How we calculate – PlanMath PA

The basics

Words you'll see on the estimator, in plain English.

Premium: The plan's monthly price. You pay it every month, whether or not you use any care.
Deductible: What you pay for covered care each year before the plan starts paying its share. Some services, such as many doctor visits with a copay, are covered before you meet it. Free preventive care never counts.
Copay: A fixed amount you pay for a service, for example $30 for a doctor visit.
Coinsurance: Your share of the price of a service, for example 20%, usually once you have met the deductible.
Out-of-pocket maximum: The most you pay for covered in-network care in a year: deductible, copays and coinsurance together. Once you reach it, the plan pays for everything else that year. Premiums don't count toward it.
Bad year: A year with a big medical event, such as a hospital stay, in which you reach the plan's out-of-pocket maximum: the most you'd pay for care that year. You still pay the premiums too.
Metal levels: Bronze plans cost the least each month and the most when you use care. Silver, Gold and Platinum cost more each month and less when you use care. Catastrophic plans are only for people under 30 or with a hardship exemption.
HMO, EPO, PPO and POS: They say which doctors and hospitals the plan pays for. HMO: only the plan's network, and usually a referral to see a specialist. EPO: only the plan's network, usually no referral. PPO: you can go out of network, but pay more. POS: like an HMO, with some out-of-network coverage. Emergencies are covered anywhere.
Off Pennie only: The plan is sold directly by the insurer, not through Pennie, so the premium tax credit and the extra Silver help (cost-sharing reductions) cannot apply to it. It is still an ACA plan, with rates approved by the PA Insurance Department. You enroll with the insurer.
Premium tax credit: Federal money that lowers your monthly premium. How much depends on your household income and size. Pennie works out the exact amount when you apply.
Cost-sharing reductions (CSR): Extra help for lower incomes (up to 250% of the poverty level): Silver plans come with a lower deductible and out-of-pocket maximum. CSR 73, 87 and 94 name the version; a higher number means more help.
Health savings account (HSA): A savings account for medical bills. Money you put in is not taxed, and you can open one only with an HSA-eligible plan, such as any Bronze or Catastrophic plan on Pennie. Skip it if you are not sure.
Summary of Benefits and Coverage (SBC): The plan's official summary of what it covers and what you pay, in a standard format every plan uses. Read it before you enroll.
Rating area: Pennsylvania is split into 9 pricing regions made of whole counties. Insurers set their prices by region, so your county decides what each plan costs.
Prescription fills: One pick-up of a prescription, usually a month's supply. One pill a day for a year is 12 fills.

The number we rank by: the average over 5 years

For each plan we add up five years of premiums and five years of what you pay for care. Your care has two parts: the care you expect or plan, which you enter and which repeats every year, and unexpected care (ER visits, urgent care, hospital stays and, if you say a baby is possible, a birth), which may or may not happen. Nobody knows which years will bring unexpected care, so we simulate 1,000 possible 5-year futures for your household and run every plan's rules on each one.

Average over 5 years = the average over the 1,000 futures. Plans are sorted from the lowest to the highest.
If things go badly: in one future out of ten, the 5-year total is higher than this amount (1 chance in 10).
A usual year: the middle value of the first year's total, premiums and care included.
Chance of reaching the out-of-pocket maximum: the share of futures with at least one year in which somebody reaches the plan's out-of-pocket maximum (or the family reaches the family maximum).
Everyone is a year older each year, so each year's premium follows the federal age curve (it stops rising at 64). People who turn 65 usually move to Medicare, which this doesn't count.
With a tax credit, it is worked out again for each year: the benchmark plan is priced for everyone's age that year, and your share stays the same part of your income, which we keep as you entered it. So the credit takes up most of what aging and premium increases add, for the benchmark plan entirely.
If you enter a yearly premium increase, every plan's premium (the benchmark's too) grows by it each year. Care prices and the plans' deductibles and out-of-pocket maximums stay at this year's amounts.

"What if things go badly?"

Under the answer, the folded section "What if things go badly?" asks how many of the next 5 years go badly. Nothing is picked until you pick a number. Picking one ranks the plans for exactly that many bad years, with a fixed method and no simulation: each of the five years is either a usual year or a bad year, and the bad years are the last ones. The page says whether that changes which plan is cheapest. "Back to my estimate" returns to the averages.

Usual year = twelve months of premiums + the cost of the care you entered.
Bad year = twelve months of premiums + the plan's out-of-pocket maximum (see below).

