Breast cancer is the most common cancer in women worldwide and is also a leading cause of cancer mortality (1). Breast cancer incidence varies internationally with a higher rate observed in Western countries however mortality rates vary less (2). Importantly, geographical inequalities exist within countries at all phases along the cancer care pathway (3-7), which has resulted in survival disparities (4).
Local control is important in breast cancer management as it can influence survival outcomes (8). Historically, mastectomy was the mainstay of primary breast cancer treatment. In the 1980s, several randomised control trials showed comparable survival outcomes for mastectomy and breast conserving therapy (BCT) which consists of breast conserving surgery (BCS) and radiotherapy. Since then, BCT has been established as the preferred modality of surgical treatment as it may lead to less morbidity and a better quality of life (9,10). More recent observational studies reported that survival outcomes were even better for BCT compared to mastectomy (11). Based on these findings, the question is raised of whether women suitable for BCT should still be offered a choice of mastectomy or BCT (9).
Women often choose mastectomy for individual reasons such as fear of cancer recurrence, worry about radiation side-effects, follow-up imaging and recall for further treatment (12). However, broader environmental factors such as travel distance to treatment facilities may also influence treatment decisions (13). This paper therefore aims to review the existing literature on geographic variations in the type of surgery received by women with early invasive breast cancer, focusing on three area-level indices—urban/rural status, accessibility (availability of and proximity to cancer services) and socioeconomic status.
The primary research question for this systematic review was whether the receipt of surgery for invasive breast cancer varies with indicators of access to care. To guide the review, specific secondary questions were: (I) are there urban-rural differences in the receipt and types of surgery for invasive breast cancer? (II) does the proximity and availability of cancer services matter in the receipt of surgery for invasive breast cancer? and (III) are there differences in the receipt and types of surgery for invasive breast cancer by area-level socio-economic status?
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement (14) was used to guide the reporting of this review.
Literature Search (information sources; search)
The literature search was carried out between 29 November and 06 December 2018, covering literature from the beginning for each included database till 2018. Searches were conducted across the following databases: Medline [1946–2018], EMBASE [1990–2018], CinhalPlus [1937–2018], Scopus [1970–2018], Cochrane [1992–2018], and Web of Science Core Collection [1900–2018]. Further international literature was sought through the use of WorldWideScience.org. Additionally, further secondary references were assessed for eligibility from the reference lists of publications screened for review.
The keywords and phrases used included breast cancer, breast carcinoma, breast tumors, rural, urban, urban-rural difference, place of residence, regional, travel distance, geographical or geographic, small-area analysis, variation, disparities, radiation therapy, surgery, and chemotherapy. The full electronic search strategy conducted on Medline is shown in Supplementary files. Note that the search terms used reflect the initial scope and focus of the review which covered all breast cancer treatments.
Eligibility criteria and study selection
Studies were included if they: (I) involved women diagnosed with early invasive breast cancer (stage I-IIIA); (II) were either population-based or institution-based; (III) assessed variations in the type of surgical treatment received based on area-level indices such as urban/rural status, accessibility (availability of and proximity to cancer services) and socioeconomic status, and (IV) published the findings in English as a full text article. Studies that merely reported small-area level or regional/state variations without considering the above-mentioned indices were excluded due to limited generalisability of the results.
All records identified through searched databases and other sources were firstly screened for duplicates. The remaining records were screened for relevance by title and abstract. Then the full text of the remaining articles were retrieved and screened for inclusion in the qualitative synthesis. Additional studies were identified from the references of screened articles. All full-text articles were identified by NC, checked by MSA and approved by STT, JZ and ME to ensure that they met the inclusion criteria and could be included for review.
Data extraction (data collection process and data items)
Data was extracted from each included article using a data collection form, and information on study setting, how the study population was identified and gathered, total patient population, databases/data sources used, area-level indices used, confounding variables controlled for, and main results were collected.
Quality assessment (risk of bias in individual studies and risk of bias across studies)
The quality and risk of bias of each study were assessed using the Newcastle-Ottawa Scale (NOS) for quality assessment of non-randomised studies in meta-analyses (15). The NOS has strong face and content validity and is endorsed by the Cochrane Collaboration. Ratings were made using a point system, where points were given for the selection of study groups; the comparability of the groups; and the assessment of outcomes. The NOS was modified for retrospective cohort studies by omitting the item assessing the presence of the outcome of interest at the beginning of the study. Eight criteria were assessed in the current systematic review: (I) representativeness of the exposed cohort; (II) selection of the non-exposed cohort; (III) ascertainment of exposure; (IV) adjustment for major confounders such as age and stage of diagnosis; (V) adjustment for other possible confounders; (VI) assessment of receipt of surgery; (VII) adequate follow-up period for receipt of surgery; and (VIII) adequacy of follow-up of cohorts. The total quality score was a maximum of eight points with higher scores indicative of higher study quality and less risk of bias.
Presentation of results
Results are presented in four tables, each addressing an indicator of care. The ‘results’ column gives the results as described in the paper. In addition, a column “key result” expresses that in a consistent fashion, as showing the association between higher use of BCS and urban setting, less distance to facilities, better facilities, and higher socio-economic status. Odds ratios are shown if they are given in the study: for the “key result” their reciprocal may be used.
Overall, the search resulted in 3,574 articles identified across Medline, EMBASE, CinhalPlus, Scopus, Cochrane, Web of Science Core Collection, and 783 articles from WorldWideScience.org. After removing duplicates, 3,109 articles remained and were subsequently screened. Seventy-two articles met the inclusion criteria after title and abstract screening. The full-text of these articles were then sought, and 39 were excluded for reasons shown in Figure 1. The remaining 33 articles were included in this review with a further 7 studies identified for addition from the reference lists of studies during eligibility screening, making a total of 40 (Figure 1).
Most studies  included in this review were from the United States (US), with studies conducted in North Carolina (16), Washington (17), Georgia (18), Kentucky (19-21), Virginia (22-24), Florida (25), South Dakota (26), New Hampshire (27), Michigan (28), and across all regions (29-46). Four other studies were conducted in Canada (47-50), one in China (51), one in Japan (52), and finally, three in Australia (53-55). All were retrospective cohort studies in design. Surgery types examined included the receipt of BCS, BCT, and mastectomy. Further details of the characteristics of the studies are presented in Tables 1-4.
