Tiny utility grew into a power giant

By Knight Ridder Tribune


NFPA 70e Training

Our customized live online or in‑person group training can be delivered to your staff at your location.

  • Live Online
  • 6 hours Instructor-led
  • Group Training Available
Regular Price:
$199
Coupon Price:
$149
Reserve Your Seat Today
In 1911, Oklahoma Gas & Electric Co. marked a milestone by unveiling Oklahoma City's most elaborate electric sign, an ostentatious marquee containing 2,160 flashing light bulbs.

Even the company annals referred to the sign as a "huge flashing affair" that alternated between the words "Oklahoma Gas and Electric" and "Light and Power." OG&E was celebrating a "coming of age," according to a company history that commemorated the utility's 100-year anniversary. The utility served little more than 10,000 customers. It had just purchased two cars and a truck to complement its horses. It had acquired the El Reno Gas Co. about 28 miles west of Oklahoma City.

By today's standards, OG&E - now the state's largest electricity provider - was in its infancy. The company had struggled to provide consistent power to its customers. And its founder, Edward H. Cooke, and George W. Wheeler, its first president, had suffered through at least one false start. Wheeler took charge of an enterprise that seemed doomed from the outset, the first Oklahoma Gas and Electric Co. The company folded in 1899 after a fire claimed its power plant. And ensuing legal battles between the company and investors drained the company coffers, leading the utility to turn off streetlights on nights when the moon was bright.

Cooke, who was an investor in the first company, later joined with Wheeler. The pair, backed by investor Harry M. Blackmer, formed OG&E, which continues to this day.

The company was incorporated in 1902, five years before Oklahoma achieved statehood. Its portfolio of assets were measured in the number of lights rather than megawatts. The fledgling power company boasted 89 carbon arc streetlamps, 120 carbon arc lamps in businesses and 7,000 incandescent lamps throughout Oklahoma City. All tied in to the company's single power plant.

OG&E continued to grow, becoming the state's largest electricity provider by 1928. It was the first utility to use combined cycle power plants, which use steam and steam turbines to generate power. Today, OG&E generates power for more than 750,000 customers in Oklahoma and Arkansas through a combination of coal, natural gas and wind power.

The company has a generation capacity of 6,100 megawatts - enough electricity to power more than 5.5 million homes. Its parent company, OGE Energy Corp. is a publicly traded, diversified power company that operates a pipeline business and employs more than 3,000 people. In a letter to OG&E, former Gov. Frank Keating pegged the utility as "one of Oklahoma's original companies" and congratulated the OG&E family on reaching its centennial before Oklahoma did.

"From an era when electricity was a rare luxury to a new millennium where electric-powered communication spans the globe at the speed of light," he wrote, "OG&E has provided the energy to build, to grow, to prosper and to thrive."

Related News

NTPC bags order to supply 300 MW electricity to Bangladesh

NTPC Bangladesh Power Supply Tender sees NVVN win 300 MW, long-term cross-border electricity trade to BPDB, enabled by 500 MW HVDC interconnection; rivals included Adani, PTC, and Sembcorp in the competitive bidding process.

 

Key Points

It is NTPC's NVVN win to supply 300 MW to Bangladesh's BPDB for 15 years via a 500 MW HVDC link.

✅ NVVN selected as L1 for short and long-term supply

✅ 300 MW to BPDB; delivery via India-Bangladesh HVDC link

✅ Competing bidders: Adani, PTC, Sembcorp

 

NTPC, India’s biggest electricity producer in a nation that is now the third-largest electricity producer globally, on Tuesday said it has won a tender to supply 300 megawatts (MW) of electricity to Bangladesh for 15 years.

Bangladesh Power Development Board (BPDP), in a market where Bangladesh's nuclear power is expanding with IAEA assistance, had invited tenders for supply of 500 MW power from India for short term (1 June, 2018 to 31 December, 2019) and long term (1 January, 2020 to 31 May, 2033). NTPC Vidyut Vyapar Nigam (NVVN), Adani Group, PTC and Singapore-bases Sembcorp submitted bids by the scheduled date of 11 January.

Financial bid was opened on 11 February, the company said in a statement, amid rising electricity prices domestically. “NVVN, wholly-owned subsidiary of NTPC Limited, emerged as successful bidder (L1), both in short term and long term for 300 MW power,” it said.

Without giving details of the rate at which power will be supplied, NTPC said supply of electricity is likely to commence from June 2018 after commissioning of 500 MW HVDC inter-connection project between India and Bangladesh, and as the government advances nuclear power initiatives to bolster capacity in the sector. India currently exports approximately 600 MW electricity to Bangladesh even as authorities weigh coal rationing measures to meet surging demand domestically.

