[PPT] - Hospital Metrics TAG April 11, 2017 PLEASE DO NOT PUT YOUR PHONE ON PowerPoint Presentation

SLIDE 1

Hospital Metrics TAG

April 11, 2017 PLEASE DO NOT PUT YOUR PHONE ON HOLD: IT IS BETTER TO HANG UP AND CALL BACK IN IF NEEDED

SLIDE 2

Welcome and Introductions

2

SLIDE 3

Agenda Overview

Updates
EDIE-sourced measure
CLABSI / CAUTI NHSN Rebase and QA on Internal

Validation

Voluntary ED Opioid Prescribing Metric (exclusions

discussion)

3

SLIDE 4

Updates

4

SLIDE 5

CCO Metrics & Scoring Committee Updates

Met on March 17th:

– Had presentations on alternative patient experience measures from the CAHPS survey and on dental metrics

Next meets on April 21st, to discuss:

– The mid-year CCO metrics report – Begin high level discussions to narrow down the list of potential 2018 CCO incentive measures

Committee meeting materials are available here:

http://www.oregon.gov/oha/analytics/Pages/Metrics- Scoring-Committee.aspx

5

SLIDE 6

Hospital Performance Metrics Advisory Committee

Met by phone on March 16, 2017

– Reviewed written public testimony on the voluntary opioid metric and the incentivized EDIE metric – Discussed Year 4 as approved by CMS – On hiatus until July (may meet earlier as needed). – Subject to consultation with the President of the Senate and the Speaker of the House, all committee members’ terms will be extended by one year – Minutes available on the Committee’s webpage: http://www.oregon.gov/oha/analytics/Pages/Hospital- Performance-Metrics.aspx

6

SLIDE 7

HTPP Year 4 Incentive Measures

Reminder that the Year 4 incentive measures

eligible for payment remain the same as in Year 3

– All-cause readmissions – Hypoglycemia with insulin – Excessive anticoagulation with warfarin – ADEs with opioids – CLABSI (switch to SIR) – CAUTI (switch to SIR) – Follow-up after hospitalization for mental illness – SBIRT – EDIE (switch to revisit measure) – HCAHPS – discharge instructions – HCAHPS – explain medication

Official Year 4 improvement targets from OHA will

be sent in August (allowing time for review and any corrections to Year 3 data

7

SLIDE 8

HTPP Year 4 Voluntary Measures – not eligible for payment / not required submissions

C-sections; opioid prescribing in the ED; c-difficile
Voluntarily submitted data will be used to assist in

benchmarking (and baseline) for possible future quality incentives

Hospitals choosing to report on these metrics will

do so directly to OHA (not via the Apprise reporting platform)

OHA will send out directions for reporting data in

the spring (OHA anticipates asking for two data submissions –

ne in approximately June, and another towards the end of the

Year 4 measurement period)

8

SLIDE 9

HTPP Year 3 Data Submission (1/2)

OHA has already received official submissions from

hospitals for the measures below.

OHA is reviewing, and may contact hospitals with

questions regarding the data submitted.

Hospitals must be able to provide documentation of data

submitted should it be requested.

9

Measure Data Source Readmissions Apprise Hypoglycemia with insulin ADEs with opioids Excessive anticoagulation with warfarin HCAHPS – discharge info HCAHPS – medication SBIRT EDIE-sourced measure CMT Follow-up after hospitalization for mental illness OHA (claims)

SLIDE 10

HTPP Year 3 Data Submission (2/2)

OHA is awaiting data for CLABSI and CAUTI.
After internal validation is complete, these data will be

pulled directly from NHSN by the state’s HAI Program and provided to the OHA Office of Health Analytics on April 28, 2017.

10

SLIDE 11

SBIRT – Counting Brief Interventions

At the March H-TAG office hour, hospitals on the call

requested that OHA consider allowing metric credit for brief interventions that are not conducted face-to-face (i.e., over the telephone).

OHA is still reviewing the literature and consulting with

billing and other experts before making a decision.

In the interim, hospitals should continue adhering to the

current measure specifications, for which brief interventions conducted via telephone are not eligible for metric credit.

11

SLIDE 12

Year 4 EDIE Measure

12

SLIDE 13

Year 4 EDIE Measure Discussion Overview

Review current Year 4 measure
Proposed specification changes to existing Year

4 metric

Proposed process to ensure patients who

LWOBS and AMA are excluded

Proposal for measure change (heard by

Committee)

13

SLIDE 14

Recap of Current Year 4 Measure (1/3)

Revised, outcome focused metric developed by

OHA in partnership with the OHLC, OAHHS, and with consultation from the H-TAG.

