Attachment_J.8_-_Priority_Measures.pdf

Priority Measures for

CMMI

Monitoring and Evaluation

Center for Medicare & Medicaid Innovation

September 2013

Page | 2

Page | 3

SUMMARY

INTRODUCTION

In an effort to enhance understanding of initiatives and to establish a consistent framework for performance measurement and quality improvement, the Research and Rapid Cycle Evaluation Group aims to align indicators used for evaluations across the Innovation Center where applicable. Appropriate measurement alignment is necessary for the Innovation Center to examine the overall impact of its initiatives on the health of populations, quality, and efficiency of care, and to compare the effectiveness of different models. This priority list is intended to identify a minimum set of meaningful measures for monitoring and evaluations of CMMI models, and we encourage additional measures within each model as necessary to fully address model-specific monitoring and evaluation needs.

MEASURE INCLUSION PROCESS

As a way to coordinate with other movements both external and internal to CMS, the list includes, but is not limited to, measures recommended by the Measure Applications Partnership, Patient-Centered Medical Home Collaborative, National Committee for Quality Assurance, CMS quality reporting programs, and Agency for Healthcare Research & Quality (AHRQ) standards. In addition, measures identified in this priority list were compared against a scan of monitoring and evaluation measures that CMS have used or considered in the past.

A majority of the measures found in this document have been endorsed by a nationally recognized consensus-based entity that employs a multi-stakeholder process to assess the reliability, validity, and usability of quality performance measures. Preference has been given to the use of measures that have been endorsed by consensus-based entities to ensure the use and availability of standardized measures. Currently, the National Quality Forum performs this function1.

We encourage the use of endorsed measures to the extent possible. In the absence of an NQF endorsed standard, we have identified indicators or recommendations from other sources that we believe are appropriate for consideration.

On a periodic basis, measures will be examined for relevance and new measures considered for inclusion. Measures may be added or removed, based on the following criteria:

1. Measure addresses an important condition/topic with a performance gap and has a strong scientific evidence base to demonstrate that the measure when implemented can lead to the desired outcomes and/or more appropriate costs.

2. Measure addresses one or more of the six National Quality Strategy Priorities (safer care, effective care coordination, preventing and treating leading causes of mortality and morbidity, person- and family-centered care, supporting better health in communities, making care more affordable).

3. Promotes alignment with specific program attributes and across CMS and HHS programs

4. Measures included meet the needs of monitoring and evaluation activities within the Innovation Center

5. Measure reporting is feasible

ALIGNMENT AND USE

Due to a multitude of diverse initiatives, unique intervention approaches, and varying availability of data, no parsimonious set will adequately serve as measures for any of our initiatives. We acknowledge that there are many measurement gaps and that not all measures are applicable to all patients and settings. Where it makes sense, we ask our contractors and intramural researchers to align with the measures in the priority list but also to tailor as appropriate to the population and models being studied.

1 Selecting Quality and Resource Use Measures: A Decision Guide for Community Quality Collaboratives. AHRQ Publication No.

09(10)-0073, May 2010. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/qual/perfmeasguide/

Page | 4

Domains and Descriptions

The following measure types and associated definitions2 reflect the organization of measures found in this document.

Measures may fall under multiple domains and can be cross-cutting.

Structure*

*To the extent possible, analyses should include a focus on vulnerable populations, and health disparities should be examined by stratification of findings using data collection standards established by Section 4302 of the Affordable Care Act for race, ethnicity, sex, primary language, and disability status. See http://minorityhealth.hhs.gov/templates/browse.aspx?lvl=2&lvlID=208

Structure: Features of a healthcare organization or clinician relevant to the capacity to provide healthcare. This may include, but is not limited to, measures that address health IT infrastructure, provider capacity, systems, and other healthcare infrastructure supports.

Process: A healthcare service provided to, or on behalf of, a patient. This may include, but is not limited to, measures that may address adherence to recommendations for clinical practice based on evidence or consensus.

Outcome: The health state of a patient (or change in health status) resulting from healthcare— desirable or adverse.

