Most people who call here have already been told to breathe. They have the app, they know the box-breathing pattern, and they can explain perfectly well that the plane will not crash. None of it has made the slightest difference at three in the morning, and by now the not-working has become its own evidence that something is wrong with them.
Advanced Counseling and Research Services treats anxiety disorders across Pennsylvania, in person at our Lancaster office or by secure telehealth. We are a psychotraumatology practice, and that shapes two things about how we work: we ask where the anxiety came from as well as what it is doing, and we take exposure seriously as a treatment rather than as something to mention and avoid.
This page is longer and more specific than most, and it tells you what the evidence does not support as clearly as what it does. If you have already been through a course of therapy that did not help, that is the part worth reading.
The anxiety disorders in the DSM-5-TR most often seen in adults are generalized anxiety disorder, panic disorder, social anxiety disorder, agoraphobia, specific phobia and separation anxiety disorder. Selective mutism is also in the chapter, and is usually identified in childhood.
Which one you have matters, because the treatment protocols genuinely differ. Panic disorder is largely a fear of the body's own sensations. Social anxiety is a fear of evaluation, and — unusually — the head-to-head evidence there slightly favours cognitive work over exposure. Generalized anxiety is worry about worry, with the content changing week to week. A page that lists the subtypes and then offers one undifferentiated approach has not really told you anything, and most of them do exactly that.
Two conditions people expect to find here are not here, and the reason is worth knowing. Post-traumatic stress disorder and obsessive-compulsive disorder are no longer classified as anxiety disorders. The DSM-5 moved them into separate chapters — trauma- and stressor-related disorders, and obsessive-compulsive and related disorders — and the DSM-5-TR keeps them there. The stated reasoning for moving PTSD is one this practice takes seriously: anxiety is only one of several emotional responses to trauma, and treating the whole response as an anxiety problem misses most of it. If that is what you are dealing with, PTSD and complex trauma and OCD have their own pages here, and their own treatments. Note also that a panic attack is a defined event that can occur across many conditions rather than a diagnosis in itself — “anxiety attack” is not a clinical term at all.
Cognitive behavioural therapy is the best-supported psychotherapy for anxiety, and you will find that stated everywhere. What you will not find is the size of the effect, so here it is. A 2018 meta-analysis of 41 randomised trials comparing CBT against a genuine placebo — not a waiting list — found a moderate effect on target symptoms overall, and the range varied a great deal by condition: the largest effects were in OCD, generalized anxiety and acute stress, while panic disorder and social anxiety sat at the bottom of the range. Those last two are among the most common reasons people call.
It gets more sobering. When the same research group restricted the analysis to placebo-controlled trials published since 2017 — ten studies, 1,250 participants — the pooled effect was small, and at six-month follow-up it was no longer statistically significant. The authors say plainly that recent effect sizes appear smaller than those in earlier meta-analyses. Their recent sample was dominated by PTSD trials, so this is not a clean estimate for generalized anxiety or panic, and we would not want to overstate a discouraging finding any more than an encouraging one.
What all of that means: therapy for anxiety works, it is the best tool available, and anyone quoting you a large number is quoting the older literature with the weaker comparison groups. We would rather you started with an accurate expectation than a marketing one.
If you have had therapy for anxiety that consisted of talking about the anxiety, learning breathing exercises, and being encouraged to challenge your thoughts — and it did not work — this section is the likeliest explanation, and it is not about you.
Exposure is a core component of every evidence-based protocol for anxiety disorders. Surveys of practising clinicians consistently find that most do not deliver it. In one survey of clinicians treating anxiety, only between 12% and 38% used exposure, even though 71% described themselves as cognitive-behavioural. In another, looking at childhood anxiety, only about a quarter of therapists said they often used it and fewer than one in sixteen said they always did.
A 2023 systematic review and meta-analysis looked at what predicts whether a therapist will use exposure. The strongest predictor was the therapist's own beliefs about it. Whether they had been trained in it mattered a great deal. The therapist's own anxiety mattered, in the direction you would expect. And years of experience predicted nothing at all — the association was effectively zero.
Two honest caveats. These are cross-sectional surveys in which therapists reported on their own practice, so they cannot prove that beliefs cause under-delivery, and self-reported practice is not observed practice. And exposure is not magic: head-to-head trials comparing exposure against cognitive therapy found no significant difference for panic, PTSD or OCD, and for social phobia the limited evidence actually favoured cognitive therapy. So the claim is not that exposure is the only thing that works. The claim is narrower and better evidenced — that it belongs in the treatment, that a full dose of it is uncomfortable to deliver, and that a great many courses of anxiety therapy quietly leave it out.
