Data and Methods

Abigail Aiken • Jeanne Altmann • Elizabeth Armstrong • Susan Fiske • Noreen Goldman • Bryan Grenfell • Tod G. Hamilton • Sara F. McLanahan • C. Jessica E. Metcalf • Germán Rodríguez • James Trussell


Jeanne Altmann’s research deals with life history approaches to behavioral ecology and with nonexperimental research design. Altmann emphasizes an integrated, holistic approach by carrying out concurrent studies of behavior, ecology, demography, genetics, and physiology at the level of individuals, social groups, and populations.


Jeanne Altmann’s current research centers on the magnitude and sources of variability in primate life histories, parental care, and behavioral ontogeny. Recently, with her collaborators, Altmann has been conducting studies that relate endocrine and genetic data to demographic and behavioral information for the same individuals in the Amboseli baboon population.


In studies of monogamous Peromyscus in captivity, Jeanne Altmann is investigating behavioral causes and consequences of inbreeding depression and of mate choice. Initial studies examined behavioral risk factors and experimentally separated effects of parental and offspring inbreeding on inbreeding depression.


With Miranda Waggoner, Elizabeth Armstrong is working on a project that examines the uses of data from the Dutch Hunger Winter. During the winter of 1944-45, Nazi forces occupied the western provinces of the Netherlands, cutting off food and fuel shipments to the area. A severe famine ensued, which came to be known as the Dutch Hunger Winter, affecting some 4-5 million people. The health consequences of the famine have been extensively studied; in particular, data on the effects of exposure to famine in utero collected through the Dutch Famine Birth Cohort Study have become paradigmatic within epidemiology and in the emerging field of epigenetics. In addition, these data have been discussed extensively in the obstetric literature, the popular press, and increasingly, in social sciences like economics. This project examines patterns of dissemination and interpretation of evidence from the Dutch Hunger Winter through time and disciplinary space.


Existing moral psychology research commonly explains certain phenomena in terms of a motivation to blame. However, this motivation is not measured directly, but rather is inferred from other measures, such as participants’ judgments of an agent’s blameworthiness. With Daniel L. Ames (Princeton University) Susan Fiske published, “Perceived Intent Motivates People to Magnify Observed Harm,” in the Proceedings of the National Academy of Sciences (PNAS). This paper introduces new methods for assessing this theoretically important motivation, using tools drawn from animal-model research. Fiske and Ames test these methods in the context of recent “harm-magnification” research, which shows that people often overestimate the damage caused by intentional (versus unintentional) harms. A preliminary experiment exemplifies this work and also rules out an alternative explanation for earlier harm-magnification results. Experiment 1 asks whether intended harm motivates blame or merely demonstrates the actor’s intrinsic blameworthiness. Consistent with a motivational interpretation, participants freely chose blaming, condemning, and punishing over other appealing tasks in an intentional-harm condition, compared with an unintentional-harm condition. Experiment 2 also measures motivation but with converging indicators of persistence (effort, rate, and duration) in blaming. In addition to their methodological contribution, these studies also illuminate people’s motivational responses to intentional harms. Perceived intent emerges as catalyzing a motivated social cognitive process related to social prediction and control.


Michael S. North (New York University) and Susan Fiske examine “Modern Attitudes Toward Older Adults in the Aging World: A Cross-Cultural Meta-Analysis” in their paper published in Psychological Bulletin. Prevailing beliefs suggest that Eastern cultures hold older adults in higher esteem than Western cultures do, due to stronger collectivist traditions of filial piety. However, in modern, industrialized societies, the strain presented by dramatic rises in population aging potentially threatens traditional cultural expectations. Addressing these competing hypotheses, a literature search located 37 eligible papers, comprising samples from 23 countries and 21,093 total participants, directly comparing Easterners and Westerners (as classified per U.N. conventions) in their attitudes toward aging and the aged. Contradicting conventional wisdom, a random-effects meta-analysis on these articles found such evaluations to be more negative in the East overall (standardized mean difference = -0.31). High heterogeneity in study comparisons suggested the presence of moderators; indeed, geographical region emerged as a significant moderating factor, with the strongest levels of senior derogation emerging in East Asia (compared with South and Southeast Asia) and non-Anglophone Europe (compared with North American and Anglophone Western regions). At the country level, multiple-moderator meta-regression analysis confirmed recent rises in population aging to significantly predict negative elder attitudes, controlling for industrialization per se over the same time period. Unexpectedly, these analyses also found that cultural individualism significantly predicted relative positivity—suggesting that, for generating elder respect within rapidly aging societies, collectivist traditions may backfire. The findings suggest the importance of demographic challenges in shaping modern attitudes toward elders—presenting considerations for future research in ageism, cross-cultural psychology, and even economic development, as societies across the globe accommodate unprecedented numbers of older citizens.


