The Joint Commission 2017 Annual Report

The report then discusses performance on our traditional core quality measures. A total of 14 core measures were retired by CMS and The Joint Commission at the end of 2015 because performance was consistently very high; this year’s report documents 2016 performance on the remaining 15 different chart-abstracted accountability measures in seven measure sets.

This year’s report shows hospitals’ continued strong performance on these measures. While the data show impressive gains in hospital quality, improvements can still be made. Some hospitals perform better than others in treating particular conditions. More than 3,200 Joint Commission-accredited hospitals contributed data. Quality and safety results for specific hospitals can be found at www.qualitycheck.org.

The key findings of the report are:

1. 470 hospitals reported eCQM data in 2016.

This represents a dramatic increase from the 34 hospitals that voluntarily submitted eCQM data in 2015. In 2017, we expect that the number of reporting hospitals will increase to more than 2,000. We recognize:

• 470 Data Contributors: Hospitals that voluntarily transmitted eCQM data for The Joint Commission’s production database.
• Nine Expert Contributors: Hospitals that advanced the evolution and utilization of eCQMs through contributions, by presenting at a Pioneers in Quality™ webinar or participating in eCQM development.
• 11 Solution Contributors: Hospitals that submitted a Proven Practice selected for inclusion in The Joint Commission’s Proven Practices Collection. Introduced in April, the Proven Practices Collection is a new resource available to Joint Commission-accredited hospitals. This new initiative recognizes hospitals that have successfully leveraged eCQMs and health IT to drive quality improvement.

The success stories of the Expert and Solution Contributors are shared via the Pioneers in Quality™ webinars, which assists hospitals on their journey toward eCQM adoption.

2.  Two voice of the customer surveys on eCQMs conducted by The Joint Commission found that awareness of eCQM reporting requirements is very high and most hospitals plan to report 2017 eCQM data to CMS. 

Compared to hospitals responding to the first survey conducted in spring 2016, hospitals participating in the second survey in fall 2016 revealed:

• More willingness to report voluntarily
• More confidence about the accuracy of their eCQM data
• Increased perceived readiness to successfully submit eCQM data
• Increased confidence in generating quality reporting document architecture (QRDA) Category 1 documents
• The ability to submit using their own electronic health records (EHR) data

3. Hospital performance on accountability measures continued to be strong, greatly enhancing the quality of care provided in Joint Commission-accredited hospitals. 

Accountability measures are evidence-based care processes closely associated with positive patient outcomes. The 2016 overall accountability composite calculation is derived from a total of 15 accountability measures from seven sets (inpatient psychiatric services, venous thromboembolism (VTE) care, stroke care, perinatal care, immunization, tobacco use treatment, and substance use care). 

In 2016, improvements on several individual measures increased as much as 9.8 percentage points. Performance on a few individual measures declined slightly. Relatively small percentage-point improvements on measures for which performance is already strong can often require as much or even more diligence than large percentage-point improvements where much room for improvement exists. All improvements are important and contribute to better care for patients.

• The 2016 inpatient psychiatric services result is 92.1 percent, up from 89.7 percent in 2012 – an improvement of 2.4 percentage points.
• The 2016 perinatal care result is 98.1 percent, up from 57.6 percent in 2012 – an improvement of 40.5 percentage points.
• The 2016 tobacco use treatment result is 87.7 percent, up from 75.8 percent in 2014 – an improvement of 11.9 percentage points.
• The 2016 substance use care result is 82.2 percent, up from 58.2 percent in 2014 – an improvement of 24.0 percentage points.

The heart attack and children’s asthma care accountability measures included in last year’s report have been retired. There are no VTE, stroke or immunization measure set composites this year because a measure set composite must have at least two measures and these measure sets are comprised of only one accountability measure. Performance on the individual measures on these clinical topics showed good improvement.

• The VTE warfarin discharge instructions measure result is 92.9 percent, up from 82.2 percent in 2012 – an improvement of 10.7 percentage points. 
• The stroke care thrombolytic therapy result is 89.6 percent, up from 77.3 percent in 2012 – an improvement of 12.3 percentage points.
• The influenza immunization measure result is 94.3 percent, up from 86.2 percent in 2012 – an improvement of 8.2 percentage points.