It is not being thrown in at the deep end, and it is not something done to you. It is planned, collaborative, and starts at a step you agree is possible — you are the one who decides the order and the pace, and the therapist's job is to keep the step small enough to take and large enough to matter.
The older way of explaining it was habituation: stay in the situation until your anxiety comes down by half, and repeat. That framing has been superseded. Craske and colleagues reframed exposure around inhibitory learning — the point is not that the fear subsides during the session but that you build a new, competing piece of learning that the feared outcome did not happen, or that you could tolerate it when it did. Practically, that changes what a session aims for: variety, unpredictability, dropping the props, and a willingness to be surprised, rather than watching a number come down.
Being accurate about this: the inhibitory-learning account is a better theory and it is well regarded, but a comparative trial found inhibitory-learning-informed therapy hardly distinguishable from CBT as it is ordinarily delivered. The reframe is not yet a demonstrated clinical advantage. We think it produces better sessions. We are not going to tell you that has been proven.
Sitting near the exit. Carrying the medication you never take. Having a friend on standby, an escape plan, a bottle of water, a rehearsed excuse. These are safety behaviours, and no competitor page in Pennsylvania that we can find so much as mentions them, which is odd, because they are the main reason anxiety persists in someone who is otherwise doing everything right. They work in the moment. That is exactly the problem: they get the credit that would otherwise go to your own capacity, so the learning never lands.
The traditional instruction is to drop all of them immediately. That instruction is more contested than it is usually presented — a randomised trial by the same researchers who criticised safety behaviours found no meaningful difference between eliminating them outright and fading them systematically. In practice we plan this with you rather than announcing it, and either route is defensible.
A large number of people arrive here having been through the medical system first — the chest pain, the racing heart, the numbness in one hand, the sense that this time it is real. Panic disorder is very frequently first presented to emergency departments and cardiology clinics rather than to a therapist, and being told “your tests are normal, it's just anxiety” is one of the least useful sentences in medicine. It is heard as nothing is wrong with you, when what it means is the thing that is wrong is treatable and is not your heart.
Two things need saying together. Getting the medical workup was the right call, and unexplained physical symptoms should be assessed by a physician rather than assumed to be anxiety — we are not in a position to rule anything out and we will not pretend otherwise. And once that has been done, the fear of the sensations is itself a treatable target with a well-defined protocol, which is a much more specific answer than “try to relax.”
Not all anxiety is trauma-related, and a practice that told you otherwise would be selling you its own specialty. But a substantial share is, and this is the part of the picture the rest of the field leaves out.
A 2016 meta-analysis restricted to prospective cohort studies — the design that avoids asking adults to recall their own childhoods — found that childhood maltreatment was associated with later anxiety, with an odds ratio of about 2.7 compared with people reporting no such history. Only eight studies met the quality bar for inclusion, which tells you how thin this literature still is. It is an association rather than a demonstrated cause, and we are not going to convert it into a claim about what proportion of anxiety trauma explains.
What it changes in the room is the sequencing. If the anxiety is downstream of childhood trauma, an attachment injury, or a specific frightening event, then exposure work aimed only at the current trigger tends to hold while you are doing it and slide back afterwards. That is when we move to processing the source directly — and it is the case in which EMDR for trauma-rooted anxiety genuinely belongs, as distinct from EMDR offered for anxiety in general, which is a different and much weaker claim. See the next section.
Anxiety and low mood also arrive together far more often than not; if both are present, our page on depression counseling and therapy covers the other half.
Every practice in Pennsylvania lists the same modalities. Here is which of them actually has evidence in anxiety disorders, and which does not.
Exposure-based CBT is the only approach on this page with a substantial anxiety-specific evidence base, and even that is moderate rather than large — see the numbers above.
EMDR has good evidence for PTSD. It does not have good evidence for anxiety disorders. A 2021 systematic review of EMDR outside PTSD found the anxiety-disorder literature to consist mostly of case series and case studies rather than controlled trials, with contradictory results, including a randomised trial concluding it should not be a first-line treatment for panic disorder. We use EMDR where the anxiety sits on trauma, which is where its evidence lives. If someone offers you EMDR as a treatment for generalized anxiety, they have taken it past where the research goes.
Somatic and body-based work has essentially no anxiety-disorder evidence base. What literature exists is in PTSD, is preliminary, and is mixed in quality. We use it as an adjunct within trauma work, and for the body-level component of panic. It is not a treatment for an anxiety disorder and we will not describe it as one.
Brainspotting has the thinnest evidence base of anything we offer. There is one randomised controlled trial, in PTSD rather than anxiety, and the VA's National Center for PTSD reviewed it in 2024 and concluded that methodologically stronger studies would be needed to support its use. For anxiety disorders specifically there is next to nothing. Some clients find it valuable and we offer it on that basis, as an adjunct — not as an evidence-based anxiety treatment, and not instead of exposure.