Care Experience: Patient and their care givers’ experience of care

2 Definitions are based on those provided by NQF and AHRQ available at: http://www.qualityforum.org/Home.aspx, and at:

http://www.qualityindicators.ahrq.gov/Downloads/Software/SAS/V21R3A/pqi_guide_rev3.pdf

HIT Utilization Measures related to use of HIT

Preventive Care Measures examining provision of preventive care

Clinical Care Measures assessing adherence to processes of care

Care Coordination Measures assessing relationship and communication between providers and patients, including plan of care development and follow-up; follow-up to tests, referrals, etc.; availability of patient information to necessary caregivers/patient/ family members; and care transition issues

Cost of Care Measures and recommendations for calculating cost of care

Readmissions Measures related to n-day readmissions

Ambulatory Care Sensitive Condition (ACSC) Admissions Measures tied to hospitalizations for which quality outpatient care can potentially prevent, or for which early intervention can prevent complications or more severe disease

ER/ED Visits:

Measures tied to utilization of the ED/ER

Patient, Care Giver Experience Measures or surveys that use feedback from patients and their families about their experience with care

Care Experience*

Cost and Resource Use* Process* Outcome*

Mortality Mortality measures including disease-specific or all-cause, reported for a specific time period

Morbidity Intermediate outcome measures that describe level of health or disease

Functional and Health Status Change Assessment tools that examine changes in patient outcomes related to functional and other health status changes

Safety Outcomes Measures assessing outcomes of poor safety practices http://minorityhealth.hhs.gov/templates/browse.aspx?lvl=2&lvlID=208 http://www.qualityforum.org/Home.aspx

Page | 5

Cost and Resource Use: Counting the frequency of units of defined health system services or resources; some may further apply a dollar amount (e.g., allowable charges, paid amounts, or standardized prices) to each unit of resource use (i.e., monetize the health service or resource use units)

Domain ID Measure Title and Description

St ru ct ur e

H

IT

U til iz at io n

NQF

1. Adoption of Medication e-Prescribing Documents whether provider has adopted a qualified e-Prescribing system and the extent of use in the ambulatory setting

NQF

2. Ability for Providers with HIT to Receive Laboratory Data Electronically Directly into their Qualified/Certified EHR System as Discrete Searchable Data

Documents the extent to which a provider uses certified/qualified EHR system that incorporates an electronic data interchange with one or more laboratories allowing for direct electronic transmission of laboratory data into the EHR as discrete searchable data elements

Pr oc es s

Pr ev en tiv e Ca re

NQF

3. Childhood Immunization Status Measure calculates a rate for each recommended vaccines and nine separate combination rates.

NQF

4. Influenza Vaccination Percentage of patients aged 6 months and older seen for a visit between October 1 and the end of February who received an influenza immunization OR patient reported previous receipt of an influenza immunization

NQF

5. Pneumonia Vaccination Status for Older adults Percentage of patients 65 years of age and older who ever received a pneumococcal vaccination

NQF

6. Measure Pair: A) Tobacco Use Assessment, B) Tobacco Cessation Intervention A) Percentage of patients who were queried about tobacco use one or more times during the two-year measurement period, B)Percentage of patients identified as tobacco users who received cessation intervention during the two-year measurement period

NQF

7. Colorectal Cancer Screening Percentage of members 50-75 years of age who had appropriate screening for colorectal cancer

NQF

8. Well-Child Visits in the First 15 Months of Life Percentage of members who turned 15 months old during the measurement year and who had the following number of well-child visits with a PCP during their first 15 months of life.

NQF

9. Well-Child Visits in the Third, Fourth, Fifth, and Sixth Years of Life Percentage of members 3–6 years of age who received one or more well-child visits with a PCP during the measurement year

NQF

10. Body Mass Index (BMI) 2 through 18 Years of Age Percentage children, 2 through 18 years of age, whose weight is classified based on BMI percentile for age and gender

NQF

11. Adult Weight Screening and Follow-Up Percentage of patients aged 18 years and older with a calculated BMI documented in the medical record AND if the most recent BMI is outside the parameters, a follow up plan is documented

Cl in ic al

C ar e

Di ab et es

12. Proportion of Days Covered: 5 Rates by Therapeutic Category Percentage of patients 18 years and older who met the proportion of days covered threshold of 80% during the measurement year. Rate is calculated separately for the following medication categories: Beta- Blockers, ACEI/ARB, Calcium-Channel Blockers, Diabetes Medication, Statins

13. Comprehensive Diabetes Care: Eye Exam Percentage of adult patients with diabetes aged 18-75 years who received an eye screening for diabetic retinal disease during the measurement year

NQF

14. Diabetes: Foot Exam Percentage of adult patients with diabetes aged 18-75 years who received a foot exam (visual inspection, sensory exam with monofilament, or pulse exam)

NQF

15. Comprehensive Diabetes Care: Medical Attention for Nephropathy The percentage of members 18-75 years of age with diabetes (type 1 and type 2) who received a nephropathy screening test or had evidence of nephropathy during the measurement year.

