Community-Based Overdose Prevention Grants

Community-Based Overdose Prevention Grants is sponsored by Maryland Department of Health Behavioral Health Administration Office of Overdose Prevention & Education (OPE). Supports local harm reduction programs, including those targeting women and youth, to reduce substance-related morbidity and mortality.

Implementation of Overdose Prevention in Maryland: Implications for Resource Allocation, Program Scale-Up and Evaluation - PMC As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. .

Author manuscript; available in PMC: 2026 Mar 4. Published in final edited form as: Health Promot Pract. 2023 Nov 22;26(2):332–341.

doi: 10.

1177/15248399231209935 Implementation of Overdose Prevention in Maryland: Implications for Resource Allocation, Program Scale-Up and Evaluation 1 Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University; Find articles by Catherine Tomko 2 Department of Health, Behavior and Society, Bloomberg School of Public Health, Johns Hopkins University; Find articles by Saba Rouhani 1 Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University; 3 Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University; Find articles by Renee M Johnson 1 Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University; Find articles by Ryoko Susukida 1 Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University; Find articles by Himani Byregowda 2 Department of Health, Behavior and Society, Bloomberg School of Public Health, Johns Hopkins University; Find articles by Taylor Parnham Kristin E Schneider , PhD 2 Department of Health, Behavior and Society, Bloomberg School of Public Health, Johns Hopkins University; Find articles by Kristin E Schneider 4 Maryland Opioid Operational Command Center; Find articles by Marianne Gibson Teresa Heath , M.

Ed.

, NCC 4 Maryland Opioid Operational Command Center; Find articles by Teresa Heath 4 Maryland Opioid Operational Command Center; Find articles by Robin Rickard 5 Maryland Department of Health; Find articles by Christine E Boyd 6 Alpert Medical School, Brown University Find articles by Ju Nyeong Park 1 Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University; 2 Department of Health, Behavior and Society, Bloomberg School of Public Health, Johns Hopkins University; 3 Department of Epidemiology, Bloomberg School of Public Health, Johns Hopkins University; 4 Maryland Opioid Operational Command Center; 5 Maryland Department of Health; 6 Alpert Medical School, Brown University PMCID: PMC12955725 NIHMSID: NIHMS2150508 PMID: 37991198 The publisher's version of this article is available at Health Promot Pract United States (U.S.) overdose mortality continues to climb, with Maryland being one of the hardest-hit states.

We summarized the implementation of overdose prevention and response programs in Maryland and identified associations between 2019 per capita opioid overdose deaths by jurisdiction and subsequent overdose program implementation by 2021.

Data on program implementation are from Maryland’s Opioid Operational Command Center (OOCC) Program Inventory, which coordinates the state’s response to overdose comprised of 145 programs across 12 domains (e.g., public health, education, judiciary), including 10 programs designed to broaden naloxone access.

The level of program implementation was dichotomized as substantial implementation versus other levels (i.e., partial, planned, none). We estimated associations between per capita opioid overdose deaths and substantial implementation of: all 145 programs; programs within each of 12 domains; and 10 naloxone programs. Program implementation and overdose mortality data are both jurisdiction-level.

Median substantial implementation was 51% across all programs and 70% among naloxone programs. Overdose mortality was associated with subsequent substantial implementation of programs within the public health domain ( p =0. 04), but not in the other 11 domains.

In conclusion, we did not find evidence that per capita overdose deaths in 2019 spurred overdose program implementation by 2021 with the exception of public health programs. The OOCC Program Inventory is a novel way to track implementation across jurisdictions and program domain.

Results and methodology described here can inform the implementation and evaluation of overdose programs in other jurisdictions across the U.S. Keywords: substance use, public health laws/policies, evaluation methods, implementation, mortality, overdose Drug overdose has been a public health crisis in the United States (U.S.) for more than two decades and data indicate a worsening trend through 2022 ( F.

Ahmad, Cisewski, Rossen, & Sutton, 2022 ; Hedegaard, Miniño, Spencer, & Warner, 2021 ).

