Journal of Surgical Oncology 2014;110:903–906
Prevalence of Breast Masses and Barriers to Care: Results from a Population-Based Survey in Rwanda and Sierra Leone FAUSTIN NTIRENGANYA, MD,1,2 ROBIN T. PETROZE, MD, MPH,3 THAIM B. KAMARA, MD,4 REINOU S. GROEN, MD, MIH,5,6 ADAM L. KUSHNER, MD, MPH,5,7,8 PATRICK KYAMANYWA, MD,2 J. FORREST CALLAND, MD,3 AND T. PETER KINGHAM, MD5,9* 1
Department of Surgery, Kigali University Teaching Hospital, Kigali, Rwanda 2 Faculty of Medicine, National University of Rwanda, Rwanda 3 Department of Surgery, University of Virginia, Charlottesville, Virginia 4 Department of Surgery, College of Medicine and Allied Health Sciences and Connaught Hospital, Freetown, Sierra Leone 5 Surgeons OverSeas, New York, NY 6 Department of Gynecology and Obstetrics, Johns Hopkins Hospital, Baltimore, Maryland 7 Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland 8 Department of Surgery, Columbia University, New York, NY 9 Memorial Sloan Kettering Cancer Center, New York, NY
Background and Objectives: Breast cancer incidence may be increasing in low‐ and middle‐income countries (LMIC). This study estimates the prevalence of breast masses in Rwanda (RW) and Sierra Leone (SL) and identifies barriers to care for women with breast masses. only. Methods: Data were collected from households in RW and SL using Surgeons Overseas Assessment of Surgical Need (SOSAS), a cross‐sectional, randomized, cluster‐based population survey designed to identify surgical conditions. Data regarding breast masses and barriers to care in women with breast masses were analyzed. Results: 3,469 households (1,626 RW; 1,843 SL) were surveyed and 6,820 persons (3,175 RW; 3,645 SL) interviewed. Breast mass prevalence was 3.3% (SL) and 4.6% (RW). Overall, 93.8% of masses were in women, with 49.1% (SL) and 86.1% (RW) in women >30 years. 73.7% (SL) and 92.4% (RW) of women reported no disability; this was their primary reason for not seeking medical attention. Overall, 36.8% of women who reported masses consulted traditional healers only. Conclusions: For women in RW and SL, minimal education, poverty, and reliance on traditional healers are barriers to medical care for breast masses. Public health programs to increase awareness and decrease barriers are necessary to lower breast cancer mortality rates in low‐ and middle‐ income countries (LMIC).
J. Surg. Oncol. 2014;110:903–906. ß 2014 Wiley Periodicals, Inc.
KEY WORDS: prevalence; barriers to care; breast mass; developing countries
INTRODUCTION The prevalence of breast cancer in developing countries is not known. However, the incidence may be rising sharply due to lifestyle changes, reproductive factors, and increased life expectancy [1]. In addition to Westernization of lifestyle, the seeming increase in incidence may be related to past underdiagnosis, a situation that is now only beginning to be alleviated through early efforts that heighten public awareness of breast cancer as a health issue. Breast cancer is a leading cause of cancer‐ related mortality among women in developing countries as well as in the developed world [2]. Case fatality rates are, however, higher in low‐ and middle‐income countries (LMICs) than in high‐income countries (HICs). This disparity in outcomes may be due to a lack of awareness of the benefits of early detection; a scarcity of adequate facilities for screening, diagnosis, and treatment; poor access to primary health care; and shortages in skilled personnel [3]. In Sub‐Saharan Africa, studies on breast cancer epidemiology or outcomes are small and retrospective hospital‐based reports. These studies report an increase in breast cancer incidence, along with late‐ stage presentation, poor prognosis, and low survival rates [4–7]. In Rwanda (East Africa) and Sierra Leone (West Africa), health policy makers are working to improve access to primary health care as well as to modern medical technologies such as mammography, CT, MRI, advanced surgical procedures, and adjuvant therapies [8]. Data indicate that breast cancer survival rates improved in the United States in the early 1970s, a period that pre‐dates widespread
ß 2014 Wiley Periodicals, Inc.