Page | 6

Domain ID Measure Title and Description

CA

D

NQF

16. ACE Inhibitor or ARB Therapy–Diabetes or LVSD Percentage of patients aged 18 years and older with a diagnosis of CAD seen within a 12 month period who also have diabetes or a current or prior LVEF <40% who were prescribed ACE inhibitor or ARB therapy

NQF

17. Antiplatelet Therapy Percentage of patients aged 18 years and older with a diagnosis of CAD seen within a 12 month period who were prescribed aspirin or clopidogrel

Pr oc es s

Cl in ic al

C ar e

CA

D

NQF

18. Beta-Blocker Therapy—Prior MI or LVSD Percentage of patients aged 18 years and older with a diagnosis of CAD seen within a 12 month period who also have prior MI or a current or prior LVEF <40% who were prescribed beta-blocker therapy

NQF

19. Lipid Control Percentage of patients aged 18 years and older with a diagnosis of CAD seen within a 12 month period who have a LDL-C result <100 mg/dL OR patients who have a LDL-C result >=100 mg/dL and have a documented plan of care to achieve LDL-C <100mg/dL, including at a minimum the prescription of a statin

H F

NQF

20. Beta-blocker Therapy for Left Ventricular Systolic Dysfunction Percentage of patients aged 18 years and older with a diagnosis of heart failure with a current or prior LVEF < 40% who were prescribed beta-blocker therapy either within a 12 month period when seen in the outpatient setting or at hospital discharge

IV

D

NQF

21. Use of Aspirin or Another Antithrombotic Percentage of patients 18 years and older with IVD who were discharged alive for AMI, CABG or PCI from January 1-November 1 of the year prior to the measurement year, or who had a diagnosis of IVD during the measurement year and the year prior to the measurement year and who had use of aspirin or another antithrombotic during the measurement year.

NQF

22. Complete Lipid Profile and LDL Control <100 Percentage of patients 18 years of age and older who were discharged alive for AMI, CABG or PCI from January 1–November 1 of the year prior to the measurement year, or who had a diagnosis of IVD during the measurement year and the year prior to measurement year, who had each of the following during the measurement year: Complete Lipid Profile and LDL-C control <100 mg/dL

AM

I

NQF

23. Fibrinolytic Therapy Received Within 30 Minutes of Hospital Arrival Percentage of AMI patients with ST-segment elevation or LBBB on the ECG closest to arrival time receiving fibrinolytic therapy during the hospital stay and having a time from hospital arrival to fibrinolysis of 30 minutes or less

NQF

24. Fibrinolytic Therapy Received Within 30 Minutes of ED Arrival Emergency Department AMI patients receiving fibrinolytic therapy during the ED stay and having a time from ED arrival to fibrinolysis of 30 minutes or less

NQF

25. Primary PCI Received within 90 Minutes of Hospital Arrival Percentage of AMI patients with ST-segment elevation or LBBB on the ECG closest to arrival time receiving primary PCI during the hospital stay with a time from hospital arrival to PCI of 90 minutes or less

NQF

26. Median Time to Transfer to Another Facility for Acute Coronary Intervention Median time from ED arrival to time of transfer to another facility for acute coronary intervention

Re sp ir at or y

NQF

27. COPD: Inhaled Bronchodilator Therapy Percentage of symptomatic patients with COPD who were prescribed an inhaled bronchodilator

PQRS

28. Asthma: Asthma Assessment Percentage of patients who were evaluated during at least one office visit for the frequency of daytime and nocturnal asthma symptoms

NQF

29. Asthma: Pharmacologic Therapy for Persistent Asthma Percentage of all patients with mild, moderate, or severe persistent asthma who were prescribed either the preferred long-term control medication (inhaled corticosteroid) or an acceptable alternative treatment

Page | 7

Dr ug

D ep en de nc e

NQF

30. Initiation and Engagement of Alcohol and Other Drug Dependence Treatment The percentage of adolescent and adult members with a new episode of alcohol or other drug (AOD) dependence who received the following.

a) Initiation of AOD Treatment: The percentage of members who initiate treatment through an IP AOD admission, OP visit, intensive OP encounter or partial hospitalization within 14 days of the diagnosis. b) Engagement of AOD Treatment: The percentage of members who initiated treatment and who had two or more additional services with a diagnosis of AOD within 30 days of the initiation visit.