|Author||Study area, country [level of geography]||Patient Identification/population or Institution||Sample size||Stage||Geographic measures||Type of surgery||Results||Confounders controlled for||NOS Score||Key result: higher BCS in urban: result or OR|
|Adelson et al.  (53)||NSW, Australia [State]||Registry/population||5,040||Local, regional, and metastatic||Regional Health Service (rural regions) vs. Area Health Service (urban regions)||Receipt of mastectomy||Adjusted OR for rural regions cf. urban regions =1.561; 95% CI, 1.218–2.001||Age, residence, hospital characteristics, private insurance status, spread of disease||8||Positive 1.56 (1.22–2.00)|
|Anderson et al.  (38)||7 States, United States [Census Tract]||Registry/population||6,505||Stage I-IIIA||Urban vs. rural areas||Receipt of BCS vs. mastectomy||Adjusted OR for women <65 in rural areas cf. urban areas =0.72; 95% CI, 0.53–0.96; adjusted OR for women ≥65 in rural areas cf. urban areas =0.68; 95% CI, 0.46–1.00||Age at diagnosis, race/ethnicity, health insurance, census tract poverty, census tract education, rural/urban residence, state of residence, surgical facility Commission on Cancer status, clinical tumour stage, tumour size, lymph node status, comorbidity, surgical approach, neoadjuvant chemotherapy||8||Positive 1.39 (1.04–1.89)|
|Answini et al.  (16)||North Carolina, United States [Counties]||Registry/population||3,349||Stage 0-II||Urban vs. rural counties||Receipt of BCS||Higher rate of BCS in urban counties than in rural counties||6||Positive|
|Azzopardi et al.  (55)||Australia [Australian Standard Geographical Classification Remoteness Structure]||Clinic/institution||21,643||Stage 0-II||Outer regions vs. cities||Receipt of BCS vs. mastectomy||Lower rates of BCS and higher rates of mastectomy in outer regions vs. cities||Geographic location, tissue diagnosis, surgical treatment, tumour size, histologic grade, number of tumours, lymph node status, lymphovascular invasion||7.5||Positive|
|Beaulieu et al.  (21)||Kentucky, United States [Counties]||Registry/population||8094||Stage I-II||Rural vs. urban counties||Receipt of BCS||Lower rates of BCS in rural counties||6||Positive|
|Craft et al.  (54)||Canberra, Australia [State]||Clinic/institution||2,081||Early Breast Cancer||Rural vs. metropolitan areas||Receipt of BCS vs. mastectomy||Lower in metro area OR 0.82 (0.64–1.07)||Age, residence, tumour size, tumour grade, nodal involvement, hormone receptor status, lymphovascular invasion, type of surgery, place of surgery||7.5||negative, n.s. 0.82 (0.74–1.07)|
|Dodgion et al.  (42)||United States [Census Tract]||Clinic/institution||4,766||Stage 0-IV||Hospitals located in metropolitan vs. non-metropolitan areas||Receipt of BCS vs. mastectomy vs. mastectomy with reconstruction||Lower rates of BCS, higher rates of mastectomy and lower rates of mastectomy with reconstruction at hospitals in non-metropolitan areas||Age, race, census median income, census educational level, tumour stage, nodal stage, oestrogen receptor, progesterone receptor, Her2 neu status, first therapeutic procedure, comorbidity, number of plastic surgeons per 100 annual breast procedures, plastic surgeons per 100,000 female population, radiation oncology services available, mean distance to nearest radiation facility, radiation oncologists per 100,000 female population, total bed size, total number of operating rooms, nursing staff ratio, hospital approval/accreditation, ambulatory surgery, chemotherapy administered, rural urban commuting area code||7.5||Positive|
|Dragun et al.  (19)||Kentucky, United States [Counties]||Registry/population||21,869||Stage 0-II||Rural vs. urban counties||Receipt of BCS vs. mastectomy||Adjusted OR of receiving mastectomy for rural counties cf. urban counties =1.073; 95% CI, 1.012–1.138||Surgery procedure, race, age at diagnosis, urban/rural status and Appalachian status, year at diagnosis, smoking history, insurance status, survival status at the end of the study, primary cancer sequence number, laterality, stage, nodes examined, ER/PR status, tumour grade, histology, and tumour size||8||1.07 (1.01–1.14)|
|Elward et al.  (22)||Virginia, United States [State]||Clinic/institution||1,512||Stage I-II||Urban hospital vs. rural hospital||Receipt of BCS||Higher rates of BCS in large urban hospitals and those with on-campus radiation services||Stage of disease||6.5||Positive|
|Fisher et al.  (50)||Alberta, Canada [State]||Registry/population||14,646||Stage I-III||Rural vs. urban regions||Receipt of BCS||Higher rates of BCS in urban centres (Calgary or Edmonton) cf. Central Alberta||Date of diagnosis, age at diagnosis, oestrogen and progesterone receptor status, cancer stage, type of surgery, geographic region of surgery, receipt of neo-adjuvant and adjuvant chemotherapy||8||Positive|
|Freeman et al.  (20)||Appalachian Kentucky, United States [State]||Registry/population||5,541||Stage I-II||Rural vs. urban areas||Receipt of mastectomy||Adjusted OR for rural location vs. urban location =2.075; 95% CI, 1.743–2.472||Surgery type, race, age at diagnosis, urban/rural status, insurance status, year at diagnosis, smoking status, stage of disease, oestrogen receptor/progesterone receptor status, tumour grade, histology, tumour size, nodes examined||8||2.08 (1.7–2.47)|
|Gilligan et al.  (33)||9 SEER regions, United States [Counties]||Registry/population||158,496||Local and regional||Rural vs. urban areas||Receipt of BCT||Lower rates of BCS whether accompanied by RT and LND in rural areas||Age, race, stage of disease, income, education, metropolitan statistical area, SEER site||8||Positive|
|Haggstrom et al.  (34)||SEER Regions, United States [Counties]||Registry/population||22,071||Stage I-II||Rural vs. metropolitan areas||Receipt of BCS||Adjusted OR for rural areas cf. metropolitan areas =0.72; 95% CI, 0.65–0.79||Race/ethnicity, age, location of residence, socioeconomic status, individual year of diagnosis, SEER region, tumour size, tumour stage, comorbidity||8||1.58 (1.26–1.97)|
|Izuo et al.  (52)||7 Regions, Japan [State]||Clinic/institution||31,686||Stage I-III||Rural vs. cities and urban regions||Receipt of BCS||Lower rates of BCS in rural regions||4.5||Positive|
|Jacobs et al.  (35)||SEER Regions, United States [Counties]||Registry/population||137,303||Stage I-III||Rural vs. urban areas||Receipt of mastectomy||Adjusted OR for rural areas cf. urban areas =1.58; 95% CI, 1.26–19.87; adjusted for density per population, and 1.69; 95% CI, 1.38–2.08; adjusted for density per area||Age, race, marital status, income, stage, employment, education, transportation of the study population, density of radiation oncologists, density of radiation technologists||8||1.58 (1.26–1.97)|
|LeMasters et al.  (41)||SEER Regions and West Virginia, United States [Counties]||Registry/population||26,917||Stage I-II||Urban vs. rural areas||Receipt of mastectomy and receipt of BCS only vs. BCT||Adjusted OR for urban areas cf. rural areas for mastectomy =0.71; 95% CI, 0.65–0.79
Adjusted OR for urban areas cf. rural areas for BCS only vs. BCT =0.73; 95% CI, 0.75–0.8
|Initial local treatment, year of diagnosis, age, frequency of Primary Care Provider visits, stage at diagnosis, ER status, PR status, tumour grade, area-level density of mammography screening and oncology treatment centres, specialisation of the treating surgeon(s), race, education, annual income, metro status||8||1.41 (1.54–1.27)|
|Liu et al.  (51)||Tianjin, China [City]||Clinic/institution||468||Stage I-II||Urban vs. rural areas||Receipt of BCT||Adjusted OR for urban areas cf. rural areas =2.14; 95% CI, 1.29–3.56||Age, educational level, occupation, family income, medical insurance, residence, source of BCT information, communication with the attending physician, tumour location, clinical stage, pathology, lymph node involvement, oestrogen receptor status||7.5||2.14 (1.29–3.56)|
|Markossian et al.  (18)||Georgia, United States [Counties]||Registry/population||23,500||Stage 0-IV||Rural vs. urban areas||Receipt of BCS||Adjusted OR for rural areas cf. urban areas =0.73; 95% CI, 0.58–0.91||Age at diagnosis, race, hormone receptor status, tumour characteristics, county of residence, AJCC stage, hormone receptor status, tumour grade, BC treatment, surgery type, county-level SES, survival status||8||1.37 (1.72–1.10)|
|Michalski, et al.  (30)||United States [Census Tract]||Registry/population||41,937||Stage I-III||Urban areas vs. rural areas||Receipt of BCS||Significant negative association with increasing rurality||Socioeconomic factors (median family income, mean proportion of college graduates, mean proportion of the population living below the poverty line, mean proportion of vacant housing units), race, stage of disease, population size of the zip code||8||Positive|
|Nattinger, et al.  (44)||United States [National]||Registry/population||36,982||Stage I-III||Non-metropolitan areas vs. metropolitan areas||Receipt of BCS||Significant positive association with increasing size of metropolitan areas||Race; population of metropolitan area, full-time house staff, geriatric services, radiation therapy facility||7||Positive|
|Samet et al.  (46)||9 SEER Regions, United States [Counties]||Registry/population||19,661||Stage I-III||City of ≥10,000 in county||Receipt of BCS||Adjusted OR for city of ≥10,000 in county cf. city of <10,000 in county =1.18; 95% CI, 1.10–1.28||Age at diagnosis, SEER region, number of physicians per 10,000 residents, percentage of residents with 16 or more years of education, percentage of families with incomes below the poverty level, presence of a city of 100,000 or larger within the county, and presence of a cancer treatment centre within the county, race||8||1.18 (1.10–1.28)|
|Smith et al.  (36)||United States [Counties]||Registry/population||56,725||Stage I-II||Metropolitan areas vs. non-metropolitan areas||Receipt of BCS||Higher rates of BCS in metropolitan areas cf. non-metropolitan areas||Age, race, comorbidity score, axillary lymph node involvement, axillary dissection, chemotherapy, screening mammography, physician visits, surgeon density, radiation oncologist density, metropolitan area, education, tumour stage||8||Positive|
“Key result” column positive means a higher BCS rate in urban areas, or the odds ratio and 95% limits for this association, if given. This is sometimes the reciprocal of that given in the paper. BCS, breast conserving surgery; BCT, breast conserving therapy; RT, radiotherapy; LND, lymph node dissection.
