SKSarah C. Kramer

Sarah C. Kramer

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Sarah C. Kramer has a total of 26 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    The limitations of non-mechanistic methods for characterizing pathogen-pathogen interactions: A simulation study

    by Sarah C. Kramer, Sarah Pirikahu, Cana Kussmaul, Lulla Opatowski, Matthieu Domenech de Cellès

    Pathogen-pathogen interactions occur when infection with one pathogen influences one’s chance of infection or disease due to another. Increasingly, evidence suggests that interactions are a common feature of infectious disease epidemiology. However, due to both the nonlinearities and stochasticity inherent to infectious disease transmission, and the frequency of confounding (e.g., by shared seasonal forcing), simple, correlative methods for characterizing interactions may be prone to failure. Here, we perform a simulation study to evaluate several more complex non-mechanistic approaches for inferring causality from time series data: generalized additive models (GAMs), Granger causality, transfer entropy, and convergent cross-mapping (CCM). Specifically, we use a two-pathogen mechanistic transmission model, calibrated to produce dynamics resembling outbreaks of influenza and respiratory syncytial virus (RSV), to generate synthetic datasets with a range of values for interaction strength and duration. We then apply each method to all synthetic datasets. We find that Granger causality, transfer entropy, and CCM all fail to consistently infer whether data contain signal of an interaction; in particular, methods tend to incorrectly identify interactions where none are modeled (average sensitivity = 80.6%, 92.1%, 72.1%, respectively; average specificity = 31.0%, 33.3%, 33.1%). Furthermore, we find little to no association between point estimates from each method and true interaction strength. In contrast, GAMs infer the existence of interactions more accurately than the other methods (sensitivity = 85.2%, specificity = 72.5%), and consistently yield larger point estimates for stronger interactions. However, their practical utility is limited by an inability to evaluate interaction asymmetry (i.e., whether the effect of pathogen A on pathogen B is identical to that of B on A). Overall performance patterns were similar when methods were applied to two real-world datasets from Hong Kong and Canada. We conclude that accurately and comprehensively characterizing pathogen-pathogen interactions based on outbreak data remains a significant challenge. For this reason, it is critical that any proposed methods be rigorously evaluated before being used to draw conclusions about interactions.

  • Open Access

    Hitchhiker bias distorts symptom-based surveillance of infectious diseases

    by Kiran Kumari, Sarah C. Kramer, Matthieu Domenech de Cellès

    Symptom-based infectious disease surveillance, which infers pathogen circulation by testing individuals who seek care for clinical symptoms, is widely used in public health but may introduce biases that distort our understanding of pathogen dynamics. One such bias, which we introduce and call “hitchhiker bias”, occurs when the observed number of hospitalizations of asymptomatic or mildly symptomatic pathogens is inflated because they coinfect individuals already being tested for more severe symptoms caused by other pathogens. Using a compartmental SEIR co-infection model, we demonstrate that co-circulation with a symptomatic pathogen significantly distorts the apparent severity and peak time of a hitchhiking virus. These distortions are amplified as the temporal overlap between pathogens increases. We develop and apply a modeling framework to adjust for hitchhiker bias, showing that it can recover the true pathogenicity of viruses that would otherwise be misinterpreted using standard symptom-based surveillance alone.

  • Open Access

    The impact of observation models on epidemic inference for emerging infectious diseases: SARS-CoV-2 as a case study

    by Matthieu Domenech de Celles, Sarah C Kramer

    Parameter estimation is often necessary to inform transmission models of infectious diseases. This estimation requires choosing an observation model, defined here as the probability distribution linking model outputs to observed data. Although potentially consequential, this choice has received little attention in the literature. Here, we compared eight observation models, including common distributions such as the Poisson, binomial, negative binomial, and normal. Using Bayesian inference, we fit an SIR-like model to daily case reports during the first wave of COVID-19 in Belgium, Finland, Germany, Sweden, Switzerland, and the UK. We found considerable differences in the log-likelihoods of the observation models, spanning three orders of magnitude between the best and the worst. Compared with the best models, the binomial, Poisson, and normal models received no support due to their rigid variance structures. Additionally, the binomial and Poisson models produced overly narrow prediction and confidence intervals, especially for key parameters such as the basic reproduction number. The other five models—each with a free dispersion parameter scaling the variance to the mean—performed much better, with the negative binomial model ranking among the top models in all countries. We conclude that flexible observation models are essential for transmission models to accurately capture all sources of uncertainty.