use of the technologies mentioned above. Authors who analyzed data from prior to 1975 attributed this improvement in survival rates to more effective breast education programs, increased breast cancer awareness, increased detection of tumors palpable with self or clinical examination, and better diagnostics [3,9,10]. Thus, while it is important to increase access to advanced technologies in the developing world, these patients’ survival rates may be more immediately improved by intensifying the awareness of breast cancer and promoting the benefits of early treatment. The objective of this study was to estimate the prevalence of untreated breast masses in the general populations of Rwanda and Sierra Leone and, further, to identify and compare barriers to medical treatment for women with breast masses in these two low‐income countries (LICs).
Grant sponsor: Surgeons OverSeas, a non‐profit organization. Conflict of Interest: The authors have no conflicts of interest to declare. *Correspondence to: T. Peter Kingham, MD Hepatopancreatobiliary Service, Memorial Sloan Kettering Cancer Center 1275 York Avenue, New York, NY 10065. Fax: þ212‐794‐5842. E‐mail: [email protected] Received 8 April 2014; Accepted 18 June 2014 DOI 10.1002/jso.23726 Published online 2 August 2014 in Wiley Online Library (wileyonlinelibrary.com).
904
Ntirenganya et al.
METHODS Data were collected and aggregated from respondents with self‐ reported breast masses, as identified by the Surgeons Overseas Assessment of Surgical Need (SOSAS) survey in the two Sub‐ Saharan African Nations [11]. SOSAS is a cross‐sectional, cluster‐based population survey designed to identify untreated surgical conditions and provide estimations of prevalence. It was conducted in Rwanda in October 2011 and in Sierra Leone in January 2012, and was administered via electronic tablets [12,13]. SOSAS version 2.0 was used in Rwanda and version 3.0 [14] was used in Sierra Leone, with almost no changes between the two versions and very few local modifications. The survey instrument contains specific questions regarding breast pathology and barriers to seeking medical treatment for breast masses. The sample size was calculated using Cochran’s formula where estimated prevalence was previously established in a pilot study of the same survey in Sierra Leone [11]. Clusters were randomly selected in two stages with a probability proportional to population size after stratification by urban and rural populations. Medical and nursing students were recruited and trained in questionnaire content, informed consent, interview techniques, sampling strategies, and electronic tablet use [14]. These surveys were done in collaboration with local physicians and surgeons. Permission was obtained from the respective national health ministries, and ethical approval was obtained from local ethical and scientific review committees as well as from the institutional review boards of the Royal Dutch Tropical Institute in the Netherlands and the University of Virginia in the United States. Written informed consent was obtained from all survey respondents in both Rwanda and Sierra Leone [12,13], and respondents did not receive a stipend. In the selected village or cluster, separate households in each cluster were randomly assigned for investigation. After the household representative was identified, he or she was asked questions from sections A,B,C, and D of the SOSAS questionnaire. After the interview with the household representative, two other household members were selected using either a random number calculator, or random number table. Both household members were interviewed with questions from sections E,F,G,H,I,J, and K, and the last section (L) for women only [14]. Collected data were screened on site by the field supervisors and feedback was provided to enumerators on the same day. Final, full assessment was done during the study period after data collection to identify inconsistencies and missing items [12]. For the present study, we used data only from questionnaire sections A,B,C,D, and L from the two countries. Data were retrieved, aggregated, and analyzed using Microsoft Excel version 12 and descriptive statistics.
TABLE I. Demographic Data for Women With Breast Masses
Median age (range) Age group 30 31–60 60 Highest level of education none primary school secondary school tertiary school Literate Illiterate Employment none home maker domestic helper farmer Self‐employed/small business Menstrual cycle last year No Yes Pregnant Breast feeding
Sierra Leone (N ¼ 57)
Rwanda (N ¼ 79)
31 (0.5–80)
43 (1–80)
29 (50.9%) 24 (42.1%) 4 (7.0%)
11 (13.9%) 48 (60.8%) 20 (25.3%)
38 3 14 2 42 15
33 (41.8%) 44 (55.7%) 2 (2.5%) 0 39 (49.4%) 40 (50.6%)
(66.7%) (5.3%) (24.6%) (3.5%) (73.7%) (26.3%) 10 7 2 26 12
9 0 0 70 0
14 (24.6%) 43 (75.4%) 5 (8.8%) 9 (15.8%)
45 (57%) 34 (43%) NA 21 (26.6%)
NA ¼ not available.