In early 2016, Noreen Goldman will be fielding a short follow-up survey of the Taiwan participants to obtain updated information on health and functional performance. A proposal is underway to use these extensive longitudinal data to determine rates of senescence and linkages with stressful experience, social status and survival. She is beginning a project in collaboration with researchers at UCLA on the relationship between occupation and disability among middle-aged and older adults in Mexico. Goldman foresees an extension of this work to examine such associations among Mexican Americans to explore what appears to be one of many paradoxes of Latino health, namely why it is that Mexicans in the U.S. experience higher disability rates than whites yet live longer. High-risk occupations and poor employment conditions may provide a partial answer.


Bryan Grenfell’s research continued to focus on combining basic developments in infectious disease dynamics with application to public health. This year’s research result is the identification of a major new disease threat from measles, deriving from the prolonged immune-modulation that follows measles infection which can increase mortality from other infections.


Bryan Grenfell and co-authors, Amy Wesolowski (Harvard School of Public Health), C. Jessica Metcalf, Nathan Eagle (Harvard School of Public Health and Northeastern University), Janeth Kombich (University of Kabianga, Kericho Country, Kenya) et al. published, “Quantifying Seasonal Population Fluxes Driving Rubella Transmission Dynamics Using Mobile Phone Data” in PNAS. This paper reviews changing patterns of human aggregation which are thought to drive annual and multiannual outbreaks of infectious diseases. However, the paucity of data about travel behavior and population flux over time has made this idea difficult to test quantitatively. Current measures of human mobility, especially in low-income settings, are often static, relying on approximate travel times, road networks, or crosssectional surveys. Mobile phone data provide a unique source of information about human travel, but the power of these data to describe epidemiologically relevant changes in population density remains unclear. Here they quantify seasonal travel patterns using mobile phone data from nearly 15 million anonymous subscribers in Kenya. Using a rich data source of rubella incidence, they show that patterns of population travel (fluxes) inferred from mobile phone data are predictive of disease transmission and improve significantly on standard school term time and weather covariates. Further, combining seasonal and spatial data on travel from mobile phone data allows us to characterize seasonal fluctuations in risk across Kenya and produce dynamic importation risk maps for rubella. Mobile phone data therefore offer a valuable previously unidentified source of data for measuring key drivers of seasonal epidemics.


In “Understanding Herd Immunity,” published in Trends in Immunology, C. Jessica Metcalf, M. Ferrari (Pennsylvania State University), A. L. Graham (Princeton University, Department of Ecology and Evolutionary Biology), and Bryan Grenfell state that individual immunity is a powerful force affecting host health and pathogen evolution. Importantly, the effects of individual immunity also scale up to affect pathogen transmission dynamics and the success of vaccination campaigns for entire host populations. Population-scale immunity is often termed 'herd immunity'. Here they outline how individual immunity maps to population outcomes and discuss implications for control of infectious diseases. Particular immunological characteristics may be more or less likely to result in a population level signature of herd immunity; we detail this and also discuss other population-level outcomes that might emerge from individual-level immunity.


In their paper, “Quantifying the Risk of Pandemic Influenza Virus Evolution by Mutation and Re-assortment,” published in Vaccine, Leslie A. Reperant (Artemis One Health Research Foundation), Bryan Grenfell, and Albert D. M. E. Osterhaus (Erasmus Medical Centre, The Netherlands) state that large outbreaks of zoonotic influenza A virus (IAV) infections may presage an influenza pandemic. However, the likelihood that an airborne-transmissible variant evolves upon zoonotic infection or co-infection with zoonotic and seasonal IAVs remains poorly understood, as does the relative importance of accumulating mutations versus re-assortment in this process. Using discrete-time probabilistic models, we determined quantitative probability ranges that transmissible variants with 1–5 mutations and transmissible re-assortants evolve after a given number of zoonotic IAV infections. The systematic exploration of a large population of model parameter values was designed to account for uncertainty and variability in influenza virus infection, epidemiological and evolutionary processes. The models suggested that immunocompromised individuals are at high risk of generating IAV variants with pandemic potential by accumulation of mutations. Yet, both immunocompetent and immunocompromised individuals could generate high viral loads of single and double mutants, which may facilitate their onward transmission and the subsequent accumulation of additional 1–2 mutations in newly-infected individuals. This may result in the evolution of a full transmissible genotype along short chains of contact transmission. Although co-infection with zoonotic and seasonal IAVs was shown to be a rare event, it consistently resulted in high viral loads of re-assortants, which may facilitate their onward transmission among humans. The prevention or limitation of zoonotic IAV infection in immunocompromised and contact individuals, including health care workers, as well as vaccination against seasonal IAVs—limiting the risk of co-infection—should be considered fundamental tools to thwart the evolution of a novel pandemic IAV by accumulation of mutations and re-assortment.


