How insurance quietly decides who can treat you, how often, and for how long

A card in your wallet is not the same as care. Coverage tells you a plan exists. It does not tell you who is allowed to treat you, how often, for how long, with which approach, or whether that clinician can afford to keep doing the work well. Most patients discover the difference only after they need help: the authorization is denied, the sessions run out, the referral list is full of clinicians who are not taking patients, or the clinician they finally trusted quietly stops taking their plan. This is the part of mental health care most patients are never shown, and it shapes the care they actually get.

This is not an argument that insurance is useless. For many people it is the only door available, which is exactly why its limits matter. When a system promises access but cannot sustain the clinicians, the continuity, the frequency, or the duration that good care requires, the patient still has coverage. What they may not have is care.

I learned this from the inside

I took insurance for several years, deliberately. I wanted to treat people who could not pay out of pocket, because those are often the people who need the work most and can reach it least. I tried hard to make it work. I learned the billing, paid for the consultants, chased the authorizations, and absorbed the audits. And I lost money. Not a little. Thousands, year after year, for doing the work and then fighting to be paid for it.

What ended it was not money alone. It was realizing that the system made good care structurally difficult, and that the cost of trying landed on the patients I most wanted to serve, the clinicians I was trying to train, and the sustainability of the work itself. I stopped taking insurance because I could not keep absorbing the losses, and that decision still bothers me. It meant stepping back from an underserved population that needed help. That failure belongs upstream, but both clinician and patient pay for it.

The patient usually experiences this as a disappearance: the clinician leaves the network, the referral list fails, or the care becomes too expensive to keep. Underneath that disappearance is not indifference. It is an economic structure.

The economics are the engine

Start with the math, because the math drives everything else. The rate an insurer pays varies by plan, region, code, and contract, but the basic problem is consistent: the paid hour is not the worked hour. You cannot bill forty clinical hours in a forty-hour week. The rest goes to notes, scheduling, billing, no-shows that no one reimburses, and the unpaid labor of getting paid at all.

Then come the obstacles, each of which costs time, the one resource a clinician cannot replace. Pre-authorization before you can begin. Audits after the fact, sometimes random, with the threat of clawbacks: money already earned and spent, demanded back. Payments that arrive late or not at all.

When I tried to get reimbursed myself, I was put on hold for hours and sent in circles. When I paid specialists to recover the reimbursement, that cost came out of the same shrinking number, and even then it did not always work. At one point my own billing specialist told me the cleanest path to being paid was to turn around and bill the clients I had already gone to the wall for.

The comparison that kept surfacing was not flattering. A more reliable hourly wage was available at In-N-Out, with none of the debt, the liability, or the years of training behind it.

The rules push against good care

Here is what the system measures, and what it does not. I once asked an auditor to call my clients and ask them about the care I provided. He went silent. The care was never the question. The codes were.

Several of the rules push directly against clinical judgment. The system rewards documentation that demonstrates acuity. Care is easier to authorize when the record emphasizes severity, risk, and impairment. Sometimes that is clinically accurate. Sometimes it creates pressure to write toward reimbursement rather than toward the whole truth of the patient. Certain diagnoses and severity profiles are easier to authorize than others, which creates a quiet incentive to make the chart fit the benefit rather than the patient. I declined to practice that way. I am naming it because the pressure is real, and resisting it is an ethical discipline, not a given.

Then there is utilization review, the running judgment about whether your care will keep being paid for, and it cuts in predictable, damaging ways. A plan may refuse to authorize more than one session in a week unless the patient is in active crisis, as if a person has to be in danger to deserve enough care to stay out of it. Prevention is cheaper than crisis, but the system often pays more reliably once crisis has already arrived. Some patients clearly needed more care before a crisis forced a higher level of care, and the coverage would not allow it. I once continued clinically necessary work after authorization ended and absorbed the loss myself, because I judged that the person needed it. What I took from it has stayed with me: when I take someone into my care, that carries a level of responsibility that does not depend on whether I am being paid. That is not something an insurer can authorize or deny.

The same logic often refuses to cover the combinations that real treatment sometimes requires: a family session and an individual session in the same week when something has to be processed, or couples work alongside individual work. And when a patient does well, coverage often stops, on the theory that improvement means the need is gone. Frequently the opposite is true. The absence of crisis is not the same as readiness to stop. The gains are new and fragile, and what protects them is maintenance, the quiet, ongoing work that keeps hard-won progress from unraveling. That is exactly the work the system is quickest to defund.

The network is gated

Even getting into a network is gated. Panels can be closed, or open only to clinicians who already carry the kind of demand that signals they do not need the panel. Larger groups are often paid more than solo clinicians for the same work, which pushes care toward volume and away from the individual practitioner who might have known you for years. And many patients run into what regulators and policy researchers call ghost networks: directories that look full on paper but list clinicians who are not taking new patients, do not accept the plan shown, or cannot be reached at all. To the patient, the network looks large until every call returns the same answer: not taking new clients, no longer in network, no response.