State and local health agencies play critical roles in effectively responding to the crisis (and other emergencies) through activities such as: funding and managing resources; engaging stakeholders across multiple sectors and at multiple levels; surveillance; and supporting, incentivizing, and monitoring implementation of prevention and intervention strategies ( Boufford et al. , 2002 ).

State agencies also provide guidance to local jurisdictions and promote coordination and collaboration across jurisdictions.

Federal agencies such as the National Institutes of Health (NIH) and Centers for Disease Control and Prevention (CDC) have invested substantial resources to support state and local response to the overdose crisis, e.g., through NIH’s Helping to End Addiction Long-Term Initiative and CDC’s Overdose Data to Action (OD2A) program. To fully understand the impact of those investments, ongoing monitoring of response strategies is needed.

Following their 2018 study on local health officials’ involvement in activities to address the overdose crisis, the National League of Cities (NLC), an advocacy organization providing support to city leaders across the U.S., highlighted the critical role of documenting local health department (LHD) activities in mounting an effective response ( National League of Cities, 2018 ).

Further, NLC recommended that LHD engage in comprehensive strategies across the continuum of prevention, harm reduction, and treatment, while also collecting and analyzing robust data on programs’ effectiveness for more targeted implementation ( National League of Cities, 2018 ).

A systematic review of published community-level plans for combatting the overdose crisis found that healthcare and law enforcement were the community partners most frequently involved in overdose mortality prevention, but fewer than 10% reported partnerships with stakeholders from other stakeholders such as religious leaders, politicians, and media ( Leece et al. , 2019 ).

Formal evaluation of these strategies and their implementation is rarer still. Maryland ranks third in the nation on opioid overdose deaths, highlighting the importance of an effective public health response ( F. B.

Ahmad, Rossen, & Sutton, 2021 ). Annual overdose fatalities in Maryland have exceeded 2,000 since 2016, driven mainly by opioids ( Opioid Operational Command Center, 2022 ). Maryland saw 2,518 opioid overdose deaths in 2020 and overdose deaths remain at a near-historic high in 2021 ( Maryland Department of Health, 2021 ).

Baltimore City, the state’s largest city, has the most overdose fatalities of any city in the U.S. ( Baltimore City Health Department, 2018 ). In 2017, Governor Larry Hogan established the Opioid Operational Command Center (OOCC) to serve as the primary state body for promoting a comprehensive response to overdose in Maryland. The OOCC works across state agencies (e.g., social services, education) and with LHDs.

OOCC’s three priority areas for local programming include prevention and education, enforcement and public safety, and treatment and recovery.

The structure of the OOCC also facilitates local engagement in overdose response: through OOCC, all 24 of the state’s jurisdictions (i.e., 23 counties and Baltimore city) have an Opioid Intervention Team (OIT), multidisciplinary bodies that promote collaboration to support overdose prevention and coordinate local efforts ( Maryland Opioid Operational Command Center, 2021 ).

Maryland’s OOCC and OIT’s multisectoral programming infrastructure adheres to NLC’s recommendations of robust data collection and analysis to, in part, support program evaluation. The OOCC developed an inventory of 145 programs across 12 domains and use it to query OITs on a quarterly basis about progress toward implementation of these programs.

However, this level of detail about state programming and data collection is rare in the peer-reviewed literature, making it challenging for LHDs and other practitioners to learn from one another.

Other examples of statewide overdose prevention strategy tracking can provide important data on jurisdiction-level program implementation in areas such as harm reduction, treatment, and justice-involved populations, but cover more limited programming than the OOCC Program Inventory ( North Carolina Department of Health and Human Services, 2022 ).

In a 2019 environmental scan of LHD approach to overdose prevention and response, the National Association of County and City Health Officials (NACCHO) reported that 95% of local health departments collected some type of data about opioid use or overdose-related indicators ( National Association of County and City Health Officials, 2019 ).

However, only 28% of these health departments reported evaluating one opioid-related program, with health departments in low-populations areas being least likely to have evaluated a program ( National Association of County and City Health Officials, 2019 ).