Bryan Grenfell along with Sinead E. Morris (Princeton University, Department of Ecology and Evolutionary Biology), Jonathan L. Zelner (Columbia University), Deborah A. Fauquier (National Marine Fisheries Service, Marine Mammal Health and Stranding Response Program, Silver Spring.MD), Teresa K. Rowles (National Marine Fisheries Service, Southeast Fisheries Science Center, Lafayette, LA) et al. published, “Partially Observed Epidemics in Wildlife Hosts: Modelling an Outbreak of Dolphin Morbillivirus in the Northwestern Atlantic, in Journal of the Royal Society Interface. Morbilliviruses cause major mortality in marine mammals, but the dynamics of transmission and persistence are ill understood compared to terrestrial counterparts such as measles; this is especially true for epidemics in cetaceans. However, the recent outbreak of dolphin morbillivirus in the northwestern Atlantic Ocean can provide new insights into the epidemiology and spatiotemporal spread of this pathogen. To deal with uncertainties surrounding the ecology of this system (only stranded animals were observed), they develop a statistical framework that can extract key information about the underlying transmission process given only sparse data. Their self-exciting Poisson process model suggests that individuals are infectious for at most 24 days and can transfer infection up to two latitude degrees (220 km) within this time. In addition, the effective reproduction number is generally below one, but reaches 2.6 during a period of heightened stranding numbers near Virginia Beach, Virginia, in summer 2013. Network analysis suggests local movements dominate spatial spread, with seasonal migration facilitating wider dissemination along the coast. Finally, a low virus transmission rate or high levels of pre-existing immunity can explain the lack of viral spread into the Gulf of Mexico. More generally, their approach illustrates novel methodologies for analyzing very indirectly observed epidemics.


Bryan Grenfell, Corinne N. Thompson (Oxford University, London School of Hygiene & Tropical Medicine), Jonathan L. Zelner (Columbia University),, Tran Do Hoang Nhu (Oxford University), My V. T. Phan (Wellcome Trust Sanger Institute) et al published, “The Impact of Environmental and Climatic Variation on the Spatiotemporal Trends of Hospitalized Pediatric Diarrhea in Ho Chi Minh City, Vietnam,” in Health & Place. In this work, they authors write that it is predicted that the integration of climate-based early warning systems into existing action plans will facilitate the timely provision of interventions to diarrheal disease epidemics in resource-poor settings. Diarrhea remains a considerable public health problem in Ho Chi Minh City (HCMC), Vietnam and they aimed to quantify variation in the impact of environmental conditions on diarrheal disease risk across the city. Using all inpatient diarrheal admissions data from three large hospitals within HCMC, they developed a mixed effects regression model to differentiate district-level variation in risk due to environmental conditions from the overarching seasonality of diarrheal disease hospitalization in HCMC. They identified considerable spatial heterogeneity in the risk of all-cause diarrhea across districts of HCMC with low elevation and differential responses to flooding, air temperature, and humidity driving further spatial heterogeneity in diarrheal disease risk. The incorporation of these results into predictive forecasting algorithms will provide a powerful resource to aid diarrheal disease prevention and control practices in HCMC and other similar settings.


Ruthie Birger (Princeton University, Department of Ecology and Evolutionary Biology), Roger Kouyos (University Hospital Zürich and University of Zürich), Jonathan Dushoff (McMaster University), and Bryan Grenfell et al. authored, “Modeling the effect of HIV coinfection on clearance and sustained virologic response during treatment for hepatitis C virus in Epidemics. The paper’s background states that HIV/hepatitis C (HCV) coinfection is a major concern in global health today. Each pathogen can exacerbate the effects of the other and affect treatment outcomes. Understanding the within-host dynamics of these coinfecting pathogens is crucial, particularly in light of new, direct-acting antiviral agents (DAAs) for HCV treatment that are becoming available. In this study, they construct a within-host mathematical model of HCV/HIV coinfection by adapting a previously published model of HCV monoinfection to include an immune system component in infection clearance. We explore the effect of HIV-coinfection on spontaneous HCV clearance and sustained virologic response (SVR) by building in decreased immune function with increased HIV viral load. Treatment is modeled by modifying HCV burst-size, and they use clinically-relevant parameter estimates. Their model replicates real-world patient outcomes; it outputs infected and uninfected target cell counts, and HCV viral load for varying treatment and coinfection scenarios. Increased HIV viral load and reduced CD4+ count correlate with decreased spontaneous clearance and SVR chances. Treatment efficacy/duration combinations resulting in SVR are calculated for HIV-positive and negative patients, and crucially, we replicate the new findings that highly efficacious DAAs reduce treatment differences between HIV-positive and negative patients. However, we also find that if drug efficacy decays sufficiently over treatment course, SVR differences between HIV-positive and negative patients reappear. In conclusion, their model shows theoretical evidence of the differing outcomes of HCV infection in cases where the immune system is compromised by HIV. Understanding what controls these outcomes is especially important with the advent of efficacious but often prohibitively expensive DAAs. Using a model to predict patient response can lend insight into optimal treatment design, both in helping to identify patients who might respond well to treatment and in helping to identify treatment pathways and pitfalls.


