Computer Algorithms Improve Timeliness of Overdose Data

By Pat Anson, PNN Editor

An automated process using computer algorithms to analyze death certificates would speed up and improve data collection on drug overdose deaths, according to a new study by UCLA researchers.

The current system used to track U.S. overdose deaths relies on medical examiners and county coroners – including some with little medical training -- to determine the cause of death and drugs involved. Death certificates are then sent to local jurisdictions or the Centers for Disease Control and Prevention, which codes them according to the International Statistical Classification of Diseases and Related Health Problems, Tenth Edition (ICD-10).

The coding process is manual and time consuming, resulting in delays of several months before the deaths are included in CDC overdose data. With drug deaths at record levels and more dangerous substances entering the illicit drug supply, researchers say that antiquated system delays an effective public health response.

"The overdose crisis in America is the number one cause of death in young adults, but we don't know the actual number of overdose deaths until months after the fact," said lead author David Goodman-Meza, MD, assistant professor of medicine in the division of infectious diseases at the David Geffen School of Medicine at UCLA.

"We also don't know the number of overdoses in our communities, as rapidly released data is only available at the state level, at best. We need systems that get this data out fast and at a local level so public health can respond. Machine learning and natural language processing can help bridge this gap."

Goodman-Meza and his colleagues used computer algorithms to analyze the text for keywords in nearly 35,500 death certificates from nine U.S. counties in 2020. The counties include major cities such as Chicago, Los Angeles, San Diego and Milwaukee.

The researchers say their automated system demonstrated “excellent diagnostic performance” in classifying the drugs involved in overdoses.

“We found that for most substances evaluated, the performance of these algorithms was perfect or near perfect. These models could be used to automate classification of unstructured free-text, thus avoiding the manual and time-consuming process of individually reading each entry and classifying them to a specific substance,” researchers reported in JAMA Network Open.

“Excellent performance was shown for multiple substances, including any opioid, heroin, fentanyl, methamphetamine, cocaine, and alcohol using models for general text. Yet for prescription opioids and benzodiazepines, there was a considerable performance gap.”

That “performance gap” is due in part to weaknesses in the drug classification system, which lumps many synthetic opioids under the same ICD-10 code, including fentanyl, fentanyl analogs, tramadol and buprenorphine – a semi-synthetic opioid used in the addiction treatment drug Suboxone.

In the past, CDC has classified all drug deaths using that code as “prescription opioid overdoses” even though the drugs may have been illicit --- which is the case for the vast majority of deaths involving fentanyl. This resulted in government estimates of prescription opioid overdoses being significantly inflated for many years.

Using the computer algorithms developed at UCLA, prescription opioids ranked far behind fentanyl, alcohol and other substances identified as the cause of death in 8,738 overdoses.

Drugs Involved in 2020 Overdose Deaths in 9 U.S. Counties

Source: JAMA Network Open

Until recently, there was a 6-month time lag in drug deaths being counted in the CDC’s monthly Provisional Drug Overdose Death Counts report. The timeliness of the reports were improved earlier this year to a 4-month delay, but Goodman-Meza says they could be improved even more.  

"If these algorithms are embedded within medical examiner's offices, the time could be reduced to as early as toxicology testing is completed, which could be about three weeks after the death," he said.

Magnetic Gel Could Someday Treat Chronic Pain

By Pat Anson, Editor

Magnet therapy has been used for thousands of years to treat arthritis, inflammation and other chronic illnesses. Today therapeutic magnets can be found in bracelets, shoes, clothing, mattresses and dozens of other products, sold by companies that claim magnets relieve pain, improve blood flow and even flush out toxins.

It's a controversial theory and there is little science to support the medical use of magnets. One critic has even called magnet therapy “a billion-dollar boondoggle.”  

But maybe there’s something to it after all.

UCLA researchers have demonstrated that a gel-like material containing tiny magnetic particles can be used to relieve chronic pain caused by disease or injury. In a study published in the journal Advanced Materials, they say the biomechanical force of magnets can be used on damaged cells to help them heal.

"Much of mainstream modern medicine centers on using pharmaceuticals to make chemical or molecular changes inside the body to treat disease," says principal investigatorDino Di Carlo, PhD, a UCLA professor of bioengineering. "However, recent breakthroughs in the control of forces at small scales have opened up a new treatment idea -- using physical force to kick-start helpful changes inside cells. There's a long way to go, but this early work shows this path toward so-called 'mechanoceuticals' is a promising one."

Di Carlo and his colleagues used magnetic particles inside a gel to manage cell proteins that control the flow of calcium ions. The proteins are on the cell's membrane and play a role in the sensations of touch and pain. When damaged by injury or disease, these “excitable” neuron cells continually send pain signals.

"Our results show that through exploiting 'neural network homeostasis,' which is the idea of returning a biological system to a stable state, it is possible to lessen the signals of pain through the nervous system," said lead author Andy Kah Ping Tay, a recent UCLA doctoral graduate. "Ultimately, this could lead to new ways to provide therapeutic pain relief."

UCLA IMAGE

To make the magnetized gel, UCLA researchers used hyaluronic acid, a gel-like material found naturally in the spinal cord and brain. Hyaluronic hydrogel can also be produced artificially and is used in cosmetics and other beauty products as a filler and moisture barrier.

The researchers put tiny magnetic particles into the gel and then grew a type of primary neural cell -- dorsal root ganglion neurons – embedded inside the gel. In laboratory tests, they applied a magnetic field to generate a pulling force on the particles, which was transmitted through the gel to the embedded neurons.

The researchers found that the magnetically induced pulling led to an increase in calcium ions in the neurons. When they increased the magnetic force steadily over time, the neurons adapted to the continuous stimulation by reducing the signals for pain. In effect, researchers created a form of neuromodulation using magnets -- an old theory put to a new use.

In addition to treating pain, researchers say the magnetic gel could be modified with different biomaterials to treat heart disease, muscle disorders and other health conditions.

The UCLA research was funded by a New Innovator Award grant from the National Institutes of Health.