"Same insurer has a better plan"

A plan gets this label when another plan from the same insurer, on the same network of doctors (the network ID in the CMS plan file), is at least as good on everything we compare: the premium you'd pay, the deductibles, the out-of-pocket maximum and every benefit (a lower or equal copay or coinsurance, never a deductible where the other has none), and costs the same or less for your household in every case we compute. Then the other plan costs the same or less whatever care you need. Plans that include dental or vision are never labeled, because we don't price those extras. The label doesn't change the order.

The monthly premium

For each person, the premium is the plan's base rate for a 21-year-old in your rating area × an age factor × a tobacco factor (only for smokers aged 21 or older, and only if the plan charges one). The age factor follows the standard federal age curve that Pennsylvania uses: 0.765 up to age 14, 1.000 at 21, 1.357 at 43 and 3.000 from 64 on. Only the three oldest children under 21 are charged. Rates come from the CMS public data files.

People 65 or older usually get Medicare instead, but we still price them on the 64+ factor.

A usual year: walking through your care

Every visit, test or prescription you enter becomes an event with a full price (the price table below). Then we replay the year in order and apply the plan's rules from its Summary of Benefits and Coverage (SBC) to each event: a copay, a share of the cost (coinsurance), no charge, or not covered. Some rules apply only after the deductible.

The order of events matters

Whether you pay a copay or the full price depends on whether you already met the deductible, so the order changes the answer. We use one fixed, repeatable order:

If you enter n of a service, the first one happens 0.5 ÷ n of the way through the year, the second at 1.5 ÷ n, the third at 2.5 ÷ n, and so on. Four primary care visits land at 12.5%, 37.5%, 62.5% and 87.5% of the year. Each service is spread out on its own.
A delivery (birth) always happens three quarters of the way through the year.
Everything is sorted by time. When two events happen at the same moment, the first person in your household goes first, then services in a fixed order.

What counts toward the deductible

Only what you actually pay counts toward the deductible. If the out-of-pocket maximum cuts a payment short, the part the plan covers instead does not count as deductible you paid. Services the plan does not cover, and visits above a plan's yearly limit (for example, physical therapy sessions past the cap), are paid in full and do not count toward the out-of-pocket maximum.

Deductibles and maximums

Individual and family deductibles and out-of-pocket maximums both apply. Everything you pay in cost-sharing counts toward both maximums.
Embedded deductible: each person stops at the individual amount. Aggregate: only the family amount counts. The SBC tells us which one a plan has.
When a plan has a separate prescription deductible, drugs use that one.
Some plans charge a copay per hospital day up to a number of days, or give the first few visits at a different price. We follow the SBC. For hospital stays these counts start again at each admission, the way the SBCs describe them (a copay per admission, or per day up to a number of days per admission).

Unexpected care: chances and simulation

This is not a medical prediction. The chances are averages for privately insured people of the same age, adjusted for the health conditions you tick. They say nothing certain about you and are not medical advice.

What we sample, for each person and year

The number of ER visits that don't lead to a hospital stay.
The number of urgent care visits.
The number of unplanned hospital stays, the nights of each stay, and whether the stay began with an ER visit (which adds that visit).
A birth, only for a person marked "A baby is possible in the next 5 years".

The chance of at least one of each comes from the person's age band and health conditions (tables below). When there is at least one, how many there are and how many nights each stay lasts are drawn from what privately insured people of that age had (MEPS 2022-2024 for ER visits and hospital stays, NHIS 2025 for urgent care). Each sampled event is priced with the price table below and goes through the plan's rules together with that year's planned care. No yearly chance is ever set above 95%.

The care you enter is a floor, not an addition

The ER visits, urgent care visits and hospital days you enter are ones you know about or plan. In each simulated year we use the larger of what you entered and what was sampled, not the two added together: with one ER visit entered and two sampled, that year has two. The hospital days you enter form one planned stay; sampled stays count only when there are more of them than that one. So a choice that already includes a flare-up ("Asthma", "Usual childhood care") isn't counted twice. All other care you enter is used as entered, every year, except a planned pregnancy (see "Births").

What the chances count

ER visits. The survey asks whether a person went to an ER at least once in the past 12 months, including visits that led to a hospital stay. We take out the share of ER visits that ended in an admission (MEPS, by age band). A sampled hospital stay adds one ER visit as often as stays began in the ER, so no admitted ER visit is counted twice.
Hospital stays. An overnight stay that is not for a delivery or a newborn. The survey level is multiplied by the share of people with a stay who had at least one stay that was not for a delivery or a newborn (MEPS, by age band, men and women together). Births come only from "A baby is possible". Scheduled inpatient surgery is part of this risk; the planned surgery you can enter is outpatient surgery, so the two don't overlap.
Urgent care visits. From NHIS, the only public survey that asks about urgent care.
Health conditions. The table's levels are for everyone in the age band. Someone with none of the eight conditions starts a little lower, so that the band's average over everyone's conditions matches the survey.