Pr oc es s

Cl in ic al

C ar e

SC

IP

31. Prophylactic Antibiotics Discontinued Within 24 Hours After Surgery End Time Surgical patients whose prophylactic antibiotics were discontinued within 24 hours after Anesthesia End Time (48 hours for CABG or Other Cardiac Surgery). The Society of Thoracic Surgeons Practice Guideline for Antibiotic Prophylaxis in Cardiac Surgery (2006) indicates that there is no reason to extend antibiotics beyond 48 hours for cardiac surgery and very explicitly states that antibiotics should not be extended beyond 48 hours even with tubes and drains in place for cardiac surgery.

NQF

32. Surgery Patients Who Received Appropriate VTE Prophylaxis Within 24 Hours Pre/post-surgery Percentage of surgery patients who received appropriate Venous Thromboembolism (VTE) Prophylaxis within 24 hours prior to surgery to 24 hours after surgery end time

M en ta l H ea lth

33. Screening for Clinical Depression Percentage of patients aged 18 years and older screened for clinical depression using an standardized tool and follow-up plan documented

NQF

34. Post-Discharge Continuing Care Plan Created Patients discharged from a hospital-based IP psychiatric setting with a continuing care plan created overall and stratified by age groups3

NQF

35. Post-Discharge Continuing Plan Transmitted to Next Level of Care Provider Upon Discharge Patients discharged from a hospital-based IP psychiatric setting with a continuing care plan provided to the next level of care clinician or entity overall and stratified by age groups 4

NQF

36. Follow-Up After Hospitalization for Mental Illness Percentage of discharges for members 6 years of age and older who were hospitalized for treatment of selected mental health disorders and who had an OP visit, an intensive OP encounter, or partial hospitalization with a mental health practitioner. Two rates are reported: 1) the percentage of members who received follow-up within 30 days of discharge, 2) the percent of members who received follow-up within 7 days of discharge

Pr en at al

37. Frequency of Ongoing Prenatal Care Measure examines the percentage of Medicaid deliveries that received various numbers of expected prenatal visits.

Ca re

C oo rd in at io n

NQF

38. 3-Item Care Transition Measure Uni-dimensional patient self-reported survey that measure the quality of preparation for care transitions (to be included in HCAHPS)

NQF

39. Care Transition Record Transmitted to Health Care Professional Percentage of patients, regardless of age, discharged from an inpatient facility to home or any other site of care for whom a transition record was transmitted to the designated health care provider for follow-up care within 24 hours.

NQF

40. Transition Record with Specified Elements Received by Discharged Patients Percentage of patients, regardless of age, discharged from an inpatient facility to home or any other site of care, or their caregiver(s), who received a transition record (and with whom a review of all included information was documented) at the time of discharge

Ca re

Co or di na tio n NQF

41. Medication Reconciliation Percentage of patients aged 65 years and older discharged from any IP facility (e.g. hospital, skilled nursing facility, or rehabilitation facility) and seen within 60 days following discharge in the office by the physician providing on-going care who had a reconciliation of the discharge medications with the current medication list in the medical record documented

3 Paired measure with HBIPS-7 4 Paired measure with HBIPS-6

Page | 8

Ou tc om e M or ta lit y NQF 0229, 0230, 1893, 0468,

42. 30-Day Mortality Rate, Risk Adjusted The measure estimates a hospital-level risk-standardized mortality rate (RSMR), defined as death from any cause within 30 days after the index admission date, for patients 18 and older discharged from the hospital. These measures are specified for HF, AMI, Pneumonia, Stroke, and COPD.

M or bi di ty

Di ab et es

43. Optimal Diabetes Care The percentage of patients 18-75 with a diagnosis of diabetes, who have optimally managed modifiable risk factors (A1c<8.0%, LDL<100 mg/dL, blood pressure<140/90 mm Hg, tobacco non-use and daily aspirin usage for patients with diagnosis of IVD) with the intent of preventing or reducing future complications associated with poorly managed diabetes.