The rules also do not sit still. Contracts change, policies change, panels open and close, and the practical result is that much high-quality care has migrated out of network entirely. For many patients, finding the right clinician increasingly means finding one their plan will not pay for. None of this is the patient’s fault, and most patients never see the machinery that produced it. They just feel themselves falling through the cracks.

The training pipeline breaks

The damage is not only to current patients. It reaches the next generation of clinicians, and through them, the patients those clinicians would have treated.

In many private-practice insurance arrangements, pre-licensed clinicians, including psychological assistants and associate therapists, are difficult or impossible to reimburse under the same terms as independently licensed clinicians. So a practice built around insurance often cannot pay them adequately for that work or give them the supervised caseload they need to train. Too often, they are told to find their own out-of-pocket clients while still unlicensed. For someone without an established name, that rarely produces enough work to live on. I watched capable people barely make ends meet inside that gap. I paid some of them out of my own pocket because I wanted them trained and I valued the work and the people, and I took the hit for it. I could not sustain it, and supervising the work properly was its own large, unpaid commitment of time.

This is how a profession thins itself out. The people who should be learning the craft cannot afford to learn it, the people who should be teaching cannot afford to teach it, and patients lose both the continuity of care now and the supply of good clinicians later.

The platform model

Into that gap came the apps, venture-backed therapy platforms promising convenience and scale. They solved a real problem, access, and created new ones. The model rewards availability and volume more easily than depth, continuity, or careful long-term work, and the pay often reflects that. Clinician reports and public pay data vary, but the pattern is clear enough: some platform work compensates licensed clinicians at rates that look closer to gig work than to sustainable professional practice, especially once independent-contractor taxes, unpaid time, benefits, and overhead are counted. Some models also compensate written messaging in ways that reward output over depth.

BetterHelp is the most visible example of the model I mean. The point is larger than one company. Therapy delivered through gig-economy rates and incentives is not built to sustain careful, long-term clinical work, the clinicians who do it, or the workforce the field needs. Someone early in their career may take it for the flexibility or the foot in the door. As the foundation of a profession, it does not hold.

Leaving is not always a choice

When a good clinician stops taking insurance, it can look like abandonment, and sometimes it feels that way even to the clinician. The patient is not wrong to feel the loss. More often it is the end of a long, failing negotiation with a system that asks for high-responsibility care at rates and under rules that make the work impossible to sustain. The departure looks like a private decision. Often it is a systemic outcome.

What the fee actually carries

So when a competent clinician charges what looks like a high hourly fee, consider what that number is actually carrying.

It carries the clinical hour and all the hours around it. It carries education, debt, licensure, malpractice coverage, office overhead, self-employment tax, health insurance, retirement, sick time, vacation, no-shows no one pays for, notes, scheduling, and billing. There is no salary, no employer, and no floor, so each of those comes out of the same number or does not exist at all. And it carries the cost of staying good. Consultation and a clinician’s own therapy are not luxuries. I pay for consultation with clinicians whose judgment I trust when a case calls for it, and I do my own work, because that is how quality is maintained. A clinician who has stopped investing in their own judgment is not someone you want holding yours.

None of this means every high fee is justified. Private pay does not automatically mean better care. The problem is not that a clinician charges a high fee. The problem is when a fee is disconnected from real skill, real availability, and real responsibility. Run the real math, and a fee that sounded steep starts to look less like a markup and more like the cost of a sustainable career. At its best, paying out of pocket buys the specific things insurance often will not authorize: the clinician’s full judgment, the right frequency, the right combination of sessions, the maintenance work, and the responsibility that does not switch off when a benefit runs out.

The part I will not pretend away

This does not resolve cleanly, and I am not going to pretend it does. If clinicians must charge enough to sustain the work, and they must, then high private fees leave people who cannot pay them with worse options, or none. I felt that directly the day I could no longer keep serving the people who most needed the work and could least afford it. I hold both of these at once. Clinicians have to be able to charge in a way that sustains the work, and the fact that this leaves vulnerable people behind is a real loss. That loss is not solved by asking individual clinicians to subsidize a broken system indefinitely, and it is not created by the clinician who sets a fair fee. It is created by a system that makes it nearly impossible to charge a sustainable fee and still serve the patients most shut out of care. Naming who is responsible matters, because the reflex is to blame the clinician for the price and never ask why the price had to be that high.

Coverage is not care

Coverage is real, and for many people it is the only door available. That is exactly why its limits deserve to be named instead of hidden. But the card is not the care. The care is the person across from you, the time you are actually given, the continuity you are able to keep, the judgment the clinician is allowed to exercise, and whether the people who provide it can afford to keep providing it well. None of that fits inside a benefit summary.

Coverage is not care. It never was.