Hospital stays are priced per admission

Each hospital stay is its own admission. When a plan charges a copay per admission, or a copay per day up to a number of days, the count starts again with every stay, the way the plans' SBCs describe it. Two short stays in one year can therefore cost more than one long stay. When you enter hospital days, they are one stay.

Births

A birth can happen only for a person marked "A baby is possible in the next 5 years". Set it on the person who would give birth; we don't ask anyone's sex.
The chance each year is the national birth rate for women of the same five-year age group (NCHS, 2024), from 15 to 49; it is highest at 30 to 34, 93.7 births per 1,000 women. 32% of births are cesarean. A sampled birth adds the same prenatal care as "Pregnant or planning a birth this year" and a delivery of that type, at most one per person per year.
A pregnancy you enter ("Pregnant or planning a birth this year", or a delivery) is planned care for the first year only in your estimate; later years get a birth only if one is sampled. "What if things go badly?" keeps its fixed method, so the usual year, pregnancy included, repeats every year.
The newborn's own care and the switch to family coverage after a birth are not included.

The same futures for every plan, and the same answer every time

We sample 1,000 household futures once and run every plan on the same ones, so the differences between plans come from the plans' rules, not from luck. The random draws start from a fixed seed computed from what changes the chances (ages, health conditions, "A baby is possible" and a planned pregnancy), so the same household always gets the same futures and the same numbers. Plans whose averages are very close could swap places with a different set of futures.

The chances

Share of privately insured people with at least one of each in a year (NHIS 2025)
Ages	ER visit	Urgent care visit	Overnight hospital stay
0-17n = 3,499 to 3,671	13.5%	30.9%	2.1%
18-25n = 1,047 to 1,048	18.3%	36.2%	3.6%
26-44n = 4,842 to 4,846	16.7%	33.0%	5.5%
45-54n = 2,439 to 2,440	15.2%	29.8%	4.7%
55-64n = 2,831 to 2,833	18.5%	26.2%	7.5%

n is the number of people asked (it differs a little between the questions). People 65 or older use the 55-64 row. The ER column includes visits that led to a hospital stay and the hospital column includes births; we take both out before sampling (see "What the chances count").

How each health condition changes the chances (rate ratios)
Condition	ER visit	Urgent care visit	Hospital stay
Diabetesn = 695 to 696 (MEPS 1,755)	1.21MEPS 1.34	1.13	1.33MEPS 1.80
High blood pressuren = 2,705 (MEPS 5,490)	1.52MEPS 1.59	1.14	1.75MEPS 2.02
Asthman = 1,656 (MEPS 3,181)	1.38MEPS 1.55	1.21	1.13MEPS 1.35
COPDn = 207 (MEPS 303)	1.60MEPS 1.69	1.29	1.75MEPS 2.38
Heart diseasen = 307 (MEPS 575)	1.41MEPS 2.18	1.10	2.87MEPS 3.07
Depressionn = 1,891 to 1,892	1.39	1.09	1.61
Anxietyn = 2,032 to 2,033	1.30	1.18	1.15
Cancern = 707 to 708 (MEPS 1,465)	1.12MEPS 1.26	1.13	1.83MEPS 2.04

n is the number of people with the condition. A ratio of 1.50 means a 50% higher chance than for someone of the same age without the condition. The ratios come from NHIS 2025 privately insured adults 18 to 64 (a weighted Poisson regression on age band, sex and all eight conditions), and we use them for children too. They describe how often people with a condition used care, not what the condition causes. Where the MEPS 2022-2024 files have the condition, their ratio is shown under it as a check; we use the NHIS value. A person's ratios multiply, and the product is capped at 4× for each kind of event.

Sources for the chances

Every number, with its sample size, standard error or interval and a note on how it was computed, is in config/base-rates.json. The method, step by step, is in docs/research/base-rates-method.md in this project's source.

A bad year

In a bad year, somebody reaches the plan's out-of-pocket maximum. By default, one person does, and everyone else uses care as in a usual year, all capped at the family maximum. To stay on the cautious side, we try each person as the one who maxes out and use the person whose bad year costs your family the most.

