Minnesota Hospital Association

Data & Reporting

Health care Associated Infections (HAI) Reporting

Under state law, hospitals must publicly report health care-associated infections.

The law references measures recommended by the National Quality Forum (NQF), and it charges the Minnesota Hospital Association and Stratis Health, an independent, nonprofit organization that leads collaboration and innovation in health care quality and safety, with the responsibility to publish the information in collaboration with infection-control practitioners.

NQF released its recommendations for seven measures in March 2008. Since then, MHA, Stratis Health and several hospital infection-control practitioners created recommended measures for reporting hospital-acquired infections in Minnesota. Those recommendations cover conditions such as appropriate hair removal for surgery patients and central-line associated bloodstream infection-prevention measures. The recommendations also address measures hospitals should not report.

Frequently asked questions for patient care staff

General Questions

Is this required for critical access hospitals?

Yes. The law applies to all hospitals.

If I have no cases, do I need to report?

Yes. We cannot assume that not reporting means you had zero cases.

What is the deadline for reporting?

The deadline is 45 days after the end of each quarter.

How will data be displayed?

There is a work group that is discussing these issues. We will likely be following similar policies and formats as data already displayed on mnhospitalquality.org.

Where will the data be publicly reported?

The results will be posted on the mnhospitalquality.org website after two quarters worth of data are collected.

Do we report SCIP Measures to MHA or will you get them from CMS?

MHA is not collecting the SCIP data at this time. The information is downloaded from HospitalCompare through CMS.

Will there be a way to upload the data from our internal software?

This will have to be developed. We cannot do this currently.

Central Line & Ventilator Bundles

Am I right that ventilator patients can be audited multiple times and that central-line patients can only be counted once?

Yes. The denominator for VAP is number of patients sampled — because the elements of the bundle must be done every day, you can sample the same patient multiple times, and each time counts as one in the denominator. For the central-line bundle, the denominator is also patients sampled, but you should survey a patient only once per admission — four out of the five elements are one-time only processes. Both of these bundles are for ICU patients only, and you should count them in the month you sampled.

If the central line is inserted in Radiology (or anywhere other than the ICU), should that be counted in the population?

No, the specifications from NQF are that it must be inserted in the ICU. All lines are important though, so any monitoring of lines placed outside the ICU is encouraged.

What is the definition of ICU?

The CDC definition of ICU/ICU-Status is: a nursing care area that provides intensive observation and diagnostic and therapeutic procedures for adults who are critically ill. This excludes bone-marrow transplant units and nursing areas that provide step-down care, intermediate care or telemetry only.

What is the standard for collection on central-line bundle — femoral or jugular?

The inserter should avoid the femoral line. If there is reason that the femoral line is used (documented or per hospital protocol), give it a pass (count as compliant).

For the optimal site: Does the MD documentation of the site in his report constitute the optimal site or does the doctor need to explain why that site was chosen? At present we don't have a policy or guidelines for what is the optimal site.

If they choose femoral they should be asked to write why. Sometimes there is an obvious reason and if it is not obvious, the MD should be asked what the reason was. If he/she has one, then it should be a pass.

Should hand hygiene be both before and after insertion of the central line or just before?

It is good practice to do both, but the standard is only before.

When auditing for daily assessment, should we check for documentation every day or just for the day we audit?

Use only the day you audit. It is a snapshot of the practice of insertion elements and one day of daily assessment. Use a patient only once. All days are important though, so any monitoring beyond this required data collection is encouraged.

Can we use administrative codes to find central-line patients to audit?

There are no administrative codes to identify someone with a central line which is placed in the ICU.

What is a “large drape?”

IHI defines a large drape as, “a cover from head to toe.”

The instructions say a central-line patient should only be audited once. What if the patient was: a) discharged and readmitted and a new central line started, or, b) taken off a central line and put back on a central line later in the same admission?

In both scenarios, the old central line was discontinued and subsequently a new one was started. These can be counted twice. A more correct statement would be that a patient should be counted once per continuous time on a central line.

Some patients are put on central lines for long-term IV therapy — doing daily assessment on these patients seems pointless. Should we be excluding these patients?