|Author||Study area, country [level of geography]||Patient identification/population or institution||Sample size||Stage||Geographic measures||Type of surgery||Results||Confounders controlled for||NOS Score||Key result: more BCS if closer|
|Baldwin et al.  (17)||Washington, United States [State]||Registry/population||1,188||Local and regional||Distance from the radiation therapy centre||Receipt of BCS and radiotherapy vs. mastectomy||Adjusted OR for women living ≥40 miles cf. women living <30 miles from radiotherapy centre =0.62 [P=0.10]; >50 miles cf. women living <30 miles =0.58 [P=0.09]||Preoperative radiation and medical oncology consultation, surgical procedure, age at diagnosis, ethnicity, marital status, median household income, disease stage at diagnosis, tobacco use, case severity, rural or urban residence, residence in nursing home||8||Positive: OR 1.61, n.s.|
|Boscoe et al.  (45)||10 States, United States [State]||Registry/population||104,730||Early stage breast cancer||Distance to surgery proximity||Receipt of mastectomy||Significant positive associations with surgery proximity||Age, year of diagnosis, race, ethnicity, census tract-level household poverty rate and rural/urban status, tumour sequence, commuting proximity||8||Positive|
|Celaya et al.  (27)||New Hampshire, United States [State]||Registry/population||2,861||Stage I-IIB||Distance to a radiation treatment facility||Receipt of BCS and radiotherapy||Significant negative associations with distance to a radiation treatment facility||Age at diagnosis, marital status, primary sequence, distance to radiation treatment facility, time of year, tumour size, stage at diagnosis||8||Positive|
|Elward et al.  (22)||Virginia, United States [State]||Clinic/institution||1,512||Stage I-II||Distance from the treating hospital||Receipt of BCS||No significant associations with distance from the treating hospital||Stage of disease||6.5||No association|
|Greenberg et al.  (37)||8 States, United States [State]||Clinic/institution||10,607||Stage I-II||Distance to radiation facility||Receipt of BCS vs. mastectomy with reconstruction||Significant positive associations of mastectomy with reconstruction, compared to BCS with increasing distance to radiation facility. OR per mile, 1.03 (1.00–1.05)||Age, systemic treatment, BMI, comorbidity score, income level, tumour size, number of positive nodes, HR status, AJCC stage, type of health insurance at presentation, employment status, number of reconstructive surgeons at each NCCN site, number of radiation oncologists on staff treating breast cancer at the beginning at each site||7.5||Positive|
|Hislop et al.  (47)||British Columbia, Canada [Census Tract]||Registry/population||942||Stage I-II||Distance to radiation treatment facility||Receipt of BCS||Significant negative association. Adjusted OR for <2 hours drive cf. 2+ hour drive =0.56; 95% CI, 0.36–0.76||Age at diagnosis, place of residence, median family income of neighbourhood, tumour size, histological type, tumour location in the breast, extent of ductal carcinoma in-situ, surgeon’s sex, caseload, graduation year from medical school, university affiliation, family physician’s graduation year from medical school and university affiliation||8||Positive 1.79 (1.32–2.78)|
|Lautner et al.  (39)||United States [national]||Registry/population||727,927||Stage I-IIIA||Distance to a treatment facility||Receipt of BCT||Adjusted OR for patients with a travel distance of less than 27.7 km =1.21; 95% CI, 1.16–1.26||Year of breast cancer diagnosis; age; race/ethnicity; comorbidity; insurance status; median income; population without high school diploma; facility type; facility location; distance from treatment facility; clinical T stage; clinical N stage; hormone receptor status||8||Positive: 1.21 (1.16-1.26)|
|Lin et al.  (26)||South Dakota, United States [Census Tract]||Registry/population||4,031||Stage I-II||Distance from a radiation therapy facility||Receipt of mastectomy||Significant positive association with distance to radiation therapy facility||Race/ethnicity, age at diagnosis, tumour grade, tumour sequence, year of diagnosis, poverty rate, urban/rural residence, travel distance to closest radiation therapy facility||8||Positive|
|Mandelblatt et al.  (40)||United States [Census Tract]||Registry/population||1,833||Stage I-II||Distance from nearest cancer centre||Receipt of mastectomy vs. BCT||For African-American patients, significant positive association. Adjusted OR for fourth quartile cf. first quartile =3.13; 95% CI, 1.22–8.08
No significant association for white patients
|Surgical treatment received, chemotherapy, tamoxifen race, age, comorbidity, stage, area percent of elderly below poverty, area percent of HS education, individual monthly income, individual education, marital status, supplemental insurance, mean urban-rural continuum score, mean number of radiology specialists per 100,000, mean number of surgeons per 100,000, median distance to cancer centre, region perceptions of racism and ageism||8||Positive: 3.13 (1.22–8.08) (Af-Amer) Np association (white)|
|Meden et al.  (28)||Michigan, United States [State]||Clinic/institution||66||Stage I-II||Travel distance to radiation oncology facility||Receipt of BCT||Significant negative association with distance to a radiation oncology facility||5.5||Positive|
|Nattinger et al.  (43)||SEER Regions, United States [Census Tract]||Registry/population||17,729||Stage I-II||Distance from a radiation therapy facility||Receipt of BCS||Significant negative association with increasing distance from a radiation therapy facility||Age, stage of disease, race, education status, distance to hospital with radiotherapy facility, population density, SEER site||8||Positive|
|Parviz et al.  (23)||Virginia, United States [State]||Clinic/institution||928||Stage 0-II||Distance from a radiation centre||Receipt of BCT||Adjusted OR for distance of >40 miles away from a radiation centre cf. distance of <40 miles away from a radiation centre =0.48; P=0.007||Age, race, insurance status, stage, year of diagnosis, treating surgeon, type of surgery||7.5||Positive, OR =2.08|
|Schroen et al.  (24)||Virginia, United States [State]||Registry/population||20,094||Stage I-III||Distance from nearest radiation facility||Receipt of mastectomy||Significant positive association with increasing distance from nearest radiation facility||Age, race, year of diagnosis, tumour extent of disease, tumour size, distance to nearest RT facility||8||Positive|
|Voti et al.  (25)||Florida, United States [Census Tract]||Registry/population||18,903||Stage I-IIA||Distance from the closest radiation therapy facility||Receipt of BCT||Significant negative association with increasing distance from the closest radiation therapy facility||Age at diagnosis, distance to radiation therapy facilities, marital status, insurance at the time of diagnosis, race/ethnicity||8||Positive|
|Xu et al.  (49)||Alberta, Canada [State]||Registry/population||21,872||Stage 0-III||Distance to the nearest radiotherapy centre||Receipt of mastectomy||Significant positive association with increasing distance to the nearest radiotherapy centre. However, association not significant for Calgary and Red Deer regions||Primary surgery type, region, driving time, period of diagnosis, age, AJCC stage, RT centre, adjuvant therapy, tumour size, lobular histology type, lymph node involvement, tumour grade, molecular subtype, comorbidities, surgical institution type, educational attainment of neighbourhood, neighbourhood annual income, re-excision, chemotherapy use, anti-endocrine use, radiotherapy use||8||Positive|
“Key result” column positive means a higher BCS rate with a lower distance to facilities, or the odds ratio and 95% limits for this association, if given. This is sometimes the reciprocal of that given in the paper. BCS, breast conserving surgery; BCT, breast conserving therapy; RT, radiotherapy.
