IHP 515 Module 7 Milestone Three Example

Milestone Three: Surveillance Improvements and Analysis Plan for Harlan County Overdose Deaths

[Student Name]

Southern New Hampshire University

IHP 515: Population-Based Epidemiology

Module Seven Milestone Three

[Instructor Name]

[Date]

Milestone Three: Surveillance Improvements and Analysis Plan for Harlan County Overdose Deaths

Harlan County's descriptive epidemiology produced four hypotheses: that fentanyl in the drug supply drove the county-wide rise, that the 2020 treatment program closure contributed, that release from jail without treatment raises risk and that later arrival of fentanyl and lower treatment access explain the faster rise among Black residents. Testing them requires better data and a clear analysis plan. This milestone describes improvements to overdose surveillance, the data linkages needed, analyses for each hypothesis, reporting standards, privacy protections and a timeline.

What Surveillance Should Do

Thacker and Berkelman (1988) characterize surveillance as sustained, methodical gathering, analysis and interpretation of health data needed to plan, carry out and evaluate public health practice, closely linked to timely sharing with those who need to act. They identified attributes by which surveillance systems can be judged, including simplicity, flexibility, acceptability, sensitivity, representativeness, timeliness and usefulness. Harlan's current overdose surveillance relies on death certificates, which take three to six months to finalize, and quarterly emergency department summaries, so it is neither timely nor detailed enough to guide rapid response.

Improving Timeliness

The county will add near-real-time syndromic surveillance of emergency department visits for suspected overdose, drawing on chief complaint and diagnosis data already sent to the state. An automated alert will notify the epidemiologist when weekly visits exceed the expected range by two standard deviations, allowing rapid warnings to harm reduction partners and hospitals when a dangerous batch appears. Emergency medical services naloxone administrations will be added as another early signal.

Improving Completeness

The medical examiner will adopt a standard toxicology panel for all suspected overdose deaths, including fentanyl analogs, so that drug involvement is recorded consistently. Death investigation forms will capture circumstances such as whether the person was alone, whether naloxone was given and recent release from incarceration, fields that are often missing today.

Data Linkage

Testing the hypotheses requires linking data sources at the individual level. Death records will be linked with county jail release records, the state prescription drug monitoring program for buprenorphine, opioid treatment program enrollment records and emergency department visits, using name, date of birth and other identifiers, with linkage performed by the state health department's data unit so that analysts receive de-identified files.

Table 1. Hypotheses, Data and Analyses

Note. Each hypothesis is matched to specific data and methods.

National Comparisons

Rudd et al. (2016) documented national increases in opioid overdose deaths and the growing role of synthetic opioids. National and state rates, standardized for age, will serve as benchmarks for Harlan's trends, allowing the county to see whether its pattern mirrors or departs from broader trends and whether local interventions coincide with divergence.

Analysis Methods

Trends will be analyzed with quarterly age-adjusted rates and, for small groups, three-year rolling averages with confidence intervals. The jail release cohort will compare twelve-month overdose death risk between those treated and untreated at release, using regression to adjust for demographic factors, earlier overdoses and whether the person had stable housing. The closure question will pit mortality among people the program had been treating against a matched group of people with opioid use disorder never enrolled, before and after the closure. Racial differences will be examined through stratified trends in fentanyl detection and treatment receipt.

Reporting Standards

Von Elm et al. (2007) developed the STROBE statement to improve reporting of observational studies, listing items such as clear descriptions of study design, setting, participants, variables, data sources, bias, study size, statistical methods, results including missing data and limitations. The county's reports on the cohort and closure analyses will follow STROBE, so that readers can judge the quality of the evidence and other health departments can compare their results.

Privacy and Ethics

Linked records include sensitive information about drug use and incarceration. Linkage will occur under a data use agreement with the state, analysts will work only with de-identified data in a secure environment and results for groups smaller than ten will be suppressed in public reports. The project will be reviewed by the health department's data governance committee and, for any publication, by an institutional review board.

Evaluating the Surveillance System

The improved system should itself be evaluated after a year, using the attributes Thacker and Berkelman describe. Timeliness will be measured as the days between an overdose surge and the first alert to partners. Completeness will be measured as the share of death records with a full toxicology panel and circumstance fields filled in. Sensitivity of the syndromic alerts will be checked by comparing flagged weeks with later confirmed increases in deaths. Usefulness will be judged by whether partners acted on alerts, for example by distributing extra naloxone or issuing warnings, and by feedback from hospitals and harm reduction groups. Acceptability will be assessed by asking emergency departments and the medical examiner whether the new reporting demands are manageable. The evaluation will guide adjustments before the system becomes routine.

Timeline and Dissemination

Syndromic alerts and standardized toxicology will begin within three months. Data linkage agreements will take about six months, with the jail cohort and closure analyses completed within a year. A monthly overdose dashboard will be shared with the board, hospitals and harm reduction partners, and a detailed annual report will present the analyses.

Community Partners

Harm reduction organizations and people with lived experience of drug use will be invited to review the dashboard and alerts, since they are often first to notice changes in the drug supply and can help interpret findings in ways data alone cannot.

Conclusion

Better surveillance and linked data will let Harlan County move from describing its overdose crisis to testing why it happened and responding faster when danger rises. Clear methods, transparent reporting and strong privacy protections make the plan both useful and trustworthy.

References

Rudd, R. A., Aleshire, N., Zibbell, J. E., & Gladden, R. M. (2016). Increases in drug and opioid overdose deaths: United States, 2000-2014. MMWR. Morbidity and Mortality Weekly Report, 64(50-51), 1378-1382. https://doi.org/10.15585/mmwr.mm6450a3

Thacker, S. B., & Berkelman, R. L. (1988). Public health surveillance in the United States. Epidemiologic Reviews, 10(1), 164-190. https://doi.org/10.1093/oxfordjournals.epirev.a036021

von Elm, E., Altman, D. G., Egger, M., Pocock, S. J., Gøtzsche, P. C., & Vandenbroucke, J. P. (2007). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. Annals of Internal Medicine, 147(8), 573-577. https://doi.org/10.7326/0003-4819-147-8-200710160-00010