 

Related News

View more

Canadians Support Tariffs on Energy and Minerals in U.S. Trade Dispute

Canada Tariffs on U.S. Energy and Minerals signal retaliatory tariffs amid trade tensions, targeting energy exports and critical minerals, reflecting sovereignty concerns and shifting consumer behavior, reduced U.S. purchases, and demand for Canadian-made goods.

 

Key Points

They are proposed retaliatory tariffs on energy exports and critical minerals to counter U.S. trade pressures.

✅ 75% support tariffs; 70% back dollar-for-dollar retaliation

✅ Consumer shift: fewer U.S. purchases, more Canadian-made goods

✅ Concerns over sovereignty and U.S. trade tactics intensify

 

A recent survey has revealed that a significant majority of Canadians—approximately 75%—support the implementation of tariffs on energy exports and critical minerals in response to electricity exports at risk amid trade tensions with the United States. This finding underscores the nation's readiness to adopt assertive measures to protect its economic interests amid escalating trade disputes.​

Background on Trade Tensions

The trade relationship between Canada and the United States has experienced fluctuations in recent years, with both nations navigating complex issues related to tariffs and energy tariffs and trade tensions as well as trade agreements and economic policies. The introduction of tariffs has been a contentious strategy, often leading to reciprocal measures and impacting various sectors of the economy.​

Public Sentiment Towards Retaliatory Tariffs

The survey, conducted by Leger between February 14 and 17, 2025, sampled 1,500 Canadians and found that 70% favored implementing dollar-for-dollar retaliatory tariffs against the U.S. Notably, 45% of respondents were strongly in favor, while 25% were somewhat in favor. This strong support reflects widespread dissatisfaction with U.S. trade policies and growing support for Canadian energy projects among voters, alongside a collective sentiment favoring decisive action. ​

Concerns Over U.S. Economic Strategies

The survey also highlighted that 81% of Canadians are apprehensive about potential U.S. economic tactics aimed at drawing Canada into a closer political union. These concerns are fueled by statements from U.S. President Donald Trump, who has suggested annexation and employed tariffs that could spike NY energy prices to influence Canadian sovereignty. Such sentiments have heightened fears about the erosion of Canada's political autonomy under economic duress. ​

Impact on Consumer Behavior

In response to these trade tensions, including reports that Ford threatened to cut U.S. electricity exports, many Canadians have adjusted their purchasing habits. The survey indicated that 63% of respondents are buying fewer American products in stores, and 62% are reducing online purchases from U.S. retailers. Specific declines include a 52% reduction in Amazon purchases, a 50% drop in fast-food consumption from American chains, and a 43% decrease in spending at U.S.-based retail stores. Additionally, 30% of Canadians have canceled planned trips to the United States, while 68% have increased their purchases of Canadian-made products. These shifts demonstrate a tangible impact on consumer behavior driven by nationalistic sentiments and support for retaliatory measures. ​

Economic and Political Implications

The widespread support for retaliatory tariffs and the corresponding changes in consumer behavior have significant economic and political implications. Economically, while tariffs can serve as a tool for asserting national interests, they also risk triggering trade wars that can harm various sectors, including agriculture, manufacturing, and technology, with experts cautioning against cutting Quebec's energy exports in response. Politically, the situation presents a challenge for Canadian leadership to balance assertiveness in defending national interests with the necessity of maintaining a stable and mutually beneficial relationship with the U.S., Canada's largest trading partner.​

As Canada approaches its federal elections, trade policy is emerging as a pivotal issue. Voters are keenly interested in how political parties propose to navigate the complexities of international trade, particularly with the United States and how a potential U.S. administration's stance, such as Biden's approach to the energy sector could shape outcomes. The electorate's strong stance on retaliatory tariffs may influence party platforms and campaign strategies, emphasizing the need for clear and effective policies that address both the immediate concerns of trade disputes and the long-term goal of sustaining positive international relations.​

The survey results reflect a nation deeply engaged with its trade dynamics and protective of its sovereignty. While support for retaliatory tariffs is robust, it is essential for policymakers to carefully consider the broader consequences of such actions. Striking a balance between defending national interests and fostering constructive international relationships will be crucial as Canada navigates these complex trade challenges in the coming years.

 

Related News

View more

Energy-hungry Europe to brighten profit at US solar equipment makers

European Solar Inverter Demand surges as photovoltaics and residential solar expand during the clean energy transition, driven by high natural gas prices; Germany leads, boosting Enphase and SolarEdge sales for rooftop systems and grid-tied installations.

 

Key Points

Rising European need for solar inverters, fueled by residential PV growth, high energy costs, and clean energy policies.