The OOCC and its unique Inventory may address this literature gap by showing one example of how to track local implementation of a diverse slate of programs to inform ongoing needs assessment and subsequent targeted implementation for overdose response and prevention. Timely, appropriate, and responsive planning and response to overdose is needed.

Ongoing assessment about how jurisdictions are responding to overdose mortality is critical (e.g., monitoring implementation of specific overdose prevention and response strategies). Data from assessment can be used to inform planning and evaluation in an ongoing way.

OD2A psupports the Maryland Department of Health in collecting, coordinating, and analyzing timely data on overdose morbidity and mortality to inform overdose-related initiatives statewide. The OOCC is one of the entities providing data and coordination toward OD2A goals in Maryland. The present analysis is an initiative of Maryland to monitor comprehensive implementation of strategies to respond to opioid use disorder and overdose.

The following aims represent key OOCC evaluation needs and were developed as part of the ongoing collaboration between the OOCC and evaluation team: 1) describe the OOCC Program Inventory and its potential usefulness in tracking overdose-related program implementation in other locations; 2) examine whether high rates of overdose were associated with future implementation of overdose programs.

Specifically, we sought to determine whether jurisdictions with opioid overdose deaths per capita above the median (30 per 100,000 residents) in 2019 were more likely to have fully implemented: a) a larger proportion of the 145 programs in the Program Inventory; b) a larger proportion of programs within the 12 domains in the Program Inventory; and c) 10 programs designed to expand access to naloxone.

Our findings can be used to assist public health officials in identifying jurisdictions with high overdose mortality and limited implementation of overdose prevention programs, so that those jurisdictions can be provided with additional support for stimulating implementation.

The OOCC Program Inventory is a list of opioid response programs supported by the OOCC and implemented at the jurisdictional level to address opioid overdose in Maryland ( Table 1 ).

The executive order which created the OOCC and OITs named several partners including the local health department, office of emergency management, social services, law enforcement, education, and private sector, non-profit, community and faith-based organizations that should participate each OIT. It is up to the discretion of the OIT if they would like to include additional partners.

In addition to meeting regularly to coordinate efforts, share information and make recommendations to improve their local response to the overdose crisis, the OITs operationalize such recommendations through grant funding provided by the OOCC and other sources.

Maryland Opioid Operational Command Center (OOCC) Program Inventory, program domains, and selected programs SUD screening at enrollment for SNAP, Medicaid, or Temporary Cash Assistance Support programs for families Representative in OIT meetings Substance use information available on campus Student recovery center on campus Representative in OIT meetings Permit transport to alternative destination Representative in OIT meetings SBIRT at ED and inpatient settings Dedicated behavioral health space Dedicated SUD inpatient space “Warm handoff” to SUD providers Naloxone distribution at discharge * Public messaging SUD campaign Pre-trial services program Pre-trial referral-to-treatment Commissioners distribute information cards about treatment Representative in OIT meetings Screening for SUD at intake Request special condition for persons with SUD Treatment monitoring program Representative in OIT meetings Composition consistent with Governor’s orders Co-chaired by a health officer and an emergency management officer Elected official participation SUD programming coordinator responsible for cross-agency coordination Police trained in naloxone * Police outreach to individuals post-overdose Community awareness programming Heroin/overdose coordinator employed by jurisdiction Heroin coordinator participates in OIT meetings Grades 3–12 prevention education Behavioral health services School nurses program, including naloxone available from nurse * “Safe place” in school for students with SUD School-based drug use prevention clubs Community-awareness programming Representative in OIT meetings SUD screening for inmates Pre-trial referral to SUD treatment SUD treatment while incarcerated, including MOUD Day Reporting Center for parole, probation, and pre-trial Facilitated re-entry programs, including naloxone at release * Access to re-entry programs Representative in OIT meetings Assertive Community Treatment Level 2.