Performance on the above three measures was included in the overall accountability composite results. Composite accountability measures have been compiled for inpatient psychiatric services, VTE and stroke care since 2011, for perinatal care and immunization since 2012, and for tobacco use treatment and substance use care since 2014. The composites for each year are calculated on measures active for the entire year; active measures can change from year to year. For more information about accountability composite results versus composite results, see “Note on Calculations and Methodology.” 

4.  The 2016 composite accountability score declined slightly, which we believe is due to the retiring of measures that had a very high performance in the past. 

The Joint Commission analyzes improvement with a “composite” result, which sums up the results of individual accountability process measures into a single summary score. While the composite performance increased for all the measure sets, the overall composite decreased slightly from 93.7 percent in 2015 to 92.4 percent in 2016. This is due to the fact that 14 measures that had been used for many years were retired. These retired measures contributed roughly half of all cases to the 2015 accountability composite rate. Thus, the apparent decrease in the composite score from 2015 to 2016 is a result of removing these measures. 

The retirement of the measures was made to reduce the burden of reporting on organizations and to allow them to focus on areas where there are still significant opportunities to improve. The report also includes performance data on two non-accountability process measures noted within the measure sets (VTE-6: Incidence of potentially preventable VTE, and PC-05: Exclusive breast milk feeding), and two outcome measures (PC-02: Cesarean section, and PC-04: Newborn bloodstream infections).

The overall composite accountability score reflects 17.3 million opportunities to perform care processes closely linked to positive patient outcomes. Since the baseline has been significantly altered by the retirement of the measures, caution should be taken when comparing the 2015 and 2016 composite scores.

Measure sets with composite performance below the overall composite rate of 92.4 percent are inpatient psychiatric services (92.1 percent), tobacco use treatment (87.7 percent), and substance use care (82.2 percent). The 92.4 percent result identifies the rate at which evidence-based core measure practice is provided – combined over all hospitals – for every 100 opportunities to do so.

The 59.6 percent result measures the percentage of hospitals achieving overall composite performance greater than 95 percent.

Key components of the Pioneers in Quality™ program include:

• Regular educational webinars focused on eCQM adoption, including continuing education units (CEUs) for live webinar participation
• Expert-to-Expert series webinars
• A comprehensive eCQM resource portal
•  The Joint Commission’s annual report, focusing on components of the program and the evolution of eCQM measurement
• Recognition for eCQM pioneers, including in the annual report
• A Pioneers in Quality™ Technical Advisory Panel
• Outreach through The Joint Commission’s Speaker’s Bureau

In 2016, 470 hospitals chose to submit eCQM data; those hospitals were asked to submit a minimum of one quarter of data. The 470 hospitals are an increase from the 34 hospitals that voluntarily submitted eCQM data in 2015. In 2017, the number of reporting hospitals is expected to increase to more than 2,000.

Pioneers in Quality™ recognizes hospitals in three categories:

• 470 Data Contributors: Hospitals that voluntarily transmitted eCQM data for The Joint Commission’s production database.
• Nine Expert Contributors: Hospitals that advanced the evolution and utilization of eCQMs through contributions, by presenting at a Pioneers in Quality™
  webinar or participating in eCQM development.
• 11 Solution Contributors: Hospitals that submitted a Proven Practice selected for inclusion in The Joint Commission’s Proven Practices Collection. 

See the 2017 Pioneers in Quality™ Expert and Solution Contributors. 

The Pioneers in Quality™: Proven Practices Collection is a new resource that will be available to Joint Commission-accredited hospitals. In spring 2017, hospitals submitted their success stories via an online application form that asked applicants to clearly link their accomplishments to the use of eCQMs and health IT for quality improvement.

While this annual report shares high-level eCQM data, The Joint Commission is not publicly reporting 2016 and 2017 eCQM data on Quality Check^® because the accuracy of eCQMs continues to be a concern. Hospitals reporting on chart-abstracted measures will continue to have their data and performance on the chart-abstracted measures reported on Quality Check^®.   

The Joint Commission aligned our eCQM reporting requirements as closely as possible to the CMS Hospital Inpatient Quality Reporting Program. During 2016, there were 23 eCQMs from which Joint Commission-accredited hospitals could select and report performance data. 