Ou tc om e

M or bi di

H

TN

NQF

44. HTN: Controlling High Blood Pressure Percentage of patients > 18 years of age with a diagnosis of hypertension in the first six months of the measurement year or any time prior with last BP < 140/90 mm Hg

M at er na l/ Ch ild

45. Elective Delivery Prior to 39 Completed Weeks Gestation Percentage of babies electively delivered prior to 39 completed weeks gestation

NQF

46. Cesarean Rate for Low-Risk First Birth Women Percentage of low-risk first birth women (aka NTSV CS rate: nulliparous, term, singleton, vertex) with a Cesarean rate that has the most variation among practitioners, hospitals, regions and states

NQF

47. Healthy Term Newborn Percent of term singleton live births (excluding those with diagnoses originating in the fetal period) who DO NOT have significant complications during birth or the nursery care

Fu nc tio na l an d H ea lth

S ta tu s

Ch an ge

Cl in ic ia n As se ss m en t

48. Continuity Assessment Record and Evaluation Tool (CARE Tool) The CARE Tool was developed for CMS for the purpose of creating a uniform patient assessment instrument at acute hospital discharge and at post acute care admission and discharge. CARE is designed to measure outcomes in physical and medical treatments while controlling for factors that affect outcomes, such as cognitive impairments and social and environmental factors. Four major domains are included in the tools: medical, functional, cognitive impairments, and social/environmental factors.

49. CARE-F and CARE-C Assessment Tools for Nursing Facilities, Day Rehabilitation Programs, and Other ambulatory Settings in the Community5

Under the DOTPA project, three versions of the assessment tool for nursing facilities, day rehabilitation programs, and other ambulatory settings in the community were developed. These assessments encompass distinct domains of patient need and utilize a set of patient self-assessment items (based on the AM-PAC) and clinician-reported items (based on the CARE tool).

Pa tie nt R ep or te d

NQF

0429,

50. Activity Measure for Post Acute Care (AM-PAC)-CMS DOTPA Short Form Public Domain Version The AM-PAC is a functional status assessment instrument developed specifically for use in facility and community dwelling post acute care patients. Unlike traditional functional outcome measures which are disease, condition, or setting-specific, the AM-PAC was designed to be used across patient diagnoses, conditions and settings where post acute care is being provided; therefore, the AM-PAC is useful for developing benchmarks and for examining functional outcomes over an episode of post acute care, as patients move across care settings.6

Sa fe

51. Patient Safety for Selected Indicators A composite measure of potentially preventable adverse events for selected indicators including pressure ulcers, iatrogenic pneumothorax, central venous catheter-related bloodstream infections, postop hip fracture, postop hemorrhage or hematoma, postop physiologic and metabolic derangements, postop respiratory failure, postop PE or DVT, postop sepsis, postop wound dehiscence, accidental puncture or laceration

Pa tie nt

Ca re

G iv er

Ex pe ri en ce

0005- 7, 0009, 0517, 0691- 0693,

52. CAHPS® surveys Consumer Assessment of Healthcare Providers and Systems (CAHPS®) surveys ask consumers and patients to report on and evaluate their experiences with health care. We recommend the usage of CAHPS surveys in every setting of care for which they are available.

5 See http://optherapy.rti.org/AssessmentTools.aspx for additional information.

6 Contact jun.li@cms.hhs.gov for additional information on the CMS DOTPA Short Form public domain version.

mailto:jun.li@cms.hhs.gov

Page | 9

Co st a nd R es ou rc e

Us e

Co st o f C ar e

To ta l Co st o f C ar e

53. Measurement Principles for Medicare Fee-for-Service Payments7

A). Price standardize for DSH, IME, and area wages. Method of pricing should be transparent and standardized when possible to reflect underlying utilization changes and not artifacts of the Medicare payment system. B). No routine truncation of extreme values except those related to obvious data errors.

If truncation is necessary, model diagnostics and sensitivity analyses are recommended. C). Recommend risk adjustment. D). Partial year observations or incomplete calendar year FFS claims should be annualized by prorating and then down weighting. For deaths, consideration should be given to annualizing partial year costs in a way that accounts for the exponential increase in monthly costs as death approaches8.

