METHODOLOGY

The Sun Journal analyzed data from Maine’s Prescription Monitoring Program, or PMP, from 2016 to 2021; and data from the U.S. Drug Enforcement Administration’s Automation of Reports and Consolidated Orders System, or ARCOS, from 2006 to 2014.

Following a court order in 2019, The Washington Post gained access to the 2006-to-2014 data from the ARCOS database. The court order unsealed only those eight years of records. The Sun Journal was able to obtain and analyze the ARCOS data because The Post made the raw data available to the public.

The Sun Journal also analyzed data from Maine’s Prescription Monitoring Program, which it was able to access via a Freedom of Access Act request and a data-sharing agreement signed with the database’s administrators, the Maine Department of Health and Human Services. Because the PMP houses patient information, the Sun Journal was only able to access records that identified the pharmacy at which a prescription was filled, but not any patient- or prescriber-identifying information.

Because court order unsealed only certain years of ARCOS data and the Maine law governing the PMP requires DHHS to destroy data older than six years, the Sun Journal was not able to access data for 2015.

The ARCOS and PMP databases are not a line-by-line exact match: The ARCOS data describes shipments of sales to buyers (pharmacies and practitioners, among others), while the PMP data describes singular transactions in which a prescription is filled and dispensed to a patient at a pharmacy. The itemized raw data describes a transaction of a single buyer.

In order to account for these differences, the Sun Journal analysis includes only transactions of tablets or capsules (pills) of oxycodone and hydrocodone at retail or chain pharmacies in Maine. Hospital pharmacies, out-of-state mail order pharmacies and practitioners, for example, were not included.

The analysis looked at how many pain pills flowed in and out of Maine pharmacies per year and calculated a pills-per-person rate by county and by individual pharmacy. In 2011, for example, a total of 6,358,650 oxycodone and hydrocodone pills went to pharmacies in Androscoggin County, which amounts to 59 pills per county resident when adjusted for population.

From 2006 to 2014, the U.S. DEA recorded 61,232,334 pain pills that went to Androscoggin County, enough for an average of 47 pills per person each year.

In addition to adjusting for population, the pills-per-person rate for individual pharmacies is adjusted for the number of pharmacies in each county per year to account for the rural areas of the state and/or areas where there may not be many pharmacies.