Jemma L. Geoghegan (University of Sydney), Le Van Tan (Oxford University), Denise Kühnert (Department of Environmental Systems Science, Zürich and Swiss Institute of Bioinformatics), Rebecca A. Halpin (J. Craig Venter Institute, Rockville, MD) and Bryan Grenfell et al. published, “Phylodynamics of Enterovirus A71-Associated Hand, Foot and Mouth Disease in Viet Nam,” in Journal of Virology). Enterovirus A71 (EV-A71) is a major cause of hand, foot, and mouth disease (HFMD) and is particularly prevalent in parts of Southeast Asia, affecting thousands of children and infants each year. Revealing the evolutionary and epidemiological dynamics of EV-A71 through time and space is central to understanding its outbreak potential. We generated the full genome sequences of 200 EV-A71 strains sampled from various locations in Viet Nam between 2011 and 2013 and used these sequence data to determine the evolutionary history and phylodynamics of EV-A71 in Viet Nam, providing estimates of the effective reproduction number (Re) of the infection through time. In addition, we described the phylogeography of EV-A71 throughout Southeast Asia, documenting patterns of viral gene flow. Accordingly, their analysis reveals that a rapid genogroup switch from C4 to B5 likely took place during 2012 in Viet Nam. We show that the Re of subgenogroup C4 decreased during the time frame of sampling, whereas that of B5 increased and remained >1 at the end of 2013, corresponding to a rise in B5 prevalence. Their study reveals that the subgenogroup B5 virus that emerged into Viet Nam is closely related to variants that were responsible for large epidemics in Malaysia and Taiwan and therefore extends our knowledge regarding its associated area of endemicity. Subgenogroup B5 evidently has the potential to cause more widespread outbreaks across Southeast Asia.


Ruthie B. Birger (Princeton University, Department of Ecology and Evolutionary Biology), Roger D. Kouyos (University Hospital Zürich), Ted Cohen (Yale School of Public Health), Emily C. Griffiths (North Carolina State University, Raleigh), and Bryan Grenfell et al. published, “The Potential Impact of Coinfection on Antimicrobial Chemotherapy and Drug Resistance in Trends in Microbiology. The authors state that across a range of pathogens, resistance to chemotherapy is a growing problem in both public health and animal health. Despite the ubiquity of coinfection, and its potential effects on within-host biology, the role played by coinfecting pathogens on the evolution of resistance and efficacy of antimicrobial chemotherapy is rarely considered. In this review, they provide an overview of the mechanisms of interaction of coinfecting pathogens, ranging from immune modulation and resource modulation, to drug interactions. They discuss their potential implications for the evolution of resistance, providing evidence in the rare cases where it is available. Overall, their review indicates that the impact of coinfection has the potential to be considerable, suggesting that this should be taken into account when designing antimicrobial drug treatments.


Ph.D. candidate Christina Faust), Jonathan Zelner (National Institutes of Health, Bethesda, MD), Philippe Brasseur (L’Institut de Recherche pour le Développement), Michel Vaillant (Centre de Recherche Public), and Bryan Grenfell et al. published, “Assessing Drivers of Full Adoption of Test and Treat Policy for Malaria in Senegal,” in American Journal of Tropical Medicine and Hygiene . The abstract states that malaria treatment policy has changed from presumptive treatment to targeted “test and treat” (T&T) with malaria rapid diagnostic tests (RDTs) and artemisinin combination therapy (ACT). This transition involves changing behavior among health providers, meaning delays between introduction and full implementation are recorded in almost every instance. They investigated factors affecting successful transition, and suggest approaches for accelerating uptake of T&T. Records from 2000 to 2011 from health clinics in Senegal where malaria is mesoendemic were examined (96,166 cases). The study period encompassed the implementation of national T&T policy in 2006. Analysis showed that adherence to test results is the first indicator of T&T adoption and is dependent on accumulation of experience with positive RDTs (odds ratio [OR]: 0.55 [P = 0.001], 95% confidence interval [CI]: 0.53–0.58). Reliance on tests for malaria diagnosis (rather than presumptive diagnosis) followed after test adherence is achieved, and was also associated with increased experience with positive RDTs (OR: 0.60 [P = 0.001], 95% CI: 0.58–0.62). Logistic models suggest that full adoption of T&T clinical practices can occur within 2 years, that monitoring these behavioral responses rather than RDT or ACT consumption will improve evaluation of T&T uptake, and that accelerating T&T uptake by focusing training on adherence to test results will reduce overdiagnosis and associated health and economic costs in mesoendemic regions.