  • Open Access

    The limitations of non-mechanistic methods for characterizing pathogen-pathogen interactions: A simulation study

    by Sarah C. Kramer, Sarah Pirikahu, Cana Kussmaul, Lulla Opatowski, Matthieu Domenech de Cellès

    Pathogen-pathogen interactions occur when infection with one pathogen influences one’s chance of infection or disease due to another. Increasingly, evidence suggests that interactions are a common feature of infectious disease epidemiology. However, due to both the nonlinearities and stochasticity inherent to infectious disease transmission, and the frequency of confounding (e.g., by shared seasonal forcing), simple, purely statistical methods for characterizing interactions may be prone to failure. Here, we perform a simulation study to evaluate several more complex non-mechanistic approaches for inferring causality from time series data: generalized additive models (GAMs), Granger causality, transfer entropy, and convergent cross-mapping (CCM). Specifically, we use a two-pathogen mechanistic transmission model, calibrated to produce dynamics resembling outbreaks of influenza and respiratory syncytial virus (RSV), to generate synthetic datasets with a range of values for interaction strength and duration. We then apply each method to all synthetic datasets. We find that Granger causality, transfer entropy, and CCM all fail to consistently infer whether data contain signal of an interaction; in particular, methods tend to incorrectly identify interactions where none are modeled (average sensitivity = 80.6%, 92.1%, 72.1%, respectively; average specificity = 31.0%, 33.3%, 33.1%). Furthermore, we find little to no association between point estimates from each method and true interaction strength. In contrast, GAMs infer the existence of interactions more accurately than the other methods (sensitivity = 85.2%, specificity = 72.5%), and consistently yield larger point estimates for stronger interactions. However, their practical utility is limited by an inability to evaluate interaction asymmetry (i.e., whether the effect of pathogen A on pathogen B is identical to that of B on A). Overall performance patterns were similar when methods were applied to two real-world datasets from Hong Kong and Canada. We conclude that accurately and comprehensively characterizing pathogen-pathogen interactions based on outbreak data remains a significant challenge. For this reason, it is critical that any proposed methods be rigorously evaluated before being used to draw conclusions about interactions. Author Summary: Pathogen-pathogen interactions occur when infection with one pathogen either increases or decreases a person’s risk of infection or illness due to a second, distinct pathogen. Because interactions affect several common human pathogens, including influenza and SARS-CoV-2, a better understanding of interactions could improve epidemic control. However, past work has shown that simple methods commonly used to study interactions can lead to inaccurate conclusions. Here, we tested four methods frequently used in other fields, including ecology and neuroscience, to see whether they may also be useful for identifying interactions. Specifically, we tested each method using simulated outbreak data generated from a mathematical model. We found that most methods struggled to correctly determine whether an interaction effect was present; in particular, methods often falsely identified interactions when none occurred. Although one of the tested methods, generalized additive models, performed comparatively well at identifying interactions, it provided relatively little additional information about the interactions. Because pathogen-pathogen interactions are so challenging to study, it is important that researchers rigorously test methods before applying them to interactions, so as not to publish potentially misleading results. More broadly, a complete understanding of interactions will likely require a variety of approaches, including both laboratory and modeling studies.

  • Open Access

    Changing socioeconomic inequalities in the incidence and case fatality rates of COVID-19 in Germany, March 2020 through May 2022: an ecological study

    by Sarah C. Kramer, Laura Andrea Barrero Guevara, Matthias An der Heiden, Benjamin Wachtler, Matthieu Domenech de Cellès

    Background: Socioeconomic disparities in COVID-19 burden were widely observed during the pandemic’s early waves, including in Germany, but studies on whether these inequalities have persisted or changed as the pandemic progressed are lacking. Methods: We used an ecological study design to assess the relationship between a range of demographic, socioeconomic, and healthcare-related predictors and COVID-19 impact in Germany. Specifically, we fit generalized additive models to cumulative, district-level (n = 400) COVID-19 incidence and case fatality rates (CFRs) for each of the first five pandemic waves, which covered the period from March 2020 through May 2022. Results: We find that associations between socioeconomic deprivation and COVID-19 impact evolved over time. Specifically, districts with higher levels of deprivation experienced lower incidence initially, but higher incidence beginning in the second half of wave 1 and persisting through wave 3. Meanwhile, more deprived districts experienced higher CFRs initially as well as during waves 3 through 5, but lower CFRs during the second half of wave 1. Finally, during the first four waves, we find that district-level CFRs scaled superlinearly with incidence, suggesting that the risk of death increased with incidence. This relationship was particularly strong during the first pandemic wave. Conclusions: The association between socioeconomic position and COVID-19 impact in Germany has been complex, with patterns changing in intensity and direction over time. Continued monitoring of socioeconomic inequalities in COVID-19 impact, in particular at the individual level, is needed to better understand if and how inequalities continue to persist. Such monitoring will be instrumental in informing more equitable control strategies.

  • Open Access

    Causal inference concepts can guide research into the effects of climate on infectious diseases

    by Laura Andrea Barrero Guevara, Sarah C. Kramer, Tobias Kurth, Matthieu Domenech de Cellès

    A pressing question resulting from global warming is how climate change will affect infectious diseases. Answering this question requires research into the effects of weather on the population dynamics of transmission and infection; elucidating these effects, however, has proved difficult due to the challenges of assessing causality from the predominantly observational data available in epidemiological research. Here we show how concepts from causal inference—the sub-field of statistics aiming at inferring causality from data—can guide that research. Through a series of case studies, we illustrate how such concepts can help assess study design and strategically choose a study’s location, evaluate and reduce the risk of bias, and interpret the multifaceted effects of meteorological variables on transmission. More broadly, we argue that interdisciplinary approaches based on explicit causal frameworks are crucial for reliably estimating the effect of weather and accurately predicting the consequences of climate change.


