Advertisement

Journal of General Internal Medicine

, Volume 34, Issue 1, pp 20–22 | Cite as

Performance of Internists and Medicine Specialists in Medicare Quality Metrics: Variation by Specialty and Other Physician Characteristics

  • Andrew B. RosenkrantzEmail author
  • Gregory N. Nicola
  • Richard DuszakJr
Concise Research Reports

KEY WORDS

performance measurement quality assessment Medicare health policy 

INTRODUCTION

As healthcare payments shift from volume- to value-based paradigms, Medicare’s Quality Payment Program (QPP) was established to accelerate that transition.1 Under QPP, most physicians will be scored and paid via the Merit-Based Incentive Payment System (MIPS). MIPS provides a wide range of quality measures for various medical specialties, and its various reporting options include traditional claims-based as well as registry-based and qualified clinical data registry (QCDR)-based reporting mechanisms.

Physicians’ performance variation under such measures is currently not well understood. For example, it remains unknown whether available measures favor or disfavor certain specialties in obtaining high scores and therefore positive payment adjustments in MIPS. Understanding how various specialty groups have performed to date could help inform the Centers for Medicare & Medicaid Services (CMS) and professional societies as they jointly develop and implement meaningful and fair metrics, and physician practices as they seek success in MIPS. We aimed to assess performance variation in CMS quality measures among internists and medicine specialists.

METHODS

The Physician Compare 2015 Individual Eligible Professional Public Reporting Performance Scores data set provides 2015 performance scores for all Medicare-participating providers.2 Crosslinking to the separate Medicare Physician Compare National Downloadable File 3, national provider identification numbers were used to identify physician characteristics, including self-reported primary specialty. National performance was computed for each measure reported by at least 100 providers. Using a previously reported methodology4, normalized z-scores were derived for all internists and medicine specialists and for all reported measures by computing the number of standard deviations from the national performance mean for each measure. Accounting for lower scores representing better performance for “inverse” measures, z-scores were then averaged across reported measures to obtain a single summary performance measure for each physician. Performance on the most commonly reported measures was identified; univariable associations with physician characteristics were assessed using analysis of variance.

RESULTS

Among 28,232 internists and medicine specialists, the six most commonly reported measures (Table 1
Table 1

Performance of Internal Medicine Specialists in Medicare Quality Measures

Reporting mechanism and performance measures

n

Ave

SD

Claims measures

 Documentation of current medications in the medical record

14,012

95.5

12.1

 Preventive care and screening: tobacco use: screening and cessation intervention

10,709

96.9

7.4

 Preventive care and screening: body mass index (BMI) screening and follow-up plan

8753

70.7

27.3

 Pneumonia vaccination status for older adults

6633

66.1

27.8

 Preventive care and screening: influenza immunization

5982

55.1

31.8

 Colorectal cancer screening

5776

63.0

31.6

 Breast cancer screening

4636

61.2

26.9

 Screening or therapy for osteoporosis for women aged 65 years and older

2990

56.9

26.8

 Preventive care and screening: screening for high blood pressure and follow-up documented

2451

90.0

21.0

 Diabetes: medical attention for nephropathy

2120

76.3

26.0

 Pain assessment and follow-up

1634

35.6

40.6

 Diabetes: foot exam

1455

49.8

35.0

 Diabetes: eye exam

1234

42.5

38.3

 Medication reconciliation

1057

96.6

10.3

 Elder maltreatment screen and follow-up plan

741

14.0

34.5

 Colonoscopy interval for patients with a history of adenomatous polyps –Avoidance of Inappropriate Use

465

98.0

9.1

 Falls: risk assessment

285

92.0

19.5

 Radiology: exposure time reported for procedures using fluoroscopy

258

98.1

8.0

 Chronic obstructive pulmonary disease (COPD): inhaled bronchodilator therapy

242

97.2

13.3

 Falls: plan of care

219

57.9

37.9

 Prevention of central venous catheter (CVC)-related bloodstream infections

185

91.8

23.9

 Appropriate follow-up interval for normal colonoscopy in average risk patients

132

93.5

14.8

 Osteoarthritis (OA): function and pain assessment

117

86.2

31.3

QCDR measures

 Photodocumentation of the cecum (also known as cecal intubation rate) all colonoscopies