5 partial hospitalization Recovery support programs Care coordination for individuals with SUD Recovery-oriented systems of care Syringe services program * Fentanyl test-strip distribution Public Service Announcements Hotline for SUD treatment SUD mobile support services * Drop boxes for unused/expired prescription medication Employer education about SUD Note: SBIRT = Screening, Brief Intervention, and Referral-to-Treatment; ED = Emergency Department; SUD = substance use disorder; MOUD = medications for opioid use disorder; OIT = Opioid Intervention Team; SNAP = Supplemental Nutrition Assistance Program; EMS = Emergency Medical Services; OEM = Office of Emergency Management * Naloxone-specific program The Inventory was developed by the OOCC in 2019 to track and assess the extent to which overdose-related programs and services are implemented in each jurisdiction.

The Inventory is divided into 12 domains which represent agencies and partners essential to addressing the opioid crisis. The Inventory is intended to be comprehensive in the programs tracked though promising programs have been added after discussion with OITs since the Inventory’s creation. Each quarter, OITs submit updates on behalf of their jurisdiction.

We used the OOCC administrative dataset for analysis. Data are from the first quarter of 2021, and therefore represent the results of implementation work that took place in 2020 and by early 2021, after jurisdictions had information about their 2019 overdose death counts.

Outcomes: substantial program implementation The following four codes represented the level of implementation, defined by the OOCC in parentheses for OITs, for each of the 145 programs: 1) substantial programming (“meets or nearly meets the needs of the jurisdiction”); 2) some programming (“some programming in place, but does not meet the needs of the jurisdiction”); 3) programming in development (“will launch within the next year”); and 4) no programming planned (“no programming in place nor within one year of launch”).

Implementation level was dichotomized as substantial versus other. Then we calculated the percentage of substantially implemented programs of all 145 programs.. Substantial implementation percentage was dichotomized based on a median split: high implementation (≥51%) versus low (<51%).

Using the same OOCC administrative dataset, we then calculated the percentage of programs substantially implemented within each of the 12 Inventory domains.

We investigated implementation of programs that included naloxone distribution, education, and training for laypeople (i.e., people who use drugs and their friends or family) and professional first responders (i.e., police, emergency medical personnel, service providers) because of naloxone’s importance to preventing overdose deaths ( Carroll, Green, & Noonan, 2018 ; Lambdin, Zibbell, Wheeler, & Kral, 2017 ).

These approaches have been demonstrated to be effective at reaching the people most likely to be able to reverse an overdose ( Kahn, Wozniak, Vest, & Moore, 2022 ; Katzman et al. , 2020 ). Increasing naloxone access is also an important priority in Maryland.

We, therefore, selected the 10 programs within the Inventory (regardless of domain) that focused on naloxone distribution and training. Implementation was dichotomized as substantial versus other for each program; we then summed the number of substantially implemented naloxone programs for each jurisdiction.

The number of substantially implemented naloxone programs was dichotomized based on a median split: high implementation (≥ 7 programs) versus low (<7). Opioid overdose mortality. Opioid overdose deaths per capita by jurisdiction were unintentional intoxication deaths (i.e., fatal overdoses excluding suicides) resulting from the recent ingestion of or exposure to any opioids, which is derived from death certificates.

Overdose counts reflect deaths occurring in that jurisdiction, rather than deaths by residents of that jurisdiction. We used deaths by location of occurrence because of their relevance for OITs determining the level of program planning and implementation needed to meet the overdose mortality burden within that jurisdiction (data of overdose deaths by residence include deaths happening outside of the state, for example).

Fatal opioid overdose statistics in 2019 were from the Maryland Department of Health’s Vital Statistics Administration. We standardized deaths per 100,000 people by dividing the opioid overdose count by county population and multiplying by 100,000. Per-capita overdose deaths by jurisdiction were dichotomized based on a median split: high (≥30.

0 per 100,000) versus low (<30. 0 per 100,000). County populations were derived from 2019 American Community Survey data ( US Census Bureau, 2019 ).

We produced descriptive statistics for each outcome and per-capita overdose deaths, in total and by jurisdiction. To identify Maryland jurisdictions ranking high in per-capita overdose deaths but low in implementation of all 145 programs, we classified jurisdictions based on 2×2 matrices of implementation by per-capita overdose deaths; we repeated this for the 10 naloxone programs.