On a few plans the SBC gives a separate per-person out-of-pocket maximum for family coverage. On those plans, when more than one person is covered, each person's cap is that per-person family amount, not the plan's individual maximum.

With the checkbox "everyone reaches the family maximum" ticked, the whole household reaches the family maximum instead. Costs the plan never covers still come on top of the maximum, in both cases.

Tax credits (full prices until you ask for them)

Step 4 asks whether to show prices with a tax credit. Until you answer, and if you choose full prices, every plan is at its full price. When you ask for the credit, we estimate the premium tax credit from your household income and the number of people in your tax household. That number starts at the people you entered and can't be lower; raise it if you claim others on your taxes.

Your share of income. We compare your income to the federal poverty level. The applicable percentage in the table below (it rises smoothly within each band) is how much of your income you are expected to put toward a benchmark plan. There is no credit under 100% or above 400% of the poverty level. At or below 138%, Pennsylvania adults likely qualify for Medical Assistance (Medicaid) instead, so we show no credit and no cost-sharing reductions there, only a note to check COMPASS or Pennie.
The benchmark. It is the second-cheapest Silver plan sold on Pennie for your family in your county, priced for the ages in your household.
The credit = benchmark premium − (income × applicable percentage ÷ 12), never below zero. We subtract it from the monthly premium of every plan except Catastrophic plans and plans sold only off Pennie, and never below $0.
Cost-sharing reductions (CSR). Above 138% and up to 250% of the poverty level, and only if you are eligible for the credit, Silver plans come with lower deductibles and out-of-pocket maximums (CSR 94 up to 150%, CSR 87 up to 200%, CSR 73 up to 250%). We swap each Pennie Silver plan for its CSR version when it is in our data; when it isn't, the plan's row says so and shows the standard plan's costs. Other metal levels have no CSR version. You can get the CSR version even when your credit works out to $0.
Job-based offers. If you tell us you have an affordable job-based offer, we assume no credit and no cost-sharing reductions.
Immigration status. Starting with plan year 2027, fewer lawfully present immigrants qualify. For those years, if you tell us you are not a U.S. citizen or green-card holder, we assume no credit and no cost-sharing reductions. Check your own status with Pennie.
The final amount is set by Pennie when you enroll. Since 2026 there is no cap on paying back extra credit if your income ends up higher than you estimated.
"If your income changes", under the answer, puts these limits in dollars for your tax household (Medicaid up to 138% of the poverty level, cost-sharing reductions up to 150%, 200% and 250%, the credit up to 400%), shows which band your income is in, and says what the next limit up would change: at 400%, the credit you would lose, worked out from the same benchmark plan.

HSA tax savings (off unless you turn it on)

The health savings account (HSA) is off unless you turn it on under "More options" in step 4. With it on, HSA-eligible plans (plans marked HSA-eligible, and, since 2026, all Bronze and Catastrophic marketplace plans) get the tax savings taken off their totals, and plans are ranked by the total after HSA tax savings, in your estimate and in "What if things go badly?". The savings in a year are

contribution × (your federal bracket + Pennsylvania's 3.07%)

The contribution is what you pay for care that year (in your estimate, each simulated year's own amount), or the yearly limit if you choose that option (the limit is for one person or a family, plus an extra amount for each adult 55 or older). The limit comes from the IRS. We leave out investment growth inside the HSA, employer contributions, and Social Security and Medicare (FICA) taxes, which contributions you make yourself do not reduce. Whether you personally can open an HSA has its own rules, for example not having other coverage that is not an HSA-qualified plan.

Default prices

These are the full allowed prices before insurance. They are estimates built from public price studies and the Medicare fee schedules, and you can change every one of them under "Price assumptions" on the main page. Prices differ by region and provider, and Philadelphia tends to be above average.

