Distributed Regression Analysis Application in Large Distributed Data Networks: Analysis of Precision and Operational Performance.

BACKGROUND: A distributed data network approach combined with distributed regression analysis (DRA) can reduce the risk of disclosing sensitive individual and institutional information in multicenter studies. However, software that facilitates large-scale and efficient implementation of DRA is limited. OBJECTIVE: This study aimed to assess the precision and operational performance of a DRA application comprising a SAS-based DRA package and a file transfer workflow developed within the open-source distributed networking software PopMedNet in a horizontally partitioned distributed data network. METHODS: We executed the SAS-based DRA package to perform distributed linear, logistic, and Cox proportional hazards regression analysis on a real-world test case with 3 data partners. We used PopMedNet to iteratively and automatically transfer highly summarized information between the data partners and the analysis center. We compared the DRA results with the results from standard SAS procedures executed on the pooled individual-level dataset to evaluate the precision of the SAS-based DRA package. We computed the execution time of each step in the workflow to evaluate the operational performance of the PopMedNet-driven file transfer workflow. RESULTS: All DRA results were precise (<10-12), and DRA model fit curves were identical or similar to those obtained from the corresponding pooled individual-level data analyses. All regression models required less than 20 min for full end-to-end execution. CONCLUSIONS: We integrated a SAS-based DRA package with PopMedNet and successfully tested the new capability within an active distributed data network. The study demonstrated the validity and feasibility of using DRA to enable more privacy-protecting analysis in multicenter studies.

Cross-Network Directory Service: Infrastructure to enable collaborations across distributed research networks.

INTRODUCTION: Existing large-scale distributed health data networks are disconnected even as they address related questions of healthcare research and public policy. This paper describes the design and implementation of a fully functional prototype open-source tool, the Cross-Network Directory Service (CNDS), which addresses much of what keeps distributed networks disconnected from each other. METHODS: The set of services needed to implement a Cross-Directory Service was identified through engagement with stakeholders and workgroup members. CNDS was implemented using PCORnet and Sentinel network instances and tested by participating data partners. RESULTS: Web services that enable the four major functional features of the service (registration, discovery, communication, and governance) were developed and placed into an open-source repository. The services include a robust metadata model that is extensible to accommodate a virtually unlimited inventory of metadata fields, without requiring any further software development. The user interfaces are programmatically generated based on the contents of the metadata model. CONCLUSION: The CNDS pilot project gathered functional requirements from stakeholders and collaborating partners to build a software application to enable cross-network data and resource sharing. The two partners-one from Sentinel and one from PCORnet-tested the software. They successfully entered metadata about their organizations and data sources and then used the Discovery and Communication functionality to find data sources of interest and send a cross-network query. The CNDS software can help integrate disparate health data networks by providing a mechanism for data partners to participate in multiple networks, share resources, and seamlessly send queries across those networks.

A Query Workflow Design to Perform Automatable Distributed Regression Analysis in Large Distributed Data Networks.

INTRODUCTION: Patient privacy and data security concerns often limit the feasibility of pooling patient-level data from multiple sources for analysis. Distributed data networks (DDNs) that employ privacy-protecting analytical methods, such as distributed regression analysis (DRA), can mitigate these concerns. However, DRA is not routinely implemented in large DDNs. OBJECTIVE: We describe the design and implementation of a process framework and query workflow that allow automatable DRA in real-world DDNs that use PopMedNet™, an open-source distributed networking software platform. METHODS: We surveyed and catalogued existing hardware and software configurations at all data partners in the Sentinel System, a PopMedNet-driven DDN. Key guiding principles for the design included minimal disruptions to the current PopMedNet query workflow and minimal modifications to data partners' hardware configurations and software requirements. RESULTS: We developed and implemented a three-step process framework and PopMedNet query workflow that enables automatable DRA: 1) assembling a de-identified patient-level dataset at each data partner, 2) distributing a DRA package to data partners for local iterative analysis, and 3) iteratively transferring intermediate files between data partners and analysis center. The DRA query workflow is agnostic to statistical software, accommodates different regression models, and allows different levels of user-specified automation. DISCUSSION: The process framework can be generalized to and the query workflow can be adopted by other PopMedNet-based DDNs. CONCLUSION: DRA has great potential to change the paradigm of data analysis in DDNs. Successful implementation of DRA in Sentinel will facilitate adoption of the analytic approach in other DDNs.