✅ Germany leads EU rooftop PV installations

✅ Enphase and SolarEdge see revenue growth

✅ High gas prices and policies spur adoption

 

Solar equipment makers are expected to post higher quarterly profit, benefiting from strong demand in Europe for critical components that convert energy from the sun into electricity, amid record renewable momentum worldwide.

The continent is emerging as a major market for solar firms as it looks to reduce its dependence on the Russian energy supply and accelerate its clean energy transition, with solar already reshaping power prices in Northern Europe across the region, brightening up businesses of companies such as Enphase Energy (ENPH.O) and SolarEdge Technologies (SEDG.O), which make solar inverters.

Wall Street expects Enphase and SolarEdge to post a combined adjusted net income of $323.8 million for the April-June quarter, a 56.7% jump from a year earlier, even as demand growth slows in the United States.

The energy crisis in Europe is not as acute as last year when Western sanctions on Russia severely crimped supplies, but prices of natural gas and electricity continue to be much higher than in the United States, Raymond James analyst Pavel Molchanov said.

As a result, demand for residential solar keeps growing at a strong pace in the region, with Germany being one of the top markets and solar adoption in Poland also accelerating in recent years across the region.

About 159,000 residential solar systems became operational in the first quarter in Germany amid a solar power boost that reflects policy and demand, a 146% rise from a year earlier, according to BSW solar power association.

Adoption of solar is also helping European homeowners have greater control over their energy costs as fossil fuel prices tend to be more volatile, Morningstar analyst Brett Castelli said.

SolarEdge, which has a bigger exposure to Europe than Enphase, said its first-quarter revenue from the continent more than doubled compared with last year.

In comparison, growth in the United States has been tepid due to lukewarm demand in states like Texas and Arizona where cheaper electricity prices make the economics of residential solar less attractive, even though solar is now cheaper than gas in parts of the U.S. market.

Higher interest rates following the U.S. Federal Reserve's recent actions to tame inflation are also weighing on demand, even as power outage risks rise across the United States.

Analysts also expect weakness in California where a new metering reform reduces the money credited to rooftop solar owners for sending excess power into the grid, underscoring how policy shifts can reshape the sector. The sunshine state accounts for nearly a third of the U.S. residential solar market.

Enphase will report its results on Thursday after the bell, while SolarEdge will release its second-quarter numbers on Aug. 1.

 

Related News

View more

Jolting the brain's circuits with electricity is moving from radical to almost mainstream therapy

Brain Stimulation is transforming neuromodulation, from TMS and DBS to closed loop devices, targeting neural circuits for addiction, depression, Parkinsons, epilepsy, and chronic pain, powered by advanced imaging, AI analytics, and the NIH BRAIN Initiative.

 

Key Points

Brain stimulation uses pulses to modulate neural circuits, easing symptoms in depression, Parkinsons, and epilepsy.

✅ Noninvasive TMS and invasive DBS modulate specific brain circuits

✅ Closed loop systems adapt stimulation via real time biomarker detection

✅ Emerging uses: addiction, depression, Parkinsons, epilepsy, chronic pain

 

In June 2015, biology professor Colleen Hanlon went to a conference on drug dependence. As she met other researchers and wandered around a glitzy Phoenix resort’s conference rooms to learn about the latest work on therapies for drug and alcohol use disorders, she realized that out of the 730 posters, there were only two on brain stimulation as a potential treatment for addiction — both from her own lab at Wake Forest School of Medicine.

Just four years later, she would lead 76 researchers on four continents in writing a consensus article about brain stimulation as an innovative tool for addiction. And in 2020, the Food and Drug Administration approved a transcranial magnetic stimulation device to help patients quit smoking, a milestone for substance use disorders.

Brain stimulation is booming. Hanlon can attend entire conferences devoted to the study of what electrical currents do—including how targeted stimulation can improve short-term memory in older adults—to the intricate networks of highways and backroads that make up the brain’s circuitry. This expanding field of research is slowly revealing truths of the brain: how it works, how it malfunctions, and how electrical impulses, precisely targeted and controlled, might be used to treat psychiatric and neurological disorders.

In the last half-dozen years, researchers have launched investigations into how different forms of neuromodulation affect addiction, depression, loss-of-control eating, tremor, chronic pain, obsessive compulsive disorder, Parkinson’s disease, epilepsy, and more. Early studies have shown subtle electrical jolts to certain brain regions could disrupt circuit abnormalities — the miscommunications — that are thought to underlie many brain diseases, and help ease symptoms that persist despite conventional treatments.

