Relationship Between Travel Distance and Utilization of Breast Cancer Treatment in Rural Northern Michigan | JAMA | JAMA Network
[Skip to Navigation]
Sign In
January 2, 2002

Relationship Between Travel Distance and Utilization of Breast Cancer Treatment in Rural Northern Michigan

Author Affiliations

Not Available

Not Available

JAMA. 2002;287(1):111. doi:10.1001/jama.287.1.111-JMS0102-5-1


Two equally efficacious treatments are available for patients with stage I or II breast cancer —either modified radical mastectomy (MRM) or breast-conserving therapy (BCT), which generally involves wide excision (lumpectomy) followed by radiation therapy.1 Utilization rates of the 2 treatments in the United States vary widely, with rates of BCT ranging from 26.7% to 55.6% around a national average of 42%.2,3 The choice of BCT vs MRM has been found to be related to a number of variables, including tumor characteristics, physician and hospital attributes, geographic region, insurance, and patients' age, race, socioeconomic status, education, and treatment perceptions.4-7 Rural patients may be especially affected by the amount of travel required for radiation therapy. We studied the association between travel distance to radiation treatment and the utilization of BCT in a rural region of Michigan, where the nearest radiation oncology center may be 150 miles from patients' homes.


We retrospectively reviewed 81 medical records of patients treated for breast cancer from 1999 to 2000 in 3 community hospitals in northern Michigan. After excluding patients with prior invasive breast cancer or breast cancer other than stage I or II, 66 cases remained for analysis. We then divided the sample into 2 groups who lived either more or less than 45 miles from to the nearest radiation oncology facility, and compared the number of patients in each group who were treated with BCT vs MRM. We also compared age, marital status, employment status, and treatment surgeon between the 2 groups. Dichotomous variables were compared with χ2test, and the Spearman rank test was used for continuous variables. We also assessed the relationship between travel distance and treatment choice in the entire sample with the Kruskal-Wallis test, and the relationship between age and travel distance with the Pearson rank correlation test. All statistical analyses were performed with Minitab statistical software version 13.30 (Minitab, Inc, State College, Pa).


The mean age was 64.2 years (range, 30-92 years), and mean travel distance was 61.6 miles (range, 0-158 miles). Age and travel distance were not associated in the sample as a whole (Pearson rank correlation -.205, P = .099). The near and far groups were not significantly different in age (P = .301) or marital status (P = .856). To test the sensitivity of the 45-mile cutoff, we also recomputed these statistics using cutoffs of 25, 35, and 55 miles and still found no significant differences. The near group, however, was more likely to be employed (58.3% of near group employed vs 31.0% of the far group; χ2 = 4.86, P = .030).

BCT was utilized by 24.2% of all patients. In the entire sample, BCT utilization was negatively correlated with travel distance (Kruskal-Wallis test z = -2.18, H = 4.77, P = .029, P = .027 adjusted for ties). Of the near group 14 patients received MRM (58.3%) and 10 received BCT (41.7%). Of the far group, 36 patients received MRM (85.7%) and 6 received BCT (14.3%) (χ2 = 6.24, P = .013). Similar results were obtained with distance cut-off values of 25, 35 and 55 miles. Treatment utilization across the 2 groups was not associated with age (t = 1.04, P = .310), marital status (χ2 = .021, P = .885), occupational status (χ2 = 2.06, P = .152) or treatment surgeon (χ2 = 5.43, P = .366).


In our sample, patients with stage I or II breast cancer who lived at greater distances from a radiation oncology facility were more likely to undergo mastectomy. The overall rate of BCT of 24.2% in this rural sample is nearly half the US average. Although age, marital status, and employment were not associated with treatment in this sample, it is possible that these variables, as well as other unmeasured clinical or demographic factors, may mediate or explain this relationship.

Travel burdens include, but are not limited to, duration of travel, added travel expenses, and driving during winter months when roads may be hazardous and at times impassable. Physicians in rural counties may wish to discuss travel burden with their patients in the context of treatment decisions for early-stage breast cancer.

Acknowledgment: The authors thank Timothy Nelson, MD, Kathleen Thompson, PhD, for her help with the statistical analysis, and Janice Lubenow.
Early Breast Cancer Trialists Collaborative Group, Effects of radiotherapy and surgery in early breast cancer: an overview of the randomized trials.  N Engl J Med. 1995;3331444- 1455Google ScholarCrossref
Bland  KMenck  HScott-Conner  C  et al.  The national cancer data base 10-year survey of breast carcinoma treatment at hospitals in the United States.  Cancer. 1998;831262- 1273Google ScholarCrossref
Lazovich  DWhite  EThomas  DMoe  RTaplin  S Change in the use of breast-conserving surgery in western Washington after the 1990 NIH consensus development conference.  Arch Surg. 1997;132418- 423Google ScholarCrossref
Stafford  DSzczys  RBecker  RAnderson  JBushfield  S How breast cancer treatment decisions are made by women in North Dakota.  Am J Surg. 1998;176515- 519Google ScholarCrossref
Samet  JHunt  WFarrow  D Determinants of receiving breast-conserving surgery.  Cancer. 1994;732344- 2351Google ScholarCrossref
Albain  KSGreen  SRLichter  AS  et al.  Influence of patient characteristics, socioeconomic factors, geography and systemic risk on the use of breast-sparing treatment in women enrolled in adjuvant breast cancer studies: an analysis of two intergroup trials.  J Clin Oncol. 1996;143009- 3017Google Scholar
Kotwall  CACovington  DLRutledge  R  et al.  Patient, hospital, and surgeon factors associated with breast conservation surgery: a statewide analysis in North Carolina.  Ann Surg. 1996;224419- 426Google ScholarCrossref