Prescription drug manufacturers and distributors of bulk and/or dosage for controlled substances are required to report inventories, acquisitions, and dispositions of all Schedule I and II substances, and Schedule III narcotic and Gamma-Hydroxybutyric Acid (GHB) substances to the Drug Enforcement Agency Automated Reports and Consolidated Ordering System (DEA ARCOS). This system provides total drug amounts (in grams) distributed to retail registrants in each state. The drug amounts are converted to morphine equivalents. This report includes data from zip codes beginning with 191. There are three main limitations associated with the ARCOS dataset. First, the data reflects the distribution of prescription drugs to pharmacies and does not indicate actual prescriptions written or filled, medications taken, or distinct individuals receiving prescriptions. Second, ARCOS data includes opioids used in veterinary medicine and thus may overestimate the amount available for human consumption. Additionally, ARCOS does not distinguish between routes of administration (i.e. oral, IV, or sublingual) which can impact a drug’s potential for abuse and/or diversion.

The treatment statistics are provided by Community Behavioral Health (CBH), a component of Philadelphia Department of Behavioral Health and disAbility Services (DBHIDS. Under the HealthChoices program, CBH provides mental health and addiction treatment services for Medicaid-eligible people in Philadelphia. CBH’s paid claims for addiction treatment services are presented. Prior to the increase in overdose deaths that started nationally in 2014, Governor Tom Corbett initiated major changes to Medicaid eligibility in Pennsylvania. Beginning in 2013, Pennsylvania’s administration introduced Healthy PA, which included a private insurance option for people enrolled in Medicaid. This resulted in a loss of eligible individuals in CBH and a subsequent reduction in members using addiction treatment services. Decreases in addiction treatment services in 2013 and 2014 reflect, in part, decreases in the Medicaid-eligible population. In 2015, Governor Tom Wolf terminated the Healthy PA initiative and Pennsylvania expanded Medicaid under the Affordable Care Act. Given the transition between Healthy PA and Medicaid expansion in 2015, SUPHR recommends caution in interpreting these treatment statistics for that time. Additionally, without data from private insurance, the statistics on Medicaid-funded addiction treatment provide a partial picture of Philadelphians in need of treatment for opioid use disorder.

The following section shows data from the Philadelphia Medical Examiner’s Office (MEO). This includes any accidental death in which drug intoxication was certified as either the underlying or contributory cause of death on the death certificate regardless of residence or incidence location. Deaths due to carbon monoxide poisoning or alcohol intoxication only were excluded. Nonopioid deaths from 2003-2015 include cases that were negative or were not tested at the MEO. The major limitation of this dataset is it can take up to 90 days for toxicology reports to be complete, thus delaying data reporting for up to three months.

Chief complaint and diagnosis code information for all individuals seen in Philadelphia area emergency departments (ED) are reported to the PDPH syndromic surveillance system. The number of drug-related ED visits are assessed daily. There are some limitations to this data. First, due to changes in hospital reporting mechanisms, there are fluctuations in total counts of drug-related ED visits over time. Second, it is often impossible to discern the drug involved in the incident, so drug-specific events are likely an undercount.

Chief complaint and diagnosis code information for all individuals seen in Philadelphia area emergency departments (ED) are reported to this system. The number of drug-related ED visits are assessed daily. There are some limitations to this data. First, due to changes in hospital reporting mechanisms, there are fluctuations in total counts of drug related ED visits over time. Second, it is often impossible to discern the drug involved in the incident, so drug specific events are likely an undercount.

Naloxone administration data is provided by Philadelphia Emergency Medical Services (EMS) and the Delaware Valley Intelligence Center (DVIC). The data presented represents suspected overdose events in which naloxone was administered by EMS, Philadelphia Police, or SEPTA Transit Police. There are a few limitations. First, this data does not reflect the number of doses administered. It also does not reflect the unique number of individuals receiving naloxone. Instead, it is the number of unique suspected overdose events in which naloxone was administered by law enforcement and first responders regardless of the number of doses administered. Additionally, naloxone administration is serving as a proxy for opioid overdose. Individuals who received naloxone may not be true overdoses. Additionally, true overdose events may not have received naloxone by EMS, Philadelphia Police, or SEPTA Transit Police, as naloxone is often administered prior to their arrival.

The National Forensic Laboratory Information System (NFLIS) is used by the DEA to systematically collect drug testing results on drugs seized by law enforcement. The laboratory drug analyses are conducted by federal, state, and local forensic laboratories across the country. There are several limitations of this dataset. First, drug seizures are enforcement driven and do not necessarily reflect the true prevalence or drug combinations found on the street. Additionally, drug testing is driven by legal proceedings, and therefore data shown does not reflect all drugs seized by law enforcement. Rather, the drug seizure data reflects submissions tested for legal proceedings. Finally, NFLIS only records the top three substances detected in the highest quantities in drug samples. Many drug samples contain more than three substances, however.