|Author||Study area, country [level of geography]||Patient Identification/population or institution||Sample size||Stage||Geographic measures||Type of surgery||Results||Confounders controlled for||NOS score||Key result: higher BCS if more access|
|Ballard-Barbash et al.  (29)||9 SEER Regions, United States [Census Tract]||Registry/population||18,704||Stage I-IIB||Number of physicians and hospitals with radiation megavolt equipment in the area||Receipt of BCS||Adjusted OR for areas with 801-930 physicians per 10,000 population =1.32, 95% CI, 1.16–1.50 cf. areas with >930 physicians per 10,000 population||Age, stage of disease at diagnosis, ethnicity, comorbidity, health care characteristics, region||8||Negative: OR 0.76 (0.67–0.86)|
|Adjusted OR for areas with >8 hospitals with radiation therapy megavolt units cf. areas with <3 hospitals with radiation therapy megavolt units =1.35, 95% CI, 1.21–1.51||Positive: 1.35 (1.21–1.51)|
|Dodgion et al.  (42)||United States [Census Tract]||Clinic/institution||4,766||Stage 0-IV||Availability of radiation oncology services within the hospital or network||Receipt of BCS vs. mastectomy vs. mastectomy with reconstruction||Lower rates of BCS, higher rates of mastectomy and lower rates of mastectomy with reconstruction with lack of availability of radiation oncology services within the hospital or networks||Age, race, census median income, census educational level, tumour stage, nodal stage, oestrogen receptor, progesterone receptor, Her2 neu status, first therapeutic procedure, comorbidity, number of plastic surgeons per 100 annual breast procedures, plastic surgeons per 100,000 female population, radiation oncology services available, mean distance to nearest radiation facility, radiation oncologists per 100,000 female population, total bed size, total number of operating rooms, nursing staff ratio, hospital approval/accreditation, ambulatory surgery, chemotherapy administered, rural urban commuting area code||7.5||Positive|
|Elward et al.  (22)||Virginia, United States [State]||Clinic/institution||1,512||Stage I-II||Presence of senior||Receipt of BCS||Higher rates of BCS in the presence of Senior Membership Programs and Women’s Health Centres in the community||Stage of disease||6.5||Positive|
|Members Hip Programmes and Women’s Health Centres in the community||Higher rates of BCS in large urban hospitals and those with on-campus radiation services||Positive|
|Type of hospital||Similar rates of BCS in hospitals that reported availability of these services through referral or transport vs. those without available services||No association|
|LeMasters et al.  (41)||SEER Regions and West Virginia, United States [Counties]||Registry/population||26,917||Stage I-II||Density of mammography and oncology treatment centres||Receipt of mastectomy vs. BCT||Adjusted OR for areas with high density of mammography screening centres cf. areas with low density, for mast vs. BCT =0.98, 95% CI, 0.86–1.12||Initial local treatment, year of diagnosis, age, frequency of Primary Care Provider visits, stage at diagnosis, ER status, PR status, tumour grade, area-level density of mammography screening and oncology treatment centres, specialisation of the treating surgeon(s), race, education, annual income, metro status||8||No association: 1.02 (0.89–1.16)|
|Adjusted OR for areas with high density of oncology treatment centres cf. areas with low density for mastectomy vs. BCT =0.87, 95% CI, 0.76–0.99||Positive: 1.15 (1.01–1.32)|
|Samet et al.  (46)||9 SEER Regions, United States [Counties]||Registry/population||19,661||Stage I-III||Physician-to-population ratio||Receipt of BCS||Adjusted OR for 26.3 or more physicians per 10,000 cf. fewer than 16.5 per 10,000=1.40, 95% CI 1.27–1.53||Age at diagnosis, SEER region, number of physicians per 10,000 residents, percentage of residents with 16 or more years of education, percentage of families with incomes below the poverty level, presence of a city of 100,000 or larger within the county, and presence of a cancer treatment centre within the county, race||8||Positive: 1.40 (1.27–1.53)|
|Presence of a cancer centre||Adjusted OR for cancer centre in county cf. no cancer centre in county =1.24, 95% CI, 1.12–1.37||Positive: 1.24 (1.12–1.37)|
|Smith et al.  (36)||United States [Counties]||Registry/population||56,725||Stage I-II||Density of radiation oncologists||Receipt of BCS||Adjusted OR for higher density of radiation oncologists cf. lower density of radiation oncologists =1.37, 95% CI, 1.07–1.75||Age, race, comorbidity score, axillary lymph node involvement, axillary dissection, chemotherapy, screening mammography, physician visits, surgeon density, radiation oncologist density, metropolitan area, education, tumour stage||8||Positive: 1.37 (1.07–1.75)|
BCS, breast conserving surgery; BCT, breast conserving therapy; RT, radiotherapy. “Key result” column positive means a higher BCS rate with better access, or the odds ratio and 95% limits for this association, if given. This is sometimes the reciprocal of that given in the paper.
|Author||Study area, country [level of geography]||Patient identification/population or institution||Sample size||Stage||Geographic measures||Type of surgery||Results||Confounders controlled for||NOS Score||Key result: more BCS if SES higher|
|Anderson et al.  (38)||7 States, United States [Census Tract]||Registry/population||6,505||Stage I-IIIA||Census tract education||Receipt of BCS vs. mastectomy||Adjusted OR for women <65 years in areas with high census tract education cf. areas with low census tract education =1.20, 95% CI, 0.93–1.54||Age at diagnosis, race/ethnicity, health insurance, census tract poverty, census tract education, rural/urban residence, state of residence, surgical facility Commission on Cancer status, clinical tumour stage, tumour size, lymph node status, comorbidity, surgical approach, neoadjuvant chemotherapy||8||Positive|
|Adjusted OR for women ≥65 years in areas with high census tract education cf. areas with low census tract education =1.54, 95% CI, 1.09–2.16|
|Ballard-Barbash et al.  (29)||9 SEER Regions, United States [Census Tract]||Registry/population||18,704||Stage I-IIB||Census tract education||Receipt of BCS||Adjusted OR for census tracts with high education levels cf. census tracts with low education levels =1.15, 95% CI, 1.05–1.28||Age, stage of disease at diagnosis, ethnicity, comorbidity, health care characteristics, region||8||Positive|
|Gilligan et al.  (33)||9 SEER Regions, United States [Counties]||Registry/population||158,496||Local and regional||County income||Receipt of BCT||Lower rates of BCS whether accompanied by RT and LND in poorer counties||Age, race, stage of disease, income, education, metropolitan statistical area, SEER site||8||Positive|
|County education||Higher rates of BCT accompanied by RT and LND in better educated counties|
|No significant association if RT, LND or both omitted|
|Goel et al.  (48)||British Columbia and Ontario, Canada [State]||Registry/population||942||Node-negative breast cancer||Area median income level||Receipt of BCS||Lower rates of BCS in areas with lower median incomes. However, only significant in British Columbia||Age, tumour size and location||8||Positive|
|Haggstrom et al.  (34)||SEER Regions, United States [Counties]||Registry/population||22,071||Stage I-II||Area socioeconomic status||Receipt of BCS||Significant positive association with increasing area socioeconomic status||Race/ethnicity, age, location of residence, socioeconomic status, individual year of diagnosis, SEER region, tumour size, tumour stage, comorbidity||8||Positive|
|Lin et al.  (26)||South Dakota, United States [Census Tract]||Registry/population||4,031||Stage I-II||Census tract federal poverty line||Receipt of mastectomy||Adjusted OR for census tracts with greater than 15% of the population under the federal poverty line cf. census tracts with less than 5% of the population under the federal poverty line =1.28, 95% CI, 1.02–1.61||Race/ethnicity, age at diagnosis, tumour grade, tumour sequence, year of diagnosis, poverty rate, urban/rural residence, travel distance to closest radiation therapy facility||8||Positive|
|Mandelblatt et al.  (40)||United States [Census Tract]||Registry/population||1,833||Stage I-II||Area percent of elderly below poverty level quartiles||Receipt of mastectomy vs. BCT||Adjusted OR for areas highest quartile of elderly residents living below poverty level cf. lowest quartile of poverty =7.59, 95% CI, 2.81–20.5||Surgical treatment received, chemotherapy, tamoxifen race, age, comorbidity, stage, area percent of elderly below poverty, area percent of HS education, individual monthly income, individual education, marital status, supplemental insurance, mean urban-rural continuum score, mean number of radiology specialists per 100,000, mean number of surgeons per 100,000, median distance to cancer centre, region perceptions of racism and ageism||8||Positive|
|McGinnis, et al.  (32)||United States [State]||Registry/population||191,714||Stage I-III||Zip code income level||Receipt of mastectomy||Higher rates of partial mastectomy with or without radiation or systemic therapy in higher income zip codes||Stage of disease, race||8||Positive|
|Higher rates of modified radical mastectomy in lower income zip codes|
|Michalski et al.  (30)||United States [Census Tract]||Registry/population||41,937||Stage I-III||Area median family income||Receipt of BCS||Adjusted OR for areas with highest median family income cf. areas with lowest median family income =1.1, 95% CI, 1.0–1.2||Socioeconomic factors (median family income, mean proportion of college graduates, mean proportion of the population living below the poverty line, mean proportion of vacant housing units), race, stage of disease, population size of the zip code||8||Positive|
|Area percentage of college graduates||Adjusted OR for areas with the greatest percentage of college graduates cf. areas with the lowest percentage of college graduates =1.2, 95% CI, 1.1–1.4||Positive|
|Area proportion of persons living below the poverty level||Adjusted OR for areas with the greatest proportion of persons living below the poverty line =0.8, 95% CI, 0.7–0.9||Positive|
|Riley et al.  (31)||United States [Census Tract]||Registry/population||28,608||Stage I-IIB||Area education level||Receipt of BCS||Adjusted OR for areas with lowest quartile education level cf. areas with highest quartile education level =0.80, 95% CI, 0.73–0.87||Age, race, state and county of residence, cancer history, year of diagnosis, percentage of adults with fewer than 12 years of education at the census tract level, stage at diagnosis, tumour size, treatment pattern||8||Positive|
|Samet et al.  (46)||9 SEER Regions, United States [Counties]||Registry/population||19,661||Stage I-III||Percent of county with college education||Receipt of BCS||Adjusted OR for areas with 24.0 or more percent of county with college education cf. areas with less than 16.5% of county with college education =1.68, 95% CI, 1.53–1.85||Age at diagnosis, SEER region, number of physicians per 10,000 residents, percentage of residents with 16 or more years of education, percentage of families with incomes below the poverty level, presence of a city of 100,000 or larger within the county, and presence of a cancer treatment centre within the county, race||8||Positive|
|Percent of county below poverty level||Adjusted OR for areas with 8.6 or more percent of county under poverty level cf. less than 6.1 percent of county below poverty level =0.72, 95% CI, 0.66–0.78||Positive|
“Key result” column positive means a higher BCS rate with higher socio-economic status, or the odds ratio and 95% limits for this association, if given. This is sometimes the reciprocal of that given in the paper. BCS, breast conserving surgery; BCT, breast conserving therapy; RT, radiotherapy; LND, lymph node dissection.