We also produced box-and-whisker plots of substantial implementation per 12 Inventory domains to understand variations in implementation by program focus. Finally, we conducted additional analyses to test the statistical significance of associations between per-capita overdose deaths and program implementation, using jurisdiction as the unit of analysis.

Pearson correlations determined the strength of the association between per-capita overdose mortality and percentage of substantial implementation of all 145 programs. Spearman correlations determined the association between per-capita overdose mortality and percentage of fully implemented programs for all 12 domains. Spearman correlations were used because of non-normal distributions of data points within each domain.

Stata/SE v. 15. 1 was used for all analyses.

Average per-capita overdose deaths for Maryland was 38. 2 per 100,000 residents in 2019 (2,106 deaths), and the median across jurisdictions was 30. 0 per 100,000 (OOCC, 2021).

Baltimore City had the greatest level of per-capita overdose deaths (139. 0 per 100,000), whereas Montgomery County had the lowest (8. 0 per 100,000).

Substantial implementation of all programs Across the 24 jurisdictions, the median percentage for substantial implementation for all 145 programs was 51% ( Supplementary Figure 1 ). Table 2 shows the 24 jurisdictions stratified by high/low mortality and substantial implementation.

Harford County had the highest percentage of substantially implemented programs (79%), whereas Calvert County had the lowest (27%); both jurisdictions had per-capita overdose deaths near the median for the state. Five jurisdictions had per-capita overdose deaths above 30. 0 per 100,000 (i.e., the state median) and were at or below the median for percentage of programs with substantial implementation.

The association between per-capita opioid overdose deaths in 2019 and program implementation by 2021 was not statistically significant.

Matrix of 24 Maryland jurisdictions by level of overall substantial implementation (2021) and per capita overdose deaths in jurisdiction per 100,000 individuals (2019) in Maryland Per Capita Opioid Overdose Deaths in Jurisdiction (2019), per 100,000 * Percentage of Overdose Programs (n=145) with Substantial Implementation (2021) Baltimore City (51%, 139. 0) Baltimore Co. (62%, 38.

0) Prince George’s (74%, 11. 0) * Designation of high and low per capita overdose deaths in jurisdiction are based on a median cut-point of 30. 0 individuals per 100,000 Substantial implementation by program domain Figure 1 shows the percentage of substantially implemented programs by program domain; there was notable variation in implementation across domains and jurisdictions.

The social services domain had the highest median percentage of substantially implemented programs (80%), whereas the higher education domain had the lowest (25%).

The median level of substantially implemented programs ranged from 41%−46% for three domains (judicial, corrections, hospital); from 50%−58% for four domains (public health, behavioral health, probation, education); and from 60%−67% for the remaining three domains (police, EMS, OIT administration).

There was a positive and statistically significant association between per-capita overdose mortality and percentage of substantially implemented programs for only one domain, public health ( ρ =0. 43, p =0. 04).

For the remaining 11 domains, the association between overdose mortality and substantial implementation was not statistically significant ( p >0. 05). Substantial implementation percentage of overdose prevention programs and association between substantial implementation percentage and per capita overdose deaths in jurisdiction per 100,000 individuals (2019), by domains Note.

Rho ( ρ ) is the statistic representing the magnitude of association between 2019 per capita overdose deaths in jurisdiction per 100,000 individuals and 2021 substantial implementation percentage of each domain. Asterisk denotes statistical significance at the p <0. 05 level.

Associations between proportion of programs with substantial implementation were not statistically significant for 11 of the 12 domains. Substantial implementation of naloxone programs Five of the 10 naloxone programs had been substantially implemented in greater than 20 of Maryland’s 24 jurisdictions ( Table 3 , Supplementary Figure 2 ).

Those five programs were: naloxone training; police trained in naloxone; naloxone distribution; naloxone available at K-12 schools; and police carrying naloxone. Mobile substance use disorder (SUD) support services and syringe services programs were the least implemented. Notably, all four jurisdictions that had substantially implemented mobile SUD support services were jurisdictions with high per capita overdose deaths.

Number of counties with full implementation of 10 naloxone programs in 2021, classified by whether counties were above or below the 2019 state median per capita overdose deaths in jurisdiction per 100,000 individuals (30. 0 per 100,000) Per Capita Opioid Overdose Deaths in Jurisdiction (2019), per 100,000* Naloxone training.