The National Institute of Health’s massive BRAIN Initiative put circuits front and center, distributing $2.4 billion to researchers since 2013 to devise and use new tools to observe interactions between brain cells and circuits. That, in turn, has kindled interest from the private sector. Among the advances that have enhanced our understanding of how distant parts of the brain talk with one another are new imaging technology and the use of machine learning, much as utilities use AI to adapt to shifting electricity demand, to interpret complex brain signals and analyze what happens when circuits go haywire.

Still, the field is in its infancy, and even therapies that have been approved for use in patients with, for example, Parkinson’s disease or epilepsy, help only a minority of patients, and in a world where electricity drives pandemic readiness expectations can outpace evidence. “If it was the Bible, it would be the first chapter of Genesis,” said Michael Okun, executive director of the Norman Fixel Institute for Neurological Diseases at University of Florida Health.

As brain stimulation evolves, researchers face daunting hurdles, and not just scientific ones. How will brain stimulation become accessible to all the patients who need it, given how expensive and invasive some treatments are? Proving to the FDA that brain stimulation works, and does so safely, is complicated and expensive. Even with a swell of scientific momentum and an influx of funding, the agency has so far cleared brain stimulation for only a handful of limited conditions. Persuading insurers to cover the treatments is another challenge altogether. And outside the lab, researchers are debating nascent issues, such as the ethics of mind control, the privacy of a person’s brain data—concerns that echo efforts to develop algorithms to prevent blackouts during rising ransomware threats—and how to best involve patients in the study of the human brain’s far-flung regions.

Neurologist Martha Morrell is optimistic about the future of brain stimulation. She remembers the shocked reactions of her colleagues in 2004 when she left full-time teaching at Stanford (she still has a faculty appointment as a clinical professor of neurology) to direct clinical trials at NeuroPace, then a young company making neurostimulator systems to potentially treat epilepsy patients.

Related: Once a last resort, this pain therapy is getting a new life amid the opioid crisis
“When I started working on this, everybody thought I was insane,” said Morrell. Nearly 20 years in, she sees a parallel between the story of jolting the brain’s circuitry and that of early implantable cardiac devices, such as pacemakers and defibrillators, which initially “were used as a last option, where all other medications have failed.” Now, “the field of cardiology is very comfortable incorporating electrical therapy, device therapy, into routine care. And I think that’s really where we’re going with neurology as well.”


Reaching a ‘slope of enlightenment’
Parkinson’s is, in some ways, an elder in the world of modern brain stimulation, and it shows the potential as well as the limitations of the technology. Surgeons have been implanting electrodes deep in the brains of Parkinson’s patients since the late 1990s, and in people with more advanced disease since the early 2000s.

In that time, it’s gone through the “hype cycle,” said Okun, the national medical adviser to the Parkinson’s Foundation since 2006. Feverish excitement and overinflated expectations have given way to reality, bringing scientists to a “slope of enlightenment,” he said. They have found deep brain stimulation to be very helpful for some patients with Parkinson’s, rendering them almost symptom-free by calming the shaking and tremors that medications couldn’t. But it doesn’t stop the progression of the disease, or resolve some of the problems patients with advanced Parkinson’s have walking, talking, and thinking.

In 2015, the same year Hanlon found only her lab’s research on brain stimulation at the addiction conference, Kevin O’Neill watched one finger on his left hand start doing something “funky.” One finger twitched, then two, then his left arm started tingling and a feeling appeared in his right leg, like it was about to shake but wouldn’t — a tremor.

“I was assuming it was anxiety,” O’Neill, 62, told STAT. He had struggled with anxiety before, and he had endured a stressful year: a separation, selling his home, starting a new job at a law firm in California’s Bay Area. But a year after his symptoms first began, O’Neill was diagnosed with Parkinson’s.

In the broader energy context, California has increasingly turned to battery storage to stabilize its strained grid.

Related: Psychiatric shock therapy, long controversial, may face fresh restrictions
Doctors prescribed him pills that promote the release of dopamine, to offset the death of brain cells that produce this messenger molecule in circuits that control movement. But he took them infrequently because he worried about insomnia as a side effect. Walking became difficult — “I had to kind of think my left leg into moving” — and the labor lawyer found it hard to give presentations and travel to clients’ offices.

A former actor with an outgoing personality, he developed social anxiety and didn’t tell his bosses about his diagnosis for three years, and wouldn’t have, if not for two workdays in summer 2018 when his tremors were severe and obvious.

O’Neill’s tremors are all but gone since he began deep brain stimulation last May, though his left arm shakes when he feels tense.

It was during that period that he learned about deep brain stimulation, at a support group for Parkinson’s patients. “I thought, ‘I will never let anybody fuss with my brain. I’m not going to be a candidate for that,’” he recalled. “It felt like mad scientist science fiction. Like, are you kidding me?”

