This is a sources sought notice from the National Institute on Drug Abuse (NIDA) to identify capable small businesses for data extraction, management, analysis, and reporting of a propensity score matched pharmacoepidemiological study. The study will focus on the effects of GLP-1 receptor agonists (GLP-1RAs) on HIV and alcohol and substance use disorder (ASUD) related outcomes in a general patient population using the Veterans Aging Cohort Study (VACS) electronic health record data. Vendors must have demonstrated experience in robust pharmacoepidemiology analyses with ASUD and HIV-related outcomes, and possess full access to VACS data and an established data analysis pipeline. The period of performance is 12 months with three 12-month option periods. Interested vendors should submit a capability statement addressing project requirements, including employee numbers, personnel qualifications, GSA schedule contracts, and company information. Responses are due electronically by 5:00 PM EST on 05/06/2026.
The capability statements are due no later than 5:00 PM EST on 05/06/2026.
The bid notice states that capable offerors must demonstrate their experience with, and ability to complete the following activities: the objective of the proposed project is to examine glp 1ras and their potential health benefits for plwh, in addition to alcohol and substance use disorders outcomes. Using electronic health record data from the department of veterans affairs va, the capable vendor will investigate the association between glp 1ras prescribed for any indication and hiv and asud related outcomes in adults, led by targets provided by nida staff. Individuals exposed to glp 1ras will be propensity score matched to up to five unexposed individuals. Multivariable linear regression models will estimate changes in hiv and asud related outcomes before and after exposure. Pharmacy records will be extracted to identify patients who did and did not receive the medication under investigation dispensed at va pharmacies. The capable vendor, after consulting with nida staff, will identify the first prescription for the medication during the study period and will require a washout period for new episodes of medication exposure. Exposure to the medication is defined as receipt of two or more fills for a minimum number of continuous days determined as part of the analysis, for any indication, from va outpatient clinics. For unexposed comparators, the capable vendor will select patients who attended at least one of the same clinics, but never received the medication, ensuring that unexposed patients came from the same source population, were exposed to an overall similar medical care, and had equal opportunities to receive the same medication. The capable vendor, with input from nida staff, will create an index date also referred to as baseline, defined as the first fill date for exposed patients and a randomly selected outpatient visit date for unexposed patients. To balance the distribution of all potential confounders across treatment groups, we will perform propensity score approaches e. g. , matching, weighting. This shall be done by first modelling the probability i. e. , propensity of receiving the medication as a function of measured covariates associated with the exposure and outcome using a multivariable logistic regression model. Variables included in the propensity score model will be guided largely by our previous work. Covariates will be summarized using descriptive statistics, stratified by treatment group. To assess balance, we will calculate absolute standardized mean differences smd between audit c categories and consider 0. 1 or less as balanced. The capable vendor will be responsible for pulling, organizing and cleaning the specific datasets. Nida irp will lead the development and preparation of the manuscript. The data are generated via the va birth cohort study and housed at the va connect medical center in west haven, ct. The capable vendor will be completing data management and analysis for this project.