The jurisdiction provides training to general public on the administration of naloxone, including recognizing the signs and symptoms of an overdose. Police trained in naloxone. The police department or sheriff’s office has arranged for universal naloxone training for all sworn staff.

Naloxone distribution. The jurisdiction distributes naloxone to people in the community at higher risk of encountering someone who might have a potential overdose. Naloxone available in schools.

Naloxone is made available to K-12 school nurses, for use by people within the schools. Police carry naloxone. All sworn law enforcement officers carry naloxone in their vehicles.

Naloxone at release. Naloxone is provided to people upon release from state prisons or jails. EMS “leave- behind.

” EMS personnel leave naloxone behind at the scene of overdose calls, either with family members or overdose survivors who refuse transport. Naloxone distribution at discharge. Local hospital provides naloxone (not just a prescription) to patients who present due to overdose or other substance-related problems.

Syringe services programs. Provide prevention services: including needle/syringe distribution and collection; overdose prevention and education; naloxone distribution; access to SUD counseling, treatment, and recovery services; HIV/Hepatitis C/STI testing and linkage to care; reproductive health education and services; wound care; and various additional services, either directly or through referral partnerships.

Mobile SUD support services. The jurisdiction operates a dedicated vehicle, such as a van or RV, that offers prevention education, counseling, naloxone, and information about treatment resources. The median number of naloxone programs substantially implemented across jurisdictions was seven ( Table 4 ).

Calvert county had the smallest number of programs with substantial implementation (three programs), whereas Washington and St. Mary’s counties had the largest (nine programs each). Seven of the 11 jurisdictions with per capita overdose deaths above the state median had substantial implementation of at least seven naloxone programs.

Of the remaining four jurisdictions with high per capita overdose deaths, three counties had substantial implementation of six naloxone programs, and two counties had substantial implementation of five.

Matrix of 24 Maryland jurisdictions by level of substantial implementation of naloxone programs (2021) and per capita overdose deaths in jurisdiction per 100,000 individuals (2019) in Maryland Per Capita Opioid Overdose Deaths in Jurisdiction (2019), per 100,000 * Number of Fully Implemented Naloxone Programs (N=10) with Substantial Implementation (2021) Baltimore City (8, 139. 0) Prince George’s (8, 11.

0) * Designation of high and low per capita overdose deaths in jurisdiction are based on a median cut-point of 30. 0 individuals per 100,000 The Maryland OOCC tracks local implementation of 145 programs designed to combat the opioid overdose crisis across the state.

We sought to assess the association between per-capita opioid overdose deaths in 2019 and substantial implementation of programs by the first quarter of 2021, with the rationale that greater per-capita overdose deaths might motivate uptake of programs.

We did not find evidence of an association between overdose deaths and program implementation, with the exception of the public health domain, where we found that higher 2019 mortality was associated with greater substantial implementation of these programs.

Although 2019 per-capita overdose deaths may have stimulated implementation within the public health domain, this association should not be interpreted as causal and additional studies with more rigorous designs are needed to confirm the association. We identified several jurisdictions with high per-capita opioid overdose deaths that had noteworthy gaps in implementation and may require additional support to increase program uptake.

In three jurisdictions, per-capita overdose deaths were above the median for the state, but substantial implementation of all programs was below the median for each respective metric. Four jurisdictions showed the inverse relationship: high implementation of programs despite low per-capita overdose deaths two years prior.

Further, there was notable variation across jurisdictions in the percentage of programs with substantial implementation across the 12 domains. We did not find a statistical association between per-capita overdose deaths in 2019 and implementation of naloxone programs by 2021. All jurisdictions had substantial, partial, or planned implementation of at least five of the 10 identified naloxone programs.

Mobile SUD support services and syringe services programs were the least frequently implemented across all jurisdictions, including those with high per capita overdose deaths.