Service	Price	Source
Primary care visit	$130	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Specialist visit	$200	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Urgent care visit	$175	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Emergency room visit	$2,500	Estimate. Context: RAND hospital price transparency study (rand.org)
Lab test	$75	Estimate. Context: Medicare Clinical Laboratory Fee Schedule (cms.gov)
X-ray	$150	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Advanced imaging (MRI/CT)	$1,000	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Physical therapy session	$120	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Mental health visit	$140	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Generic prescription (monthly fill)	$15	Estimate: illustrative planning default, edit to match your medications. Context: CMS Medicare Part D spending by drug
Preferred brand prescription (monthly fill)	$500	Estimate: illustrative planning default, edit to match your medications. Context: CMS Medicare Part D spending by drug
Non-preferred brand prescription (monthly fill)	$800	Estimate: illustrative planning default, edit to match your medications. Context: CMS Medicare Part D spending by drug
Specialty prescription (monthly fill)	$6,000	Estimate: illustrative planning default, edit to match your medications. Context: CMS Medicare Part D spending by drug
Outpatient surgery, facility fee	$5,600	Estimate. Context: RAND hospital price transparency study (rand.org)
Outpatient surgery, surgeon and anesthesia fees	$1,400	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Hospital stay, facility fee (per day)	$5,000	Estimate. Context: RAND hospital price transparency study (rand.org)
Hospital stay, physician fees (per day)	$1,000	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Prenatal visit	$200	Estimate. Context: Medicare Physician Fee Schedule (cms.gov)
Vaginal delivery, hospital fee	$10,000	Estimate. Context: Peterson-KFF Health System Tracker, cost of childbirth
Vaginal delivery, physician fee	$3,000	Estimate. Context: Peterson-KFF Health System Tracker, cost of childbirth
Cesarean delivery, hospital fee	$19,000	Estimate. Context: Peterson-KFF Health System Tracker, cost of childbirth
Cesarean delivery, physician fee	$5,000	Estimate. Context: Peterson-KFF Health System Tracker, cost of childbirth

Plans sold only off Pennie

Some insurers sell ACA plans only directly, not through Pennie. We list them next to the Pennie plans, markedOff Pennie only, so you can compare every plan you could buy in your county. You enroll in them with the insurer, not on pennie.com.

Every plan we list is an ACA individual plan whose rates the PA Insurance Department approved. Other products sold off Pennie, such as short-term plans or health-care sharing ministries, are not ACA plans and are never listed.
The premium tax credit and the extra Silver help (cost-sharing reductions) only apply to plans bought through Pennie. An off-Pennie plan always shows its full price, and the credit is never subtracted from it.
An off-Pennie plan can still cost less than a Pennie plan even with the credit. On Pennie, Silver prices often carry an increase that pays for the extra help ("silver loading"); an off-Pennie Silver plan may not carry it. The credit applies only to Pennie plans, so when you ask for it, compare the prices shown.
Prices come from the insurer's rate filing, which reproduces the official Pennie file to the cent for plans that appear in both. Benefits come from the plan's SBC when it has one. Where the insurer publishes none, we use the most credible official source, and say so on the plan.

Where the numbers come from

Every number about a plan traces to a public source: the CMS public data files, the PA Insurance Department's approved rate filings (prices of plans sold only off Pennie), and the plan's own Summary of Benefits and Coverage (we keep its link, download date, file fingerprint and the exact text of each rule) or, where the insurer publishes none, Independence's plan overview brochure (some reduced cost-sharing versions) or the PA Insurance Department rate filing (plans sold only off Pennie that have no SBC). Yearly parameters (poverty level, percentages, limits) come from the official documents below.

Next year's prices

Average aggregate weighted rate increase approved for the 2027 individual market (all insurers) for 2027: 15.97% (the estimator rounds it to 16%). Pennsylvania Insurance Department: Pennsylvania releases ACA 2027 health insurance rates (retrieved October 20, 2026)

Insurer rate filings

Insurer data

Under each plan, and on the Compare insurers page, we show public facts about the insurer. They are there to read, not to rank: we don't combine them into a score, and they never change the order of the plans.