For more information on Pioneers in Quality™ or the Proven Practices Collection, visit the Pioneers in Quality™ web portal, which includes the 2017 eCQM Data Contributors being recognized by The Joint Commission.

The Joint Commission continues to be a leader in performance measurement. The data displayed on the CMS Hospital Compare website reflects many measures that The Joint Commission and CMS have in common. A large percentage of that data comes from The Joint Commission via its well-established performance measure data network. Today, this network comprises approximately 31 measurement systems, all under contract to The Joint Commission, and is the source of quality-related data on The Joint Commission’s Quality Check^® website (www.qualitycheck.org). 

America’s Hospitals: Improving Quality and Safety – The Joint Commission’s Annual Report 2017 presents the overall performance of Joint Commission-accredited hospitals on quality of care for chart-based measures relating to inpatient psychiatric services, venous thromboembolism (VTE) care, stroke care, perinatal care, immunization, tobacco use treatment, and substance use care. These measures were chosen because they provide concrete data about the best kinds of treatments or practices for common conditions for which Americans enter the hospital and seek care. 

The results are important, because they show that hospitals have improved their care quality. The results identify opportunities for further improvement, and support continual measurement and reporting. Quality improvement in hospitals contributes to saved lives, better health, and quality of life for many patients, as well as lower health care costs. 

2016 ORYX^® performance measure reporting requirements

During 2016, Joint Commission-accredited hospitals had continued flexibility in meeting the ORYX^® performance measure requirements for reporting on a minimum of six measure sets. Only one measure set – perinatal care – was mandatory as one of the six measure sets for hospitals. The threshold for mandatory reporting on the perinatal care measure set was reduced to 300 or more live births per year (previously, it was 1,100 live births per year). Accredited hospitals had the flexibility of meeting ORYX^® reporting requirements through one of three options:

• Option 1: Vendor submission of quarterly data on six of nine sets of chart-abstracted measures.
• Option 2: Vendor submission of data on six of eight sets of eCQMs.* 
• Option 3: Vendor submission of data on six measure sets using a combination of chart-abstracted measures and eCQMs.*  

*For 2016, hospitals could report on as few as one eCQM in an eCQM set and it was counted as an eCQM set.

A special focus on accountability measures

Accountability measures are evidence-based care processes closely linked to positive patient outcomes. These measures are most suitable for use in programs that hold providers accountable for their performance to external oversight entities and to the public. There has been an evolution of such oversight programs – including those for value-based purchasing, accreditation, certification, and public reporting – and they are often used to demonstrate quality and cost-efficient performance, to drive market share, and to determine appropriate reimbursements. 

Each accountability measure meets four criteria that evaluate whether or not evidence-based care processes associated with the measures lead to positive patient outcomes. As new measures are introduced, they are evaluated against the criteria. 

For more information about accountability measures, see the New England Journal of Medicine article “Accountability Measures – Using Measurement to Promote Quality Improvement,” for which Mark R. Chassin, MD, FACP, MPP, MPH, president and chief executive officer of The Joint Commission, was the lead author. 

Also see the Annals of Internal Medicine article, “Holding Providers Accountable for Health Care Outcomes,” by Dr. Chassin and lead author David W. Baker, MD, MPH, FACP, executive vice president in the Division of Health Care Quality Evaluation at The Joint Commission. This latter article suggests a national critical look is needed on how to assess the validity of outcome measures used by public accountability programs. Outcome measures are intended to quantify the end results of a health care service or intervention. Yet, criteria for assessing whether they are accurate and valid enough to use for public reporting, payment and other accountability programs are not well defined.

It’s important to note that where a patient receives care makes a difference. Not all hospitals deliver the same level of quality; some hospitals perform better than others in treating particular conditions and in achieving patient satisfaction. This variability has been known within the hospital industry for a long time. Designation as an accountability measure is included in the information on Quality Check® (www.qualitycheck.org).

How quality measures are determined 

The Joint Commission worked closely with clinicians, health care providers, hospital associations, performance measurement experts, and health care consumers across the nation to identify the quality measures. This collaborative process identified measures that reflect the best “evidence-based” treatments relating to inpatient psychiatric services, VTE care, stroke care, perinatal care, immunization, tobacco use treatment, and substance use care. Current measures are the product of The Joint Commission’s Hospital Core Measure Initiative that sought to create sets of standardized national measures that would permit comparisons across organizations. Subsequently, The Joint Commission collaborated with CMS to align common measures to ease data collection efforts by hospitals and to allow the same data sets to be used to satisfy multiple data requirements. 

