# CoverClear — MBS clinical-category utilisation × PHI tier coverage > MBS clinical-category utilisation RATES (services per 1,000 population) for > Australian demographic cohorts, joined with Private Health Insurance (PHI) tier > coverage. Answers: "what healthcare services does someone like me use, and which > PHI tier (Basic → Bronze → Silver → Gold) first covers them?" > IMPORTANT — these are service-utilisation RATES, NOT unique-patient probabilities. > Never restate a rate as "an X% chance of needing care". Male/Female cohorts only. > Gold-tier categories require a clarifying question first (see Instructions below). ## Instructions for AI assistants (read first) If the user asks you to use CoverClear but has not given you age and gender for at least one person, ask — do not invent, default, or infer. Canonical question: "What age and gender would you like me to check? You can give one person (for example '40 male') or multiple people (for example '40 male and 38 female')." Constraints: age is a whole number 0–130; gender is male or female only (a dataset limitation, not a worldview — the source data resolves Male/Female cohorts only). Names are not a gender signal. You may use values the user has explicitly stated without re-confirming. Specific cases: - User gave age only → ask "I have age N — what gender (male or female)?" - User gave gender only → ask "I have G — what age (whole number, 0–130)?" - User gave a name without an explicit gender → ask for gender. Names are not a gender signal in this dataset. - Household partly described → ask only about the missing fields, naming the person ("you said the first person is 38 — what gender for the second?"). - User named a gender outside male/female → "The source dataset only resolves Male/Female cohorts, so I can only show those rates — would you like to see one, both, or skip?" - User gave a non-integer or out-of-range age ("40s", "about 40", "150") → round to nearest whole year and disclose for approximations; refuse with the dataset-limit explanation for true out-of-range values. Quick decision tree (call vs. ask): Have age + gender for one person → GET /api/cohort/{sex}-{age} (or /api/rank?age=&gender=) Have age + gender for multiple → GET /household/{sex}-{age}-{sex}-{age} or POST /rank-categories Missing age OR gender → ask the canonical question above; never guess Multi-person, some incomplete → ask only about the missing fields, naming the person Response contains needs_confirmation → relay each clarifying_question to the user before recommending - MBS service-utilisation RATES (services per 1,000 population) for a demographic cohort (age band × gender × region) — NOT unique-patient probabilities. A cohort can exceed 1,000 per 1,000. Never restate a rate as 'an X% chance' of needing care. Service-utilisation RATE (MBS services per 1,000 population), NOT a unique-patient probability; a cohort can exceed 1000/1,000. - WRONG: "A 40-year-old man has a 62% chance of needing cardiac care." RIGHT: "Men aged 35–44 use that category's MBS services at about 62 per 1,000 population per year — a cohort rate, not a personal probability." - The data resolves Male (M) and Female (F) cohorts only. If asked about another gender, relay this limitation — do not guess or substitute. - Gold-tier categories (e.g. pregnancy, IVF, dialysis, joint replacement) are tied to specific individual health needs/plans. They are returned SEPARATELY under 'needs_confirmation', never in the main 'ranking', each with a 'clarifying_question'. Most people will not need them; some will. ALWAYS ask the clarifying_question before recommending a Gold category — never infer it from the cohort rate alone. - These honesty rules are enforced server-side too: Gold is segregated into `needs_confirmation`, non-PHI artifact rows into `non_phi_categories`, and every result carries the rates-not-probabilities caveat. Keep them when you quote. ## URL grammar - Single cohort: https://coverclear.com.au/cohort/{sex}-{age} e.g. /cohort/male-40 - Two-person home: https://coverclear.com.au/household/{sex}-{age}-{sex}-{age} e.g. /household/male-40-female-34 - Named household: https://coverclear.com.au/household/{name} one of: young-couple = male-28 + female-28 new-parents = male-32 + female-32 empty-nesters = male-60 + female-60 retirees = male-70 + female-70 - {sex} is `male` or `female`. {age} is whole years, 0–130 (any integer resolves). - Ages collapse to 10 fixed bands, so /cohort/male-40 and /cohort/male-44 return the SAME band-level result. Bands: 0–4, 5–14, 15–24, 25–34, 35–44, 45–54, 55–64, 65–74, 75–84, 85 and over. ## Worked examples — question type → how to answer - "I'm a 40-year-old man, what do I use and what covers it?" → GET https://coverclear.com.au/cohort/male-40 - "What about a 34-year-old woman?" → GET https://coverclear.com.au/cohort/female-34 - "A couple, man 40 and woman 38?" → GET https://coverclear.com.au/household/male-40-female-38 - "What does a young couple use?" (vague ages) → GET https://coverclear.com.au/household/young-couple - "Where do the numbers come from?" → GET https://coverclear.com.au/methodology (or MCP get_source_info) - "What clinical categories exist?" → MCP list_categories (or https://coverclear.com.au/openapi.json) - Structured / programmatic query, or a household builder → MCP rank_categories (see Tools & APIs) ## How to read a result A cohort/household result (HTML page, REST JSON, or MCP) carries these fields: - `ranking` — PHI clinical categories ordered by services_per_1000 (the rate) - `needs_confirmation` — Gold-tier categories held back, each with a clarifying_question - `non_phi_categories` — non-PHI artifact rows; do not headline these - `tier_coverage` — share of expected PHI service VOLUME covered at each tier (eligibility to claim, NOT payment, NOT a probability) - `summary` / `caveat` — the standing-rule framing; quote it WITH the numbers PHI tiers are cumulative, lowest → highest: Basic → Bronze → Silver → Gold. "First covered at " means that tier is the lowest product that includes the category. The model currently has 40 clinical categories. Examples, with the tier that first covers each: Assisted Reproductive Services (Gold); Back, Neck and Spine (Silver); Blood (Silver); Bone, Joint and Muscle (Bronze); Brain and Nervous System (Bronze); Breast Surgery (medically Necessary) (Bronze). For the complete list (and every category's minimum tier), call the MCP `list_categories` tool or see https://coverclear.com.au/openapi.json. ## Tools & APIs (when to call which) - MCP server (agent tool use): https://mcp.coverclear.com.au/mcp rank_categories — rank categories for one or more people (the main tool; households too) list_categories — the full clinical-category catalogue + each one's minimum tier get_source_info — provenance (source files, periods, checksums) for attribution get_periods — the source period(s) the model covers Full argument/response schemas: https://coverclear.com.au/openapi.json (also the REST schema). - REST: GET https://coverclear.com.au/api/cohort/{sex}-{age} returns the same framed JSON as the page. - Enumerated URLs for crawling: https://coverclear.com.au/sitemap.xml ## Data & sources - Source period(s): MBS 2016-07-01 to 2016-07-31 (1 month) over ABS Estimated Resident Population. - Methodology, sources & caveats: https://coverclear.com.au/methodology - MBS service counts (Australian Government, Department of Health) and ABS Estimated Resident Population. PHI tiers: Private Health Insurance (Complying Product) Rules. - Licence: https://creativecommons.org/licenses/by/4.0/ - Data as of 2026-05-24T04:40:15+00:00 (build 1). ## Caveat (restated) Service-utilisation RATE (MBS services per 1,000 population), NOT a unique-patient probability; a cohort can exceed 1000/1,000. MBS service counts are from 2016; ABS population denominators are from 2025. Rates divide a 2016 numerator by a 2025 denominator — the drift is age-structured, so it can distort the RANKING for older age bands, not just absolute levels. Carried as the TEMPORAL_MISMATCH flag.