  • Open Access

    Characterizing the interactions between influenza and respiratory syncytial viruses and their implications for epidemic control

    by Sarah C. Kramer, Sarah Pirikahu, Jean-Sébastien Casalegno, Matthieu Domenech de Cellès

    Pathogen-pathogen interactions represent a critical but little-understood feature of infectious disease dynamics. In particular, experimental evidence suggests that influenza virus and respiratory syncytial virus (RSV) compete with each other, such that infection with one confers temporary protection against the other. However, such interactions are challenging to study using common epidemiologic methods. Here, we use a mathematical modeling approach, in conjunction with detailed surveillance data from Hong Kong and Canada, to infer the strength and duration of the interaction between influenza and RSV. Based on our estimates, we further utilize our model to evaluate the potential conflicting effects of live attenuated influenza vaccines (LAIV) on RSV burden. We find evidence of a moderate to strong, negative, bidirectional interaction, such that infection with either virus yields 40-100% protection against infection with the other for one to five months. Assuming that LAIV reduces RSV susceptibility in a similar manner, we predict that the impact of such a vaccine at the population level would likely depend greatly on underlying viral circulation patterns. More broadly, we highlight the utility of mathematical models as a tool to characterize pathogen-pathogen interactions.


  • Open Access

    Assessing the effect of social contact structure on the impact of pneumococcal conjugate vaccines

    by Anabelle Wong, Sarah C. Kramer, Daniel M. Weinberger, Matthieu Domenech de Cellès

    Although pneumococcal conjugate vaccines (PCVs) have greatly reduced diseases caused by vaccine-targeted serotypes (VT) of Streptococcus pneumoniae , vaccine impact may be eroded by the increase in rates of disease caused by non-vaccine serotypes (NVT). Here, we investigated the effect of social contact patterns on the dynamics of vaccine impact in carriage. We developed a neutral, age-structured, Susceptible–Colonized model incorporating VT-NVT co-colonization and verified it against real-world carriage data in children. Using contact matrices from 34 countries, we assessed the impact of contact patterns on the time required to eliminate VT (i.e., 95% reduction in VT proportion in carriage). Finally, we quantified the contribution of various parameters—such as vaccine efficacy, coverage, immunity waning, and population susceptibility—to the dynamics of VT elimination. Our model recapitulated the observed decline of VT carriage and showed that varying the contact structure alone led to different time-to-elimination. We found that higher total contact rate and assortativity in children under 5 accelerated VT elimination. Additionally, higher vaccine efficacy and coverage, and slower immunity waning led to shorter time-to-elimination. These findings illuminate the mechanisms controlling the dynamics of vaccine impact and may help predict the impact of PCVs in communities with different contact patterns.

  • Open Access

    Characterizing the interactions between influenza and respiratory syncytial viruses and their implications for epidemic control

    by Sarah C. Kramer, Sarah Pirikahu, Jean-Sébastien Casalegno, Matthieu Domenech de Cellès

    Pathogen-pathogen interactions represent a critical but little-understood feature of infectious disease dynamics. In particular, experimental evidence suggests that influenza virus and respiratory syncytial virus (RSV) compete with each other, such that infection with one confers temporary protection against the other. However, such interactions are challenging to study using common epidemiologic methods. Here, we use a mathematical modeling approach, in conjunction with detailed surveillance data from Hong Kong and Canada, to infer the strength and duration of the interaction between influenza and RSV. Based on our estimates, we further utilize our model to evaluate the potential conflicting effects of live attenuated influenza vaccines (LAIV) on RSV burden. We find evidence of a moderate to strong, negative, bidirectional interaction, such that infection with either virus yields 40-100% protection against infection with the other for one to five months. Assuming that LAIV reduces RSV susceptibility in a similar manner, we predict that the impact of such a vaccine at the population level would likely depend greatly on underlying viral circulation patterns. More broadly, we highlight the utility of mathematical models as a tool to characterize pathogen-pathogen interactions.


  • Open Access

    Using LASSO regression to estimate the population-level impact of pneumococcal conjugate vaccines

    by Anabelle Wong, Sarah C. Kramer, Marco Piccininni, Jessica L. Rohmann, Tobias Kurth, Sylvie Escolano, Ulrike Grittner, Matthieu Domenech de Cellès

    The pneumococcal conjugate vaccines (PCVs) protect against diseases caused by Streptococcus pneumoniae , such as meningitis, bacteremia, and pneumonia. It is challenging to estimate their population-level impact due to the lack of a perfect control population and the subtleness of signals when the endpoint – like all-cause pneumonia – is non-specific. Here we present a new approach to estimate PCVs’ impact – using LASSO regression to predict the counterfactual outcome for vaccine impact inference. We first used a simulation study to test the performance of LASSO regression and established methods including the synthetic control (SC) approach. We found that LASSO achieved accurate and precise estimation, even in complex simulation scenarios where the association between outcome and all control variables was non-causal. We then applied LASSO to real-world data and found that it yielded estimates of vaccine impact similar to SC. The LASSO method is accurate, easily implementable, and can be applied to study the impact of PCVs and of other vaccines.