Related quality reporting efforts of other organizations 
The CMS Hospital Compare website (www.hospitalcompare.hhs.gov) reports quality information from over 4,000 Medicare-certified U.S. hospitals, including treatments relating to cataracts, colonoscopy, heart attack, emergency department care, preventative care (immunization), stroke care, blood clot prevention, perinatal care, and medical imaging. Hospital Compare also includes information on patient experiences, readmissions, complications, deaths, and payment and value of care. 

In addition, CMS in 2013 began receiving data on The Joint Commission’s perinatal care elective delivery measure, which was adopted for use in the CMS Hospital Inpatient Quality Reporting Program, and now around 3,300 hospitals are submitting data to CMS on this measure. 

Joint Commission-developed measures also have been adopted into a number of CMS quality reporting programs. Today, Joint Commission/CMS common measures and Joint Commission-only measures are used in the CMS Hospital Inpatient Quality Reporting Program, Hospital Outpatient Quality Reporting Program, Hospital Value-Based Purchasing Program, Inpatient Psychiatric Facility Quality Reporting (IPFQR) Program and the Medicare & Medicaid EHR Incentive Program for eligible Hospitals/Critical Access Hospitals. The Joint Commission-developed hospital-based inpatient psychiatric services (HBIPS) measures were adopted as the initial set of measures for the CMS IPFQR Program with other Joint Commission-developed measures subsequently adopted (i.e., tobacco use treatment and substance use care). 

Consumers can use Hospital Compare to compare care of local hospitals to state and national averages. Unlike Quality Check^®, Hospital Compare includes data from organizations accredited by CMS-recognized accrediting organizations other than The Joint Commission and some unaccredited organizations. Hospital Compare does not currently include Department of Defense and Indian Health Service hospitals. 

The National Quality Forum’s National Quality Partners (NQP) engages its members – including The Joint Commission – in health care quality issues of national importance.

Data collection and reporting requirements 

For 2016, The Joint Commission required most hospitals to select six measure sets. Hospitals chose sets best reflecting their patient population and reported on all the applicable measures in each of the sets they choose. Hospitals submitted monthly data on a quarterly basis on all measures of performance within specific sets they choose to third-party vendors, which compiled and provided data to The Joint Commission. Hospitals can obtain feedback reports through The Joint Commission’s Connect™ extranet. 

Accountability process measure. An accountability process measure is a quality measure that meets four criteria designed to identify measures that produce the greatest positive impact on patient outcomes when hospitals demonstrate improvement. The four criteria are: research, proximity, accuracy and adverse effects (see Understanding the Quality of Care Measures section for an explanation of the criteria). Accountability measures are a subset of core measures (see core measure).

Admission screening. Evaluating a patient for violence risk, substance use, psychological trauma history and patient strengths within the first three days of admission to an inpatient psychiatric facility. 

Antenatal steroids. Medication given to a mother in premature labor before delivery to promote lung development in the baby. 

Antithrombotic therapy. Pharmacologic agents (oral or parenteral) that prevent or interfere with the formation of a blood clot. 

Cesarean section. A surgical procedure in which an abdominal incision is made to deliver the infant. 

Composite measure. A measure that combines the results of two or more process measures into a single rating. A composite is a summary of a related set of measures, which could be a condition specific set, all accountability measures, or accountability and non-accountability measures. However, accountability composites are restricted to accountability measures; non-accountability measures are excluded. 

Continuous variable measure. A type of measure in which the value of each measurement can fall anywhere along a continuous scale (e.g., the time [in minutes] from hospital arrival to administration of a medication).

Core measure. A core measure is a standardized quality measure with precisely defined specifications that can be uniformly embedded in different systems for data collection and reporting. A core measure must meet Joint Commission-established attributes, such as: targets improvement in population health, precisely defined and specified, reliable, valid, interpretable, useful in accreditation, under provider control, and public availability. 

Elective delivery. A delivery occurring between 37 and 39 weeks of gestation, without a medical reason.