Out of the 40 studies, 22 performed analyses at the small area level (e.g., census tract, counties, and cities), 16 at the state level, and two at the national level.
Overall, study quality was good, with many studies  receiving quality scores of 7 or 8. The population-based studies were scored 1 for the first criterion (representativeness of the exposed cohort) and institution-based studies were scored 0.5. Most studies used registry and surgical records to examine exposures and outcomes, and controlled for important confounding factors such as age and tumour stage, along with various other factors such as median family income, ethnicity, year of diagnosis, and insurance status.
Results of individual studies
Twenty-three articles examined variations in the receipt of different surgical treatments among women living in urban and rural regions (16,18-22,30,33-36,38,41,42,44,46,48,50-55), and were conducted in the US , Australia , Canada , China , and Japan  (Table 1).
All but one of these studies found that patients living in urban areas were significantly more likely to receive BCS or other more selective surgery, compared to mastectomy, and in the 11 studies where odds ratios are given, the urban excesses were always significant. Only one study (54) found a lower frequency of BCS in an urban area, in Australia, although this was not significant: the study found however that in rural areas radiotherapy was less often used after BCS. The odds ratios reported depend on the categorisations used and may not be comparable: they varied from small variations up to doubling of the odds of receiving BCS in studies in urban areas in Kentucky (20) and in China (51).
Proximity to health care services
Fifteen studies in the US  and Canada  examined the association between travel distance and the type of surgery received by breast cancer patients (17,22-28,37,39,40,43,45,47,49) (Table 2). Fourteen studies found significant differences in the receipt of BCS/BCT versus mastectomy, with women living further away from radiation treatment facilities and treatment centres less likely to receive BCS and more likely to receive mastectomy (17,23-28,37,39,40,43,45,47,49). Only one study, in Virginia, US, found no significant associations with increasing distance from a treating hospital, but that study did find that large urban hospitals had significantly higher rates of BCT than smaller hospitals (22). However, in a US study of 1,833 women (40) the association was seen for African Americans, in whom it was strong, and no significant association was found in white patients. In one US study, it was initially found that those who lived further away from a radiation therapy facility were more likely to receive mastectomy (24), and when a new radiation facility was opened in the rural area, the rate of mastectomy fell from 61% to 45% among patients who lived within 15 miles of the new facility (24).
Other measures of the availability of health care facilities
Six studies, all conducted in the US, identified variations in the type of breast cancer surgery received by women according to measures other than distance in the availability of health care facilities (22,29,36,41,42,46) (Table 3). Most studies showed that women living in areas with more health facilities were more likely to have BCS rather than mastectomy. Thus, higher rates of BCS were seen in areas with a high density of radiation oncologists and a greater number of physicians (36,46), areas with a cancer care centre (46), and a high density of oncology treatment centres (41). Similarly, older patients who lived in areas with the presence of Senior Membership Programmes and Women’s Health Centres had higher rates of BCS (22). Areas with less availability of radiation oncology services within hospitals or hospital networks had higher rates of mastectomy and lower rates of mastectomy with reconstruction (42). Women from areas with a greater number of hospitals with radiation therapy megavolt equipment were more likely to receive BCS (29). Only a few findings were inconsistent: in this last study (29), women were more likely to receive BCS if they were in areas with a lower rate of physicians per 10,000 population. One study (41) also found that women in areas with a higher density of mammography screening centres are more likely to receive RT after BCS (41).
Area level socioeconomic status
Eleven studies, 10 in the US and one in Canada, examined and identified differences in the type of surgery received by breast cancer patients residing in areas of different socioeconomic status (26,29-34,38,40,46,48) (Table 4). All 11 studies found that women living in higher socioeconomic areas were more likely to receive BCS. In areas with a greater proportion of college graduates and higher education levels, patients were significantly more likely to receive BCS/BCT than mastectomy (29-31,33,38,46). Similarly, patients living in areas with a higher median income were more likely to receive BCS (30-34). Conversely, patients living in areas with a greater proportion of people living below the poverty line and in areas with lower median incomes were less likely to receive BCS and were significantly more likely to receive a mastectomy (26,30,32,33,40,46,48).
This systematic review found that significant geographic variations existed in the receipt of different types of surgery in women diagnosed with early breast cancer. The evidence was generally consistent across the US as well as in the few studies from Canada, Australia, China, and Japan.
To our knowledge, this is the first systematic review on geographic variations assessed by area-level indicators in the types of surgery received by breast cancer patients. The review was limited to articles published in English only. As the majority of studies were published in the US (31 out of 40 studies included for review), the generalisability of our results may be limited as the healthcare systems are vastly different across countries, both in terms of funding and structure. As this review was based on peer-reviewed studies, we did not consider grey literature such as cancer registry reports or studies based on individual level data. We also excluded the studies that reported geographic differences only within a small area. We used the NOS scale to assess quality; this may not be sufficiently discriminating, as most studies received a high score. Several studies showed the size of these differences by odds ratios, which generally suggested variations of 20–50%, but the odds ratios were not comparable as they depended on the categorisation of the factors considered. Therefore, we do not present a quantitative meta-analysis.
We identified geographic variations in surgery types, in terms of urban and rural settings, distance and other aspects of accessibility to cancer care, and area-level socio-economic status. In general, women living in urban areas, in close proximity to cancer care facilities, particularly radiation centres, and in more affluent neighbourhoods were more likely to receive BCS, and less likely to receive mastectomy. Similar findings have been reported in other narrative reviews on this topic (3,6,56). Ayanian et al. reported that patients living in urban and metropolitan areas were more likely to receive BCS than patients living in rural areas, with the likelihood of receipt of BCS being strongly associated with the size of the metropolitan region (6). Two other reviews also found that breast cancer patients were less likely to receive BCS if they had greater travel distance from cancer treatment centres and radiotherapy facilities (3,56). Other reviews have identified geographic disparities across the breast cancer care continuum, from diagnosis to survival (4,6,7,56). Overall, the evidence to date shows that area of residence and accessibility to cancer care play an important role in the types of cancer treatment received by breast cancer patients.