Michael J. Mina (Princeton University, Department of Ecology and Evolutionary Biology; Emory University), C. Jessica Metcalf, Rik L. de Swart (Erasmus University Medical Center), , A. D. M. E. Osterhaus (Erasmus Medical Centre, The Netherlands), and Bryan Grenfell et al. published, “Vaccines. Long-term Measles Induced Immunomodulation Increases Overall Childhood Infectious Disease Mortality,” in Science. Immunosuppression after measles is known to predispose people to opportunistic infections for a period of several weeks to months. Using population-level data, they show that measles has a more prolonged effect on host resistance, extending over two to three years. They find that nonmeasles infectious disease mortality in high-income countries is tightly coupled to measles incidence at this lag, in both the pre- and post-vaccine eras. They conclude that long-term immunologic sequelae of measles drive interannual fluctuations in nonmeasles deaths. This is consistent with recent experimental work that attributes the immunosuppressive effects of measles to depletion of B and T lymphocytes. Their data provide an explanation for the long-term benefits of measles vaccination in preventing all-cause infectious disease. By preventing measles-associated immune memory loss, vaccination protects polymicrobial herd immunity.


C. Jessica Metcalf, V. Andreasen (Universitetsvej, Denmark), O. N. Bjørnstad (Pennsylvania State University), K. Eames (London School of Hygiene & Tropical Medicine), and Bryan Grenfell et al. published, “Seven Challenges in Modelling Vaccine Preventable Diseases,” in Epidemics which states that vaccination has been one of the most successful public health measures since the introduction of basic sanitation. Substantial mortality and morbidity reductions have been achieved via vaccination against many infections, and the list of diseases that are potentially controllable by vaccines is growing steadily. The authors introduce key challenges for modeling in shaping our understanding and guiding policy decisions related to vaccine preventable diseases.


Thomas P. Van Boeckel (Princeton University, Department of Ecology and Evolutionary Biology), Charles Brower (Center for Disease Dynamics, Economics and Policy, Washington, DC) Marius Gilbert (Universite Libre de Bruxelles, Belgium; Fonds National de la Recherche Scientifique, Belgium), Bryan Grenfell, and Simon A. Levin (Princeton University, Department of Ecology and Evolutionary Biology; Beijer Institute of Ecological Economics,Sweden; and Resources for the Future, Washington, DC) et al. published, “Global Trends in Antimicrobial Use in Food Animals,” in Proc Natl Acad Sci. Demand for animal protein for human consumption is rising globally at an unprecedented rate. Modern animal production practices are associated with regular use of antimicrobials, potentially increasing selection pressure on bacteria to become resistant. Despite the significant potential consequences for antimicrobial resistance, there has been no quantitative measurement of global antimicrobial consumption by livestock. They address this gap by using Bayesian statistical models combining maps of livestock densities, economic projections of demand for meat products, and current estimates of antimicrobial consumption in high-income countries to map antimicrobial use in food animals for 2010 and 2030. They estimate that the global average annual consumption of antimicrobials per kilogram of animal produced was 45 mg•kg-1,148 mg•kg-1, and 172 mg•kg-1 for cattle, chicken, and pigs, respectively. Starting from this baseline, we estimate that between 2010 and 2030, the global consumption of antimicrobials will increase by 67%, from 63,151 ±1,560 tons to 105,596 ±3,605 tons. Up to a third of the increase in consumption in livestock between 2010 and 2030 is imputable to shifting production practices in middle-income countries where extensive farming systems will be replaced by large-scale intensive farming operations that routinely use antimicrobials in subtherapeutic doses. For Brazil, Russia, India, China, and South Africa, the increase in antimicrobial consumption will be 99%, up to seven times the projected population growth in this group of countries. Better understanding of the consequences of the uninhibited growth in veterinary antimicrobial consumption is needed to assess its potential effects on animal and human health.


Bryan Grenfell, Saki Takahashi (Princeton University, Department of Ecology and Evolutionary Biology), C. Jessica Metcalf, Matthew J. Ferrari (Pennsylvania State University), and William J. Moss (Johns Hopkins), et al published, “Reduced Vaccination and the Risk of Measles and Other Childhood Infections Post-Ebola,” in Science. In this paper they state that the Ebola epidemic in West Africa has caused substantial morbidity and mortality. The outbreak has also disrupted health care services, including childhood vaccinations, creating a second public health crisis. We project that after 6 to 18 months of disruptions, a large connected cluster of children unvaccinated for measles will accumulate across Guinea, Liberia, and Sierra Leone. This pool of susceptibility increases the expected size of a regional measles outbreak from 127,000 to 227,000 cases after 18 months, resulting in 2000 to 16,000 additional deaths (comparable to the numbers of Ebola deaths reported thus far). There is a clear path to avoiding outbreaks of childhood vaccine-preventable diseases once the threat of Ebola begins to recede: an aggressive regional vaccination campaign aimed at age groups left unprotected because of health care disruptions.


