Quality Intelligence
A readmission rate tells you the score. It tells you nothing about the game.
Knowing you are at 15.8% does not tell you which discharges are coming back, what those patients have in common, or which of the three things you could do about it would actually work.
Works from discharge and admission records you already produce. Penalty exposure quantified while the measurement year is still open.
The problem
The rate is reported. The pattern underneath it is not.
Every hospital tracks readmissions, and most track them well. The rate is calculated, compared to a benchmark, discussed at the quality committee, and reported upward. All of that is competent, and none of it identifies a cause.
Readmissions are not evenly distributed. They cluster — around a discharge destination, a specific pathway, a follow-up that does not happen within a particular window, a medication reconciliation step that gets skipped when the discharge is on a Friday. Those clusters are visible in data the organization already holds, and finding them means cutting discharge data by a dozen dimensions and looking for concentration.
That analysis takes an analyst several days, so it gets done once a year or when a penalty lands. In between, the organization runs readmission reduction programmes designed against last year's understanding of the problem, and is surprised when the rate does not move.
- The readmission rate is reported monthly and investigated annually
- Nobody can say which discharges are coming back beyond broad diagnosis groups
- Penalty exposure is understood after CMS confirms it, not while it can be influenced
- Readmission reduction effort is applied broadly because the concentration is unknown
- Follow-up compliance is tracked separately from the readmissions it is meant to prevent
- Risk scores are calculated at discharge and nobody checks whether they predicted anything
Capabilities
What Vizier surfaces in readmission data
Vizier evaluates discharges and returns continuously, and raises where readmissions concentrate — with what the affected group shares and what the exposure is worth.
Where returns concentrate
Which discharge cohorts are coming back, defined by more than diagnosis — destination, pathway, timing, follow-up status and the combinations of those.
What the returning group shares
The common factor across a cluster, which is usually a process step rather than a clinical characteristic and is therefore fixable.
The window that matters
Where in the post-discharge period returns happen, since early and late readmissions have different causes and different interventions.
Follow-up effectiveness
Whether follow-up is happening, whether it is happening inside the window that matters, and whether it is associated with fewer returns for that cohort.
Penalty exposure, early
Where measure performance is heading against the programme thresholds while the measurement period is still open.
Deterioration before the rate moves
A cohort or unit drifting while the organization-level rate still looks stable, which is the only point where prevention is cheap.
What a finding looks like
The organization's overall readmission rate had moved by half a point — noticeable, not alarming, and not obviously actionable.
Underneath it, one destination and two days of the week were producing a ten-point gap, driven by follow-up that was not being arranged before the weekend. Comparable acuity rules out a sicker-patient explanation. That is a scheduling fix with a penalty number attached to it.
Returns concentrate in discharges to one destination type, and only when discharge falls late in the week.
- 30-day readmission for this cohort is 24.1% against 14.6% for comparable discharges elsewhere.
- The gap appears only for discharges on Thursday and Friday; midweek discharges to the same destination track the baseline.
- Documented follow-up within seven days occurs for 38% of the late-week group and 79% of the midweek group.
- Clinical acuity at discharge is comparable across both groups.
Review how follow-up appointments are arranged for late-week discharges to this destination, and what changes in availability over the weekend.
The reporting gap
Why the rate hides the thing you can act on
A readmission rate is a single number describing a heterogeneous population. Nearly everything useful about readmissions is a property of subgroups, and a single number is the format guaranteed to remove subgroup structure.
- Organization-level rates absorb concentration in specific cohorts and pathways.
- Diagnosis grouping is too coarse — the cause is usually a process factor cutting across diagnoses.
- Monthly reporting is too slow to expose day-of-week and timing effects.
- Follow-up compliance and readmissions are tracked in separate reports and rarely joined.
- Penalty programmes are confirmed retrospectively, so exposure is known when it is fixed.
- Risk stratification output is rarely evaluated against what actually happened.
What this replaces
This replaces the annual readmission deep-dive
Most quality teams have done a readmission analysis. It typically takes weeks, produces a genuinely good understanding of last year's pattern, and informs a programme built on that understanding.
The pattern then changes — a destination's capacity shifts, a follow-up pathway is reorganised, a service line's discharge practice moves — and the programme keeps running against a problem shape that no longer exists.
Vizier does that analysis continuously and raises the clusters as they form, so intervention design tracks the current pattern rather than last year's.
