David Fong

Recently Published

Vaccination effectiveness - AztraZeneca and reductions in COVID-19 cases and death
Is Astra-Zeneca doses administered (at least a single dose) as a proportion of *all* COVID vaccination administered related to a national reduction in COVID-19 cases and death?
Cancer Screening 2021
Tamarind Project SAFE
GPstat! user manual
User documentation for GPstat!. GPstat! is a 'near-future' tool to opportunistically find opportunities for screening and chronic disease management.
Results marked ‘non-urgent discussion’ - rate of notification without a result notification procedure
Without any procedure to ‘recall’ patients who have a result marked ‘non-urgent discussion’, how many results are notified over the following eleven weeks?
Pharmacy in Practice
Does a pharmacist working in general practice see high-risk patients, improve patient safety and increase practice revenue?
serum uric acid
Spur Afrika Mobile Clinic Report 2020
Spur Afrika mobile clinics were held in January 2020, seeing children from several schools in Kibera, Kenya, and children sponsored by the Spur Afrika Elimsha program. Health activities included medical and dental checks, dental hygiene education and medication education. 885 children were seen, ranging in ages from seven to eighteen years.
Editable DataTables (DT) package for shiny apps. Originally created by Jason Bryer. Data columns to view or edit can be selected. Different datatypes can be edited. Valid input choices and interaction with persistent storage can be defined through callbacks. In the case of limited selection choices for data columns with ‘dropdown’ inputs, the choices can be defined and dynamically changed. Available in general-purpose and module versions.
Relationship between Morbidity potentially attributable to air pollution and Gender in Kibera
The relationship between gender and symptoms and conditions potentially aggravated by air pollution is explored amongst school-age children seen during a screening clinic in Kibera, an informal settlement of Nairobi, Kenya. The interim results of this study finds female children had an estimated odd-ratio of 2.23 (95% confidence interval 1.17 to 4.39) of having potentially air-pollution related conditions or symptoms (such as cough, asthma or eye irritation) compared to male children.
dMeasure - an interface to the Best Practice electronic medical record
Interface to Best Practice Electronic Medical Record database to provide information about opportunities for preventative health activities, screening and chronic condition cycles of care and care planning and billings. Part of the “Daily Measure” software suite and the back-end to browser-based dashboard “GP Stat!”.
World Population 7,777,777,777
When, according to currently available population data from, will the world population reach 7,777,777,777? for ENVR-101 Introduction to Sustainability and Environmental Management
Cervical screening intensive telephone-based recall
Regular cervical cancer screening reduces the risk of cancer incidence by approximately a factor of 3 to 4 compared to no screening (for women aged 35 to 79 years), and reduces mortality by a factor of 4 to 10. Intensive telephone-based recall, with a particular interest in effectiveness with refugee/asylum-seeking populations (where English is potentially not the first language)
Polypharmacy Project 2020
GPstat! installation and basic configuration
Installation and configuration of GPstat!, a screening, preventative health activity, chronic disease management, care planning and billing aid for the Best Practice Electronic Medical Record.
ENVR E210 Assignment 9 ANOVA
ANOVA chicken feed and carbon dioxide
ENVR E-210 Assignment 6 Multiple regression
Migratory bird data. Murder rates in the states of the United States
Framingham Risk Equation : an implementation in R
Framingham Risk Equation, using the equations from Anderson et al "Cardiovascular Disease Risk Profiles" (1991), and producing the same results as the online calculator at (National Vascular Disease Prevention Alliance)
ENVR-E-210 : Assignment 1a - Sustainability data
Greenhouse gases and methane. Assignment 1a of Critical Analysis of Environmental Systems (Harvard Extension School)
Colorectal cancer screening, Kensington coHealth 2019
Colorectal cancer screening at Kensington site of coHealth, 2019. A proposal to increase the rate of surveillance
Improving Diabetes cycles of care - Kensington 2019
Methods and results of improvement in the monitoring and promotion of recommended interventions among patients with diabetes who visit the coHealth Kensington general practice.
Predicting Stock Returns with Cluster-Then-Predict
Predicting stock returns in December based on preceding monthly stock returns. Using a cluster-then-predict methodology. (MITx homework assignment)
Separating Spam from Ham
Nearly every email user has at some point encountered a “spam” email, which is an unsolicited message often advertising a product, containing links to malware, or attempting to scam the recipient. In this homework problem, we will build and evaluate a spam filter. (MITx "Analytics Edge" homework assignment)
Automating Reviews in Medicine - analyzing abstracts
Reviews of medical literature are often performed manually, with multiple people reviewing each search result, this is tedious and time consuming. In this problem, we will see how text analytics can be used to automate the process of information retrieval.
Predicting vandalism in Wikipedia
Wikipedia is a free online encyclopedia that anyone can edit and contribute to. It is available in many languages and is growing all the time. On the English language version of Wikipedia: One of the consequences of being editable by anyone is that some people vandalize pages. In this assignment we will attempt to develop a vandalism detector that uses machine learning to distinguish between a valid edit and vandalism.
Investigating Enron e-mails
Predicting 'responsiveness' of e-mails from Enron during investigation into electricity price manipulation
Turning Tweets into Knowlege
Predicting sentiment in tweets about Apple. An example from MITx "The Analytics Edge"
Predicting Earnings from Census Data
The United States government periodically collects demographic information by conducting a census. In this problem, we are going to use census information about an individual to predict how much a person earns – in particular, whether the person earns more than $50,000 per year. This data comes from the UCI Machine Learning Repository. The file census.csv contains 1994 census data for 31,978 individuals in the United States. (assignment for MITx “The Analytics Edge")
Letter Recognition
Identifying letters from images. Assignment for MITx "The Analytics Edge"
Diabetes outcomes
Improving diabetes outcomes. Daily case-finding
Understanding Why People Vote
Predicting why people vote using logistic regression and CART models. An experiment in various types of social pressure. (Assignment for MITx The Analytics Edge)
Parole Violation Prediction
Predicting parole violations based on potential parolee characteristics. (Assignment for MITx Analytics Edge)
Predicting Loan Repayment
Predicting loan repayment from borrower and loan characteristics using logistic regression and receiver operator characteristic curve (ROCR). (assignment for MITx Analytics Edge)
Claims Data Medicare and Medicaid
Predicting Medicare (USA) and Medicaid claims using Classification and Regression Trees (CART). Part of Analytics Edge (MITx) exercise
Judge, Jury, Classifier
Predicting Judge Steven's decisions in the Supreme Court of the United States
Zostavax (herpes zoster) immunisation, coHealth Kensington
Improving Zostavax immunization coverage among eligible population at coHealth Kensington.
Weather Variability to Estimate the Response of Savings to Transitory Income in Thailand
Replicating the data analysis of "Using Weather Variability to Estimate the Response of Savings to Transitory Income in Thailand (1992)" by Christina Paxson
Results yet to be notified, coHealth Kensington 2015-2017
Proportion of results not yet notified (of results marked 'Discuss') at the Kensington Clinic of coHealth, 2015-2017
SpurAfrika 2018 short slideshow - pitch
for John Hopkins Data Products Course
Population and General Practitioner numbers in Australia
by Primary Health Network areas Population - 2014 data General Practitioners - 2016 data
Storm impacts on public health and economy
Analysis of U.S. National Oceanic and Atmospheric Administration (NOAA) storm database 2007 to 2011, at the request of John Hopkins University Reproducible Data training course.