T. Alex Perkins (University of Notre Dame; National Institutes of Health, Bethesda, MD), C. Jessica Metcalf, Bryan Grenfell, and Andrew J. Tatem (University of Southampton; Flowminder Foundation, Sweden) published, “Estimating drivers of autochthonous transmission of chikungunya virus in its invasion of the Americas,” in PLOS Currents Outbreaks. As background, they authors share that Chikungunya is an emerging arbovirus that has caused explosive outbreaks in Africa and Asia for decades and invaded the Americas just over a year ago. During this ongoing invasion, it has spread to 45 countries where it has been transmitted autochthonously, infecting nearly 1.3 million people in total. The methods used here made use of weekly, country-level case reports to infer relationships between transmission and two putative climatic drivers: temperature and precipitation averaged across each country on a monthly basis. To do so, they used a TSIR model that enabled them to infer a parametric relationship between climatic drivers and transmission potential, and they applied a new method for incorporating a probabilistic description of the serial interval distribution into the TSIR framework. They found significant relationships between transmission and linear and quadratic terms for temperature and precipitation and a linear term for log incidence during the previous pathogen generation. The lattermost suggests that case numbers three to four weeks ago are largely predictive of current case numbers. This effect is quite nonlinear at the country level, however, due to an estimated mixing parameter of 0.74. Relationships between transmission and the climatic variables that they estimated were biologically plausible and in line with expectations. Their analysis suggests that autochthonous transmission of Chikungunya in the Americas can be correlated successfully with putative climatic drivers, even at the coarse scale of countries and using long-term average climate data. Overall, this provides a preliminary suggestion that successfully forecasting the future trajectory of a Chikungunya outbreak and the receptivity of virgin areas may be possible. The results also provide tentative estimates of timeframes and areas of greatest risk, and their extension of the TSIR model provides a novel tool for modeling vector-borne disease transmission.


Sinead E. Morris (Princeton University, Department of Ecology and Evolutionary Biology), Virginia E. Pitzer (Yale School of Public Health and National Institutes of Health, Bethesda, MD), Cécile Viboud (National Institutes of Health, Bethesda, MD), C. Jessica Metcalf, Ottar N. Bjørnstad (Pennsylvania State University), and Bryan Grenfell published, “Demographic Buffering: Titrating the Effects of Birth Rate and Imperfect Immunity on Epidemic Dynamics, in J R Soc Interface. The authors write that host demography can alter the dynamics of infectious disease. In the case of perfectly immunizing infections, observations of strong sensitivity to demographic variation have been mechanistically explained through analysis of the susceptible–infected–recovered (SIR) model that assumes lifelong immunity following recovery from infection. When imperfect immunity is incorporated into this framework via the susceptible–infected–recovered–susceptible (SIRS) model, with individuals regaining full susceptibility following recovery, they show that rapid loss of immunity is predicted to buffer populations against the effects of demographic change. However, this buffering is contrary to the dependence on demography recently observed for partially immunizing infections such as rotavirus and respiratory syncytial virus. They show that this discrepancy arises from a key simplification embedded in the SIR(S) framework, namely that the potential for differential immune responses to repeat exposures is ignored. They explore the minimum additional immunological information that must be included to reflect the range of observed dependencies on demography. They show that including partial protection and lower transmission following primary infection is sufficient to capture more realistic reduced levels of buffering, in addition to changes in epidemic timing, across a range of partially and fully immunizing infections. Furthermore, their results identify key variables in this relationship, including R0.


Bryan Grenfell along with Virginia E.Pitzer Yale School of Public Health and National Institutes of Health, Bethesda, MD), Cécile Viboud (National Institutes of Health, Bethesda, MD), Wladimir J.Alonso (National Institutes of Health, Bethesda, MD), Tanya Wilcox (National Institutes of Health, Bethesda, MD) et al. published, “Environmental Drivers of the Spatiotemporal Dynamics of Respiratory Syncytial Virus in the United States,” in PLOS Pathogens. The authors write that epidemics of respiratory syncytial virus (RSV) are known to occur in wintertime in temperate countries including the United States, but there is a limited understanding of the importance of climatic drivers in determining the seasonality of RSV. In the United States, RSV activity is highly spatially structured, with seasonal peaks beginning in Florida in November through December and ending in the upper Midwest in February-March, and prolonged disease activity in the southeastern US. Using data on both age-specific hospitalizations and laboratory reports of RSV in the U.S., and employing a combination of statistical and mechanistic epidemic modeling, they examined the association between environmental variables and state-specific measures of RSV seasonality. Temperature, vapor pressure, precipitation, and potential evapotranspiration (PET) were significantly associated with the timing of RSV activity across states in univariate exploratory analyses. The amplitude and timing of seasonality in the transmission rate was significantly correlated with seasonal fluctuations in PET, and negatively correlated with mean vapor pressure, minimum temperature, and precipitation. States with low mean vapor pressure and the largest seasonal variation in PET tended to experience biennial patterns of RSV activity, with alternating years of ‘‘early-big’’ and ‘‘late-small’’ epidemics. Their model for the transmission dynamics of RSV was able to replicate these biennial transitions at higher amplitudes of seasonality in the transmission rate. This successfully connects environmental drivers to the epidemic dynamics of RSV; however, it does not fully explain why RSV activity begins in Florida, one of the warmest states, when RSV is a winter-seasonal pathogen. Understanding and predicting the seasonality of RSV is essential in determining the optimal timing of immunoprophylaxis.