But over time, the idea became less radical, as O’Neill spoke to DBS patients and doctors and did his own research, and as his symptoms worsened. He decided to go for it. Last May, doctors at the University of California, San Francisco surgically placed three metal leads into his brain, connected by thin cords to two implants in his chest, just near the clavicles. A month later, he went into the lab and researchers turned the device on.

“That was a revelation that day,” he said. “You immediately — literally, immediately — feel the efficacy of these things. … You go from fully symptomatic to non-symptomatic in seconds.”

When his nephew pulled up to the curb to pick him up, O’Neill started dancing, and his nephew teared up. The following day, O’Neill couldn’t wait to get out of bed and go out, even if it was just to pick up his car from the repair shop.

In the year since, O’Neill’s walking has gone from “awkward and painful” to much improved, and his tremors are all but gone. When he is extra frazzled, like while renovating and moving into his new house overlooking the hills of Marin County, he feels tense and his left arm shakes and he worries the DBS is “failing,” but generally he returns to a comfortable, tremor-free baseline.

O’Neill worried about the effects of DBS wearing off but, for now, he can think “in terms of decades, instead of years or months,” he recalled his neurologist telling him. “The fact that I can put away that worry was the big thing.”

He’s just one patient, though. The brain has regions that are mostly uniform across all people. The functions of those regions also tend to be the same. But researchers suspect that how brain regions interact with one another — who mingles with whom, and what conversation they have — and how those mixes and matches cause complex diseases varies from person to person. So brain stimulation looks different for each patient.

Related: New study revives a Mozart sonata as a potential epilepsy therapy
Each case of Parkinson’s manifests slightly differently, and that’s a bit of knowledge that applies to many other diseases, said Okun, who organized the nine-year-old Deep Brain Stimulation Think Tank, where leading researchers convene, review papers, and publish reports on the field’s progress each year.

“I think we’re all collectively coming to the realization that these diseases are not one-size-fits-all,” he said. “We have to really begin to rethink the entire infrastructure, the schema, the framework we start with.”

Brain stimulation is also used frequently to treat people with common forms of epilepsy, and has reduced the number of seizures or improved other symptoms in many patients. Researchers have also been able to collect high-quality data about what happens in the brain during a seizure — including identifying differences between epilepsy types. Still, only about 15% of patients are symptom-free after treatment, according to Robert Gross, a neurosurgery professor at Emory University in Atlanta.

“And that’s a critical difference for people with epilepsy. Because people who are symptom-free can drive,” which means they can get to a job in a place like Georgia, where there is little public transit, he said. So taking neuromodulation “from good to great,” is imperative, Gross said.


Renaissance for an ancient idea
Recent advances are bringing about what Gross sees as “almost a renaissance period” for brain stimulation, though the ideas that undergird the technology are millenia old. Neuromodulation goes back to at least ancient Egypt and Greece, when electrical shocks from a ray, called the “torpedo fish,” were recommended as a treatment for headache and gout. Over centuries, the fish zaps led to doctors burning holes into the brains of patients. Those “lesions” worked, somehow, but nobody could explain why they alleviated some patients’ symptoms, Okun said.

Perhaps the clearest predecessor to today’s technology is electroconvulsive therapy (ECT), which in a rudimentary and dangerous way began being used on patients with depression roughly 100 years ago, said Nolan Williams, director of the Brain Stimulation Lab at Stanford University.

Related: A new index measures the extent and depth of addiction stigma
More modern forms of brain stimulation came about in the United States in the mid-20th century. A common, noninvasive approach is transcranial magnetic stimulation, which involves placing an electromagnetic coil on the scalp to transmit a current into the outermost layer of the brain. Vagus nerve stimulation (VNS), used to treat epilepsy, zaps a nerve that contributes to some seizures.

The most invasive option, deep brain stimulation, involves implanting in the skull a device attached to electrodes embedded in deep brain regions, such as the amygdala, that can’t be reached with other stimulation devices. In 1997, the FDA gave its first green light to deep brain stimulation as a treatment for tremor, and then for Parkinson’s in 2002 and the movement disorder dystonia in 2003.

Even as these treatments were cleared for patients, though, what was happening in the brain remained elusive. But advanced imaging tools now let researchers peer into the brain and map out networks — a recent breakthrough that researchers say has propelled the field of brain stimulation forward as much as increased funding has, paralleling broader efforts to digitize analog electrical systems across industry. Imaging of both human brains and animal models has helped researchers identify the neuroanatomy of diseases, target brain regions with more specificity, and watch what was happening after electrical stimulation.