1061

96.1

7.0

 Photodocumentation of the cecum (also known as cecal intubation rate) screening colonoscopies

1046

97.0

6.2

 Adequacy of bowel preparation

1046

94.2

7.0

 Documentation of history and physical rate—colonoscopy

1003

98.6

4.0

 Hypertension (HTN): blood pressure (BP) management

967

85.5

9.8

 Appropriate indication for colonoscopy

842

88.2

11.2

Registry measures

 Care plan

5131

69.9

34.7

 Preventive care and screening: unhealthy alcohol use †“Screening

2678

75.5

25.3

 Use of high-risk medications in the elderly*

2628

11.6

10.9

 Adult kidney disease: blood pressure management

1393

62.3

28.0

 Urinary incontinence: assessment of presence or absence of urinary incontinence in women aged 65 years and older

643

53.6

36.0

 Preventive care and screening: screening for clinical depression and follow-up plan

518

41.5

36.0

 Adult kidney disease: laboratory testing (lipid profile)

377

54.7

34.1

 Atrial fibrillation and atrial flutter: chronic anticoagulation therapy

326

57.6

36.3

 Stroke and stroke rehabilitation: discharged on antithrombotic therapy

253

60.1

36.0

 Diabetes mellitus: diabetic foot and ankle care, peripheral neuropathy –Neurological Evaluation

239

35.2

32.7

 Chronic obstructive pulmonary disease (COPD): spirometry evaluation

182

75.5

26.7

 Screening colonoscopy adenoma detection rate measure

155

48.9

22.6

*Only listed for measures reported by at least 100 internal medicine specialists

) were all claims-based (rather than registry- or QCDR-based) and included documentation of care in the medical record (average score 95.5%), tobacco use screening and cessation intervention (96.9%), body mass index screening and follow-up plan (70.7%), pneumonia vaccination status for older adults (66.1%), influenza immunization (55.1%), and colorectal cancer screening (63.0%). In the full cohort, average ± SD z-score was 0.132 ± 0.747. Performance improved significantly (Table 2
Table 2

Overall Performance Among Internal Medicine Specialists, Stratified by Physician Cohorts. Differences Considered Statistically Significant at p < 0.001

Cohort

Number of physicians reporting at least one measure with an associated z-score

Average z-score

SD z-score

Gender (p = 0.003)

 Female

7259

0.154

0.770

 Male

20,972

0.124

0.739

Years in practice (p < 0.001)*

 < 10

3453

0.065

0.819

 10–24

12,334

0.126

0.759

 25+

12,246

0.155

0.712

Group practice size (p < 0.001)*

 < 10

5535

0.259

0.696

 10–49

6749

0.142

0.742

 50–99

2814

0.107

0.692

 100+

9616

− 0.022

0.768

Geographic region (p < 0.001)**

 Midwest

5826

0.051

0.706

 Northeast

5940

0.125

0.810

 South

11,705

0.148

0.751

 West

4727

0.197

0.695

Specialty (p < 0.001)***

 Hospitalist

834

0.307

0.744

 Allergy/immunology

503

0.225

0.716

 Hospice/palliative care

74

0.195

0.641

 Pulmonary disease

1502

0.181

0.677

 Internal medicine

12,718

0.176

0.792

 Gastroenterology

3367

0.158

0.680

 Hematology/oncology

946

0.153

0.703

 Critical care (intensivists)

371

0.128

0.816

 Nephrology

1172

0.101

0.697

 Hematology

55

0.099

0.671

 Addiction medicine

11

0.076

0.964

 Endocrinology

706

0.044

0.725

 Cardiovascular disease (cardiology)