- The annual readmission deep-dive, and the consulting spend when it is outsourced
- Analyst days spent cutting discharge data to find where returns concentrate
- Broad readmission reduction programmes applied where concentration was never established
- Waiting for CMS confirmation to understand penalty exposure
- Manual joining of follow-up compliance data to readmission outcomes
Who this is for
One cluster of returns. Three decisions it drives.
Population Health / Quality
Which discharges are coming back, and what do they share?
- Care gaps and rising-risk cohorts prioritized by impact
- Measure performance variation against peer benchmarks
- Programs ranked by what the evidence says will actually move outcomes
CFO / Finance
What is our penalty exposure, and which intervention reduces it most?
- Revenue leakage surfaced with the exposure quantified
- Reimbursement and payer performance movement, early
- Financial impact ranked so the biggest number gets attention first
COO / Operations
Which process step is producing the returns, and can we change it?
- Throughput and patient flow constraints identified by location and service line
- Performance variability between sites made visible rather than averaged away
- Operational deterioration flagged while it is still a trend, not a crisis
Getting your data in
Start from discharge records you already produce.
Readmission analysis needs less than most people expect — admissions and discharges with enough identifier continuity to link a return to its index admission.
- An admission and discharge extract covering at least twelve months.
- Discharge destination and follow-up scheduling data where recorded.
- Diagnosis and acuity coding, so clinical explanations can be tested and ruled out.
01
Upload
CSV, Excel, or an export you already produce. Drop it in and Vizier reads it. This is where most organizations start, and it is enough to see real findings against your own numbers.
02
Scheduled
A recurring feed over secure transfer, on whatever cadence your team already runs. No one re-uploads anything by hand, and nothing about your source systems has to change.
03
Connected
Direct read-only connectivity to your EHR or source systems via FHIR R4, HL7 v2, or vendor APIs. Vizier reads; it never writes back.
Connect your EHR when you’re ready. See supported systems.
Security and governance
The page your CIO will ask for
Security questions get answered before a demo, not after procurement stalls.
HIPAA compliant
PHI handled under HIPAA Security Rule safeguards.
BAA included
Executed within one business day, on every plan.
Encrypted throughout
AES-256 at rest, TLS 1.3 in transit.
Read-only access
Vizier reads from source systems. It never writes back.
Role-based access control
Scoped permissions with SSO available.
Audit logging
Every query logged with account, timestamp and result size.
Tenant isolation
Your data is segregated from every other customer's.
SOC 2 Type II audit underway
Not yet certified. Report available under NDA on completion.
FAQ
Questions buyers ask
Does this predict which patients will be readmitted?
It surfaces where returns concentrate and what those discharges share, which is a different and generally more useful thing. Individual risk prediction has a long history in this field and a mixed record — a score at discharge tells you a patient is high risk, which the clinician usually already knew, and does not tell you which process step is producing returns across forty patients. We would rather show you the pattern that is fixable than a score that is hard to act on.
How do you handle CMS readmission programme methodology?
Findings reflect the measure definitions your organization is held to, including the exclusions and windows that apply. Where our calculation and a payer's confirmed figure differ, the payer's is authoritative — the value here is knowing where performance is heading while the measurement period is open, not producing a competing official number.
Can it separate avoidable readmissions from unavoidable ones?
It can identify where returns concentrate around process factors rather than clinical ones — a follow-up that did not happen, a destination, a timing pattern — and test whether acuity explains the difference. Whether a specific readmission was avoidable is a clinical judgement and stays with your team. The finding above is the useful shape: comparable acuity, different follow-up rates, one destination, two days of the week.
We already have a readmission risk model. Does this conflict?
No, and they answer different questions. A risk model scores patients; this finds patterns across discharges. If anything the two work well together — Vizier can show whether your model's high-risk group actually readmitted more, which is a question surprisingly few organizations have checked and one that occasionally produces uncomfortable answers.
How far back do you need data?
Twelve months is a good starting point — enough to establish seasonal and day-of-week patterns and to give a return window room. Less will work for recent-trend detection but weakens the pattern analysis, which is the part that produces the actionable findings.
Related reading
See which discharges are actually coming back.
Bring twelve months of admission and discharge records. Thirty minutes, and the concentration is usually already visible.
Start with the data you already have. Connect your EHR when you’re ready.