The Health Care Workplace Indexes are two staff measures written for rural health care organizations: critical access hospitals, rural health clinics, and long term care. One asks whether people would recommend the organization as a place to work. The other asks what it is actually like to work there, across six areas that matter on a clinical floor. We wrote them because a twenty-five bed hospital should be able to measure its own workforce without paying to license somebody else's name.
Both staff indexes are built and running today in RuralHub, our employee intranet for rural health care. This page is the open specification. A hospital, a consultant, or another software vendor may implement the same measures under the license near the foot of the page. Two further indexes for patients are specified here and are not built.
Why we published this
A hospital can ask its staff these questions on paper for free. What costs money is the workflow around them: a survey window that opens and closes on schedule, anonymity that holds up on a two-person night shift, results that roll up an org tree, a way for a leader to reply to a comment without learning who wrote it, and history that does not disappear when somebody leaves.
Software that runs that workflow under a licensed third-party brand usually prices that license into the suite. A twenty-five bed hospital cannot carry that, and it should not have to. We wrote our own indexes, put them in RuralHub, and published the method here so that rural software can include these tools without that premium.
Built and shipping
The two indexes in this section are implemented in RuralHub today. The full specification for each follows: the question text, the grouping or dimensions, the formula, and a worked example a person can reproduce on paper.
Health Care Workplace Recommendation Index™ (WRI)
One question about intent. It is asked once per survey window and is never reworded between windows, because a trend across a changed question is two numbers on one line pretending to be one.
The question, exactly:
How likely are you to recommend this organization as a place to work?
The answer is a whole number from 0 to 10, presented as plain numbers. Each answer is placed in one of three groups.
| Answer | Group |
|---|---|
| 9 or 10 | Advocate |
| 7 or 8 | Neutral |
| 0 through 6 | Critic |
Formula
WRI = (advocates − critics) ÷ everybody who answered × 100
The result ranges from minus 100 to plus 100. Report it as a whole number, rounded to the nearest integer.
Neutrals count in the denominator and nowhere else. That is what makes the number move sharply on a small team, and a rural hospital needs to know it before reading a twelve point swing as a crisis. On forty respondents, one person moving from advocate to critic moves the index five points. One person moving from advocate to neutral moves it two and a half. A swing that would be noise at a two thousand bed system can be two people at a critical access hospital. Read the count of responses next to the index every time.
When nobody answered, the index is not reported. It is not reported as zero. Zero is a real result, and quite a different one: it means advocates and critics were evenly matched.
Worked example
A hospital closes its January window with forty responses.
| Group | Answers | Count | Share of 40 |
|---|---|---|---|
| Advocate | 9 or 10 | 18 | 45 percent |
| Neutral | 7 or 8 | 14 | 35 percent |
| Critic | 0 through 6 | 8 | 20 percent |
WRI = (18 − 8) ÷ 40 × 100 = 25
In July the same hospital again closes with forty responses. Two people who answered 9 in January answer 6 in July, and everybody else answers as before. Advocates are now 16, neutrals 14, critics 10.
WRI = (16 − 10) ÷ 40 × 100 = 15
The index fell ten points. Two people changed their minds. Both facts are true, and the second one is the one to act on.
Health Care Workplace Culture Index™ (WCI)
Six dimensions, two statements each, twelve statements in total. Every statement is answered on a five point agreement scale.
| Answer | Label |
|---|---|
| 1 | Strongly disagree |
| 2 | Disagree |
| 3 | Neither agree nor disagree |
| 4 | Agree |
| 5 | Strongly agree |
The six dimensions are asked in the order below. Each entry gives what a low score on that dimension actually means, followed by its two statements.
| Dimension | A low score means | Statements |
|---|---|---|
| Voice and safety | Somebody cannot raise a problem without it costing them. | I can raise a concern here without worrying about the consequences for me. When somebody reports a problem, something is done about it. |
| Staffing and tools | There are not enough people, or not enough working equipment, to do the job properly. | We have enough staff on my shift to do the work safely. I have the equipment and supplies I need to do my job. |
| Leadership trust | People do not believe what they are told, or do not believe it is decided honestly. | I trust the people who lead this organization to do the right thing. I am told about decisions that affect my work before they happen. |
| Teamwork and handoffs | Work does not move between people and shifts without falling on the floor. | My team works well together. Information I need is passed on to me at handoff. |
| Workload and schedule | The job is not sustainable at the pace it is actually run at. | My workload is manageable most weeks. My schedule lets me have a life outside work. |
| Respect and belonging | People are not treated decently, by each other or by the organization. | I am treated with respect at work. I feel like I belong here. |
Why these six
These dimensions were chosen for rural clinical work, not for an office. A generic engagement list asks about career growth and recognition programs. A critical access hospital needs to know whether the night shift is safely staffed. Staffing and tools is first among equals here. The pair about speaking up is the one that predicts whether anything else on the list ever gets reported; if people will not raise a concern, every other dimension is measuring what they are willing to say rather than what is happening.