IHI says this in their How-to Guide, October 2008: “Define an appropriate timeframe for regular review of necessity, such as weekly, when central lines are placed for long-term use (e.g., chemotherapy, extended antibiotic administration, etc.). Daily review was designed for the intensive care population and may not be appropriate when long-term use over weeks or months is planned.”

Therefore, you should exclude patients intended to have long-term use of central lines.

Are there exclusions for VAP for kinetic beds?

If someone is on a kinetic bed, give it a pass (count as compliant) for that element.

If my facility has a small volume of ventilator days, should I report on every day the patient is on the ventilator?

Yes. You can count the same patient. You should report at least 15 day audits or as many as you have (e.g. if you had one patient vented for 9 days you would report audits for 9 days).

Some of the elements of the ventilator bundle may be not applicable for medical reasons, but the data collection tool doesn’t let me change the denominator for the individual elements. What should I do?

If a bundle element is not applicable, it should be counted as met. Therefore, the denominator for individual elements will always be the same as the overall bundle volume. Remember to increase your numerator by the number of audits where the elements were not applicable.

Surgical Site Infections

Do I count 30 days (or 12 months) from the procedure date or the discharge date for SSIs?

You need to look for infections for 30 days after the PROCEDURE date for vaginal hysterectomies, and 12 months for total knee arthroplasties. The denominator should be the number of procedures in a given month, regardless of discharge date.

When entering data, where do I attribute the infection — the date it was identified or the date of the original procedure (exposure)?

When entering data, you should attribute the infection to the date of the original procedure (exposure).

If we have no SSIs, do we need to do the risk indexing?

Yes. In order to have statewide comparisons, it is necessary to know what the risk score was on every case.

How do I risk-adjust the surgeries?

The SSI Risk index is a score used to predict a surgical patient’s risk of acquiring a surgical site infection. This system was derived from the Scenic report published decades ago.

The risk index score, ranging from 0 to 3, is the number of risk factors present among the following:

  • a patient with an American Society of Anesthesiologists’ (ASA) physical status classification score of 3, 4, 5
  • operation classified as contaminated or dirty/infected, and
  • an operation lasting longer than the duration cut point hours for a specific surgical procedure.

Definitions of the risk index elements may be found in key terms section of the NHSN Manual which starts on page 75.

How often does risk indexing for SSI need to be done?


Is there sampling for SSI like there is with SCIP?

No. All cases must be monitored.

For TKAs, the look-back period is 12 months. Are we supposed to enter the data for surgeries that occurred starting from Jan. 1, 2008?

No. We are starting with surgeries that were performed on Jan. 1, 2009 and after.

If I report to NHSN, can I use that data for this reporting requirement?

No. Each NHSN operative procedure category consists of a group of ICD-9-CM codes. When monitoring a specific NHSN operative procedure category, all the ICD-9-CM codes within that category must be included. The public reporting indicators for Minnesota are subsets of the NHSN procedure category. *NHSN facilities that enter the ICD-9-CM codes into NHSN when entering procedure data can generate a report for the specified public reporting procedures.

If a TKA is done elsewhere and the infection is found at my facility, what should be done for public reporting?

Do not count the SSI in your rates if not done in your facility. We encourage you to notify the facility where the procedure was done

If a hysterectomy or elective total knee has additional surgical procedures done in the same OR episode, is that case excluded from the population?

The indicator procedure should be included.

In the statement “Infection is identified on operative admission, readmission to operative procedure facility or other facility,” what is the definition of facility — is it just a hospital or can it be a clinic also?

Facility refers to a setting where the patient is diagnosed with the infection. Clinics do count, but these generally require readmission.

What are the surgery duration times we should use for assessing risk?

We have listed a 2-hour cut-off time, but NHSN has newly published times (Nov 08 AJIC) for both surgical procedures – TKAs are 122 minutes and vaginal hysterectomies are 132 minutes.

NHSN combines some risk categories when it reports out its results. Should we report our risk categories combined or separate?

You should report numerators and denominators for each risk category separately. It is possible that when the data are displayed that some risk categories will be aggregated, but that will be done by MHA.