Another key step has been the shift from open-loop stimulation — a constant stream of electricity — to closed-loop stimulation that delivers targeted, brief jolts in response to a symptom trigger. To make use of the futuristic technology, labs need people to develop artificial intelligence tools, informed by advances in machine learning for the energy transition, to interpret large data sets a brain implant is generating, and to tailor devices based on that information.

“We’ve needed to learn how to be data scientists,” Morrell said.

Affinity groups, like the NIH-funded Open Mind Consortium, have formed to fill that gap. Philip Starr, a neurosurgeon and developer of implantable brain devices at the University of California at San Francisco Health system, leads the effort to teach physicians how to program closed-loop devices, and works to create ethical standards for their use. “There’s been extraordinary innovation after 20 years of no innovation,” he said.

The BRAIN Initiative has been critical, several researchers told STAT. “It’s been a godsend to us,” Gross said. The NIH’s Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative was launched in 2013 during the Obama administration with a $50 million budget. BRAIN now spends over $500 million per year. Since its creation, BRAIN has given over 1,100 awards, according to NIH data. Part of the initiative’s purpose is to pair up researchers with medical technology companies that provide human-grade stimulation devices to the investigators. Nearly three dozen projects have been funded through the investigator-devicemaker partnership program and through one focused on new implantable devices for first-in-human use, according to Nick Langhals, who leads work on neurological disorders at the initiative.

The more BRAIN invests, the more research is spawned. “We learn more about what circuits are involved … which then feeds back into new and more innovative projects,” he said.

Many BRAIN projects are still in early stages, finishing enrollment or small feasibility studies, Langhals said. Over the next couple of years, scientists will begin to see some of the fruits of their labor, which could lead to larger clinical trials, or to companies developing more refined brain stimulation implants, Langhals said.

Money from the National Institutes of Mental Health, as well as the NIH’s Helping to End Addiction Long-term (HEAL), has similarly sweetened the appeal of brain stimulation, both for researchers and industry. “A critical mass” of companies interested in neuromodulation technology has mushroomed where, for two decades, just a handful of companies stood, Starr said.

More and more, pharmaceutical and digital health companies are looking at brain stimulation devices “as possible products for their future,” said Linda Carpenter, director of the Butler Hospital TMS Clinic and Neuromodulation Research Facility.


‘Psychiatry 3.0’
The experience with using brain stimulation to stop tremors and seizures inspired psychiatrists to begin exploring its use as a potentially powerful therapy for healing, or even getting ahead of, mental illness.

In 2008, the FDA approved TMS for patients with major depression who had tried, and not gotten relief from, drug therapy. “That kind of opened the door for all of us,” said Hanlon, a professor and researcher at the Center for Research on Substance Use and Addiction at Wake Forest School of Medicine. The last decade saw a surge of research into how TMS could be used to reset malfunctioning brain circuits involved in anxiety, depression, obsessive-compulsive disorder, and other conditions.

“We’re certainly entering into what a lot of people are calling psychiatry 3.0,” Stanford’s Williams said. “Whereas the first iteration was Freud and all that business, the second one was the psychopharmacology boom, and this third one is this bit around circuits and stimulation.”

Drugs alleviate some patients’ symptoms while simultaneously failing to help many others, but psychopharmacology clearly showed “there’s definitely a biology to this problem,” Williams said — a biology that in some cases may be more amenable to a brain stimulation.

Related: Largest psilocybin trial finds the psychedelic is effective in treating serious depression
The exact mechanics of what happens between cells when brain circuits … well, short-circuit, is unclear. Researchers are getting closer to finding biomarkers that warn of an incoming depressive episode, or wave of anxiety, or loss of impulse control. Those brain signatures could be different for every patient. If researchers can find molecular biomarkers for psychiatric disorders — and find ways to preempt those symptoms by shocking particular brain regions — that would reshape the field, Williams said.

Not only would disease-specific markers help clinicians diagnose people, but they could help chip away at the stigma that paints mental illness as a personal or moral failing instead of a disease. That’s what happened for epilepsy in the 1960s, when scientific findings nudged the general public toward a deeper understanding of why seizures happen, and it’s “the same trajectory” Williams said he sees for depression.

His research at the Stanford lab also includes work on suicide, and obsessive-compulsive disorder, which the FDA said in 2018 could be treated using noninvasive TMS. Williams considers brain stimulation, with its instantaneity, to be a potential breakthrough for urgent psychiatric situations. Doctors know what to do when a patient is rushed into the emergency room with a heart attack or a stroke, but there is no immediate treatment for psychiatric emergencies, he said. Williams wonders: What if, in the future, a suicidal patient could receive TMS in the emergency room and be quickly pulled out of their depressive mental spiral?

