Introduction | Overview | Covariate settings function | Example function | Covariate construction function | Function inputs | Function outputs | Using the custom covariate builder

서론 | 개요 | 공변량 설정 함수 | 함수 예제 | 공변량 구성 함수 | 함수 입력 | 함수 출력 | 사용자 지정 공변량 빌더 사용

Introduction | Covariate settings | Using the default set of covariates | Using prespecified analyses | Creating a set of custom covariates | Temporal covariates | Constructing covariates for a cohort of interest | Configuring the connection to the server | Creating a cohort of interest | Creating per-person covariates for a cohort of interest | Per-person covariate output format | Saving the data to file | Removing infrequent covariates, normalizing, and removing redundancy | Creating aggregated covariates for a cohort of interest | Aggregated covariate output format | Creating a table 1 | Comparing two cohorts

서론 | 공변량 설정 | 기본 공변량 집합 사용하기 | 사전 지정된 분석 사용 | 사용자 공변량 집합을 생성 | 시간 공변량 | 관심 코호트에 대한 공변량 구성 | 서버 연결 설정 | 관심 코호트 생성 | 관심 코호트에 대한 1인당 공변량 생성 | 사용자 공변량 출력 형식 | 데이터를 파일로 저장 | 간헐적인 공변량 제거, 정규화 및 중복 제거 | 코호트에 대한 집계 공변량 생성 | 집계 공변량 출력 형식 | 테이블 만들기 1

Overview | Minimum Detectable Relative Risk (MDRR) | What it checks | Method | Interpretation | Example | Role in blinding | Pre-Exposure Gain | Why it matters | Event-Dependent Observation | Expected Absolute Systematic Error (EASE) | When it runs | Tiered Blinding | Running Diagnostics | Customizing thresholds | Selecting specific diagnostics | Inspecting failures

Setup and Data | Running a Basic Analysis | Understanding Risk Windows | Using Custom Cohorts | Diagnostics and Tiered Blinding | Empirical Calibration | Multi-Analysis Workflow | Working with the Results Database

Overview | Minimum Detectable Relative Risk (MDRR) | What it checks | Method | Interpretation | Example | Role in blinding | Pre-Exposure Gain | Why it matters | Event-Dependent Observation | Expected Absolute Systematic Error (EASE) | When it runs | Tiered Blinding | Running Diagnostics | Customizing thresholds | Selecting specific diagnostics | Inspecting failures

Setup and Data | Running a Basic Analysis | Understanding Risk Windows | Using Custom Cohorts | Diagnostics and Tiered Blinding | Empirical Calibration | Multi-Analysis Workflow | Working with the Results Database

Introduction | Running PhenotypeDiagnostics | Database diagnostics | Codelist diagnostics | Cohort diagnostics | Modify windows, event in windows, and episodes in window | Population diagnostics | Save the results | Visualisation of the results

Introduction | Setup | Examples | Target Baseline Covariates | Risk Factor Covariates | Case Series Covariates | Dechallenge Rechallenge | Time to Event | Run Multiple

Using the OhdsiReportGenerator Package | Example Result Database | Extracting Cohorts | Extracting Characterization Results | Extracting Prediction Results | Extracting Estimation Results

Introduction | Operating the Patient Designer | Saving Test Sets for testthat Integration | Integrating with testthat

Introduction: Create descriptions | Database Descriptions | Clinical Descriptions | Generate clinical descriptions using LLMs

Introduction: Run ShinyDiagnostics | Shiny App Overview | Special cases

Comparing phenotype diagnostic results against phenotype expectations | Creating phenotype expectations | Using an LLM to draft your phenotype expectations

Introduction | Start with a mock CDM | Keep a target concept set | Exclude directly related concepts | Control which domains are always retained | Apply subsetting after building a CDM

Introduction | Run achilles analysis | Differences with the Achilles R package | Execution Model | Small Cell Suppression | Observation Period Consistency

R Installation | Note | Executing Data Quality Checks | Viewing Results | View checks

Time-to-event | Inputs | Output | Worked Example | Example Inputs | Example Data Image | Example Data Table | Dechallenge-rechallenge | Example Data Plot | Intermediary Table | Intermediary Plots | Summary | Target Baseline Covariates | Outputs | Example Data | Results | Risk Factor Analysis | Intermedeiary Tables | Case Series

Introduction | Data extraction | Configuring the connection to the server | Preparing the exposures and outcome(s) | Extracting the data from the server | Saving the data to file | Defining the study population | Propensity scores | Fitting a propensity model | Propensity score diagnostics | Using the propensity score | Evaluating covariate balance | Inspecting select population characteristics | Generalizability | Follow-up and power | Outcome models | Fitting a simple outcome model | Adding interaction terms | Adding covariates to the outcome model | Inspecting the outcome model | Kaplan-Meier plot | Time-to-event plot | Acknowledgments

