While geographic variation in the use of surgery has long been recognized, not all surgical procedures are equally variable. For example, colon resection (colectomy) exhibits the same low variation pattern seen with hospitalization rates for hip fracture (Chapter Three). Other procedures, such as coronary artery bypass grafting, are highly variable.

What distinguishes low variation from high variation surgery? In general, low variation procedures are non-discretionary; they are used to treat clinical conditions for which physicians agree on the most appropriate treatment strategy. In addition, patient and doctor preferences are aligned – both parties have the same goals. Conversely, high variation procedures involve physician discretion; the variability reflects underlying problems in medical decision making that occur because of inadequate science and failure to take patient preferences into account.

• Sometimes, medical science is inadequate to provide definitive information on which treatment is likely to provide the best outcome for a given patient. In these cases, procedure rates vary because physicians disagree about the effectiveness of surgery.
  
  
• Sometimes, the scientific evidence regarding outcomes is adequate, but the available treatments have different risks and benefits, which only the patient can assess. The fact that patient preferences are unevenly incorporated into treatment decisions results in high variation in procedure rates.

In this chapter, we describe how these two factors are reflected in the variation profiles of common surgical procedures. The eight high variation procedures are all performed on members of BCBSM. We then examine the “surgical signatures” which demonstrate the sometimes striking variations in the amount of surgery provided between neighboring hospital service areas. The study is restricted to the 48 hospital service areas where the BCBSM adult membership exceeds 10,000.

To read chapter
To view and print the complete chapter, The Surgical Treatment of Common Diseases in Adobe Acrobat PDF format, click on the link further below. You will need Adobe Acrobat Reader to view and print this file.

To download this file to your computer's hard drive, follow the directions applicable to your computer system.

• For Windows users running Microsoft Internet Explorer, click on the mouse's right button and select "Save Target As . . . "
  
  
• For Windows users running Netscape, click on the mouse's right button and select "Save Link As . . . "
  
  
• For Macintosh users running Microsoft Internet Explorer, control-click and select "Download Link to Disk . . . "
  
  
• For Macintosh users running Netscape, control-click and select "Save this Link as . . . "
  

To get Adobe Acrobat Reader
If you need Adobe Acrobat Reader, you can get a free copy from the Adobe Web site.

Back to top

NOTICE TO ATLAS READERS
While not giving answers, the atlas raises questions about health care service use that merit careful consideration. "High" rates of use are not necessarily bad and "low" volumes good (or vice versa). Our goal is to move toward rates that are consistent with high quality health care, which need to be determined with local clinical, community and patient discussion and dialogue.

The atlas is not a physician or hospital report card. When reviewing data, note that the Hospital Service Areas in the atlas were defined by the atlas author. They may differ significantly from what a hospital considers its market area.