Electronic Clinical Quality Measure (eCQM). A clinical quality measure that is specified in a standard electronic format and is designed to use structured, encoded data present in the electronic health record.

Evidence-based care. Using current best evidence in making decisions about the care of individual patients or in the delivery of health services. 

Exclusive breast milk feeding. An infant receives only breast milk during the hospital stay, with no additional food or drink, including water. 

Fibrinolytic therapy. Medication that dissolves blood clots. Breaking up blood clots increases blood flow to the heart. If blood flow is returned to the heart muscle quickly during a heart attack, the risk of death is decreased. 

Health care-associated infections in newborns. An infection acquired during a newborn’s stay in a hospital.

Inpatient psychiatric services. Inpatient psychiatric services include care provided to a patient for a mental disorder while hospitalized in a psychiatric unit of an acute care hospital or a free-standing psychiatric hospital. Services rendered to outpatients or “day treatment” patients are not considered inpatient psychiatric services. 

Median. The value in a set of observations whose values are arranged from smallest to largest that divides the data into two parts of equal size (e.g., if looking at the time [in minutes] from hospital arrival to administration of a medication and the ranked observations were  5, 10, 20, 30 and 40 minutes, the median would be 20 minutes).

Multiple antipsychotic medications. Antipsychotic medications are drugs prescribed to treat mental disorders; if two or more medications are routinely administered or prescribed, it is considered multiple medications. 

Outcomes measure. A measure that focuses on the results of the performance or nonperformance of a process. (See process measure.) 

Overlap therapy. Administration of parenteral (intravenous or subcutaneous) anticoagulation therapy and warfarin to treat patients with VTE.

PCI therapy. PCI stands for “percutaneous coronary interventions.” PCI therapy is a coronary angioplasty procedure performed by a doctor who threads a small device into a clogged artery to open it, thereby improving blood flow to the heart. A lack of blood supply to the heart muscle can cause lasting heart damage. PCI therapy is used as an alternative treatment to coronary artery bypass graft surgery (CABG). 

Percentage points. This is the difference on a percentage scale between two rates expressed as percentages. For example, the difference between a performance rate of 85 percent and a performance rate of 92 percent is 7 percentage points. 

Perinatal. The period shortly before and after birth.

Perioperative. This generally refers to 24 hours before surgery and lasts until the patient leaves the recovery area. 

Physical restraint. A physical restraint is any manual or physical or mechanical device, material, or equipment that immobilizes a patient or reduces the ability of a patient to move his or her arms, legs, body or head freely. A physical restraint is used as a restriction to manage a patient’s behavior or restrict the patient’s freedom of movement and is not a standard treatment for the patient’s medical or psychiatric condition. 

Process measure. A measure that focuses on one or more steps that lead to a particular outcome. (See outcomes measure.) 

Prophylaxis. Any medical intervention designed to preserve health and prevent disease.

Range. The smallest and largest values in a set of data (e.g., if looking at the time [in minutes] from hospital arrival to administration of a medication and the values from the ranked observations were 5, 10, 20, 30 and 40 minutes, the range would be 5,40 minutes). The range can also be defined as a single number, the difference between the smallest and largest values (e.g., 40 – 5 = 35 minutes in the example).

Rehabilitation assessment. Evaluation of the need for or receipt of rehabilitation services. Rehabilitation is a treatment or treatments designed to facilitate the process of recovery from injury, illness or disease to as normal a condition as possible.

Seclusion. Seclusion is the involuntary confinement of a patient alone in a room or an area where the patient is physically prevented from leaving. 

Statin. A class of pharmaceutical agents that lower blood cholesterol. Specifically, the agents modify LDL-cholesterol by blocking the action of an enzyme in the liver which is needed to synthesize cholesterol, thereby decreasing the level of cholesterol in the blood. Statins are also called HMG-CoA reductase inhibitors. 

Test measure. A measure being evaluated for reliability of the individual data elements or awaiting National Quality Forum endorsement. 

Thrombolytic therapy. Administration of a pharmacological agent intended to cause lysis of a thrombus (destruction or dissolution of a blood clot). 