Our review of peer-reviewed literature showed that most studies were from the US. The US health care system has greater disparities in access to care than other developed countries (57). The few studies from other countries suggest their situation may be different. The only study not showing an urban-rural difference was from Australia (54), although the other Australian study did show an urban effect (53); in Ontario, Canada, differences were non-significant (48), although urban increases in BCS were seen in British Columbia and in Alberta (48,50). However, higher rates of BCS were seen in higher socio-economic groups in British Columbia, as in the several US studies (48). In a related study, women in the lowest socio-economic groups in Ontario and in California were compared, and the Ontario women were more likely to receive BCS (58); this study was not included as it did not give results comparing different socio-economic groups within either area. Our review found no studies from Europe; perhaps the question is regarded as unimportant there, or studies were published only in grey literature or non-English language journals. There are also no studies from low or middle-income countries. It would be valuable to have more information from other countries; specifically, studies of health systems in which these variations in care may potentially not occur would be useful.
The provision of BCS plus RT as an alternative to mastectomy requires a more comprehensive and costly service. A surgeon working alone or in a small practice can provide mastectomy. To give BCS plus RT requires a team approach including a radiotherapist, RT facilities, and extended care over several weeks. Thus, the indicators of urban residence and closeness to large facilities or to more sophisticated services used in these studies likely all relate to whether the women gets her care at a facility offering a true choice of treatments. The other key issue is patient choice, and distance from facilities and socio-economic status are probably indicators of the patient’s ability to accept the weeks of regular RT and cope with the demands on her family and work commitments and the costs (59). The other key influence on choice of treatment is how the options are viewed by the doctors involved and by the patient, and how that discussion is framed and how the decision is made (59). The use of BCS rather than mastectomy has been used sometimes as a clinical care quality indicator, but ideally patient choice and shared decision-making should be also considered (59,60).
Geographical variations and specifically urban-rural differences in care provision are a major challenge for all governments worldwide, and are a priority in cancer control plans and service provision. For example, in 2019 the American Society for Clinical Oncology announced a new task force to address the “rural cancer care gap” in the US (61). Inequities related to access to care exist in participation in screening programs, stage distribution and late diagnosis, and the use of most types of treatment. Radiotherapy is the most difficult treatment to provide equitably, as it requires expensive equipment, dedicated premises, and a specialised workforce. It is inevitable that many patients will live far from the nearest radiotherapy centre. The national radiotherapy plan for England proposes that all patients should live within 45 min travel time to their nearest centre, and so new satellite centres have been developed (62). Surgical services are more widely distributed, but quality varies, and there is good evidence that surgeons with higher cancer workloads give better outcomes (63), so access to the best service may be difficult. In Australia, BCS was less frequently used by women whose surgeons have a low case load, independently of the effect of rural residence (64). In northern Italy, lower use of BCS was seen in women treated at hospitals with low surgical volumes, as well as in those living far from radiotherapy facilities (65). Chemotherapy services can be provided locally if there is a system of training specialised nurses, and ensuring protocols used in the main centres are followed. Systems of giving primary care physicians extra training to enable them to supervise such services can be beneficial. Such a program exists in British Columbia, Canada, where rural patients were less likely to have BCS, but did not have reduced chemotherapy or hormonal therapy (66). Even primary care services for cancer have difficulties, and education and support for rural practitioners can be beneficial (67).
It seems likely that variations in the type of breast surgery used are largely driven by the requirement for radiotherapy after BCS and hence the issues of access to radiotherapy, which normally requires multiple visits over many weeks. New developments such as intra-operative radiotherapy, where radiotherapy is given only at the time of surgery, could improve access, but may be only appropriate for selected patients (68). It is understandable that women in rural areas more often opt to have mastectomy, thus avoiding the need for radiotherapy. However, the size of this variation may seem surprising. In Queensland, the use of BCS in very low access areas was less than half of that in the best access areas (13). In a study in 10 states in the US, the use of BCS was reduced by 30% in women living more than 75 km from a radiotherapy centre (45), with similar findings in South Dakota; but in New Jersey, similar disparities were seen in those living more than 15 km or 19 min travel time away (69). Distance from the centres relates to travel time, cost, and inconvenience. In New Hampshire, a northern US state with severe winters, the use of BCS was lower with greater travel time, and also lower in winter. In Northern England, the use of BCS was not related to travel time in general, but was lower for women living in areas with no regular bus service (70).
It would be expected that variations in treatment by distance would be more marked for less affluent patients, in terms of direct cost, but also in terms of difficulties in employment and childcare. Few studies have appeared to look at this interaction, but several have reported lower BCS use in the lower socio-economic groups. In Florida, a reduction in BCS with increasing distance was seen in older women, but not in those under 50 years (71).
While these access issues have been well described, there seem to have been few interventions to improve them. In one area of New South Wales, the use of BCS showed no change after a free transport service to the nearest radiotherapy facilities (68 and 86 km distant) was introduced, but increased significantly when a local free radiotherapy service was started, with the largest increase for patients over age 70 (72).
This systematic review found that for women with early invasive breast cancer, higher rates of breast-conserving surgery rather than mastectomy were consistently associated with urban location, closeness to facilities, more advanced facilities, and higher socio-economic status. These variations were seen in many studies from the US, several having good control for clinical factors such as stage of disease, although our review did not address patient choice. Our findings highlighted that inequalities in care exist which may be substantial. The lack of studies from countries other than the US limits the ability to understand the extent of inequalities in other countries. There are no studies from low- or middle-income countries. Variations in care given, if not for clinical reasons or for informed patient choice, need to be documented and addressed in cancer care planning.
Medline (1946-2018) - 455 studies/articles identified
1. Search terms used
1. Breast cancer.mp.
2. Breast carcinoma.mp.
3. Breast neoplasm*.mp.
4. Breast Neoplasms/
5. Breast tumor*.mp.
6. Breast tumour.mp.
7. 1 or 2 or 3 or 4 or 5 or 6
10. Urban-rural difference*.mp.
15. Remote population.mp.
18. Geographic pattern.mp.
19. Geographic loca*.mp.
21. Spatial analysis/
22. Spatial analys*.mp.
23. Place of residency.mp.
24. Place of residence.mp.
25. Rural Population/
26. Rural population.mp.
27. Urban population/
28. Urban population.mp.
29. Rural health/
30. Rural health services/
31. Hospitals, rural/
32. Urban health/
33. Hospitals, urban/
34. Travel distance.mp.
35. Travel time.mp.
36. Travel burden.mp.
37. Health service access*.mp.
38. Healthcare access*.mp.
39. Health service* accessibility.mp.
40. Health services accessibility/
44. Health service utili#ation.mp.
45. Geographic information system*.mp.
46. Geographic information systems/
47. Geospatial analysis.mp.
48. Geospatial analyses.mp.
49. Geographic mapping/
50. Small-area analysis/
51. Small area.mp.
52. Small area analys#s.mp.
53. Cancer map.mp.
54. 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42 or 43 or 45 or 46 or 47 or 48 or 49 or 50 or 51 or 52 or 53
58. Healthcare Disparities/
59. Health care disparities.mp.
65. 55 or 56 or 57 or 58 or 59 or 60 or 61 or 62 or 63 or 64
68. Drug Therapy/
70. Systemic treatment*.mp.
71. Systemic therapy.mp.
72. Systemic therapies.mp.
73. Local treatment*.mp.
74. Local therapy.mp.
75. Local therapies.mp.
76. hormon* therapy.mp.
77. hormon* therapies.mp.
78. Locoregional treatment*.mp.
79. Breast cancer surgery.mp.
82. Breast conserving surgery.mp.
83. Mastectomy, Segmental/
85. Axillary surgery.mp.
86. General Surgery/
88. Biological treatment*.mp.
89. Targeted therapy.mp.
90. Targeted therapies.mp.
92. 66 or 67 or 68 or 69 or 70 or 71 or 72 or 73 or 74 or 75 or 76 or 77 or 78 or 79 or 80 or 81 or 82 or 83 or 84 or 85 or 86 or 87 or 88 or 89 or 90 or 91
96. Patient Compliance/
97. “Treatment Adherence and Compliance”/
98. 93 or 94 or 95 or 96 or 97/
99. 7 and 54 and 65 and 92 and 98
Funding: Support for this review was provided by the generous contribution of the University of Auckland School of Medicine Foundation and Auckland Cancer Society Research Centre.
