Research methodology

Introduction

The Center analyzed Medicare claims data obtained from the Centers for Medicare and Medicaid Services (CMS). For the analysis on the four cancer screening tests, the Center used a subset of the data submitted by physicians, hospitals and clinics from 2003 to 2008, the last year available at the time the data were acquired. Denied claims were excluded.

For privacy purposes and other reasons, the Center was limited to a 5 percent sample of national Medicare Part B data that contain claims for medical procedures, such as routine screenings, and used mainly by researchers and consultants. Besides limitations of sampling, the data have only the quarter in which a procedure was performed, not actual dates. And a permanent federal injunction prevents naming individual doctors who received payment for the claims.

Results from the 5 percent sample were multiplied by 20 to give a national scope to analyzed trends, an accepted survey research technique. However, even with a sample this large, it is impossible to account for all types of errors in the data. This means all calculations are estimates and rounded and must be considered imprecise.

The analysis of Medicare claims for prostate, cervical, colon and breast cancer screenings was based on procedure codes obtained from documents and guidelines put out by CMS and the U.S. Preventive Services Task Force, a panel of medical experts. Some codes were also confirmed by professionals in the field.

The codes represent a number of different procedures for each type of cancer screening. Colon cancer screenings include screening colonoscopies, sigmoidoscopies, barium enemas and fecal occult blood tests. Cervical cancer screenings include both Papanicolaou smears and liquid-based cytology. Breast cancer screenings include both film and digital mammograms, and prostate cancer screenings include digital rectal exams as well as prostate-specific antigen tests.

The Preventive Services Task Force notes that high risk patients over 65 are excluded from its recommendation against getting routine cervical cancer screenings. After consulting with doctors and available documentation from CMS, we excluded any cervical cancer screening claims for which the high risk code ‘V1589’ was included in the diagnosis.

The Center relied on the number of claims billed to indicate trends and frequency for the corresponding test. However, a claim cannot be equated with a procedure. Certain procedures can be billed in multiple claims. This likely results in a higher number of claims than procedures, making it difficult if not impossible to calculate cost per procedure.

To protect patients’ identities, some information, such as the date of a claim, was removed from the data, making it impossible to isolate procedures.

The overall price for each test included the payments Medicare made to physicians, hospitals, clinics and laboratories. According to CMS’s Research Data Assistance Center (ResDAC), some claims (about 5 percent of the claims analyzed) incorrectly include a small amount of interest on the first line of payment. Therefore the payment amounts, like all else in the data, cannot be considered exact.

For the total mammography cost, the Center included the amount of money Medicare paid for computer-aided detection (CAD), an optional add-on that highlights for radiologists areas on a screening image that may contain tumors.