Top 10 percent. For measure reporting, this indicates the value (number) at which one-tenth of the recorded values are at this value or better (e.g., if looking at the time [in minutes] from hospital arrival to administration of a medication and the values from the ranked observations, a top 10 percent value of 151 would indicate that one-tenth of reporting hospitals have a measure value of 151 or less). 

UFH monitoring. Using a protocol or nomogram to ensure that UFH (unfractionated heparin) achieves a sufficient level of anti-coagulation. 

VTE. VTE stands for venous thromboembolism and is when a blood clot forms in a deep vein in the body, such as in the leg. VTE is a common complication at surgery, and hospitalized medical patients – particularly those who have decreased mobility – are at risk for development of VTE.

Recent changes to The Joint Commission’s ORYX^® performance measurement requirements are the result of the transition to eCQMs, as well as efforts to maintain close alignment with the Centers for Medicare & Medicaid Services (CMS) Hospital Inpatient Quality Reporting Program.

Why are eCQM rates different from chart-abstracted measure rates?

Due to the differences in how eCQMs and chart-abstracted measures are calculated, it is not surprising that we see apparent differences in performance rates. Performance rates on eCQM measures appear to be lower than expected when compared to the rates of chart review measures. 

There are several reasons why eCQM rates are different from chart-abstracted measure rates:

• Specifications for eCQMs and chart-abstracted specifications are different: The representation of data elements and inclusions and exclusions are constrained by the standards used to represent eCQMs, as well as by the information that is captured in a structured and encoded fashion in an EHR system. For example, a chart-abstracted data element may be represented by multiple data elements in the eCQM.
• Data sources for eCQMs are more limited than data sources used for chart-abstracted measures: eCQMs rely solely on data that is captured in a structured and encoded fashion in the EHR. In addition, eCQMs typically rely on a single structured data field in the EHR for a given data element. Discrepancies in rates often happen when data is not consistently captured in the field selected for data extraction.
• Release schedules and updates for eCQM specifications and chart-abstracted specifications are not always aligned: While there are continued efforts to keep eCQMs and chart-abstracted measure specifications as closely aligned as possible, eCQM specifications updates are released on a different schedule than the chart-abstracted measures manual. Updates for eCQMs are published once a year in early spring, whereas the chart-abstracted measures manual is released twice a year, in January and July.

Voice of the customer survey on eCQMs

During 2016, The Joint Commission conducted two voice of the customer surveys on eCQMs — one in the spring and another in the fall. The surveys found that awareness of reporting requirements is very high and for 2017 most hospitals plan to report eCQMs to CMS, as required. 

Compared to responses to the first survey, hospitals participating in the second survey showed more willingness to report voluntarily, more confidence about the accuracy of their eCQM data, increased perceived readiness to successfully submit eCQM data, increased confidence in generating quality reporting document architecture (QRDA) Category 1 documents, and greater ability to submit EHR data.

Comments from accredited hospitals included requests for more alignment with CMS to make data submission more efficient, and that changes to workflow and processes were necessary for eCQM reporting. Specific comments included:

• “Align with CMS so we are doing the same thing for both.”

• “There is a ton of work to be done to prepare for eCQMs that include workflow changes, documentation changes, education, and follow up on measures.” 

Another customer pointed out the advantage of using electronic methods to measure quality, so that “efforts can be focused on improvement rather than obtaining data.” Other customers requested support from The Joint Commission via best practices, webinars, and other educational offerings and resources. The Pioneers in Quality™ program has provided this needed education. “We have appreciated the forum to ask questions and discuss concerns,” one commented.

Table 1: Number of eCQM sets submitted for 2016

Graph 2: Most frequently reported eCQMs for 2016

These topic areas are in alignment with CMS eCQMs. The top three areas (eED, eVTE and eSTK) are eCQMs that hospitals have been reporting for the longest time.

Table 2: Summary of values for eCQMs reported in 2016

The rate (%) for the proportion measures listed reflects the percentage of time that recommended care was provided. The value (minutes) for the two eED measures reflects the time patients spend in the emergency department from their arrival until admitted to the hospital, and the time it takes for a patient to be admitted to the hospital after being seen in the emergency department. 

No hospitals had cases to report for eAMI-7a: Fibrinolytic therapy within 30 minutes. Also, PC-05: Exclusive breast milk feeding, and PC-05a: Exclusive breast milk feeding considering mother’s choice, are counted as one measure.

