450-JCI Library of Measures Consultant Practicum Sept 2011 - Copy

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Library of Measures and Data Validation as

Required by Joint Commission International

Zakaria Zaki Al Attal

PhD, CPHQ

JCI consultant

[email protected]

[email protected]

00971558818777

1

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

What is the JCI Library of Measures?

The Library of Measures consists of a list of10

disease specific population groups identified as

measure sets.

Each measure set consists of at least 2 to 8

process and/or outcome measures. A total of 36

2

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

International Library of Measures-

Measure Sets

1) Acute Myocardial Infarction (AMI) 6

2) Heart Failure (HF) 3

3) Stroke (STK) 4

4) Children’s Asthma Care (CAC) 2

5) Hospital-Based Inpatient Psychiatric Service (HBIPS) 2

6) Nursing-Sensitive Care (NSC) 3

7) Perinatal Care (PC) 3

8) Pneumonia (PN) 3

9) Surgical Care Improvement Project (SCIP) 8

10)Venous Thromboembolism (VTE) 2

3

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

4

122, 39%

191, 61%

JCI Accredited Hospitals (313) Respond to Library of Measures Survey

Survey Completed Survey Not Completed

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

5

46, 6% 37, 5%

37, 5%

70, 9%

44, 6%

75, 10%

439, 59%

International Hospitals Selections of Library Measures (36 measures) Top 6 Measures Selected to Date

I-AMI-1 ASA on arrival

I-AMI-2 ASA on discharge

NSC-2 Pressure ulcers

I-NSC-4 All falls

I-NSC-5 Falls with injuries

I-SCIP-1 Antibiotics within 1 hr.

Other

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Measurement Requirements in Standards

QPS.3.1 Standard, Measurable Element (ME) 2,

requires JCI accredited hospitals to select at least 5 of 36

measures from the Library of Measures.

Hospitals may select all 5 measures from one measure

set, or a total of 5 measures from different measure sets.

6

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Additional Information and Resources

Identified resources developed now include:

– Sampling methodology

– Measure calculation

– Validation methodology

– Data abstraction tools

– Data element dictionary

– Initial eligible population criteria- ICD codes, or

diagnosis or clinical description

7

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Library Measure Selection and Data

Abstraction

8

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Clinical Practice Guidelines

Drive the Library of Measures

QPS.2.1 Clinical practice guidelines, clinical

pathways, and/or clinical protocols are used to guide

clinical care.

This standard addresses the creation and adoption of

guidelines, pathways or protocols, and their use in a

leadership driven prioritizing process.

9

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Example: Prevention of Surgical Site Infections

Guidelines Antimicrobial Prophylaxis

Recommendations of Antimicrobial Prophylaxis

‒ Administer prophylactic antimicrobial agents only

when indicated, and select in accordance with

published recommendations as delineated in

national guidelines

‒ Administer by the intravenous route the initial dose

of prophylactic antimicrobial agent

o Prophylactic antibiotic should be received within one hour

prior to surgical incision

10

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Library of Measures: Measure Information

I-SCIP-Inf-1d

Measure Name ‒Prophylactic antibiotics received within one hour prior to

surgical incision for Hip Arthroplasty patients

Rationale

–A goal of prophylaxis with antibiotics is to establish

bactericidal tissue and serum levels at the time of skin incision

Numerator

–Number of surgical patients (hip arthroplasty) with

prophylactic antibiotics initiated within one hour prior to

surgical incision

Denominator

– All selected surgical patients (hip arthroplasty) with no

evidence of prior infection and who are > = 18 years.

11

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

ICD Principal

Procedure for Hip

Arthroplasty

ICD

Principal Diagnosis

of Infection

Other

Surgeries

Surgical

Incision Date

YES

NO

Valid Date

Case NOT in the

I-SCIP-Inf-1d

Initial Population

Case not in Measure

Population-Excluded (B)

NO

YES

YES

Unable to

determine

Run inpatient cases with a Principal Procedure

Code or Principal Procedure of Hip Arthroplasty

on Appendix A, Table 5.04

ICD Principal Diagnosis Code or Principal

Diagnosis on Appendix A, Table 5.09, Infections

Documentation that “Other procedures”

requiring general or spinal/epidural anesthesia

that occurred within three days prior to or after

the principal procedure during this hospital stay.

Documented Principal Procedure Surgical Date:

dd/mm/yyyy

I-SCIP-Inf-1d Prophylactic Antibiotic Received Within One Hour Prior to

Surgical Incision - Hip Arthroplasty

Case Not in Measure

Population-Excluded (B)

Infection Prior

to Anesthesia

Case not in Measure

Population-Excluded (B) YESCheck if there is documentation that the patient

had an infection prior to the Principal Procedure

Patient Age (in years ) =

Admission Date – Birthdate

Run case for patients = >18 years old

Case NOT in the

I-SCIP-Inf-1d

Initial Population

= >18 years

Patient Age <18 years

NO

NO

Case Failed Measure

and is in the

Measure Population (D)

START

I-SCIP-

Inf-1d

K

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Surgical

Incision Time

Valid time

Unable to

determine

Documented Principal Procedure Surgical

Incision Time

Hours and minutes

Case Failed Measure

and is in the

Measure Population (D)

Antibiotic

Timing 1

Received Within 1 hour

Prior to Surgical

Incision Time

Check if the documented antibiotic was

received within one hour (or 2 hours if

Vancomycin, Appendix C, Table 3.8) prior to

Surgical Incision Time: Answer Yes

YES

= > 0 minutes and

=< 60 minutes

for at least

one antibiotic

Case Failed Measure

and is in the

Measure Population (D)