State and Local Chronic Disease Surveillance Using Electronic Health Record Systems.

OBJECTIVES: To assess the feasibility of chronic disease surveillance using distributed analysis of electronic health records and to compare results with Behavioral Risk Factor Surveillance System (BRFSS) state and small-area estimates. METHODS: We queried the electronic health records of 3 independent Massachusetts-based practice groups using a distributed analysis tool called MDPHnet to measure the prevalence of diabetes, asthma, smoking, hypertension, and obesity in adults for the state and 13 cities. We adjusted observed rates for age, gender, and race/ethnicity relative to census data and compared them with BRFSS state and small-area estimates. RESULTS: The MDPHnet population under surveillance included 1 073 545 adults (21.8% of the state adult population). MDPHnet and BRFSS state-level estimates were similar: 9.4% versus 9.7% for diabetes, 10.0% versus 12.0% for asthma, 13.5% versus 14.7% for smoking, 26.3% versus 29.6% for hypertension, and 22.8% versus 23.8% for obesity. Correlation coefficients for MDPHnet versus BRFSS small-area estimates ranged from 0.890 for diabetes to 0.646 for obesity. CONCLUSIONS: Chronic disease surveillance using electronic health record data is feasible and generates estimates comparable with BRFSS state and small-area estimates.

Software-Enabled Distributed Network Governance: The PopMedNet Experience.

INTRODUCTION: The expanded availability of electronic health information has led to increased interest in distributed health data research networks. DISTRIBUTED RESEARCH NETWORK MODEL: The distributed research network model leaves data with and under the control of the data holder. Data holders, network coordinating centers, and researchers have distinct needs and challenges within this model. SOFTWARE ENABLED GOVERNANCE POPMEDNET: The concerns of network stakeholders are addressed in the design and governance models of the PopMedNet software platform. PopMedNet features include distributed querying, customizable workflows, and auditing and search capabilities. Its flexible role-based access control system enables the enforcement of varying governance policies. SELECTED CASE STUDIES: Four case studies describe how PopMedNet is used to enforce network governance models. ISSUES AND CHALLENGES: Trust is an essential component of a distributed research network and must be built before data partners may be willing to participate further. The complexity of the PopMedNet system must be managed as networks grow and new data, analytic methods, and querying approaches are developed. CONCLUSIONS: The PopMedNet software platform supports a variety of network structures, governance models, and research activities through customizable features designed to meet the needs of network stakeholders.

Real-time surveillance for tuberculosis using electronic health record data from an ambulatory practice in eastern Massachusetts.

OBJECTIVE: Electronic health records (EHRs) have the potential to improve completeness and timeliness of tuberculosis (TB) surveillance relative to traditional reporting, particularly for culture-negative disease. We report on the development and validation of a TB detection algorithm for EHR data followed by implementation in a live surveillance and reporting system. METHODS: We used structured electronic data from an ambulatory practice in eastern Massachusetts to develop a screening algorithm aimed at achieving 100% sensitivity for confirmed active TB with the highest possible positive predictive value (PPV) for physician-suspected disease. We validated the algorithm in 16 years of retrospective electronic data and then implemented it in a real-time EHR-based surveillance system. We assessed PPV and the completeness of case capture relative to conventional reporting in 18 months of prospective surveillance. RESULTS: The final algorithm required a prescription for pyrazinamide, an International Classification of Diseases, Ninth Revision (ICD-9) code for TB and prescriptions for two antituberculous medications, or an ICD-9 code for TB and an order for a TB diagnostic test. During validation, this algorithm had a PPV of 84% (95% confidence interval 78, 88) for physician-suspected disease. One-third of confirmed cases were culture-negative. All false-positives were instances of latent TB. In 18 months of prospective EHR-based surveillance with this algorithm, seven additional cases of physician-suspected active TB were detected, including two patients with culture-negative disease. A review of state health department records revealed no cases missed by the algorithm. CONCLUSIONS: Live, prospective TB surveillance using EHR data is feasible and promising.