*A lower score reflects better performance for this measure.

Results are determined by the number of times the hospital met the measure divided by the number of opportunities (eligible patients for the measure) the hospital had during the year. Results are expressed as a percentage. 

All improvements or decreases in performance are statistically significant. Many of the smaller percentage improvements occurred within large patient populations, meaning that significantly more patients received a treatment. In some cases, performance was already quite high and there was less room for improvement. 

Composite measures combine the results of all individual process measures on a similar medical condition into a single percentage rating calculated by adding up the number of times recommended evidence-based care was provided to patients and dividing this sum by the total number of opportunities to provide this care. 

Composite for all measures: The composite for all measures calculation is derived from the accountability measures for each measure set. These composite results have historically been provided in previous annual reports, allowing them to be tracked from year to year. Any exclusions to the composite are noted with the tables.

See Glossary for definitions.

Table 5: Inpatient psychiatric services measure results

As in the other measure sets, high rates are preferred in this measure set for two of the measures. The overall measure and rates are indicated in bold; the stratified measures (by specific age ranges of patients) are indicated in regular type. 

Note: Admission screening became an accountability measure in 2014; it was a test measure in previous reports. 

Since implementation in 2009, the average number of hospitals reporting data was 718 and ranged from 244 to 2,076.

* The full name of the measure is “Multiple antipsychotic medications at discharge with appropriate justification — overall rate.” 

 Test measure; not included in the composite.

The following maps show measure performance from the first full year that data was reported compared to 2016 performance.

Table 7: Venous thromboembolism (VTE) care measure results

Since implementation in 2010, the average number of hospitals reporting data was 913 and ranged from 59 to 2,639.

 Test measure; not included in the composite. Also, a lower score reflects better performance for this measure, so the negative performance point difference is favorable.

The following maps show measure performance from the first full year that data was reported compared to 2016 performance.

 Table 8: Stroke care measure results

Since implementation in 2010, the average number of hospitals reporting data was 972 and ranged from 105 to 2,508.

The following maps show measure performance from the first full year that data was reported compared to 2016 performance.

Table 9: Perinatal care measure results

Since implementation in 2011, the average number of hospitals reporting data was 1,268 and ranged from 151 to 2,985.

* For this measure, a decrease in the rate is desired, so a negative percentage point difference is favorable.

** This measure was included in the composite for 2012, but not subsequently. 

This measure is an outcome measure and is not included in the composite. Only proportion process measures are included in the composite.

The following maps show measure performance from the first full year that data was reported compared to 2016 performance.

Table 10: Immunization measure results

Since implementation in 2012, the average number of hospitals reporting data was 1,313 and ranged from 78 to 2,741.

The following maps show measure performance from the first full year that data was reported compared to 2016 performance.

Table 11: Tobacco use treatment measure results

Since implementation in 2014, the average number of hospitals reporting data was 914 and ranged from 68 to 2,011.

The following maps show measure performance from the first full year that data was reported compared to 2016 performance.

Table 12: Substance use care measure results

Since implementation in 2014, the average number of hospitals reporting data was 271 and ranged from 130 to 513.

Table 13: Percentage of hospitals achieving 95 percent or greater performance

The following table shows percentage of hospitals achieving the annual targeted performance of 95 percent or more compliance on a measure. The last column is reported as percentage points – the difference on a percentage scale between two rates, in this case 2015 performance versus 2016 performance.

* For this measure, a decrease in the rate is desired, so the percentage represented is the percent of hospitals with percentage of 5 percent or less.

 Test measure; not included in the composite.

Table 6: Inpatient psychiatric services rate measure results

The following table includes two rate measures: physical restraint hours per 1,000 patient hours and seclusion hours per 1,000 patient hours. In addition, these two measures are stratified by age groups 1-12 years, 13-17 years, 18-64 years, and age 65 and above. Lower rates reflect better performance. 

The overall measure and rates are indicated in bold; the stratified measures (by specific age ranges of patients) are indicated in regular type.

Since implementation in 2009, the average number of hospitals reporting data was 718 and ranged from 244 to 2,076.

* A lower ratio is preferred for this measure. Also, it is not included in the composite results because the denominator represents patient days rather than patients, and therefore cannot be combined with the other measures.