IP

E pi so de

54. Medicare Spending Per Beneficiary, Risk-adjusted and Price Standardized

MSPB measure evaluates hospitals’ efficiency relative to the efficiency of the median hospital. Specifically, the MSPB Measure assesses the cost to Medicare of services performed by hospitals and other healthcare providers during an MSPB episode, which comprises the period immediately prior to, during, and following a patient’s hospital stay

Re ad m is si on s NQF

55. Hospital All-Cause Unplanned Readmissions, Risk Adjusted Hospital-wide, all-cause, risk standardized readmission rate (RSRR) following hospitalization for all conditions and procedures

AC

SC

56. Diabetes Long-term Complications The number of discharges for long-term diabetes complications per 100,000 population Age 18 Years and Older in a Metro Area or county in a one year time period

NQF

57. Chronic Obstructive Pulmonary Disease This measure is used to assess the number of admissions for COPD per 100,000 population

NQF

58. Congestive Heart Failure Admission Rate Percent of county population with an admissions for CHF

NQF

59. Bacterial Pneumonia Number of admissions for bacterial pneumonia per 100,000 population

NQF

60. Urinary Tract Infection Admission Rate Number of discharges for urinary tract infection per 100,000 population age 18 years and older in a metro area or county in a one year time period

NQF

61. Adult Asthma Number of admissions for asthma in adults per 100,000 population

ER

/E

D

62. Hospital ED Visit Rate that did not Result in Hospital Admission, by Condition Numerator: All beneficiaries attributed to the Group Practice Reporting Option (GPRO) group with a given CCW indicator, sum the number of ED visits identified in the Outpatient SAF as specified by ResDac9.

Denominator: Count number of beneficiaries attributed to the GPRO group with a given CCW flag.

Note: Please also consider ED observation unit visit rates

7 See Appendix A for more details 8 Lubitz J, Riley G. Trends in Medicare Payments in the Last Year of Life. N Engl J Med 1993; 328: 1092-1096 9 http://www.resdac.org/resconnect/articles/144

Page | 10

Appendix A

Measurement Principles for Medicare Fee-for-Service Payments In many CMMI models, a key outcome of interest is the impact of interventions on Medicare fee-for-service (FFS) payments. The following principles developed by the Research and Rapid Cycle Evaluation Group serve as a guide to meaningful measurement of fee-for-service Medicare Part A and B expenditures when evaluating FFS Innovation Center models.

Price Standardization: To adjust for payment system variation that is not directly related to utilization decisions.

Recommendation: Standardize for area wages (WI, GPCI, etc.), disproportionate share hospital (DSH), and Indirect Medical Education (IME). Other standardization criteria may also apply, such as for site of service differentials, COLA, and CAHs depending on the research question of interest. For trend analysis and comparisons over multiple years, constant dollar adjustment shall be considered to account for inflations.

Rationale: The basic unit of analysis should be related to changes in utilization and not to artifacts of the Medicare payment system.

Truncation: To remove extreme values in data, but not to obscure the effects of outliers Recommendation: Do not routinely truncate extreme values, except those related to obvious data errors. If truncation is necessary, a variety of model diagnostics and sensitivity analyses is recommended to ascertain whether spurious data points are driving the conclusions that are being drawn from the model. In that case, censoring observations may be justified.

Rationale: The inclusion of all values including high-cost patients will provide a more reliable measure of the cost to the Medicare program.

Risk Adjustment: To reduce the variation in expenditures that stem from differences in risk factors Recommendation: Risk adjustment techniques should be applied to data when appropriate for the analysis.

Depending on the analysis and time periods involved, a demographic model leading up to a full HCC model may be considered. HCCs would be an appropriate risk adjuster in most cases, and is available as a prospective or concurrent adjuster, depending on the research question being addressed.

Rationale: In order to account for inherent population risk differences, comparisons between different populations will generally require risk adjustment. However, in cases where the population characteristics can be controlled or propensity matched, less reliance may be made in applying risk adjustment models. For initiatives where the objective is to reduce risk factors, risk adjustment must be carefully applied so as to not adjust away the benefits of the intervention.

Partial year beneficiary data: To account for partial year participation Recommendation: Beneficiaries with partial year claims history should not be routinely excluded from analysis. In most cases, partial year costs can be annualized, and those observations assigned lesser weights in the analysis, based on the proportion of the year during which claims experience was observed. In the case of deaths, consideration should be given to annualizing partial year costs in a way that accounts for the exponential increase in monthly costs as death approaches.

Rationale: Partial year beneficiary data can be informative, and can skew the analysis if routinely ignored. Deaths represent a partial year situation of special importance because of the high costs and heavy use of services in the last months of life. Please see: Lubitz J, Riley G. Trends in Medicare Payments in the Last Year of Life. N Engl J Med 1993; 328: 1092-1096