12 months. 73.7% of Sierra Leonean and 92.4% of Rwandan women respondents reporting breast mass stated that no disability was associated with these masses. This absence of disability was the reason reported most commonly for not seeking medical treatment. Another common reason for not seeking medical care was lack of money, as reported by 35.1% of women in Sierra Leone and 11.4% of women in Rwanda. Additionally, some women reported that they did not consult a medical practitioner because of a lack of trust in the health care system, long distance required to reach the provider, or stigma associated with having a breast problem. For respondents in both countries combined, 36.8% of women with breast masses consulted traditional healers instead of going to health centers, and over 75% of women had no schooling or stopped after primary school. Because SOSAS is a cross‐sectional, cluster‐based population survey, the results may be extrapolated to the general populations of Rwanda and Sierra Leone as estimates, and with limitations.
DISCUSSION RESULTS For SOSAS in Sierra Leone, we analyzed data from 1,843 (98%) of the 1,875 households targeted and from 3,645 respondents [12]. For Rwanda, we analyzed data from 1,626 (99%) of the 1642 households targeted and from 3,175 respondents [13]. Data from a total of 6,820 subjects were analyzed [12,13]. In Sierra Leone, the breakdown of the respondents was 46% men and 54% women; in Rwanda it was 43% men and 57% women. Combined, the ratio of women to men was 1.25. The prevalence of breast mass reported was 3.3% (2.9% women and 0.4% men) and 4.6% (4.4% women and 0.2% men) in Sierra Leone and Rwanda, respectively. Combined, this prevalence is close to 4%. The majority (93.8%) of these masses were reported in women. Overall, 70.6% of breast masses were reported in women aged more than 30 years (Table I). However, we found that in Sierra Leone, 49.1% of women with breast masses were over 30 years of age, compared to 86.1% in Rwanda. All women reporting breast masses described a single lump (Table II). Of these women, 88.6% and 66.7% in Rwanda and Sierra Leone, respectively, reported that the mass was present for more than Journal of Surgical Oncology
In this study, the number of women complaining of breast masses is unconfirmed by physical exam and there is no information about TABLE II. Characteristics of Breast Masses (Women Only)
Number of masses One More than one Mass discovered 12 months Unknown Present currently Not present currently NA ¼ not available.
Sierra Leone
Rwanda
57 (100%) 0
79 (100%) 0
6 12 38 1 30 27
1 7 70 1
(10.5%) (21.0%) (66.7%) (1.8%) (52.6%) (47.4%)
(1.3%) (8.9%) (88.6%) (1.3%) NA NA
Breast masses in Rwanda and Sierra Leone whether the masses are benign or malignant. Most of the women did not undergo medical evaluations. For that reason, the term “breast mass” here applies equally to all breast lesions, i.e., fibroadenomas, cysts, normal “lumpy” breast tissue, and cancer. The breakdown of benign versus malignant disease in settings such as Rwanda and Sierra Leone is not known. In both countries, there was effectively little or no breast cancer care. There was little capacity to perform biopsy, process pathology, perform mastectomy and axillary node dissection, or to administer chemotherapy, hormone therapy, or radiotherapy. These circumstances can be an additional reason for little motivation on the part of women to seek medical help. The pattern of care for women with breast masses, however, may be slowly changing, at least in Rwanda, where there are some unpublished data suggesting that women may be seeking care earlier. Although these data are preliminary and not published, the fact of the studies’ existence emphasizes that cancer care in LMICs is slowly being recognized as a major priority for global health efforts. In patients with breast cancer, Gakwaya et al. in Uganda reported a 5‐year survival rate of 56%, which is lower than the rates for both South African blacks (64%) and North American whites (82–88%). In the same study, only 23% of patients presented in early stages (0‐IIB); 26% presented with distant metastasis, and 75% underwent mastectomy and axillary clearance [7]. It is difficult to plan optimally effective oncologic public health initiatives to improve these outcomes because prospectively obtained baseline data on the prevalence of cancers is lacking. SOSAS was developed to provide policy makers with prospective baseline data to support the position that surgical conditions constitute a substantial and important part of global