Introduction | Algorithm Code Structure | Set | Fit | Predict | VarImp | Algorithm Example | Variable importance | Acknowledgments

Introduction | Avialable datasets | Download a dataset | Create a cdm reference of a mock dataset

Introduction | Classic Local Control | Nearest neighbors Local Control | Survival Local Control | Case study: Framingham heart patients | Patient level prediction/heterogeneity of treatment effect | Conclusion

Introduction | Subset Operators | Subset Definition | Creating cohort subset definitions | Cohort Definition Set | Subset operators | Create the subset definition | Adding subsets to Cohort Definition Set | Generating subsets | Visualizing subset membership | Saving and loading subset definitions | Saving to packages/directories | Writing json objects

Guide for generating cohorts using CohortGenerator | Basic Example | Loading a cohort definition set from ATLAS | Loading an example cohort definition set | Saving in a study package | Advanced Options | Cohort Statistics (Inclusion Rule Statistics) | Incremental Mode

Introduction | Limitations of this approach | Basic SQL templates | Validating custom sql cohorts | Built in large scale definitions | Drug ingredient cohorts | ATC Base cohorts | SNOMED condition cohorts | Creating custom cohort templates | Generating the cohorts | On execution order | Conclusion

Introduction | Basic usage | Timing options | Measurement cohorts | Stratifications | Other arguments

Introduction | Defining a codelist of measurements | Measurement diagnostics | Visualise results | Stratifications | Estimates | Histogram estimates | Other arguments

Introduction | Create diagnostics results | Tables | Plots | Measurement summary | Numeric-value summary | boxplot | densityplot | barplot | Concept-value summary | Visualisation with other packages | Shiny Apps with OmopViewer | Customisation of plots and tables with visOmopResults | Application of MeasurementDiagnostics in PhenotypeR

Add a New Check to Data Quality Dashboard | Steps | Write the SQL Query for Your Check | Format the Query for the Data Quality Dashboard | Add the Query to Check Descriptions File | Add Hooks to Field/Table/Concept Check CSV File

Introduction | Concept based cohort creation | Demographic based cohort creation | Measurement Cohort | Death cohort

Introduction | Obtaining drivers | The JAR folder | Obtaining drivers for SQL Server, Oracle, PostgreSQL, PDW, Spark, RedShift, Snowflake, BigQuery, and Synapse | Obtaining drivers for Netezza, Impala and InterSystems IRIS | Obtaining drivers for SQLite or DuckDb | Creating a connection | Using Windows Authentication for SQL Server | Connecting to a SQLite database | Connecting with Windows authentication from a non-windows machine

Keep only the first record per person | Keep only the last record per person | Keep only a range of records per person | Keep only records within a date range | Keep only if entry lasts a given duration | Applying multiple cohort requirements | Keep only records from cohorts with a minimum number of individuals

Restrict cohort by age | Restrict cohort by sex | Restrict cohort by number of prior observations | Applying multiple demographic requirements to a cohort

Introduction | Exit at Specific Date | exitAtObservationEnd() | exitAtDeath() | Cohort Entry or Exit Based on Other Date Columns | entryAtFirstDate() | entryAtLastDate() | exitAtFirstDate() | exitAtLastDate() | Trim Dates Functions | trimDemographics() | trimToDateRange() | Pad Dates Functions | padCohortStart() | padCohortEnd() | padCohortDate() | Cohort ID argument

Introduction | General approach | Preparation for the example | Preparing the exposures and outcome(s) | Specifying hypotheses of interest | Specifying analyses | Covariate balance | Executing multiple analyses | Restarting | Retrieving the results | Diagnostics | Empirical calibration and negative control distribution | Exporting to CSV | Acknowledgments

Introduction | Fields with minimum values | Tables

Sampling with CohortGenerator | Sampling method | Using the sampler functions

Introduction | Create a mock CDM | Database characteristics | Summarise Characteristics | Selecting tables to characterise | Stratifying by Sex | Stratifying by Age Group | Filtering by date range and time interval | Sample the CDM | Including Concept Counts | Other arguments | Visualise the characterisation results | Customise the Shiny App | Disconnect from CDM

Introduction | Create a mock cdm from a cohort table | Create a mock CDM from drug_exposure

Restrictions on cohort presence | Restrictions on concept presence | Restrictions on presence in clinical tables

Introduction | stratifyCohorts | yearCohorts

Introduction | Collaboration | Cohorts | Cohort counts and overlap | Performance | By definition | By domain | Cohort stratification

Introduction | Create mock data | Use matchCohorts() to create an age-sex matched cohort | matchSex parameter | matchYear parameter | ratio parameter | Generate matched cohorts simultaneously across multiple cohorts