And a hole in the numbers rather than the treatments: there is no Pennsylvania-specific prevalence figure for diagnosed anxiety disorders. The state health surveys do not measure it — BRFSS has no anxiety-diagnosis module, and the federal state-level estimates cover any mental illness rather than anxiety specifically. Any Pennsylvania anxiety percentage you encounter is either a national figure relabelled or a symptom-screen result described as a diagnosis. The most-cited national prevalence figures, meanwhile, come from a survey completed in 2003 under criteria in which PTSD and OCD still counted as anxiety disorders. They no longer do, and no comparable survey has re-estimated it since.
You may have filled one out, or found one online. It is a seven-item questionnaire, free to use, and useful for tracking whether something is shifting over time. It is not a diagnosis, and its limits are larger than most people are told. A 2025 Cochrane review pooling 48 studies and more than 19,000 people found that at the standard cutoff, the GAD-7 identifies only about 64% of generalized anxiety disorder and about 48% of anxiety disorders overall, while correctly clearing about 91% of people who do not have one. In plain terms: a high score is worth taking seriously, and a low score rules nothing out.
For completeness, the US Preventive Services Task Force recommended in June 2023 that adults aged 19 to 64, including during pregnancy and after birth, be screened for anxiety disorders. For adults 65 and over it concluded the evidence was insufficient either way — a distinction frequently misreported, including by sources that should know better.
We start where the evidence is strongest and work outward. That means cognitive and behavioural work with real exposure in it, built with you rather than sprung on you, and dosed properly rather than gestured at. Where avoidance is entrenched or the fear traces to a specific event, we use a formal exposure protocol. Where the anxiety sits on trauma, we process the trauma rather than continuing to manage its downstream effects. Body-level work and Brainspotting are adjuncts, used for what they are good at and described honestly. Sessions can be individual or in a group — and for social anxiety in particular, a group is not the lesser option, it is closer to the target. DBT skills are available where distress arrives as overwhelm. Our team's credentials are on the clinicians page.
The best-evidenced approach for anxiety, and the one that carries the exposure work.
A formal exposure protocol for entrenched avoidance and event-linked fear.
For anxiety that sits on trauma — where its evidence actually is.
An adjunct for the body-level component of panic. Not a standalone treatment.
An adjunct some clients value. The evidence base is thin and we say so.
Two barriers usually sit between someone with an anxiety disorder and a first appointment, and only one of them is money. Naming both is fairer than pretending the second does not exist.
Take the money first, because it is the more tractable. What we accept is set out on the fees and insurance page, and a Good Faith Estimate means the number is known before you commit to anything. If you are uninsured or your income has dropped, Medical Assistance — Pennsylvania's Medicaid programme — covers mental health care through Behavioral HealthChoices; you can apply at compass.dhs.pa.gov or call 1-866-550-4355. Pennie, at pennie.com or 1-844-844-8040, is the state's official marketplace and the only route to subsidised coverage. And if none of that resolves it, the free drop-in anxiety group in Lancaster listed further down is a real option rather than a consolation prize.
Now the other barrier. For a good number of people reading this, making the phone call is a recognisable instance of the exact problem they want treated — an unpredictable conversation with a stranger, about something embarrassing, with no script. That is worth saying out loud, because the usual advice at this point is a cheerful “just reach out,” which is precisely the kind of encouragement that has never once helped anybody do a hard thing.
So: the first conversation is ten minutes, it is free, and you are allowed to have your questions written down in front of you. You can email instead of calling if that is genuinely easier, though it is worth noticing whether the email is the accessible option or the avoidant one — that distinction is most of what the work is about, and you are not expected to have it sorted out before you get here.
Some of what helps is free, and some of it is not therapy. Each of these is described as its own organisation describes it, and all were verified in August 2026.
You have almost certainly tried, and the fact that it did not work is not evidence about your intelligence or your willpower — it is evidence about how anxiety is maintained, which is by what you do rather than by what you believe. That is workable, and it is workable in a way that has an actual protocol behind it rather than a suggestion to breathe. The first conversation is ten minutes and costs nothing. You are welcome to spend it asking what exposure would look like for your particular version of this before you commit to anything.
Call us at (717) 394-3994
Clinically reviewed by the ACRS clinical team. Last reviewed August 11, 2026. Helpline and organisation details verified August 11, 2026. Every research finding above is reported with its design and its limitations, including where those limitations are unflattering; where an approach lacks an evidence base in anxiety disorders, this page says so rather than omitting it. Diagnostic terms follow DSM-5-TR. General information only — not a substitute for individual clinical assessment, and not medical advice. Decisions about medication belong with a prescriber, and physical symptoms belong with a physician.