Titled "Mortality and Economic Impact of a Police Sweep on People Who are Unsheltered and Who Use Opioids" and colloquially called "Sweeps", this project uses the RESPOND model to model the mortality and health-economic impacts of a street sweep on a cohort of individuals who are unhoused and use opioids.
RESPOND (Researching Effective Strategies to Prevent Opioid Death) is a population-based state transition model that simulates a population with opioid use disorder (OUD), with characteristics including the natural history of OUD, movement on to and off medication for opioid use disorder (MOUD) and with outcomes such as overdose and cost. The model can be used to evaluate the effectiveness and cost-effectiveness of different strategies for offering MOUD, and is informed by data from the Massachusetts Public Health Data Warehouse, clinical trials, and national insurance claims data. We calibrated the model using Massachusetts-specific estimates of disease prevalence, admissions to treatment facilities, and overdose deaths. To learn more about the model and view our figures and user guide check out our Model Materials.
https://www.syndemicslab.org/respond
R
-> Make sure Rscript is on PATH
Git
1. Clone this repository
- On GitHub.com, navigate to the main page of the repository. Above the list of files, click <> Code.
- Copy the URL for the repository.
- Open Git Bash, and navigate the working directory to the location where you want the cloned directory.
- Type git clone, and then paste the URL you copied earlier.
git clone https://github.com/SyndemicsLab/Addressing-Encampments.git
- Press Enter to create your local clone.
2. Clone the RESPONDv1 repository into the newly made Sweeps Repository Follow the same instructions as #1, with the RESPONDv1 repository.
git clone https://github.com/SyndemicsLab/RESPONDv1.git
3. Use the RESPONDv1 repository to generate inputs 1 - 1,000
- Navigate to RESPONDv1.
- For Windows, open Command Prompt as an administrator. For Unix,
- Navigate to the 'ec_base_inputs' folder.
cd C:\YOURFILEPATH\Addressing-Encampments\ec_base_inputs
- Run buildbase.bat using the following command. This will create 1,000 input sets and copy them into 'Addressing-Encampments/ec_base_inputs'.
buildbase.bat
- This will take a while! Maybe go for a walk or make yourself a little snack.
4. Update the age structure
- For Windows, go back to your Command Prompt. Make sure you're still in the 'ec_base_inputs' folder, and run the age_chunks_5_to_2.R script to update the age structure of your base 1,000 files.
- For Unix,
Rscript age_chunks_5_to_2.R
- This will also take a while. Do you have a pet in need of some attention?
5. Run the Status Quo batch file to run that strategy
- Remain in the 'ec_base_inputs' folder. Run the Status Quo batch script.
Run_SQ.bat
- This will do a number of things. First it will read the base inputs, update them to reflect the status quo strategy, and populate them into the 'sq_inputs' folder. Then it will update the relevant R files (user_inputs.R and input_file_paths.R) for each input file. Finally, it will run the RESPOND model and populate the outputs in the 'sq_outputs' folder.
6. Run the results script to calculate results
- Navigate to the 'SQ_outputs' folder:
cd C:\YOURFILEPATH\Addressing-Encampments\SQ_outputs
- Run the Results script:
Rscript Results_SQ.R
This will calculate a number of outcomes and provide the results in CSV form in the output folder.
7. Repeat steps 5-8 for the other strategies: Sweep, Housing with MOUD Required, and Housing with MOUD Choice
- Run_SQ.bat will run the Status Quo strategy (from the 'ec_base_inputs' folder).
- Run_SW.bat will run the Sweep strategy (from the 'ec_base_inputs' folder).
- Run_HM.bat will run the Housing (MOUD Required) strategy (from the 'ec_base_inputs' folder).
- Run_HC.bat will run the Housing (MOUD Choice) strategy (from the 'ec_base_inputs' folder).