Referral Patterns to Neurosurgeons and Cost Analysis of Primary Care Physicians in Patients Presenting with Low Back Pain



Nicole A. Swallow, Class of 2005,
Student Box 228,

Faculty Advisor

Dr. Stephen K. Powers, Department of Neurosurgery,

Research Period

May 20, 2002 to October 30, 2002


  1. To determine the patterns of referral from selected physicians through physician interviews and surveys.
  2. To utilize a cost analysis to determine the actual cost to the medical system as compared to an ideal cost based upon “best practice” referrals, diagnostic exams, and treatment protocols.
  3. To use the results of this study to educate physicians who initially see these patients in the proper protocol for referrals for patients with lower back pain.



Low back pain is a common problem among adult Americans (1-9), and can be a symptom of many different diseases and disorders, including herniated discs, pinched nerves, compression fractures, spinal stenosis, and less commonly infections and tumors of the vertebrae (6,7). Up to 90% of patients suffering from an acute attack of low back pain will recover within 3 months with only conservative therapy, overseen by their primary care physician (2,5,7). Conservative therapy is defined as treatment with any of the following in single use or in combination: stretching, exercise, analgesic and/or anti-inflammatory agents, narcotics if pain is severe, and physical therapy (3,7). Despite this statistic, a significant percentage of patients are referred to specialists for consultation, many prior to the completion of a course of conservative therapy. A limited number of patients with specific signs and symptoms need to be seen by a specialist (7), and a small percentage will actually need some sort of invasive treatment (i.e. surgery). While many studies have looked at patient outcomes when various specialists are involved in care of the patient with acute low back pain (1,8), there have been few studies to look strictly at referral trends in primary care physicians. This study will survey primary care physicians who refer their patients to the Division of Neurosurgery at the Milton S. Hershey Medical Center to determine trends in referral patterns among these physicians. These patterns will then be compared to a guideline developed through a literature review, and undergo a statistical analysis to determine the cost effectiveness of these practices.


Initial preparation for this project will involve the compilation of the names of primary care physicians who frequently refer patients to the Department of Neurosurgery at PSHMC. These physicians and others who work in the sane practice will then receive a survey by mail. The focus of the survey will be to see what factors influence these physicians’ referrals of patients within their practices who present with acute lower back pain. Factors may include severity of pain, time of onset, activity at time of onset, radiation of pain to legs, or demonstrable neurological deficit, including loss of motor or sensory function in lower extremities or loss of bladder and/or bowel control. The survey will also look at extent of physical exam, depth of history taking, time of conservative therapy prior to referral, and other factors. The information gleaned from these surveys will be used to determine if the factors that are influencing referrals from these physicians are indeed adequate, or if a more ideal guideline can be developed and implemented. If necessary, this guideline would be an aid to these physicians in their referrals, ensuring that the patient is optimally served by both their primary care physician and any necessary specialists.

The cost analysis of the referral patterns is a comparison between the actual cost and the ideal "best practice" cost. The best practice cost will be determined by looking at the ideal protocol for diagnosis, treatment, and referral of patients presenting to primary care physicians with acute lower back pain. This ideal protocol will be determined through a thorough literature review on the topic, as well as by looking at insurance company guidelines for referrals. Through a survey of four local insurance companies and Medicare, an average reimbursement will be assigned to each of the following: PCP office visits, neurosurgeon office visits, diagnostic tests (including plain radiograph, CAT scan, and diagnostic MRI), and various treatment options (PT, analgesics, surgery). These assigned values will then be used to compare actual costs incurred by referral practices and the costs that would be incurred if the ideal protocol was followed. If there is a discrepancy between the actual and ideal costs, this would infer a discrepancy between the actual referral practices and the ideal referral protocol, and through education of primary care physicians we may be able to induce them to follow the ideal protocol for referrals and thereby reduce the cost to the medical profession.

IRB approval and waiver of informed consent was granted – protocol # 2002-114EP.

Student's Responsibilities

  1. Formulate and distribute surveys to selected physicians.
  2. Compile groupings of patients by referring physician.
  3. Collect patient data from chart review on number of PCP office visits, specialist office visits, diagnostic tests, and treatments for patients in each PCP grouping.
  4. Determine referral patterns for groups of physicians and chiropractors through analysis of the survey results.
  5. Prepare an ideal protocol for referral through interviews and a thorough literature review.
  6. Prepare a cost analysis of actual cost to the medical community based on actual referral patterns compared to referrals based on the ideal protocol.
  7. Prepare research paper.


Sponsor's Responsibilities

  1. Aid in the determination of the ideal protocol based upon the data collected from the surveys and his work experience in this area
  2. Help with the cost analysis and interpretation if needed
  3. Evaluate and critique research paper



  1. Carey, Timothy S., Joanne Garrett, Anne Jackman, Curtis McLaughlin, John Fryer, Douglas R. Smucker, The North Carolina Back Pain Project.  1995.  The Outcomes and Costs of Care for Acute Low Back Pain among Patients Seen by Primary Care Practitioners, Chiropractors, and Orthopedic Surgeons.  The New England Journal of Medicine.  333(14):913-917.
  2. Croft, Peter R., Gary J. Macfarlane, Ann C. Papageorgiou, Elaine Thomas, Alan J. Silman.  1998.  Outcome of low back pain in general practice: a prospective study.  British Journal of Medicine. 316(7141):1356-1359.
  3. Deyo, Richard A.  1996.  Acute low back pain: a new paradigm for management: Limited imaging and an early return to normal activities.  British Medical Journal. 313(7069):1343-1344.
  4. Henley, Eric.  2000.  Understanding and Treating Low Back Pain in Family Practice.  The Journal of Family Practice.  49(9):793-795.
  5. Little, Paul, Lisa Smith, Ted Cantrell, Judith Chapman, John Langridge, Ruth Pickering.  1996.  General practitioner’s management of acute back pain: a survey of reported practice compared with clinical guidelines.  British Medical Journal. 312(7029):485-488.
  6. Lurie, Jon D., Paul D. Gerber, Harold C. Sox.  2000.  A Pain in the Back.  The New England Journal of Medicine.  343(10):723-726.
  7. Malanga, Gerard A. MD, Scoot F. Nadler, DO.  1999.  Nonoperative Treatment of Low Back Pain.  Mayo Clinic Proceedings.  74(11):1135-1148.
  8. Reis, Shmuel MD, Doron Hermoni, MD, Jeffrey M. Borkan, MD, Aya Biderman, MD, Chava Tabenkin, MD, Avi Porat, MD.  1999.  A New Look at Low Back Complaints in Primary Care; A RAMBAM Israeli Family Practice Research Network Study.  The Journal of Family Practice.  48(4):299-303.
  9. Suarez-Almazor, Maria E. MD, MSc, PhD, Elaine Belseck, BSCN, Anthony S. Russell, MB, BCH, FRCPC, John V. Mackel, MB, CCFP, FCFP.  1997.  Use of Lumbar Radiographs for the Early Diagnosis of Low Back Pain: Proposed Guidelines Would Increase Utilization.  The Journal of the American Medical Association.  277(22):1782-1786.