CD107a as a functional marker for the identification of natural killer cell activity.

Natural killer (NK) cells are a subset of lymphocytes that play a central role in the innate immune response to tumors and infections. An important limitation in the field of NK research is attributable to the deficit of assays available for the detection of the functional activity of NK cells. Recently, lysosomal-associated membrane protein-1 (LAMP-1 or CD107a) has been described as a marker of CD8+ T-cell degranulation following stimulation. Here we describe CD107a as a marker of NK cell functional activity using multi-parameter flow cytometry. CD107a is significantly upregulated on the surface of NK cells following stimulation with MHC devoid targets. Additionally, CD107a expression correlates with both cytokine secretion and NK cell-mediated lysis of target cells. However, as well as being coordinately expressed on nearly all cytokine secreting cells, CD107a was also expressed on a large subset of NK cells that did not secrete cytokine following stimulation. These data suggest that employing CD107a as a marker of NK cell functional activity may allow for the identification of a large fraction of activated NK cells that may degranulate in the absence of cytokine secretion. Cumulatively, the data presented here demonstrate that CD107a is a sensitive marker of NK cell activity.

Increased natural killer cell activity in viremic HIV-1 infection.

NK cells are a subset of granular lymphocytes that are critical in the innate immune response to infection. These cells are capable of killing infected cells and secreting integral cytokines and chemokines. The role that this subset of cytolytic cells plays in HIV infection is not well understood. In this study, we dissected the function of NK cells in viremic and aviremic HIV-1-infected subjects, as well as HIV-1-negative control individuals. Despite reduced NK cell numbers in subjects with ongoing viral replication, these cells were significantly more active in secreting both IFN-gamma and TNF-alpha than NK cells from aviremic subjects or HIV-1-negative controls. In addition, NK cells in subjects with detectable viral loads expressed significantly higher levels of CD107a, a marker of lysosomal granule exocytosis. The expression of CD107a correlated with NK cell-mediated cytokine secretion and cytolytic activity as well as with the level of viral replication, suggesting that CD107a represents a good marker for the functional activity of NK cells. Finally, killer Ig-related receptor+ NK cells were stable or elevated in viremic subjects, while the numbers of CD3-/CD56+/CD94+ and CD3-/CD56+/CD161+ NK cells were reduced. Taken together, these data demonstrate that viremic HIV-1 infection is associated with a reduction in NK cell numbers and a perturbation of NK cell subsets, but increased overall NK cell activity.

HIV-1 Nef is preferentially recognized by CD8 T cells in primary HIV-1 infection despite a relatively high degree of genetic diversity.

OBJECTIVE: To compare the magnitude, breadth and protein specificity of HIV-1-specific CD8 T-cell responses against the clade B consensus sequence during primary and chronic HIV-1 infection and to analyze the impact of viral diversity on the localization of detected responses. METHODS: HIV-1-specific CD8 T-cell responses against the clade B consensus sequence in individuals with acute (n = 10), early (n = 19) and chronic (n = 10) infection were longitudinally assessed using an interferon-gamma EliSpot assay. RESULTS: CD8 T-cell responses against clade B consensus sequences were preferentially directed against central regions of Nef during primary HIV-1 infection, despite a relatively higher degree of genetic diversity compared with other subsequently targeted regions. In subjects with acute and early infection, Nef-specific CD8 T-cell responses against the consensus Nef sequence represented 94 and 46% of the total magnitude of HIV-1-specific CD8 T-cell responses, respectively. Subjects with untreated chronic infection exhibited broadly diversified CD8 T-cell responses against more conserved viral regions, with only 17% of virus-specific T-cell responses targeting Nef. The initial immunodominance of Nef persisted in individuals with treated acute infection, but shifted rapidly to Gag, Env and Pol in subjects with continuous antigen exposure. CONCLUSION: These data show that despite relatively high sequence variability, viral regions within the clade B consensus sequence of Nef are preferentially recognized during primary HIV-1 infection. Later diversification of responses to other proteins during prolonged antigen exposure provides evidence of the initial preferential immunogenicity of Nef epitopes compared to similarly conserved regions within other viral proteins.