Aim to reduce ED revisits for patients frequently

treated at the same facility.

Measures the number of patients readmitted to

the ED within 30 days of their fifth visit to the same facility in 12 months.

14

SLIDE 15

Recap of Current Year 4 Measure (2/3)

Notes:

– Count of events, specifically the fifth visit in 12 months. – A patient on their 40th visit will not show up on the report – Intent is to focus on high utilizers before they become super utilizers

15

SLIDE 16

Recap of Current Year 4 Measure (3/3)

CMT uses discharge disposition “discharge to

home or self-care” to identify whether the visit is counted in the metric.

However, this field is sometimes blank, and

some hospitals do not share this field with CMT. If there is no discharge disposition, CMT:

1. Reviews the discharge date. If visit < 12 hours,

it is not counted.

2. If neither discharge disposition or date, CMT

looks for evidence of in-hospital transfer. Such visits are not counted.

16

SLIDE 17

Concerns About Exclusions Related to Current Measure – LWBS / AMA (1/2)

Some hospitals (mainly Epic) reported

inconsistencies between EDIE report and what is displayed in Epic for discharge disposition.

This can occasionally result in a Left Without

Being Seen (LWBS) or Left Against Medical Advice (AMA) coming through to CMT as a discharge to home/self-care

Potential fix: adding ED discharge disposition

field to the existing ADT field

17

SLIDE 18

Concerns About Exclusions Related to Current Measure (LWBS / AMA 2/2)

If a hospital has concerns that this is an issue,

the hospital can send CMT files with such patient visits identified, and CMT will exclude them.

This would happen twice: once in Spring 2017

(this month) for use in baseline, and in early 2018 (for final 2017 results).

OHA will send instructions in late April.

18

SLIDE 19

Concerns About Exclusions Related to Current Measure (inpatients – 1/2)

Some hospitals have reported visits that led to an

inpatient stay being included in an EDIE report.

This is generally explained by a missing

message/ADT data point from the hospital to CMT indicating that an inpatient admission or transfer

ccurred.
One way to address this broadly is to amend the

specifications and exclude any patients with an ED visit of >12 hours

19

SLIDE 20

Concerns About Exclusions Related to Current Measure (inpatients 2/2)

CMT has provided test data showing how the 12-

hour limit would impact hospitals in terms of aggregate counts (see packet materials)

Imperfect solution: Will result in some legitimate

ED patients with visits >12 hours being excluded from the denominator

Thoughts?

20

SLIDE 21

Public Testimony to Change Year 4 Metric

See full testimony in materials
Concern that hospitals that make efforts to target high

utilizers very early in the process may limit the number of patients who reach five visits in 12 months to those who are most difficult to serve.

As the denominator shrinks to the more difficult to serve

population, it may be harder for the hospital to achieve the benchmark.

Benchmark would remain the 90th percentile (of HTPP Year

2 performance on these revised specification)

21

SLIDE 22

Proposed EDIE Metric Shift

Emergency Department Information Exchange April 11, 2017 Justin Keller, Director of Network Operations Larry Finch, HTPP Report Writer

SLIDE 23

HTPP Measure – Potential Shift

Current measure is focused on actionable outcome of

reducing the high utilizer population:

– Focuses on just the 5th visit at the same facility for the denominator – Focuses on just the 6th “revisit” within 30 days of the denominator visit

Potential shift would focus on the proportion of a hospital’s

full patient volume that are “5 in 12” high utilizers

– Denominator would be all unique patients who visit the ED at least once during the performance year – Numerator would be all patients who visit 5 or more times during the performance year – Outstanding question: unclear from proposal whether it would be 5 or more times at the same facility or just any facility

SLIDE 24

Initial Data

Performance Year 2 and Year 3 data was pulled
Two sets of numbers: taken with and without

the criteria that the 5+ visits had to occur at the same facility

Data looks at all HTPP-participating hospitals

SLIDE 25

Data Highlights – Same Facility

Denominator range:
Numerator range:
Proportion range:

Year 2 Year 3 5,574 4,941 62,736 65,317 129 2,827 102 3,103 Year 2 Year 3 2.16% 6.07% 2.06% 6.4% Year 2 Year 3

SLIDE 26

Data Highlights – Any Facility

Denominator range:
Numerator range:
Proportion range:

Year 2 Year 3 5,574 4,941 62,736 65,317 672 5,890 600 6,235 Year 2 Year 3 4.82% 17.41% 5.78% 16.75% Year 2 Year 3

SLIDE 27

Things to Note

The time frame for determining high utilizers

is required to be set at the performance period: Jan 1, 2017 – Dec 31, 2017

– 12 month look back can no longer be rolling

Status reports will not be as meaningful until

closer to the end of the year

– 3-in-12 reports and 4-in-12 reports can continue but will still only alert on events, not total high utilizers

SLIDE 28

Things to Note (cont.)