Researchers are also actively investigating the brain biology of addiction. In August 2020, the FDA approved TMS for smoking cessation, the first such OK for a substance use disorder, which is “really exciting,” Hanlon said. Although there is some nuance when comparing substance use disorders, a primal mechanism generally defines addiction: the eternal competition between “top-down” executive control functions and “bottom-up” cravings. It’s the same process that is at work when one is deciding whether to eat another cookie or abstain — just exacerbated.

Hanlon is trying to figure out if the stop and go circuits are in the same place for all people, and whether neuromodulation should be used to strengthen top-down control or weaken bottom-up cravings. Just as brain stimulation can be used to disrupt cellular misfiring, it could also be a tool for reinforcing helpful brain functions, or for giving the addicted brain what it wants in order to curb substance use.

Evidence suggests many people with schizophrenia smoke cigarettes (a leading cause of early death for this population) because nicotine reduces the “hyperconnectivity” that characterizes the brains of people with the disease, said Heather Ward, a research fellow at Boston’s Beth Israel Deaconess Medical Center. She suspects TMS could mimic that effect, and therefore reduce cravings and some symptoms of the disease, and she hopes to prove that in a pilot study that is now enrolling patients.

If the scientific evidence proves out, clinicians say brain stimulation could be used alongside behavioral therapy and drug-based therapy to treat substance use disorders. “In the end, we’re going to need all three to help people stay sober,” Hanlon said. “We’re adding another tool to the physician’s toolbox.”

Decoding the mysteries of pain
Afavorable outcome to the ongoing research, one that would fling the doors to brain stimulation wide open for patients with myriad disorders, is far from guaranteed. Chronic pain researchers know that firsthand.

Chronic pain, among the most mysterious and hard-to-study medical phenomena, was the first use for which the FDA approved deep brain stimulation, said Prasad Shirvalkar, an assistant professor of anesthesiology at UCSF. But when studies didn’t pan out after a year, the FDA retracted its approval.

Shirvalkar is working with Starr and neurosurgeon Edward Chang on a profoundly complex problem: “decoding pain in the brain states, which has never been done,” as Starr told STAT.

Part of the difficulty of studying pain is that there is no objective way to measure it. Much of what we know about pain is from rudimentary surveys that ask patients to rate how much they’re hurting, on a scale from zero to 10.

Using implantable brain stimulation devices, the researchers ask patients for a 0-to-10 rating of their pain while recording up-and-down cycles of activity in the brain. They then use machine learning to compare the two streams of information and see what brain activity correlates with a patient’s subjective pain experience. Implantable devices let researchers collect data over weeks and months, instead of basing findings on small snippets of information, allowing for a much richer analysis.

 

Related News

View more

Ottawa sets out to protect its hydro heritage

Ottawa Hydro Substation Heritage Designation highlights Hydro Ottawa's 1920s architecture, Art Deco facades, and municipal utility history, protecting key voltage-reduction sites in Glebe, Carling-Merivale, Holland, King Edward, and Old Ottawa South.

 

Key Points

A city plan to protect Hydro Ottawa's 1920s substations for architecture, utility role, and civic electrical heritage.

✅ Protects five operating voltage-reduction sites citywide

✅ Recognizes Art Deco and early 20th century utility architecture

✅ Allows emergency demolition to ensure grid safety

 

The city of Ottawa is looking to designate five hydro substations built nearly a century ago as heritage structures, a move intended to protect the architectural history of Ottawa's earliest forays into the electricity business, even as Ottawa electricity consumption has shifted in recent years.

All five buildings are still used by Hydro Ottawa to reduce the voltage coming from transmission lines before the electricity is transmitted to homes and businesses, and when severe weather causes outages, Sudbury Hydro crews work to reconnect service across communities.

Electricity came to Ottawa in 1882 when two carbon lamps were installed on LeBreton Flats, heritage planner Anne Fitzpatrick told the city's built heritage subcommittee on Tuesday. It became a lucrative business, and soon a privately owned monopoly that drew public scrutiny similar to debates over retroactive charges in neighboring jurisdictions.

In 1905, city council held a special meeting to buy the electrical company, which led to a dramatic drop in electricity rates for residents, a contrast with recent discussions about peak hydro rates for self-isolating customers.

The substations are now owned by Hydro Ottawa, which agreed to the heritage designations on the condition it not be prevented from emergency demolitions if it needs to address incidents such as damaging storms in Ontario while it works to "preserve public safety and the continuity of critical hydro electrical services."

Built in 1922, the substation at the intersection of Glebe and Bronson avenues was the first to be built by the new municipal electrical department, long before modern battery storage projects became commonplace on Ontario's grid.