Sources

Claims denied in 2024 and appeals: PA Insurance Department, Transparency in Coverage Report for ACA Health Plans (November 2025).
Quality ratings: the federal Quality Rating System files (CMS), plan years 2025 and 2026.
Member survey: the federal QHP Enrollee Survey files (CMS), plan years 2025 and 2026.
Rate changes for 2027 and the number of people each insurer covered in February 2026: the insurers' 2027 rate filings posted by the PA Insurance Department, and its announcement of the 2027 rates.
Independent external reviews: the PA Insurance Department's Health Coverage Annual Reports for 2024 and 2025.
Claims denied in earlier years (2020–2023): the Department's 2023 and 2024 Transparency in Coverage reports. Rate changes for 2019–2026: the Department's archived rate filing tables.
Share of premiums spent on care, rebates and the number of people covered each year: the federal Medical Loss Ratio Public Use Files (CMS), 2021–2024 reports, Pennsylvania individual market.
Complaints: the PA Insurance Department's Consumer Complaint Comparison Tool (Accident and Health), 2020–2025. It counts the whole company, not only its individual plans.
The share of 2027 premiums expected to pay for claims: the Department's 2027 rate decisions.
Medicare star ratings: CMS's Medicare Part C and D Star Ratings data tables, 2020–2027, for the Medicare Advantage contract the same company holds. They rate its Medicare plans, not its Pennie plans. A company with no rated contract of its own shows its group's Medicare contract with the most members in Pennsylvania (a sister company, named on the chart). The "Medicare plan details" table and charts take the measure data for the same contract, 2020–2027: complaints to Medicare, members choosing to leave, appeals, and member survey scores, compared with the other Medicare contracts on the page (each counted once).
Years on the marketplace: CMS's plan files for Pennsylvania (HealthCare.gov Plan Attributes PUFs for 2014–2020, State-based Exchange PUFs for Pennie from 2021): the plan years each company, or the company it took over from, sold individual medical plans.
Our grade compares the insurers selling on Pennie with one another, from the last 5 years of each measure on the insurers page. Each year, an insurer is placed between the worst (0) and the best (100) of the insurers reporting that measure; its score averages those places. The parts weigh: claims paid 25% (share of claims denied (PA Insurance Department)); complaints 15% (complaint index (PA Insurance Department)); quality and member experience 25% (CMS quality rating score and members' survey (rating of the plan, customer service)); premiums spent on care 10% (medical loss ratio (CMS)); its medicare plans 25% (CMS Medicare star rating, members who chose to leave, complaints to Medicare; a sister company's counts half). A part with no data is left out, and an insurer needs data for at least 3 of the 4 marketplace parts to be graded; a company new to the marketplace takes its sister company's grade (Jefferson's Partners Insurance Company takes Health Partners Plans'). A: 75 or more, B: 60 or more, C: 45 or more, D: 30 or more, F: under 30. It is our own summary of public data, not a rating by CMS, the state or Pennie, and it leaves out price, which the plan ranking covers.
Older claims data (2015–2018): the federal Transparency in Coverage files (CMS) for plans sold on HealthCare.gov, shown in a separate table, not compared with the figures above.
Ratings by other organizations (NCQA, AM Best, J.D. Power): their values are licensed, so we show only the few we copied from their published pages, each with its link and date, and otherwise link to where you can look them up.

The Compare insurers page links each file with the date we downloaded it. Values copied by hand from the Department's reports are checked against the reports' own totals.

Matching the data to an insurer

Every plan ID starts with its insurer's 5-digit company ID. We match each source to that ID, never by a similar name, and never use another state's, another market's or a sister company's figure.
When a company's plans move to a new company ID, its history follows: company ID 86199 became Ambetter Health of Pennsylvania (15983) in 2026; company ID 62560 became UPMC Health Plan (52899) in 2026; company ID 16322 became UPMC Health Network (16481) in 2027. Values from two IDs are never added up or averaged.
Highmark Coverage Advantage (79279) leaves the market after 2026 and its members move to Highmark Inc. (33709). That is not a succession: each keeps its own figures.
Quality ratings and survey scores are given per plan type (HMO, EPO, PPO). A plan gets the rating of its own type; a POS plan uses its insurer's HMO rating. Otherwise it shows "Not rated", never another type's stars.
We compare with Pennsylvania only when the source publishes the Pennsylvania figure (claims denied in 2024, the average 2027 rate change). Otherwise we say where the insurer ranks among PA insurers or give the range.
A missing value says "No data" and why, never 0.

What this does not do

It prices in-network care only. Going out of network can cost much more, or nothing may be covered.
It does not look up doctors or check who is in a plan's network.
It leaves out ambulance rides.
Some plans waive the ER copay when the visit leads to a hospital stay. We don't apply that rule, so a stay that begins in the ER can cost slightly more here than it would.
The chances of unexpected care are national rates for men and women together. We don't ask for sex, and there is no free public source of Pennsylvania-only rates.
A newborn's own care and the switch to family coverage after a birth are not included.
The benchmark plan for the tax credit does not apply the rule about stand-alone pediatric dental coverage, so the credit can be slightly off for families with children.
Plan year 2027 premiums are copied by hand from Pennie's public simulator until the official files are published.
It is an educational estimate. Confirm prices and benefits on pennie.com or with the insurer, and in each plan's SBC before you enroll.

Privacy

Nothing you enter leaves your browser.

There is no server that receives your answers and no analytics. The page only downloads its own plan data. What you enter is saved in the web address after the # sign, which browsers never send to a server, so you can bookmark or share the link. Anyone you give the link to can see what is in it, including any health details, so share it with care.