Anxiety

Your heart starts racing before anything has actually happened. No fire. No threat. But your body is already responding like something is.

This episode is about anxiety disorders, the most common category of mental illness in the world. We're working from Penninx, Pine, Holmes, and Reif, Anxiety Disorders, published in The Lancet in 2021.

This episode covers what actually separates fear from anxiety, why anxiety disorders are so often comorbid with each other and with depression, the genetics behind why some people's alarm systems are simply built more sensitive than others, the specific brain circuitry behind why the alarm sometimes can't tell smoke from fire, what brain imaging can and can't actually tell us, and a full breakdown of the treatment evidence, including why less than 25 percent of people with an anxiety disorder ever receive adequate care, despite treatments that genuinely work.

For anyone whose alarm system won't power down. And for anyone who loves someone whose alarm system won't power down.

Transcript

Copy Transcript Copied

SPEAKER_00 0:00

Do you know that moment when your heart starts racing before anything has actually happened? There's no threat in the room, nobody's chasing you, nothing is actually wrong, but your body is already responding like something is. That's the difference between fear and anxiety. And it matters more than people think. Fear is the response to something happening right now. Anxiety is your system bracing for something that hasn't happened yet and might not happen at all. Today we're talking about anxiety disorders, not the occasional nervousness everyone gets, the kind that doesn't turn off, even when there is nothing actually in the room. And to understand what's happening, I want you to picture a smoke detector. But before that. Anxiety is the anticipation of a future threat, something that might happen, not something that is happening. Both are completely normal, both are adaptive, they exist because they help us survive. Plenty of fear and anxiety in childhood and adulthood is just normal life. Stranger anxiety in toddler, performance anxiety before something important, feeling unsettled during a big life transition, none of that is a disorder. It becomes clinical when it's disproportionate to the actual threat or when it starts disrupting how someone actually lives their life. Anxiety disorders are the most common group of mental health disorders that exist. The World Health Organization ranks them as the ninth leading cause of health-related disability worldwide. In Europe alone, they account for over 74 billion euros a year in costs. There isn't just one anxiety disorder. There are several, and which ones someone has depends on what exactly the fear attaches itself to. Separation anxiety, where the fear is about losing an attachment figure, specific phobias, where fear of a particular object or situation, social anxiety disorder, fear of being judged or scrutinized by other people, agoraphobia, fear of being trapped or unable to escape, panic disorder, fear of panic attacks themselves, and generalized anxiety disorder, persistent worry that isn't tied to one specific thing. And here's something worth knowing: anxiety disorders are also highly core morbid with each other. Somewhere between 48 and 68% of adults with one anxiety disorder meet criteria for another at the same time. These aren't always cleanly separate conditions. A lot of the time they're overlapping. I think this is the part that people misunderstand the most. Anxiety isn't the problem. Anxiety is supposed to be there. The problem is a system that can't tell the difference between an actual threat and the memory of one or the idea of one. That distinction is going to matter a lot for the rest of this episode. Okay, so between the ages of 15 and 25, the cumulative prevalence of all anxiety disorders combined is somewhere between 20 and 30%. That means somewhere between 1 in 5 and 1 in 3 young people will have had an anxiety disorder by that point in their life. In adulthood, 10 to 14% of people meet criteria for an anxiety disorder within any given year. The most common is specific phobia, followed by social anxiety disorder than panic disorder or agoraphobia. Women are anywhere from 1.3 to 2.4 times more likely to have an anxiety disorder than men. That gap becomes more pronounced after adolescence. This isn't usually something that resolves quickly either. Two-year remission rates are about 70% for panic disorder and generalized anxiety disorder on their own, but they drop to 50 to 55% for social anxiety disorder and down to 43% for people with more than one anxiety disorder at once. And the health impact goes beyond anxiety itself. Risk of death from natural causes is increased by 1.4 times and from unnatural causes by 2.5 times. Anxiety has also been consistently linked to a higher risk of cardiovascular disease later on. Despite how common this is, less than 25% of people with an anxiety disorder receive any treatment at all. And even among people who actively want care, less than 25% get something that's actually adequate. That number means something. This is the most common category of mental illness that exists. And the vast majority of people dealing with it are getting no help or help that isn't enough. Not because the treatments don't