Partial Hospitalization (PHP): Hours, Cost and How It Differs from IOP

Partial hospitalization is day treatment. A person attends a structured clinical programme for most of the working day, typically five days a week, and goes home at night. In the ASAM Criteria — the framework most clinicians and insurers use to match people to levels of care — it sits at Level 2.5, between intensive outpatient below and residential above.

The name causes most of the confusion. "Partial hospitalization" sounds like a partial hospital stay, which suggests beds and admission. There are no beds. The word "hospitalization" is a billing and regulatory inheritance, not a description of the experience. What you actually get is a full clinical day — group therapy, individual sessions, psychiatric review, medication management, family work — delivered in the same intensity a hospital programme would deliver, minus the overnight stay.

That distinction matters more than it sounds. Everything PHP does clinically, it does during daylight hours. Everything that happens at night — cravings at 2 a.m., a housemate drinking in the next room, the argument that used to trigger a relapse — happens without staff present. Which is precisely why the question of where a person sleeps turns out to be the deciding one.

Levels of care are separated by intensity, measured in structured clinical hours per week. The numbers below are the working thresholds in the ASAM framework. They are not arbitrary — they mark the points at which the amount of supervision and clinical contact changes what is realistically treatable.

This is the single most common question about this level of care, and most answers reduce it to a number of hours. The hours are real, but they are the consequence of a clinical judgement, not the judgement itself.

Here is the part that rarely appears in comparison tables, and it matters more than the hour count. PHP delegates the unsupervised hours to the environment. Sixteen hours of every day, and the whole weekend, are held not by clinicians but by wherever the person goes home to.

So the clinical question is not only "does this person need twenty hours of structure?" but "are the other hundred and forty-eight safe?" A person with a stable home, a supportive partner, and no substances in the house gets a genuine advantage from PHP over residential: they practise recovery in the actual environment they will live in, with staff available the next morning to work through what went wrong. That is real therapeutic value, not a compromise.

Reverse those conditions and the same programme becomes a liability. If the household is actively using, if housing is unstable, if the evenings are unsupervised in a way that has previously led straight back to use, then PHP is the wrong level however well the hours fit — and the honest recommendation is residential, or PHP combined with sober living. A programme that never asks in detail about your living situation before recommending PHP has skipped the most predictive question in the assessment.

Programmes vary, but the shape is consistent: a clinical day built around group work, with individual and medical contact threaded through it. The value is in the accumulation — the same people, the same structure, five days running, which is what makes patterns visible in a way weekly therapy cannot.

This is where clinical suitability and coverage eligibility come apart, and it catches families out repeatedly.

The ASAM Criteria is a clinical framework. It places PHP at Level 2.5 and describes it by intensity and by patient need across six dimensions. Clinicians use it to answer: what does this person require?

Medicare operates a regulatory definition instead. Partial hospitalization is covered as a distinct benefit with its own conditions — including a requirement that a physician certify the person would otherwise require inpatient care, an individualised treatment plan, and a defined minimum of weekly programme hours. That is a narrower and differently-shaped test than "this is the clinically appropriate level".

The consequence is practical. A person can be clinically well-suited to PHP and still fail the coverage test, because the certification framing asks whether they would need hospitalisation without it — not whether PHP is simply the best fit. Commercial insurers apply their own medical-necessity criteria, often ASAM-derived but not identical. So the question to ask any programme is not just "do you take my insurance?" but "which criteria does my plan apply to PHP, and how do you document against them?"

PHP costs less per week than residential care, and that is how it is usually presented. The comparison is incomplete, because the two levels consume different resources.

Residential bundles everything: clinical hours, accommodation, meals, 24-hour staffing. PHP bills the clinical hours and leaves you to cover living costs — which you were covering anyway, so that genuinely is a saving. But PHP also consumes the working day for several weeks, and that cost lands on income and career rather than on the invoice. For someone salaried with leave available, the trade is favourable. For someone self-employed or hourly, five days a week away from work for a month can exceed the difference in programme fees.

Two further practicalities. Transport — five round trips a week for several weeks is a real cost, and a real barrier: attendance is the mechanism by which PHP works, and a programme ninety minutes away tends to produce gaps. Sober living — where PHP is clinically right but the home environment is not, supported housing alongside the programme is often the correct combination, and the two costs together should be compared against residential rather than against PHP alone.

Partial hospitalization did not originate in addiction treatment. It has been used for years in psychiatry — for depression, anxiety, bipolar disorder and eating disorders — and a large share of PHP programmes treat mental health conditions either alongside substance use or on their own. This matters practically, because roughly the same structure serves both, and the two so often travel together.

Where both are present, the important question is whether a programme treats them together or sequentially. Sequential treatment — get sober first, then address the depression — has a poor record, because for many people the substance use is doing work that the untreated condition created. Integrated treatment addresses both in the same programme, with one team.

Level of care is matched across all six ASAM dimensions, not by severity of use alone. Below is the shape of the judgement in plain terms.

PHP is rarely the whole treatment. It is the middle of a sequence: most people arrive from detox or residential care, and leave into intensive outpatient, then standard outpatient, with recovery supports running alongside throughout.

The step-down should be designed in the first week, not the last. There are two reasons. First, the drop in structure is significant — going from twenty-plus hours a week to nothing is the pattern under which relapse concentrates, and the tapered path through IOP exists precisely to prevent that cliff. Second, authorisation for the next level takes time to arrange, and a gap of even a fortnight between levels is a gap in which momentum is lost.

Alongside the clinical step-down, three things deserve explicit decisions rather than vague intentions: medication for alcohol or opioid use disorder, which should be settled during PHP rather than deferred; recovery supports in the community, which take weeks to become genuinely useful and therefore should start while the programme still provides a safety net; and the living situation, since whatever made the home environment workable during PHP needs to hold once the daily structure disappears.