Scoring
Each five point answer maps in a straight line to a score from 0 to 100. An answer of 1 becomes 0. An answer of 3 becomes 50. An answer of 5 becomes 100.
statement score = (answer − 1) ÷ 4 × 100
A statement's score for a group is the average of the mapped answers from that group. A dimension's score is the average of its two statements. The WCI is the average of the six dimension scores. Report the WCI and each dimension as whole numbers, rounded to the nearest integer.
The obvious alternative is to multiply the answer by 20. That is the most common mistake in this kind of instrument, and it is worth writing down why. Multiplying by 20 makes the lowest possible answer score 20. No organization can ever be below 20, and every result is flattered by the same invisible amount. Under this specification the floor is 0, so a bad result looks bad.
A dimension nobody was asked about reports as unmeasured, never as zero. Averaging a zero into the index would drag the whole number down by a sixth because somebody removed a question. If a window asks five dimensions, the WCI for that window is the average of five.
Worked example
The same forty people answer the first Staffing and tools statement, "We have enough staff on my shift to do the work safely."
| Answer | Mapped score | People | Points |
|---|---|---|---|
| 1 | 0 | 4 | 0 |
| 2 | 25 | 12 | 300 |
| 3 | 50 | 14 | 700 |
| 4 | 75 | 8 | 600 |
| 5 | 100 | 2 | 200 |
statement score = 1,800 ÷ 40 = 45
The second Staffing and tools statement scores 35 by the same method. The dimension is the average of the two.
Staffing and tools = (45 + 35) ÷ 2 = 40
The other five dimensions are scored the same way. Suppose the window closes with these results.
| Dimension | Score |
|---|---|
| Voice and safety | 62 |
| Staffing and tools | 40 |
| Leadership trust | 58 |
| Teamwork and handoffs | 72 |
| Workload and schedule | 50 |
| Respect and belonging | 78 |
WCI = (62 + 40 + 58 + 72 + 50 + 78) ÷ 6 = 360 ÷ 6 = 60
The headline is 60. The number to act on is 40.
The two indexes are never derived from each other
This is a design commitment, not a convenience. The WRI and the WCI measure different things, and the entire point of having both is that they can disagree. A workforce that would recommend the place and reports poor staffing is telling you something specific and actionable. One blended score hides it. A product that guessed the second number from the first would be inventing detail it does not have.
An implementation of this specification computes the WRI only from answers to the WRI question, and the WCI only from answers to the twelve culture statements. Neither number is an input to the other.
Specified, not built
Status: specified on this page, not built, no date.
Two indexes for people who receive care are specified here. Neither is implemented in RuralHub, and there is no date for either. They are not available and should not be read as a feature.
Patient Recommendation Index (PRI)
Whether somebody would recommend the organization for care. This is the relationship question: does the town still trust the hospital.
Visit Recommendation Index (TRI)
Whether somebody would recommend the organization based on one visit. This is the encounter question.
The two are kept apart on purpose. A hard night in the emergency department is not the same thing as whether the town still trusts the hospital, and a single index that blended them would mistake one for the other.
What they need before they can be built
- A way to reach people who do not have an account on the intranet.
- A feed of visits to attach a question to.
- An agreement covering the handling of what comes back.
None of that is required for the two staff indexes, which is why those two are not waiting on these.
Protections that are part of the specification
The items in this section are not fine print. They are part of the specification. An implementer who skips them has not implemented it.
Anonymity is enforced by not capturing, never by hiding
An anonymous response stores no respondent identifier at all. There is no hidden column, no hashed lookup, and no audit table that could be joined back. There is nothing to find later, because nothing was written.
A confidential response is different, and the difference is stated to the respondent before they answer. It stores the person so that a leader can reply to what somebody wrote without being told who wrote it. No read path anywhere in the system returns that identifier to a leader, a report, or an export.
A reporting floor
Results for any group are withheld until enough people have answered. The specification requires a floor; the organization sets the number. When a group is below the floor, its answers roll up to the next level of the org tree and are reported there.