3371

0.022

0.665

 Sleep medicine

51

0.015

0.962

 Rheumatology

626

− 0.014

0.693

 Interventional cardiology

674

− 0.019

0.617

 Geriatric medicine

131

− 0.021

0.793

 Medical oncology

245

− 0.040

0.766

 Infectious disease

467

− 0.087

0.886

 Cardiac electrophysiology

361

− 0.090

0.702

 Preventative medicine

22

− 0.271

1.308

 Advanced heart failure and transplant cardiology

24

− 0.457

0.568

*Characteristic not known for all physicians

**Excluded when outside of the 50 states or Washington DC

***Specialties listed in descending order of overall performance

) with increasing years in practice (z-score range 0.065 to 0.155; p < 0.001) and decreasing group practice size (range − 0.022 to 0.259; p < 0.001). Physician performance was highest (p < 0.001) in the West (0.197) and lowest in the Midwest (0.051). Specialties with best overall performance, in order, were hospitalists (0.307), allergists (0.225), hospice and palliative care physicians (0.195), pulmonologists (0.181), and general internists (0.176). The specialties with worst performance, in order, were transplant cardiologists (− 0.457), preventative medicine physicians (− 0.271), cardiac electrophysiologists (− 0.090), infectious disease physicians (− 0.087), and medical oncologists (− 0.040).

DISCUSSION

Overall, internists and medicine specialists reported a diverse set of quality measures to CMS, with no single measure reported by a majority of physicians. Overall performance across measures was associated with various physician characteristics, including experience, group size, and geography. Of note, performance varied widely among specialties, aligning with recent criticism of MIPS by the highly influential Medicare Payment Advisory Commission5 and suggesting that any fair comparison of disparate specialties inside the QPP will be difficult. Medicine practices should be aware of such variation in selecting measures to report to MIPS.

Policymakers and national specialty societies should continue to develop comprehensive measure sets, encompassing measures of relevance to all Medicare-participating specialties. As current measures may advantage or disadvantage physicians purely on the basis of their specialty, QPP normalization calibration may be necessary to create a level programmatic playing field.

Notes

Funding Information

Drs. Rosenkrantz and Duszak are supported by research grants from the Harvey L. Neiman Health Policy Institute.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they do not have a conflict of interest.

References

  1. 1.
    Centers for Medicare & Medicaid Services. Quality Payment Program. https://qpp.cms.gov/. Accessed on July 18, 2018.
  2. 2.
    Centers for Medicare & Medicaid Serivces. Physician Compare 2015 Individual EP public reporting. https://data.medicare.gov/Physician-Compare/Physician-Compare-2016-Individual-EP-Public-Report/5qwk-yzai/data. Accessed on July 18, 2018.
  3. 3.
    Centers for Medicare & Medicaid Services. Physician Compare National Downloadable File. https://data.medicare.gov/Physician-Compare/Physician-Compare-National-Downloadable-File/mj5m-pzi6. Accessed on July 18, 2018.
  4. 4.
    Rosenkrantz AB, Niocla GN, Duszak R Jr. Characteristics of high-performing radiologists within Medicare quality programs. Journal of the American College of Radiology. 2018;15(6):842–849.CrossRefGoogle Scholar
  5. 5.
    Medicare Payment Advisory Commission. Report to the Congressl Medicare Payment Policy. Chapter 15. March 2018. http://www.medpac.gov/docs/default-source/reports/mar18_medpac_entirereport_sec.pdf? Accessd on July 18, 2018.

Copyright information

© Society of General Internal Medicine 2018

Authors and Affiliations

  • Andrew B. Rosenkrantz
    • 1
    Email author
  • Gregory N. Nicola
    • 2
  • Richard DuszakJr
    • 3
  1. 1.Department of RadiologyNYU School of Medicine, NYU Langone Medical CenterNew YorkUSA
  2. 2.Hackensack Radiology GroupHackensackUSA
  3. 3.Department of Radiology and Imaging SciencesEmory University School of MedicineAtlantaUSA

Personalised recommendations