In Tod Hamilton’s paper titled, “The Healthy Immigrant (Migrant) Effect: In Search of a Better Native-Born Comparison Group,” published in Social Science Research, he evaluates whether immigrants’ initial health advantage over their U.S.-born counterparts results primarily from characteristics correlated with their birth countries (e.g., immigrant culture) or from selective migration (e.g., unobserved characteristics such as motivation and ambition) by comparing recent immigrants’ health to that of recent U.S.-born interstate migrants (“U.S.-born movers”). Using data from the 1999–2013 waves of the March Current Population Survey, he finds that, relative to U.S.-born adults (collectively), recent immigrants have a 6.1 percentage point lower probability of reporting their health as fair or poor. Changing the reference group to U.S.-born movers, however, reduces the recent immigrant health advantage by 28%. Similar reductions in the immigrant health advantage occurs in models estimated separately by either race/ethnicity or education level. Models that examine health differences between recent immigrants and U.S.-born movers who both moved for a new job—a primary motivation behind moving for both immigrants and the U.S.-born—show that such immigrants have only a 1.9 percentage point lower probability of reporting their health as fair or poor. Together, the findings suggest that changing the reference group from U.S.-born adults collectively to U.S.-born movers reduces the identified immigrant health advantage, indicating that selective migration plays a significant role in explaining the initial health advantage of immigrants in the United States.


Sara McLanahan is the principal investigator of the Fragile Families and Child Wellbeing Study (FFS), a longitudinal, birth cohort study of approximately 5,000 parents and their children, including a large oversample of unmarried parents. Mothers and fathers were interviewed shortly after the birth of their child, and follow-up interviews were conducted with both parents one, three, five and nine years after the child’s birth. The study is currently in the field collecting data from mothers and children fifteen years after child’s birth. The 9-year interview collected saliva samples from mothers and children to be used for genetic and epigenetic analyses. The 15-year interview is collecting new saliva samples from teens to be used to study changes in epigenetic markers and biomarkers. The study is supported by grants from the National Institute of Child Health and Human Development (NICHD), National Science Foundation (NSF), the Ford Foundation, the Robert Wood Johnson Foundation and a host of other local and national foundations. The data are a valuable resource to the Princeton community of postdocs, graduate students and undergraduates as well as to the broader research community.


Germán Rodríguez contributed an article on “Multilevel Models in Demography” to the International Encyclopedia of the Social and Behavioral Sciences, Second Edition, edited by James D. Wright. Following some historical remarks he introduced multilevel models in the context of a classical analysis of contraceptive use in various countries by Mason and Wong, describing random-intercept and random-slope models, cross-level interactions, fixed and random effects, and subject-specific and population-average probabilities. He then turned to an extended analysis of infant and child survival in Kenya using a three-level hazard model with family and community random effects, focusing on interpretation of the parameters and translation of the results into probabilities of infant and child death.


Drawing inspiration from online information aggregation systems like Wikipedia and from traditional survey research, Matthew Salganik and Karen E.C. Levy (New York University) proposed a new class of research instruments called wiki surveys. Just as Wikipedia evolves over time based on contributions from participants, Salganik and Levey envisioned an evolving survey driven by contributions from respondents. In a paper published in PLoS ONE, “Wiki Surveys: Open and Quantifiable Social Data Collection” they developed three general principles that underlie wiki surveys: they should be greedy, collaborative, and adaptive. Building on these principles, they developed methods for data collection and data analysis for one type of wiki survey, a pairwise wiki survey. Using two proof-of-concept case studies involving their free and open-source website, All Our Ideas (http://www.allourideas.org/), they show that pairwise wiki surveys can yield insights that would be difficult to obtain with other methods. The replication data and code can be downloaded from the Office of Population Research (OPR) data archive (http://opr.princeton.edu/archive/ws/).


Led by Matthew Salganik, All Our Ideas (www.allourideas.org), is a research project that seeks to develop a new form of social data collection by combining the best features of quantitative and qualitative methods. Using the power of the web, they are creating a data collection tool that has the scale, speed, and quantification of a survey while still allowing for new information to "bubble up" from respondents as happens in interviews, participant observation, and focus groups. Launched in February 2010 All Our Ideas is currently hosting 7,940 wiki surveys with 451,124 ideas and 11.3 million votes. Current contributors are Karen Levy (New York University) and Luke Baker (Agathon Group).


