Citizen Portal
Sign In

Get Full Government Meeting Transcripts, Videos, & Alerts Forever!

Get email alerts on the Opioid Response Ems topic

No spam. Unsubscribe anytime.

Lowell sees decline in overdose calls; health department pilots machine-learning classification

AI-Generated Content: All content on this page was generated by AI to highlight key points from the meeting. For complete details and context, we recommend watching the full video. so we can fix them.

Summary

Lowell Health Department reported a substantial drop in opioid-related calls compared with recent peaks and described a machine‑learning project with Tufts to automate case classification; EMS response times and staffing were also discussed.

Lowell Health Department staff told the Board of Health on March 5 that opioid-related calls and Narcan administrations have fallen significantly from the epidemic peak and that the department is piloting machine‑learning to help categorize overdose reports.

“We had 460 overdoses in the city last year,” said Mr. Kelly, who presented the January opioid report and a year‑end recap covering 2011–2024. He said the city’s overdose volume has fallen toward earlier levels and that the drop was evident in January reporting. “The numbers are so low,” he said, after the report was re-run to confirm the figure.

Mr. Kelly described an existing internal program in which a staff member, “Joe,” reviews and categorizes opiate reports and said Tufts School of Medicine data scientists have trained models to reproduce that work. “Right now they’re up to about 94% accuracy,” Mr. Kelly said. The department has applied for a CDC grant to expand the automated system to five agencies nationwide to test whether the approach generalizes beyond Lowell.

The department also presented the fourth-quarter EMS and response-time report. Staff said total call volume remained roughly steady while some incidents compressed into multiple simultaneous calls, and that out‑of‑service or “out-of-shoot” times are a major factor in extended response times. “We’re not fully staffed,” Mr. Kelly said, explaining that workload and fewer trucks on the road increase on-scene or turnaround time. The department has begun running a daily manual ‘‘shoot’’ report and distributing it to supervisors to address patterns in delayed response times.

The board asked for small edits to the published report (for example, correcting a mislabeled quarter column and clarifying where intubation entries apply to Lowell versus the company overall). Staff said they would correct the packet for the next meeting.

No formal action was taken on the reports; the Board received the presentations and asked staff to return corrected materials.