The largest of the substations being protected dates back to 1929 and is found at the corner of Carling Avenue and Merivale Road. It was built to accommodate a growing population in areas west of downtown including Hintonburg and Mechanicsville.

The substation on Holland Avenue near the Queensway is different from the others because it was built in 1924 to serve the Ottawa Electric Railway Company. The streetcar company operated from 1891 to 1959, and urban electrical infrastructure can face failures such as the Hydro-Québec manhole fire that left thousands without power.

This substation on King Edward Avenue was built in 1931 and designed by architect William Beattie, who also designed York Street Public School in Lowertown and the substation on Carling Avenue. 

The last substation to be built in a 'bold and decorative style' is at 39 Riverdale Ave. in Old Ottawa South, according to city staff. It was designed in an Art Deco style by prominent architect J. Albert Ewart, who was also behind the Civic Hospital and nearby Southminster Church on Bank Street.

 

Related News

View more

Manitoba Hydro's burgeoning debt surpasses $19 billion

Manitoba Hydro Debt Load surges past $19.2B as the Crown corporation faces shrinking net income, restructuring costs, and PUB rate decisions, driven by Bipole III, Keeyask construction, aging infrastructure, and rising interest rate risks.

 

Key Points

Manitoba Hydro Debt Load refers to the utility's escalating borrowings exceeding $19B, pressuring rates and finances.

✅ Debt rose to $19.2B; projected near $25B within five years.

✅ Major drivers: Bipole III, Keeyask, aging assets, restructuring.

✅ Rate hikes sought; PUB approved 3.6% vs 7.9% request.

 

Manitoba Hydro's debt load now exceeds $19 billion as the provincial Crown corporation grapples with a shrinking net income amid ongoing efforts to slay costs.

The utility's annual report, to be released publicly on Tuesday, also shows its total consolidated net income slumped from $71 million in 2016-2017 to $37 million in the last fiscal year, mirroring a Hydro One profit drop as electricity revenue fell.

It said efforts to restructure the utility and reduce costs are partly to blame for the $34 million drop in year-over-year income.

These earnings come nowhere close, however, to alleviating Hydro's long-term debt problem, a dynamic also seen in a BC Hydro deferred costs report about customer exposure. The figure is pegged at $19.2 billion this fiscal year, up from $16.1 billion the previous year and $14.2 billion in 2016.

The utility projects its debt will grow to about $25 billion in the next five years. Its largest expenses include finishing the Bipole III line, working on the Keeyask Generating System that is halfway done and rebuilding aging wood poles and substations, the report said.

"This level of debt increases the potential financial exposure from risks facing the corporation and is a concern for both

the corporation and our customers who may be exposed to higher rate increases in the event of rising interest rates, a prolonged drought or a major system failure," outgoing president and CEO Kelvin Shepherd wrote.

The income drop is primarily a result of the $50 million spent in the form of restructuring charges associated with the utility's efforts to streamline the organization and drive down costs, amid NDP criticism of Hydro changes related to government policy.

Those efforts included the implementation of buyouts for employees through what the utility dubbed its "voluntary departure program."

Among the changes, Manitoba Hydro reduced its workforce by 800 employees, which is expected to save the utility over $90 million per year. It also reduced its management positions by 26 per cent, a Monday news release said, while Hydro One leadership upheaval in Ontario drove its shares down during comparable governance turmoil.

To improve its financial situation, Hydro has applied for rate increases, even as the Consumers Coalition pushes to have the proposal rejected. The Public Utilities Board offered a 3.6 per cent average rate hike, instead of the 7.9 per cent jump the utility asked for.

In May, when the PUB rendered its decision, it made several recommendations as an alternative to raising rates, including receiving a share of carbon tax revenue and asking the government to help pay for Bipole III.

Hydro is projecting a net income of $70 million for 2018-2019, which includes the impact of the recent rate increase. That total reflects an approximately 20 per cent reduction in net income from 2017-18 after restructuring costs are calculated.

 

Related News

View more

Sign Up for Electricity Forum’s Newsletter

Stay informed with our FREE Newsletter — get the latest news, breakthrough technologies, and expert insights, delivered straight to your inbox.

Electricity Today T&D Magazine Subscribe for FREE

Stay informed with the latest T&D policies and technologies.
  • Timely insights from industry experts
  • Practical solutions T&D engineers
  • Free access to every issue

Live Online & In-person Group Training

Advantages To Instructor-Led Training – Instructor-Led Course, Customized Training, Multiple Locations, Economical, CEU Credits, Course Discounts.

Request For Quotation

Whether you would prefer Live Online or In-Person instruction, our electrical training courses can be tailored to meet your company's specific requirements and delivered to your employees in one location or at various locations.