Introduction | Build observation_period table in GiBleed | First to extraction | First to last | Inpatient | Collapse 180 | Collapse+Persistence 180 | Collapse+Persistence 365 | Pediatric | Comparison of the different definitions | Final remarks

Introduction | Installing psycopg2 | Using a virtualenv | Using conda or system python installs | Usage within functions

Introduction | Exposures, covariates of interest, and controls | Exposures-outcome-sets, analysis IDs and models | Fields with minimum values | Tables

Introduction | Not Applicable

DQD Failure Thresholds | DQD Control Files | Step 1: Find and copy the control files | Step 2: Turn On/Off Checks and Change Thresholds | Step 2a: Documenting Metadata on Updated Thresholds | Step 3: Run the DQD Using the Updated Thresholds

Description | Generating the "Incremental Insert" DQD SQL | (OPTIONAL) Execute queries

Simulate data | Fit a model locally | Approximate the likelihood function at one site | Normal approximation | Adaptive approximation | Approximate at all sites | Synthesize evidence | Fixed-effects | Visualization | Random-effects

Introduction | Learning bias distributions from negative control outcomes | Perform Bayesian adaptive bias correction

Introduction | Study specification | Problem definition 1: Stroke in atrial fibrilation patients | Problem definition 2: Angioedema in ACE inhibitor users | Study population definition | Model development settings | Example 1: Stroke in Atrial fibrillation patients | Study Specification | Study implementation | Cohort instantiation | ATLAS cohort builder | Custom cohorts | Study script creation | Data extraction | Additional inclusion criteria | Splitting the data into training/validation/testing datasets | Preprocessing the training data | Model Development | Example 2: Angioedema in ACE inhibitor users | Spliting the data into training/validation/testing datasets | Study package creation | Internal validation | Discrimination | Smooth Calibration | Other functionality | Demos | Acknowledgments | Appendix 1: Study population settings details

Introduction | Creating the learning curve | Parallel processing | Demo | Publication | Acknowledgments

Introduction | Simulating some data | Approximating the likelihood at each site | Generating approximations at all sites | Combining evidence across sites | Fixed-effects model. | Bayesian random-effects model. | Choice of prior | Forest plot

Introduction

Introduction | Load packages | Mock data

Introduction | Loading Necessary Libraries and Data | Subsetting the Data | Generating the Shiny App | Generated shiny app | Panels generated (panelDetails) | Understanding Panel Details (panelDetails) | Panel Structure (omopViewerPanels) | Customise panelStructure

Introduction | General approach | Preparation for the example | Specifying exposures-outcome sets | Specifying analyses | Executing multiple analyses | Restarting | Retrieving the results | Diagnostics summary | Negative control distribution | Acknowledgments

Introduction | Terminology | Installation instructions | Overview | Studies with a single drug | Configuring the connection to the server | Preparing the exposure and outcome of interest | Extracting the data from the server | Saving the data to file | Creating the study population | Defining a simple model | Model fitting | Adding a pre-exposure window | Including seasonality, and calendar time | Removing the COVID blip | Considering event-dependent observation time | Studies with more than one drug | Adding a class of drugs | Adding all drugs | Diagnostics for the main SCCS assumptions | Rare outcome assumption diagnostic | Event-exposure independence assumption diagnostic | Event-observation independence assumption diagnostic | Modeling assumptions diagnostic | Additional diagnostics | Power calculations | Time from exposure start to event | Ages covered per subject | Acknowledgments

Building a cohort set by domain/ clinical table | Deriving study cohorts from base cohorts | Considerations when building cohorts

Introduction | Important concepts. | Creating a cluster | Single-node cluster | Executing in parallel | Stopping the cluster | Debugging, logging, and error handling | Support for Andromeda objects

Introduction | Create a cdm object | Instantiate two cohorts in the cdm reference | Generate a sequence cohort | No specific requirements | Important Observations | Specified study period | Specified study period and prior history requirement | Specified study period, prior history requirement and washout period | Specified study period, prior history requirement and combination window | Specified study period, prior history requirement and index marker gap

Introduction | Obtain sequence ratios

Introduction | Table output of the sequence ratio results | Modify type | Plot output of the sequence ratio results | Modify onlyASR and colours

Introduction | Obtaining temporal symmetry | Modify the cohort based on cohort_definition_id | Modify timescale

Introduction

Introduction | Creating a model design | Model design example 1 | Model design example 2 | Model design example 3 | Running multiple models | Validating multiple models | Viewing the results | Acknowledgments

Best practice publications using the OHDSI PatientLevelPrediction framework

Clinical models developed using the OHDSI PatientLevelPrediction framework

Introduction | Software Prerequisites | Windows Users | Mac/Linux Users | Installing the Package | Installing PatientLevelPrediction using remotes | Creating Python Reticulate Environment | Installation issues | Common issues | python environment Mac/linux users: | Acknowledgments