work. We're gonna get to the treatments later in this episode, and they do work. It's just an access problem, not an evidence problem. Before we get into the brain itself, I want to talk about where the sensitivity of the system actually comes from, because some of it is built in from the start. Heritability estimates for anxiety disorder sit around 35% for generalized anxiety disorder, and closer to 50% for social anxiety disorder, panic disorder, and agoraphobia. That's a real genetic contribution, though notably lower than what we saw with ADHD, 70 to 80%. Like ADHD, this isn't one gene that's different. It's a complex picture with many genetic variants of small effect interacting with environmental risk factors. Researchers spent a long time looking at specific candidate genes, mainly ones involved in serotonin and stress hormones. A meta-analysis of those candidate genes found an association between panic disorder and a gene called TMEM132D. And in some subsamples, a few other genes. But the larger genome-wide studies that came after that didn't find much more. The first genome-wide study found a few or no significant genetic loci at all. What the genetic research has found clearly though is a strong overlap between anxiety, depression, and a personality trait called neuroticism. The genetic correlation between them is over 0.6, which suggests there's a shared underlying genetic risk that shows up as anxiety in some people, depression in others, and often both in the same person. There's also early research on epigenetics, meaning changes to how genes are expressed rather than the genes themselves. Some studies have found differences in genes like MAOA, CRHR1, and OXTR in people with panic disorder or social anxiety, but the studies are still small, and none of this is usable for diagnosis yet. It's a direction for future research, not a clinical tool as of today. Okay, so this section is what I've been building up to. Picture a smoke detector mounted on the ceiling of the house. Its only job is to sense danger and respond fast enough to keep the people inside safe. And in most houses, it does exactly that. It stays quiet until there's actual smoke and then it goes off immediately. In the brain, the part of the system that works the most like a smoke detector is the amygdala. It reacts almost instantly to anything that resembles danger. That's your fear response. Fast, automatic, built for something close and immediate. But not all threats are a fire right in front of you. Some are more like smelling something faintly off in the air and not being sure yet if it's actually dangerous. There's a separate part of the system for that too. And that system is called the bed nucleus of the Stria terminalis or BNS. Instead of one sharp alarm, this stays quietly armed for longer or more uncertain threats. This is closer to what anxiety actually is. Not a single alarm going off, but a system that stays on alert because it isn't sure the danger has passed. Then there's the hippocampus, which works almost like the part of the system that checks context. Has a specific situation actually been dangerous before or does it just resemble something that was? In anxiety disorder, this context check doesn't filter well. A lot of situations get treated as matches to old threats, even when they aren't actually similar. And finally, the prefrontal cortex and the region called the insula act as something like the override switch, the part of the system that's actually supposed to walk over, assess what's happening, and then reset the alarm once it's confirmed that there's no real danger. This is called extinction learning. Basically, the brain unlearning a fear once it's safe to do so. In anxiety disorders, this override function often isn't working efficiently at all. So, picture the whole system together. A sensor that fires fast and easily, a secondary system that stays alarmed long after it should have stood down, a context checker that keeps mismatching the present from the past, and an override switch that isn't resetting things the way it should. That's four separate problems. That's one smoke detector system that isn't calibrated the way it's supposed to be. So, how confident are we about all of this? I want to be honest with you about where the evidence is strong and where it's still developing. Studies looking at brain structure, meaning the physical size and shape of these regions, have found some differences in anxiety disorders, mainly in temporal, prefrontal, and cingulate regions. But the effect sizes are small, somewhere between 0.1 and 0.3, which means structural brain imaging on its own isn't something we can use clinically yet. Functional imaging, meaning what the brain is actually doing in the moment a threat shows up, tells a clearer story. When researchers deliberately evoke a threat response in the scanner, people with anxiety disorder show altered activity in temporal and prefrontal regions, and the effect sizes are considerably larger, between 0.5 and 1. So the smoke detector system doesn't always look that different sitting still. It looks different when it's actually responding to something. There's also a measure called error-related negativity, which is basically how strongly the brain reacts to making a small mistake. In the people with anxiety disorder, this response