In a paper published in the Journal of Clinical Epidemiology, Matthew Salganik along with co-authors Richard G. White (London School of Hygiene and Tropical Medicine), Kate Orroth (London School of Hygiene and Tropical Medicine), Avi J. Hakim (U.S. Centers for Disease Control and Prevention), Michael W. Spiller (U.S. Centers for Disease Control and Prevention), et al. carried out a systematic review of Respondent Driven Sampling (RDS) studies and present Strengthening the Reporting of Observational Studies in Epidemiology for RDS Studies (STROBE-RDS), a checklist of essential items to present in RDS publications, justified by an explanation and elaboration document. The resultant paper, “Strengthening the Reporting of Observational Studies in Epidemiology for Respondent-Driven Sampling Studies: STROBE-RDS Statement,” found that RDS has been used in over 460 studies from 69 countries, including the USA (151 studies), China (70), and India (32). STROBE-RDS includes modifications to 12 of the 22 items on the STROBE checklist. The two key areas that required modification concerned the selection of participants and statistical analysis of the sample. STROBE-RDS seeks to enhance the transparency and utility of research using RDS. If widely adopted, STROBE-RDS should improve global infectious diseases public health decision making.


Matthew Salganik’s paper, "Diagnostics for respondent-driven sampling", co-authored with Krista J. Gile (University of Massachusetts, Amherst) and Lisa G. Johnston (Tulane University and University of California, San Francisco), appeared in the Journal of the Royal Statistical Society, Series A. Respondent-driven sampling (RDS) is a widely used method for sampling from hard-to-reach human populations, especially populations at higher risk for human immunodeficiency virus or acquired immune deficiency syndrome. Data are collected through a peer referral process over social networks. RDS has proven practical for data collection in many difficult settings and has been adopted by leading public health organizations around the world. Unfortunately, inference from RDS data requires many strong assumptions because the sampling design is partially beyond the control of the researcher and not fully observable. Here Salganik and his co-authors introduce diagnostic tools for most of these assumptions and apply them in 12 high risk populations. These diagnostics empower researchers to understand their RDS data better and encourage future statistical research on RDS sampling and inference.


In October of this year, a revised version of Matthew Salganik’s working paper, "Generalizing the Network Scale-Up Method: A New Estimator for the Size of Hidden Populations" was posted to the arXiv. Co-authored with Ph.D. candidate Dennis Feehan, the new version has major improvements in the exposition and the framework for sensitivity analysis (October 2015). This paper shows how the network scale-up method enables researchers to estimate the size of hidden populations, such as drug injectors and sex workers, using sampled social network data. The basic scale-up estimator offers advantages over other size estimation techniques, but it depends on problematic modeling assumptions. Salganik and Feehan propose a new generalized scale-up estimator that can be used in settings with non-random social mixing and imperfect awareness about membership in the hidden population. Further, the new estimator can be used when data are collected via complex sample designs and from incomplete sampling frames. However, the generalized scale-up estimator also requires data from two samples: one from the frame population and one from the hidden population. In some situations these data from the hidden population can be collected by adding a small number of questions to already planned studies. For other situations, they develop interpretable adjustment factors that can be applied to the basic scale-up estimator. They conclude with practical recommendations for the design and analysis of future studies.


Brandon Stewart has been developing new quantitative statistical methods for applications across the social sciences. Methodologically his focus is on tools which facilitate automated text analysis and model complex heterogeneity in regression. Many recent applications of these methods have centered on using large corpora of text to better understand propaganda in contemporary China.


Recently, Brandon Stewart has been working on Latent Factor Regressions which provide a general framework for modeling dependent data. The framework covers numerous data types including grouped/multilevel, time-series cross-sectional, spatial and network data, all with a single approach. While previous proposals in the literature can take days to estimate a single model, estimation under his framework often takes less than a second. He will be releasing an R package implementing these new methods.


In the Original research article “Do as We Say, Not as We Do: Experiences of Unprotected Intercourse” reported by members of the Society of Family Planning, Abigail Aiken and James Trussell examined the lifetime and past-year prevalence and circumstances of unprotected intercourse among members of the Society of Family Planning (SFP), a professional reproductive health organization in the United States. They invited the membership of SFP (n=477) via email to participate in an anonymous online survey. The response rate was 70% (n=340). Respondents were asked whether they had ever and in the past year had unprotected vaginal intercourse when not intending a pregnancy and, if so, how many times, under what circumstances, and at what age the first time. Then they were asked about unprotected vaginal, anal, or oral intercourse ever and in the past year under three different scenarios relating to sexually transmitted infections (STIs): (1) partner STI status unknown, respondent STI-free; (2) partner known infected, respondent STI-free; (3) partner STI-free, respondent STI status unknown or known infected. Each scenario included questions about the number of times, applicable circumstances, and age at first time.

Forty-six percent of respondents had ever had unprotected vaginal intercourse when not intending pregnancy, 7% within the past year. Sixty percent had ever had unprotected vaginal, anal, or oral intercourse with a partner whose STI status was unknown, 12% within the past year. Four percent had ever had unprotected intercourse with a partner known to have STI, and 8%, with an STI-free partner when they themselves either had an STI or did not know their STI status.

Ever having taken a risk with respect to pregnancy and/or STIs is common among the sample of reproductive health professionals. Most reproductive healthcare professionals in the sample have taken sexual risks in their lifetime and a small proportion has done so in the past year. These findings could inform counseling by encouraging healthcare professionals to reflect upon their own experiences when developing strategies to promote safe sex among their patients.