods on infant feeding group; age, sex, race or ethnicity, and initial weight status; maternal education and language; and family poverty. Sex-stratified piecewise linear spline mixed models and Poisson regression models were fit to the data. ResultsNo differences across periods were observed for children in the Fully Breastfed group. Boys in the Mostly Breastfed (relative risk [RR]=1.27, 95% CI=1.17 to 1.38), Mostly Formula Fed (RR=1.07, 95% CI=1.02 to 1.13), and Fully Formula Fed (RR=1.13, 95% CI=1.06 to 1.20) groups in period 1 had higher obesity risk than their counterparts in period 3; girls in the Mostly Breastfed group in period 1 had a higher obesity risk than those in period 3 (RR=1.17, 95% CI=1.07 to 1.28). ConclusionsThe association between the WIC food package change and obesity varied by type of infant food package received, with the strongest associations observed for those who were mostly breastfed.

Adaptive detection of FOREX repetitive chart patterns

The global financial ecosystem has evolved and matured along with the ever-changing world economy that grew increasingly complicated due to globalisation. As traders are often inundated with information from various sources when formulating trading strategies, numerous analysis methods have been developed to ease the decision-making process. However, factors such as prior experience and knowledge of the trader as well as various psychological factors often influence the final trading decision. Focusing on charting-based analysis, it still suffers from drawbacks due to the time-warping properties of the chart patterns and the reliance on a large number of pre-defined chart patterns. Hence, in order to address the gaps within the FOREX research, the paper endeavours to propose a novel chart detection algorithm. The auto-segmentation implementation within the algorithm utilises piecewise linear regression to detect chart patterns within the FOREX historical data. By successfully extracting the repetitive chart patterns and subsequently establishing its similarities using Agglomerative Hierarchical Clustering, the information provided could potentially be used to assist traders in solidifying their investment decisions. The experimental results obtained show that repetitive chart patterns can indeed be successfully detected and extracted from the FOREX historical data.

Global Estimation of the Effect of Fine Particulate Air Pollution on Life Expectancy at the Age of 60

Background: Previous studies evaluating the association between particulate matter less than 2·5 μm in diameter (PM2·5) and life expectancy were conducted in a single country, assuming a linear relationship. In our study, we evaluated the association using global country-level data accounting for a non-linear relationship between PM2·5 concentration and life expectancy. Methods: A total of 17 years (from 2000 to 2016) data of country-level population-weighted PM2·5 concentration and life expectancy at age 60 were extracted from the Global Burden of Disease Study 2017 database and the World Health Organization Global Health Observatory. A generalized estimating equation with a compound symmetry covariance structure was used to evaluate the linear association between a reduction in annual PM2·5 concentration and an increase in life expectancy. Non-parametric analyses using generalized additive mixed model and piecewise linear regression were conducted for analyses of repeated measures data with non-linear assumption. Findings: Seventeen years of data from 176 countries were included in our analysis. With a linear assumption, 10 μg/m3 reduction in annual PM2·5 concentration was associated with 0·34 years (95%CI, 0·20, 0·49) of increase in life expectancy. By assuming a non-linear relationship, life expectancy increased by 0·72 years (95%CI: 0·49, 0·96) per 10 μg/m3 reduction in 0-24·0 μg/m3 PM2·5 concentration range. However, life expectancy increased by 0·01 years (95%CI: -0·06, 0·08) per 10 μg/m3 reduction in PM2·5 above 24·0 μg/m3 concentration. Interpretation: There was a non-linear relationship between country-level population-weighted annual PM2·5 concentration and life expectancy at age 60. Expected life expectancy gain for a unit decrease in PM2·5 was greater at low PM2·5 concentration range (0-24·0μg/m3) than that over 24·0 μg/m3. Future individual and multi-level studies are needed to confirm our ecologic study findings. Funding Statement: Ministry of Science, ICT, and Future Planning, Korea Centre for Disease Control and Prevention, and the Centre for Environmental Health, Ministry of Environment, Korea Declaration of Interests: The authors declare no competing interests. Ethics Approval Statement: Not required.