Introduction | Data Splitting Function Code Structure | Example | Create function | Implement function | Acknowledgments

Introduction | Feature Engineering Function Code Structure | Example | Create function | Implement function | Acknowledgments

Introduction | Sample Function Code Structure | Example | Create function | Implement function | Acknowledgments

Benchmark Tasks For Large-Scale Empirical Analyses

How to use the PhenotypeLibrary R package | The full set of predictor phenotypes

Integration of GIS Data into OHDSI Model Building | Motivation | Step-by-Step Process | Step 1: Create Target & Outcome Cohorts | Step 2: Create Generic PLP Lasso Logistic Regression Model in R | Step 3: Split plpData object to train/test, augment labels with EXPOSURE_OCCURRENCE values | Step 4: Reference augmented label objects in custom feature engineering function | Step 5: Apply new train and test datasets to runPlp and evaluate output

Introduction | Useful publication | Main steps for running a network study | Step 1 – developing the study | Step 2 – implementing the study part 1 | Step 3 – implementing the study part 2 (make sure the package is functioning as planned and the definitions are valid across sites) | Step 4 – Publication | Package Skeleton - File Structure

Introduction | Permanence | Technology | Creating an Andromeda object | Closing Andromeda objects | Temporary file location | Querying data from an Andromeda object | Simple meta-data | Using SQL | Dates and times | Batch operations | Safe mode | Saving and loading Andromeda objects | Using Andromeda in your packages | Import dplyr | Import .data from rlang | Beware of variable name confusion

As a diagram

Introduction | SQL parameterization | Substituting parameter values | Default parameter values | If-then-else | Translation to other SQL dialects | Functions and structures supported by translate | String concatenation | Bitwise operators | Table aliases and the AS keyword | Temp tables | Implicit casts | Case sensitivity in string comparisons | Schemas and databases | Optimization for massively parallel processing | Debugging parameterized SQL | Developing R packages that contain parameterized SQL | Spark SQL

Introduction | Terminology | Creating a console logger | Shorthand | Creating a file logger | Creating an error report logger | Creating an e-mail logger | Messages, warnings and fatal errors | Logging when parallel processing | Creating loggers in functions | Shiny log viewer

Introduction | Overview | Populating the cohort_attribute and attribute_definition tables | Example | Creating the cohort attributes and attributes definitions | Using the attributes as covariates

Introduction | Assumptions | Creating the required file structure | In an R package | Using folder structure | Adding a migration | Adding migrator | Unit testing | Common issues | Supporting all database platforms | SQLite column types | Non-existent data

Introduction | Problem statement | Package name | Package purpose | Scope and Intended Use | System Features and Requirements | General | Package dependencies | Data Migration Manager (DMM) Class | Class definition | Example implementation for package | Proposed structure of a migration SQL script | Naming convention | Required parameters | Example SQL | Adding script to list of executed migrations | Results Data model class | Utilities | Add migration function | Create DDL design | Create new DDL | Limitations and scope

Introduction | Creating a schema definition file | Creating a schema | Uploading results

Introduction | Creating the export manager for a package | Saving large results sets with a batch operation | Setup | Exporting a database query result | Performing R operations | Exporting an Andromeda result in batch | Creating a results manifest file

Purpose | Basic usage | Adding replacement variables at runtime

Introduction | Overview | Populate a table with the cohorts to be used for covariate construction. | Example | Creating the cohort attributes and attributes definitions | Using the cohort as covariate

Introduction | Querying | Querying using Andromeda objects | Querying different platforms using the same SQL | Inserting tables | Logging all queries

Introduction | Connecting | Querying | Using dbplyr | Date functions | Allowed table and field names in dbplyr | Temp tables | Temp tables in dbplyr | Cleaning up emulated temp tables | Closing the connection

Introduction | Basic usage | Creating an instance | Pooled connections | Querying a database

Introduction | Simulated data | MaxSPRT | Log likelihood ratio | Critical value | MaxSPRT and empirical calibration | Likelihood profiles | Fitting a null distribution using likelihood profiles | Computing a calibrated critical value | Demonstrating type 1 error with and without calibration | Computing log likelihood ratios and (uncalibrated and calibrated) critical values | Computing overall type 1 error

Introduction | Negative controls | Positive controls | Plot control effect sizes | Systematic error model | Fitting the empirical null, and specifying an assumption | Fitting the systematic error model using positive controls | Evaluating the calibration | Confidence interval calibration | Calibrating the confidence interval | References

Introduction | Negative controls | Plot negative control effect sizes | Empirical null distribution | Fitting the null distribution | Evaluating the calibration | Plotting the null distribution | P-value calibration | Calibrating the p-value | Computing the credible interval | References