> ## Documentation Index > Fetch the complete documentation index at: https://getprescience.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Account data > Read aggregate employer account data and enrollment roster data. Once a group is enrolled, two read-only endpoints return employer account and roster data: * **`GET /groups/{groupId}/account`**: funding, premium-equivalent totals, spend, and savings. * **`GET /groups/{groupId}/members`**: the enrollment roster of who's invited, active, and on COBRA. **Aggregate and de-identified by design.** The account payload never contains named claims, diagnoses, or any individual's health data, and spend categories are suppressed (k-anonymity) when a category could be traced to a small number of members. You can't accidentally render PHI, because the API never returns it. See [Compliance](/resources/compliance). Both endpoints return `404 not_found` until the group is enrolled. Pre-enrollment, use [`GET /census`](/guides/census#reading-the-census-back) for roster display. ## The account payload, in force Acme Inc, two months after their 2026-09-01 start, activated to `prod`: ```json theme={null} { "groupId": "grp_8c2f41d09a3e", "asOf": "2026-11-03T16:20:00Z", "company": { "displayName": "Acme Inc", "state": "prod", "startDate": "2026-09-01", "renewalDate": "2027-09-01" }, "plan": { "name": "Prescience Diamond", "deductibleIndividualCents": 150000, "deductibleFamilyCents": 300000, "oopMaxIndividualCents": 705000, "oopMaxFamilyCents": 1410000, "employeePremiumCents": 0 }, "population": { "coveredLives": 19, "enrolledEmployees": 12, "dependents": 7 }, "funding": { "fundBalanceCents": 990144, "monthlyFundingCents": 660096, "upcomingPulls": [ { "date": "2026-12-01", "amountCents": 660096, "status": "scheduled" } ], "recentPulls": [ { "date": "2026-11-01", "amountCents": 660096, "status": "completed" }, { "date": "2026-10-01", "amountCents": 660096, "status": "completed" } ] }, "premiums": { "monthlyPremiumEquivalentCents": 660096, "pepmCents": 55008, "employeeContributionCents": 0, "breakdown": [ { "label": "Claims paid", "cents": 408540 }, { "label": "HSA funding", "cents": 132019 }, { "label": "Risk protection & reserves", "cents": 119537 }, { "label": "Admin fees", "cents": 0 } ] }, "spend": { "mtdCents": 23418, "ytdCents": 802340, "categories": [ { "category": "Prescriptions", "cents": 240702, "pct": 30 }, { "category": "Specialty care", "cents": 192562, "pct": 24 }, { "category": "Primary care", "cents": 168491, "pct": 21 }, { "category": "Imaging & labs", "cents": 120351, "pct": 15 }, { "category": "Behavioral health", "cents": 80234, "pct": 10 } ], "suppressed": false }, "savings": { "baselineAnnualCents": 11520000, "savingsYtdCents": 599808 } } ``` Note the internal consistency you can assert on: `funding.monthlyFundingCents` equals `premiums.monthlyPremiumEquivalentCents`, the `breakdown` sums to it exactly, and `savings.savingsYtdCents` is derived from the active baseline comparison. ## Field-by-field UI mapping | UI element | Field(s) | Display guidance | | ----------------------------- | ----------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------- | | Tab header / plan badge | `plan.name`, `company.state` | "Prescience Diamond". Show an "Onboarding" badge while `state` is `sandbox`. | | Plan dates | `company.startDate`, `company.renewalDate` | "Plan year Sep 1, 2026 – Sep 1, 2027". | | Covered population | `population.*` | "19 covered lives: 12 employees + 7 dependents". | | **Funding balance** | `funding.fundBalanceCents` | Headline number: "\$9,901.44 in plan fund". Sub-label months-on-hand: balance ÷ `monthlyFundingCents` → "≈1.5 months". | | **Upcoming pulls** | `funding.upcomingPulls[]` | "Next pull: \$6,600.96 on Dec 1" with `status` chip (`scheduled` / `completed`). `recentPulls` is the history table. | | **Monthly cost** | `premiums.monthlyPremiumEquivalentCents`, `pepmCents` | "\$6,600.96/month · \$550.08 per employee". | | Employee premium contribution | `premiums.employeeContributionCents` | Integer cents. | | Premium breakdown | `premiums.breakdown[]` | Label/value rows or a chart. Preserve zero-value rows if your UI shows the full breakdown. | | Spend summary | `spend.mtdCents`, `spend.ytdCents` | "Claims spend: \$234.18 this month · \$8,023.40 plan-year-to-date". | | Spend by category | `spend.categories[]` | Bar list with `pct`. Sorted by size already. | | **Suppressed state** | `spend.suppressed`, `spend.suppressionNote` | See below; this state must be designed, not ignored. | | **Savings** | `savings.*` | `baselineAnnualCents` is prior-plan cost projected over the plan year; `savingsYtdCents` is current plan-year savings against that baseline. | All money is integer cents. Format with your standard currency utility, never floats. ## The suppressed-spend case For small groups (or early plan years), category-level spend could let an employer guess at an individual's care. When the k-anonymity check trips, the API withholds categories rather than thin them: ```json theme={null} { "spend": { "mtdCents": 23418, "ytdCents": 802340, "categories": [], "suppressed": true, "suppressionNote": "Spend categories are hidden for small groups to protect member privacy." } } ``` Render the totals (always safe), replace the category chart with the `suppressionNote`, and treat it as a normal state, not an error or an empty-data bug. Categories reappear automatically once the population supports them. ## The roster `GET /groups/{groupId}/members` powers the people table, with enrollment-admin data only: ```json theme={null} { "members": [ { "memberId": "mem_c4a91f27e83d", "externalId": "emp_0001", "firstName": "Jordan", "lastName": "Reyes", "email": "jordan@acme.com", "status": "active", "employmentType": "full_time", "hireDate": "2024-03-01", "dependents": 0, "enrollment": { "invited": true, "accountCreated": true } }, { "memberId": "mem_f7b3d59e02a4", "externalId": "emp_0008", "firstName": "Grace", "lastName": "Okafor", "email": "grace@acme.com", "status": "active", "employmentType": "full_time", "hireDate": "2023-07-15", "dependents": 1, "enrollment": { "invited": true, "accountCreated": false } }, { "memberId": "mem_92e4b7a1d0c6", "externalId": "emp_0009", "firstName": "Tom", "lastName": "Becker", "email": "tom@acme.com", "status": "cobra", "employmentType": "full_time", "hireDate": "2022-01-10", "dependents": 1, "enrollment": { "invited": true, "accountCreated": true }, "cobra": { "until": "2028-05-31" } } ], "count": 12 } ``` (Array truncated here; the API returns all members.) | Column | Field | Notes | | ------------ | ------------------------------------------------- | ------------------------------------------------------------------------------------------------ | | Name / email | `firstName`, `lastName`, `email` | Join to your employee record via `externalId`. | | Status | `status` | `active` / `terminated` / `cobra`; show COBRA with `cobra.until` ("COBRA through May 31, 2028"). | | Dependents | `dependents` | A count, deliberately; no dependent details post-enrollment. | | App status | `enrollment.invited`, `enrollment.accountCreated` | The useful HR nudge: invited-but-no-account members are who to ping during open enrollment. | `dependents` is a count and `enrollment` is two booleans on purpose: the roster tells the employer who is covered and onboarded, never anything about anyone's health. ## Freshness `asOf` timestamps every account read. The data behind it updates as pulls settle and claims process. Cache for minutes, not days; on-load fetching with a short server-side cache is the pattern our own portal uses. Both endpoints fall under the standard [120 req/min rate limit](/resources/rate-limits).