Given their effectiveness in reaching high-risk populations and delivering naloxone, implementation of syringe services programs should be a priority, especially for jurisdictions with high per-capita overdose deaths, yet only 8 jurisdictions currently have an operational program ( Carroll et al. , 2018 ; Feuerstein-Simon et al. , 2020 ).

Three counties with planned implementation in the first quarter of 2021 now have operational programs; however, these jurisdictions have per-capita overdose deaths below the state median.

Ultimately, increasing naloxone “saturation,” or the number of distributed naloxone doses that result in a substantial decrease in mortality, has been shown to be effective at reducing overdose rates in settings across the U.S. by concentrating naloxone in the hands of people who use drugs and likely bystanders to overdose ( Keane, Egan, & Hawk, 2018 ; Madah-Amiri, Gjersing, & Clausen, 2019 ).

Additional considerations That we did not observe an association between per-capita overdose deaths and subsequent program implementation raises several questions: If overdose deaths do not drive uptake of programs, what are the factors that do? Why do some jurisdictions have high implementation but not others?

Research has shown several county-level characteristics that may explain the combination of high overdose mortality and low implementation of programs, including political attitudes, low investment in public health programs, and geographic characteristics (e.g., metro- vs. micropolitan, rurality, etc.) ( Haffajee, Lin, Bohnert, & Goldstick, 2019 ).

Future studies should examine whether these county-level characteristics could potentially explain the association between overdose mortality and implementation.

Other factors such as financial constraints, supply chain issues that disrupt harm reduction supply availability, human resources, political will, or local or state policies pose challenges to implementation, making an examination and understanding of these factors critical to address programmatic gaps ( Bennett & Elliott, 2021 ; Lambdin et al. , 2017 ; Smart et al. , 2020 ).

The emergence of the COVID-19 pandemic had catastrophic effects on nearly every type of healthcare delivery system, as well as on the public health infrastructure broadly. Public health and emergency management officers and agencies charged with opioid response were reassigned to COVID response, which likely slowed progress in program implementation during 2020.

Alternatively, it is plausible that counties had real-time information on overdose trends in 2020, which may not have been associated with 2019 trends. Future work could examine whether county-level overdose deaths observed in 2019 were a good proxy to 2020 trends, or whether there were counties that responded to newer information that was made available internally. These findings are subject to some important limitations.

First, the OOCC does not provide detailed definitions of implementation levels which may introduce a reporting bias that masks associations between mortality and implementation. The OOCC is currently working to standardize these definitions for future evaluations to capture the most accurate implementation data.

Second, implementation data is self-reported by OITs to the OOCC and may be subject to desirability bias or overreporting of implementation progress (OITs report implementation to the OOCC, from which they receive funding). Third, as we only used 2019 mortality data from the VSA, we caution that relying on per-capita deaths from one year means that data from jurisdictions with smaller populations may fluctuate greatly from year to year.

Finally, some strategies or programs in the Inventory are not evidence-based. This analysis aimed to describe the implementation of all programs as they currently exist; a subsequent step is to determine the evidence base of the strategies in the Inventory.

Implications for practice and research The Program Inventory is a unique way to track overdose programs and their implementation across jurisdictions and between multiple public and private sectors. The Inventory may bolster overdose prevention and response efforts in other state and local health departments by showing a breadth of programming in multiple sectors with diverse stakeholders that may spur potential expansion.

The Inventory can also inform efforts to track implementation over time and across jurisdictions in support of program evaluation, a critical but lagging area of LHDs’ overdose response ( National Association of County and City Health Officials, 2019 ).

Specifically, our methods show one way to evaluate implementation progress and its association with relative need across jurisdictions by integrating VSA mortality data with OOCC administrative data. This method can guide highly targeted implementation support, whether by geographic area, inventory domain, or specific programs.

The Inventory may also be a useful template by which other states track overdose-related program implementation, establishing its utility beyond Maryland. Despite the strengths of the Inventory, there are several ways in which it may be improved for future implementation tracking.

The Inventory may benefit from additional contextual information (e.g., program complexity, number of people served) about implementation beyond a discrete classification of implementation status. Next, incorporating a racial justice perspective into program implementation tracking would meet calls for increasing racial equity in responses to the