NO

<0 or >60 minutes for ALL antibiotic doses

Case Met Measure

and is in the

Numerator Population (E)

Antibiotic

Administration

Date

Antibiotic

Administration

Time

Documentation of the date the patient

received intravenous antibiotics (IVAB)

closest to and before the principal procedure

incision time

dd/mm/yyyy

Documentation of the time the patient

received intravenous antibiotics (IVAB)

closest to and before the prinicipal procedure

incision time

Hour and minutes

Case Failed Measure

and is in the

Measure Population (D)

Case Failed Measure

and is in the

Measure Population (D)

Unable to

determine

Unable to

determine

Valid date

Valid Time

Antibiotic Name

YES

Case Failed Measure

and is in the

Measure Population (D)

NO

Documentation that the intravenous antibiotic

that the patient received perioperatively was

on the Appendix C, Antibiotic Medications

Table 2.1

Antibiotic Timing 1 = Surgical Incision Date and

Surgical Incision Time –(minus) Antibiotic

Administration Date and Antibiotic Administration Time

I-SCIP-

Inf-1d

K

STOP

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Most Frequently Asked Measurement

Questions What if the Library Measures are not applicable to

the hospital’s clinical service groups/specialties?

– the majority of JCI accredited hospitals

provide clinical services to the measures’

specific population groups of patients

• If you are a specialty hospital and/or need additional

help with measure selection you may contact

[email protected] for assistance.

• If an exception to QPS.3.1, ME.2 is granted, the

survey team will be notified.

– The organization granted an authorized exception

from a Library measure, is NOT exempt from

selecting and gathering data for the relevant QPS.3.1

clinical measures.

14

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Most Frequently Asked Measurement

Questions(continued)

How should a hospital approach selection of measures

from the Library?

– The hospital’s leaders should identify targeted

areas for measurement and improvement based on:

• The hospital’s clinical service areas or patient populations

served,

• high volume patient populations ( diagnoses or

procedures),

• high utilization of resources,

• high risk patients(neonatal, diabetic, etc) and/or

• problematic or newly implemented patient care process.

15

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Most Frequently Asked Measurement

Questions (continued)

How does a hospital identify the measures’ initial

eligible patient population if the hospital does not

have a coding system?

– A hospital abstractor should strive to identify

the measure’s quarterly discharge medical

records using a documented diagnosis or

procedure description.

• Descriptions are located on the code tables next to

the code included in each measure’s initial

population criteria.

• If your hospital uses a different description, contact

[email protected] for assistance..

16

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Most Frequently Asked Measurement

Questions (continued)

What if we have problems or make mistakes with

data abstraction or validation? Hospitals should

strive for data completeness with the understanding

there is an expected learning curve for all hospitals;

some hospitals may need more time and assistance

than others.

17

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Measure Overview Table and QPS Clinical

Areas (continued)

When hospitals determine which (5) measures to

select from the Library of Measures they may consider:

– relevant measures related to one or more of the

(11) “Clinical Areas” identified in the QPS intent

statement.

• Selecting a Library measure related to one or more of the

“Clinical Areas” may

– reduce an unnecessary additional data abstraction burden

for abstractors, since hospitals would be able to count the

related measure as both a Library measure and a “Clinical

Area” measure to meet the QPS.3.1 ME1 and ME2

requirements.

18

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Data Validation

What dose it mean?

Data validation is most important when:

A new measure is implemented

Data will be made public

A change has been made to

the existing measure

The data source has been changed

Introduction of new technology

or new process of care related to the issue of measures

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Just an example

Benchmark

" ومن أحياها فكأنما أحيا الناس جميعا"

Target

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Coming Soon:

Library Specifications, Version 2.0

Validation Table required to use as of January 2013

discharges and may use during the transition stage in

year 2012

21

Quarterly Number

of Medical Records

Originally

Abstracted

Validation Sampling

Requirement

180 records or

greater

At least 5%

<180 records At least 9 sampled

records

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Example of Random Sampling Using the

Version 2.0 Sample Table

Number of a measure’s abstracted “quarterly”

discharge medical records = 120 cases

Sampling Requirement at least 9 medical records

120/9 = 13

Sampling interval number = 13

Select starting point

Then, select every 13th medical record to be included

in the validation sample until you reach the 9 required

records

22

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Credible Data Validation Process Criteria

QPS.5

Re-collection of the original data by a second abstractor

Use a statistical valid sample number of records as defined

in the following table

Compare the original abstracted data with the re-collected

data

Calculating the accuracy by dividing the number of data

elements found to be the same by the total number of data

elements and multiplying by 100.

A 90% accuracy level is a good benchmark

Data elements found not to be the same (do not match with

the original results- take corrective action).

After corrective action implemented, take a new sample and

re- abstract data for accuracy

23

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Possible sources of the inaccuracy of

data

A change in the data collection tool

A change in the formula

A change in the definitions

A change in the source of data collection

Changing the individuals who are involved in the data

collection !

A change in the benchmarking definitions

" ومن أحياها فكأنما أحيا الناس جميعا"

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

What to do in case of the accuracy level

is less than 90%

When data elements are found not to be the same.

Noting the reasons (for example, unclear data definitions)

and taking corrective actions.

Collecting a new sample after all corrective actions

have been implemented to ensure the actions resulted in

the desire accuracy level.

" ومن أحياها فكأنما أحيا الناس جميعا"

© J

oin

t C

om

mis

sio

n Inte

rnatio

nal

Questions & Answers

26