Treats all high utilizers (5+ visits) as a similar

population

– Less guidance for hospitals on how to take action and address these populations—people are going to the ED frequently for varying reasons

There will be much greater variability in the

success hospitals have depending upon their size, geographic location, and population density

SLIDE 29

H-TAG Thoughts?

29

SLIDE 30

Re-baselining in NHSN

Roza Tammer, MPH, CIC HAI Reporting Epidemiologist, HAI Program Hospital Metrics Technical Advisory Group (H-TAG) Meeting April 11, 2017

SLIDE 31

Agenda

NHSN re-baselining
HTPP re-baselining for CLABSI and

CAUTI

Intersection of NHSN and HTPP re-

baselining

Internal validation Q&A
Reporting your quarterly SIR data

31

SLIDE 32

What is the SIR?

The standardized infection ratio (SIR) is a statistical measure

comparing observed and predicted healthcare-associated infections (HAI)

Used to track HAI at national, state, or facility level over time
Adjusts for the fact that each facility treats different patient

populations

Ratio of observed to predicted events
Observed HAI

– The number of infections observed in a facility and reported into NHSN during a certain time period

Predicted HAI

– The number calculated based on the national SIR baseline – Statistically adjusted for risk factors known to be statistically associated with incidence of infection

Observed Predicted SIR =

SLIDE 33

Interpreting the SIR

SIR = 1: Number of observed infections is the same as the number
f predicted infections
SIR < 1: Fewer infections observed than predicted
SIR > 1: More infections observed than predicted
Statistical significance assessed using the 95% confidence interval

33

SLIDE 34

34

National SIR Baselines

HAI Measure Original Referent Period CLABSI Jan 2006 – Dec 2008 (ACH) Jan 2013 – Dec 2013 (LTACH) CAUTI Jan 2009 – Dec 2009 (ACH) Jan 2013 – Dec 2013 (LTACH) SSI Jan 2006 – Dec 2008 MRSA bacteremia Jan 2010 – Dec 2011

C. difficile infection

Jan 2010 – Dec 2011

An HAI incidence rate for a referent time period
In other words, how many HAI occurred and were reported into

NHSN nationally during a certain time period

SLIDE 35

35

NHSN: Why Re-baseline?

Significant changes to system use, definition, and

protocol changes since original baseline periods

– More – and different types – of facilities, users, and location types entering data and using NHSN – More data being entered each year, sometimes using automated methods – Removal/addition of selected measures, event types, locations, and individual variables – Changes in data collection methods – Introduction and refinement of definitions for identifying HAIs (e.g., fungal CAUTI no longer reportable)

SLIDE 36

36

NHSN: Benefits of Re-baselining

New baselines account for 2015’s major changes to HAI

definitions and criteria

A single referent (time) period results in more consistent

methods for calculating predicted infections

Using 2015 data allows NHSN to create an updated risk

modeling strategy

Re-baselining makes more SIR analysis output options

available in NHSN

SLIDE 37

37

NHSN: Impact of Re-baselining

Data reported to NHSN for 2015 used as the new baseline for future

SIRs

Risk adjustment methods and risk models may vary from those

generated using original baselines

All new risk models implemented into the NHSN application in the

form of new SIRs

NHSN users with data analysis rights can access SIR outputs using

both the new and old baselines, depending on time period

SLIDE 38

38

Summary of New Measures

Separate SIR for acute care hospitals, critical access

hospitals, long-term acute care hospitals, inpatient rehabilitation facilities

– CLABSI, CAUTI, CDI LabID Event, MRSA Bacteremia LabID Event

MBI LCBI SIR
Separate SIR for adult and pediatric SSI
VAE SIR

– Total VAE – IVAC Plus

Standard Utilization Ratios (SUR) for all device types

SLIDE 39

39

Statistical Implications for the SIR

The 2015 baseline is a new “starting/referent point”

from which to measure future progress

Therefore, SIRs will shift closer to 1, particularly for

the 2015 SIRs calculated with the 2015 baseline

This is because we have had less time to make

progress since the new baseline measurement period (calendar year 2015 data)

What will this look like?

SLIDE 40

40