is measurably larger. Even minor errors seem to register as more threatening than they do for someone without an anxiety disorder. And there's consistent evidence of attention bias towards threat. People with anxiety disorders tend to notice and fixate on potentially threatening information faster and more persistently than other people do. The alarm system isn't just more sensitive when triggered, it's also scanning more actively for things to be triggered. One honest caveat. For things like blood markers and physiological measures, effect sizes rarely go above 0.5, meaning none of this is reliable enough yet to diagnose anxiety from a blood test or a brain scan. Diagnosis still comes from clinical conversations, not a lab result. So we've got the full picture: a sensor that's often too sensitive, a secondary system that stays alarmed too long, a context checker that overmatches, and an override switch that underperforms. So what actually helps? The honest answer from this paper is that psychotherapy and medication work about equally well as first-line treatment options. Neither one is automatically the better choice. It comes down to side effects, access, and what the person actually prefers. Let's start with psychotherapy. The strongest evidence by far is for cognitive behavioral therapy, or as most people know it, CBT. It's a short-term therapy, usually 8 to 20 sessions. The core technique is exposure, which means gradually facing the feared situation instead of avoiding it, either in real life or through imagined scenarios, until the brain learns the threat doesn't actually materialize. Think of it as a supervised fire drill. You're turning the smoke detector off, you're walking it through the same smoke deliberately and safely enough times that it stops treating that specific smoke as an emergency. Effect sizes for CBT are large when compared to a waiting list, but only small to moderate when compared to standard care or a placebo where CBT has shown similar effectiveness to face-to-face therapy, which matters a lot in terms of access. Okay, now let's move on to medication. SSRIs and SNRIs are the same broad medication classes used for depression. The first line pharmacological option. Effect sizes are in about 0.37 to 0.44 range. There are newer options that tend to be preferred because of a better side effect profile, such as Certrale or Venlafaxcine. These medications work by adjusting the chemical thresholds at which the alarm goes off in the first place, mainly through the serotonin system. They typically take two to four weeks before you notice an effect, and the guidance is to start at a low dose and increase slowly. Benzodiazepines are also effective and fast, but they only work acutely. They carry a risk of dependency, and symptoms tend to come back after you stop the medication. Because of that, they're generally used very cautiously and only for short window while another treatment has time to start working. And I want to say something here because I think the messaging around anxiety medication can get confusing. Adjusting the chemical sensitivity of the alarm is not the same as turning it off completely. The goal isn't to feel nothing. The goal is for the system to stop firing at things that aren't actually dangerous. That distinction matters, especially if you've ever, if you've ever worried that taking medication means losing your edge or your instincts. And the strongest result comes from combining the two. Research shows that adding psychotherapy to medication outperforms medication on its own. The override switch and the chemical thresholds both get addressed at the same time. Which brings us back to the treatment gap from earlier in the episode. Less than 25% of people with an anxiety disorder get any treatment at all. And that's despite the fact that what we just walked through actually works. This isn't a case of not knowing what helps, it's a case of access, not matching the evidence. Okay, so again, that was a lot of information. Let's put it all together. The smoke detector, a sensor built partly through genetics, to be more or less sensitive from the start. A secondary system that stays alarmed for threats that haven't fully resolved, a context checker that keeps mismatching the present from the past, an override switch that isn't resetting things efficiently, and a chemical threshold that can be adjusted alongside a system that can be retained through repeated safe exposure to things it's afraid of. It's no secret that anxiety disorders are the most common mental illness in the world. Highly comorbid, often chronic if left untreated, and genuinely responsive to treatment that already exists and already works. And here's what I want to leave you with. If your alarm system has been going off at things that aren't actually on fire, that's not a personal failure. That's a system that's over-calibrated. And systems that are overcalibrated can be calibrated through therapy, through medication, often through both. You stop asking people to just calm down when you understand that their alarm system was never given the chance to learn it was safe to power down. And remember, things can be redesigned. They just have to be built brick by brick. Today's episode helped. Leave a review if you have a minute, follow the account, and I'll see you in the next one.