When a small department cannot see its own numbers, that is the protection working, not a fault. A department of three that could see its own results would identify its own critic by subtraction.
Response rates work even on an anonymous window
An organization needs to know how many people answered, and it needs that count even when it cannot know who said what. The specification handles this by keeping the fact that somebody answered as a record separate from the response itself, and by keeping that record to the day and never to the moment.
The reason is precise. If the completion record carried a timestamp, it could be lined up against the timestamps on the responses to work out who wrote which one. A completion recorded only as a date, with no time, cannot be lined up against anything. The organization gets its response rate. The respondent keeps their anonymity.
The question text is frozen for the life of a trend
The WRI question and the twelve WCI statements are not reworded between windows. Changing the wording breaks comparability, and a trend line across changed wording is not a trend. An implementation that needs different wording starts a new trend and says so.
Cadence
The recommended default is twice a year: a window opening on the first of January and another on the first of July, each open for two weeks.
Quarterly is more than a small organization can keep up with. Four windows a year is four rounds of chasing people to answer and four sets of results somebody has to read and act on. A program nobody reads is worse than a shorter one somebody acts on.
Twice a year rather than annually because two points make a direction and one point makes nothing. A hospital that surveys once a year learns where it stands. A hospital that surveys twice a year learns which way it is moving, and can see by July whether what it did after January made any difference.
Open specification and license
This specification, meaning the question text, the dimensions and their statements, the grouping and scoring methods, the protections, and the cadence guidance on this page, is published under the Creative Commons Attribution 4.0 International License.
We chose that license over a custom one for the method because counsel at a hospital or a vendor already knows what it says, and because it expressly does not grant trademark rights. The names are handled separately, below.
Anybody may implement these indexes in their own product, whether commercial or not, and may reproduce the question text and statements to do it. Attribution to visuaFUSION is asked for. A suitable form:
Health Care Workplace Indexes specification by visuaFUSION LLC, visuafusion.com/Resources/HealthCareWorkplaceIndexes, licensed under CC BY 4.0.
Using the names
Health Care Workplace Recommendation Index™, Health Care Workplace Culture Index™, and the abbreviations WRI and WCI are trademarks of visuaFUSION LLC. They are not covered by the Creative Commons license. We want them used anyway. A score called a WRI in one product and a WRI in another is a standard. A score called something different in every product is a set of numbers nobody can compare.
The names are therefore licensed free of charge, to anybody, under the Health Care Workplace Indexes Trademark License. The license is accepted by using the names. In short:
- Royalty-free, for software vendors, health care organizations, consultants, and anyone else.
- The product must implement the specification in full, including the protections and the frozen question text. A product that omits or changes a required element may still use the method under CC BY 4.0. It may not call its scores WRI or WCI or describe itself using the names.
- The names may be used to label scores and to state that a product implements this specification. They may not be used in a product name, a company name, or a domain name, and a product may not suggest that it is visuaFUSION's or endorsed by visuaFUSION.
- The names carry the trademark symbol on first or prominent use, with a short trademark notice in a place a reader can find. The exact notice is in the license.
- The license passes downstream. A hospital using a conforming vendor's product may use the names without an agreement of its own with visuaFUSION.
- When this specification is revised, an implementer has twelve months from the publication date of the new version to conform to it. The existing implementation stays licensed in the meantime.
One request that is not a license term. When you describe the indexes, describe them by their names and by reference to this specification, not by comparison to any other measurement product. The names only work as a standard if they are used on their own.
The method is open. The names are ours, and they are free to use on a full implementation.
Revision history
The version number of this specification is the anchor for the conformance window in the trademark license. Every change to a question, a statement, a formula, or a protection produces a new version and a line here.
- 26.9.14.1, published September 14, 2026. Initial publication.
Legal notice
visuaFUSION LLC is not affiliated with, sponsored by, or endorsed by any other measurement vendor, and no license to any third-party trademark or service mark is claimed or implied by this page or by any product implementing this specification.
The Health Care Workplace Indexes are internal workforce measures. They are not a substitute for HCAHPS or for any federal patient experience reporting requirement, and they are not a clinical quality measure.
Health Care Workplace Recommendation Index™ and Health Care Workplace Culture Index™ are trademarks of visuaFUSION LLC. RuralHub is a product of visuaFUSION LLC.
Questions about the specification, or about implementing it, can be sent to info@visuafusion.com. To see the two built indexes in a product, see RuralHub.