Provenance and Peer Review: This article was commissioned by the Guest Editors (Peter Baade, Susanna Cramb) for the series “Spatial Patterns in Cancer Epidemiology” published in Annals of Cancer Epidemiology. The article has undergone external peer review.
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at http://dx.doi.org/10.21037/ace.2020.02.01). The series “Spatial Patterns in Cancer Epidemiology” was commissioned by the editorial office without any funding or sponsorship. The authors have no other conflicts of interest to declare.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.
- World Health Organization. Breast Cancer. 2019. Available online: https://www.who.int/cancer/prevention/diagnosis-screening/breast-cancer/en/
- Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015;136:E359. [Crossref] [PubMed]
- Bigby J, Holmes MD. Disparities across the Breast Cancer Continuum. Cancer Causes Control 2005;16:35-44. [Crossref] [PubMed]
- Purushotham AD, Pain S, Miles D, et al. Variations in treatment and survival in breast cancer. Lancet Oncol 2001;2:719-25. [Crossref] [PubMed]
- Murage P, Crawford SM, Bachmann M, et al. Geographical disparities in access to cancer management and treatment services in England. Health Place 2016;42:11-8. [Crossref] [PubMed]
- Ayanian JZ, Guadagnoli E. Variations in breast cancer treatment by patient and provider characteristics. Breast Cancer Res Treat 1996;40:65-74. [Crossref] [PubMed]
- Maddison AR, Asada Y, Urquhart R. Inequity in access to cancer care: a review of the Canadian literature. Cancer Causes Control 2011;22:359-66. [Crossref] [PubMed]
- Wong JS, Harris JR. Importance of local tumour control in breast cancer. Lancet Oncol 2001;2:11-7. [Crossref] [PubMed]
- Engel J, Kerr J, Schlesinger A, et al. Quality of Life Following Breast-Conserving Therapy or Mastectomy: Results of a 5-Year Prospective Study. Breast J 2004;10:223-31. [Crossref] [PubMed]
- Johns N, Dixon JM. Should patients with early breast cancer still be offered the choice of breast conserving surgery or mastectomy? Eur J Surg Oncol 2016;42:1636-41. [Crossref] [PubMed]
- Onitilo AA, Engel JM, Stankowski RV, et al. Survival Comparisons for Breast Conserving Surgery and Mastectomy Revisited: Community Experience and the Role of Radiation Therapy. Clin Med Res 2015;13:65-73. [Crossref] [PubMed]
- Dorval M, Maunsell E, Deschênes L, et al. Type of mastectomy and quality of life for long term breast carcinoma survivors. Cancer 1998;83:2130-8. [Crossref] [PubMed]
- Baade PD, Dasgupta P, Youl PH, et al. Geographical Inequalities in Surgical Treatment for Localized Female Breast Cancer, Queensland, Australia 1997-2011: Improvements over Time but Inequalities Remain. Int J Environ Res Public Health 2016;13:ijerph13070729.
- Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. [Crossref] [PubMed]
- Wells GA, Shea B, O'Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. 2019. Available online: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp
- Answini GA, Woodard WL III, Norton HJ, et al. Breast conservation: Trends in a major southern metropolitan area compared with surrounding rural counties. Am Surg 2001;67:994-8. [PubMed]
- Baldwin LM, Taplin SH, Friedman H, et al. Access to multidisciplinary cancer care: is it linked to the use of breast-conserving surgery with radiation for early-stage breast carcinoma? Cancer 2004;100:701-9. [Crossref] [PubMed]
- Markossian TW, Hines RB. Disparities in Late Stage Diagnosis, Treatment, and Breast Cancer-Related Death by Race, Age, and Rural Residence Among Women in Georgia. Women Health 2012;52:317-35. [Crossref] [PubMed]
- Dragun AE, Huang B, Tucker TC, et al. Increasing Mastectomy Rates Among all Age Groups for Early Stage Breast Cancer: A 10‐Year Study of Surgical Choice. Breast J 2012;18:318-25. [Crossref] [PubMed]
- Freeman AB, Huang B, Dragun AE. Patterns of Care With Regard to Surgical Choice and Application of Adjuvant Radiation Therapy for Preinvasive and Early Stage Breast Cancer in Rural Appalachia. Am J Clin Oncol 2012;35:358-63. [Crossref] [PubMed]
- Beaulieu JE, Massey CS, Tucker TC, et al. Rural-urban variation in breast-conserving surgery in Kentucky. J Ky Med Assoc 2003;101:455. [PubMed]
- Elward KS, Penberthy LT, Bear H, et al. Variation in the use of breast-conserving therapy for Medicare beneficiaries in Virginia: clinical, geographic, and hospital characteristics. Clin Perform Qual Health Care 1998;6:63. [PubMed]
- Parviz M, Cassel JB, Kaplan BJ, et al. Breast conservation therapy rates are no different in medically indigent versus insured patients with early stage breast cancer. J Surg Oncol 2003;84:57-62. [Crossref] [PubMed]
- Schroen AT, Brenin DR, Kelly MD, et al. Impact of Patient Distance to Radiation Therapy on Mastectomy Use in Early-Stage Breast Cancer Patients. J Clin Oncol 2005;23:7074-80. [Crossref] [PubMed]
- Voti L, Richardson LC, Reis IM, et al. Treatment of local breast carcinoma in Florida: the role of the distance to radiation therapy facilities. Cancer 2006;106:201-7. [Crossref] [PubMed]
- Lin Y, Wimberly MC, Da Rosa P, et al. Geographic access to radiation therapy facilities and disparities of early-stage breast cancer treatment. Geospat Health 2018;13:622. [PubMed]
- Celaya MO, Rees JR, Gibson JJ, et al. Travel Distance and Season of Diagnosis Affect Treatment Choices for Women with Early-Stage Breast Cancer in a Predominantly Rural Population (United States). Cancer Causes Control 2006;17:851-6. [Crossref] [PubMed]
- Meden T, St John-Larkin C, Hermes D, et al. Relationship Between Travel Distance and Utilization of Breast Cancer Treatment in Rural Northern Michigan. JAMA 2002;287:111. [Crossref] [PubMed]
- Ballard-Barbash R, Potosky AL, Harlan LC, et al. Factors Associated With Surgical and Radiation Therapy for Early Stage Breast Cancer in Older Women. Journal of the National Cancer Institute 1996;88:716-26. [Crossref] [PubMed]
- Michalski TA, Nattinger AB. The influence of black race and socioeconomic status on the use of breast-conserving surgery for Medicare beneficiaries. Cancer 1997;79:314. [Crossref] [PubMed]
- Riley GF, Potosky AL, Klabunde CN, et al. Stage at Diagnosis and Treatment Patterns Among Older Women With Breast Cancer: An HMO and Fee-for-Service Comparison. JAMA 1999;281:720-6. [Crossref] [PubMed]
- McGinnis LS, Menck HR, Eyre HJ, et al. National Cancer Data Base survey of breast cancer management for patients from low income zip codes. Cancer 2000;88:933-45. [Crossref] [PubMed]
- Gilligan MA, Kneusel RT, Hoffmann RG, et al. Persistent Differences in Sociodemographic Determinants of Breast Conserving Treatment despite Overall Increased Adoption. Medical Care 2002;40:181-9. [Crossref] [PubMed]
- Haggstrom DA, Quale C, Smith‐Bindman R. Differences in the quality of breast cancer care among vulnerable populations. Cancer 2005;104:2347-58. [Crossref] [PubMed]
- Jacobs LK, Kelley K, Rosson G, et al. Disparities in Urban and Rural Mastectomy Populations. Ann Surg Oncol 2008;15:2644-52. [Crossref] [PubMed]
- Smith GL, Xu Y, Shih YT, et al. Breast-Conserving Surgery in Older Patients with Invasive Breast Cancer: Current Patterns of Treatment Across the United States. J Am Coll Surg 2009;209:425-33.e2. [Crossref] [PubMed]
- Greenberg CC, Lipsitz SR, Hughes ME, et al. Institutional variation in the surgical treatment of breast cancer: a study of the NCCN. Ann Surg 2011;254:339-45. [Crossref] [PubMed]
- Anderson RT, Morris CR, Kimmick G, et al. Patterns of locoregional treatment for nonmetastatic breast cancer by patient and health system factors. Cancer 2015;121:790-9. [Crossref] [PubMed]
- Lautner M, Lin H, Shen Y, et al. Disparities in the Use of Breast-Conserving Therapy Among Patients With Early-Stage Breast Cancer. JAMA Surg 2015;150:778-86. [Crossref] [PubMed]
- Mandelblatt JS, Kerner JF, Hadley J, et al. Variations in breast carcinoma treatment in older medicare beneficiaries: is it black or white. Cancer 2002;95:1401-14. [Crossref] [PubMed]
- LeMasters T, Madhavan SS, Sambamoorthi U. Comparison of the Initial Loco-Regional Treatment Received for Early-Stage Breast Cancer between Elderly Women in Appalachia and a United States - Based Population: Good and Bad News. Global J Breast Cancer Res 2016;4:10. [Crossref] [PubMed]
- Dodgion CM, Lipsitz SR, Decker MR, et al. Institutional Variation in Surgical Care for Early Stage Breast Cancer at Community Hospitals. J Surg Res 2017;211:196-205. [Crossref] [PubMed]
- Nattinger AB, Kneusel RT, Hoffmann RG, et al. Relationship of Distance From a Radiotherapy Facility and Initial Breast Cancer Treatment. J Natl Cancer Inst 2001;93:1344-6. [Crossref] [PubMed]
- Nattinger AB, Gottlieb MS, Veum J, et al. Geographic Variation in the Use of Breast-Conserving Treatment for Breast Cancer. N Engl J Med 1992;326:1102-7. [Crossref] [PubMed]
- Boscoe FP, Johnson CJ, Henry KA, et al. Geographic proximity to treatment for early stage breast cancer and likelihood of mastectomy. Breast 2011;20:324-8. [Crossref] [PubMed]
- Samet JM, Hunt WC, Farrow DC. Determinants of receiving breast-conserving surgery. The Surveillance, Epidemiology, and End Results Program, 1983-1986. Cancer 1994;73:2344-51. [Crossref] [PubMed]
- Hislop TG, Olivotto IA, Coldman AJ, et al. Variations in Breast Conservation Surgery for Women with Axillary Lymph Node Negative Breast Cancer in British Columbia. Can J Public Health 1996;87:390-4. [PubMed]
- Goel V, Olivotto IA, Hislop TG, et al. Patterns of initial management of node-negative breast cancer in two Canadian provinces. CMAJ 1997;156:25-35. [PubMed]
- Xu Y, Bouchard-Fortier A, Olivotto I, et al. ‘Driving’ Rates Down: A Population-Based Study of Opening New Radiation Therapy Centers on the Use of Mastectomy for Breast Cancer. Ann Surg Oncol 2018;25:2994-3003. [Crossref] [PubMed]
- Fisher S, Gao H, Yasui Y, et al. Treatment variation in patients diagnosed with early stage breast cancer in Alberta from 2002 to 2010: a population-based study. BMC Health Serv Res 2015;15:35. [Crossref] [PubMed]
- Liu JJ, Zhang S, Hao X, et al. Breast-conserving therapy versus modified radical mastectomy: Socioeconomic status determines who receives what—Results from case–control study in Tianjin, China. Cancer Epidemiol 2012;36:89-93. [Crossref] [PubMed]
- Izuo M, Ishida T. Changing practices in the surgical treatment of breast cancer in Japan: A nationwide survey by the Japanese Breast Cancer Society. Surg Today 1994;24:133-6. [Crossref] [PubMed]
- Adelson P, Lim K, Churches T, et al. Surgical treatment of breast cancer in New South Wales 1991, 1992. Aust N Z J Surg 1997;67:9-14. [Crossref] [PubMed]
- Craft PS, Buckingham JM, Dahlstrom JE, et al. Variation in the management of early breast cancer in rural and metropolitan centres: Implications for the organisation of rural cancer services. Breast 2010;19:396-401. [Crossref] [PubMed]
- Azzopardi J, Walsh D, Chong C, et al. Impact of geographic location on surgical outcomes of women with breast cancer. ANZ J Surg 2014;84:735-9. [Crossref] [PubMed]
- Ambroggi M, Biasini C, Del Giovane C, et al. Distance as a Barrier to Cancer Diagnosis and Treatment: Review of the Literature. Oncologist 2015;20:1378-85. [Crossref] [PubMed]
- Dickman SL, Himmelstein DU, Woolhandler S. Inequality and the health-care system in the USA. Lancet 2017;389:1431-41. [Crossref] [PubMed]
- Gorey KM, Luginaah IN, Hamm C, et al. Breast cancer care in the Canada and the United States: Ecological comparisons of extremely impoverished and affluent urban neighborhoods. Health Place 2010;16:156-63. [Crossref] [PubMed]
- Gu J, Groot G, Boden C, et al. Review of factors influencing women's choice of mastectomy versus breast conserving therapy in early stage breast cancer: a systematic review. Clin Breast Cancer 2018;18:e539-54. [Crossref] [PubMed]
- McCahill LE, Privette AR, Hart MR, et al. Are mastectomy rates a reasonable quality measure of breast cancer surgery? Am J Surg 2009;197:216-21. [Crossref] [PubMed]
- American Society of Clinical Oncology. Implementing solutions to close the rural cancer care gap. 2019. Available online: https://www.ajmc.com/newsroom/implementing-solutions-to-close-the-rural-cancer-care-gap
- Hoskin PJ, Forbes H, Ball C, et al. Variations in radiotherapy delivery in England - evidence from the national radiotherapy dataset. Clin Oncol (R Coll Radiol) 2013;25:531-7. [Crossref] [PubMed]
- Gooiker GA. A systematic review and meta-analysis of the volume-outcome relationship in the surgical treatment of breast cancer. Are breast cancer patients better of with a high volume provider? Eur J Surg Oncol 2010;36:S27-35. [Crossref] [PubMed]
- Roder D, Zorbas H, Kollias J, et al. Factors predictive of treatment by Australian breast surgeons of invasive female breast cancer by mastectomy rather than breast conserving surgery. Asian Pac J Cancer Prev 2013;14:539-45. [Crossref] [PubMed]
- Rosato R, Sacerdote C, Pagano E, et al. Appropriateness of early breast cancer management in relation to patient and hospital characteristics: a population based study in Northern Italy. Breast Cancer Res Treat 2009;117:349-56. [Crossref] [PubMed]
- Olson RA, Nichol A, Caron NR, et al. Effect of community population size on breast cancer screening, stage distribution, treatment use and outcomes. Can J Public Health 2012;103:46-52. [Crossref] [PubMed]
- Girgis A, Doran CM, Sanson-Fisher R. Perceived training and support needs of general practitioners treating breast cancer patients living in rural versus urban areas. Cancer Strategy 2000;2:86-90.
- Coombs NJ, Coombs JM, Vaidya UJ, et al. Environmental and social benefits of the targeted intraoperative radiotherapy for breast cancer: data from UK TARGIT-A trial centres and two UK NHS hospitals offering TARGIT IORT. BMJ Open 2016;6:e010703. [Crossref] [PubMed]
- Goyal S, Chandwani S, Haffty BG, et al. Effect of travel distance and time to radiotherapy on likelihood of receiving mastectomy. Ann Surg Oncol 2015;22:1095-101. [Crossref] [PubMed]
- Sauerzapf VA, Jones AP, Haynes R, et al. Travel time to radiotherapy and uptake of breast-conserving surgery for early stage cancer in Northern England. Health Place 2008;14:424-33. [Crossref] [PubMed]
- Acharya S, Hsieh S, Michalski JM, et al. Distance to Radiation Facility and Treatment Choice in Early-Stage Breast Cancer. Int J Radiat Oncol Biol Phys 2016;94:691-9. [Crossref] [PubMed]
- Lam J, Cook T, Foster S, et al. Examining Determinants of Radiotherapy Access: Do Cost and Radiotherapy Inconvenience Affect Uptake of Breast-conserving Treatment for Early Breast Cancer? Clin Oncol (R Coll Radiol) 2015;27:465-71. [Crossref] [PubMed]
Cite this article as: Chabba N, Tin ST, Zhao J, Abrahimi S, Elwood JM. Geographic variations in surgical treatment for breast cancer: a systematic review. Ann Cancer Epidemiol 2020;4:2.