public health. Given the availability of only small, retrospective, hospital‐based reports, the prevalence in LMICs of breast cancer, specifically, is not known. Considering breast alone, our study revealed that 3.3% and 4.6% of the survey populations in Sierra Leone and Rwanda, respectively, reported masses in the breast that require surgical evaluation. Of note is that Obaikol et al. reported a similar rate (4.8%) of incidental finding of breast mass, with data from a limited sample of 320 female university students in Kampala, Uganda [15]. The Sub‐Saharan African population is relatively young compared to HIC populations with breast cancer. In addition, the incidence of breast cancer in LMICs is expected to increase due to changes in lifestyle, reproductive patterns, and life expectancy, as Westernization increases. In Sierra Leone, the majority (50.9%) of breast masses were found in women aged less than or equal to 30 years; however, an additional 42.1% of breast masses were found in women aged greater than 30 and less than 60. All of these masses require evaluation by a trained medical professional to determine the need for surgical evaluation. Of the women reporting breast masses in Sierra Leone, 66.7% stated that the mass was present for more than 12 months, and 73.7% stated that the mass was not disabling. With the young age, the length of time the masses have been present, and the low percentage that are disabling, we suspect that many of these masses are benign. It is, however, also possible that many of these masses are malignant, as breast cancer develops at a younger age in LMICs. Thus, there is a large population of women with breast masses in Sierra Leone and Rwanda that require, at the least, examination by a trained medical professional. By July 2013, Rwanda had an estimated population of 12,012,589, growth rate of 2.7%, and male‐to‐female ratio of 0.99. The fact that this is a household study is reflected in the relative oversampling of women by our surveys in both Sierra Leone, and Rwanda. Rwanda has a young population: 61.4% is aged below 25 years and only 6.4% is above 55 years [16,17]. In this study, 4.6% of the surveyed Rwandan population reported a breast mass. An extrapolation to the general Rwandan population would equate to 552,579 individuals, 528,554 women and 24,025 men, with breast masses needing surgical evaluation. A similar extrapolation to the general population of Sierra Leone, using the corresponding survey result (3.3%), estimates that 185,219 individuals have breast masses that should be evaluated. Admittedly, Journal of Surgical Oncology
905
these numbers could be viewed as an overestimation for malignant breast pathology since breast masses frequently can be benign, but the numbers do support the possibility of significant unmet health care needs for women with breast masses. The substantial difference in age distribution between the two countries is due to the Rwandan genocide in 1994. The major loss of life during that period means, that there is a large number of women in their late 20s that will be entering the age groups where breast cancer is increasingly common. For that reason Rwanda, in particular, must prepare for a rapid rise in breast cancer presentation due to the young genocide survivors coming of age. Rwanda and Sierra Leone are among the most improved performers in the 2013 United Nations human development index [8] as health policy makers are working to improve access to primary health care as well as to modern technologies such as mammography, CT, MRI, advanced surgical procedures, and adjuvant therapies. For breast cancer diagnosis alone, a modern health care infrastructure is required for proper evaluation of breast lumps. In addition to the availability of a trained practitioner to perform a physical examination, equipment and training are required for ultrasound and/or mammogram, biopsy, and pathology. Also required is the availability of a trained pathologist to interpret the biopsy. Nevertheless, because implementing improvements in access to primary health care and modern technologies will take years, breast cancer survival rates may be more immediately improved by intensifying the awareness of breast cancer and the stressing that early diagnosis can lead to curative treatment [18,19]. Several barriers limiting access to care were identified in this study. The primary reason for not seeking medical treatment was that the masses were not painful. Financial hardship and consultation with a traditional healer (instead of a medical professional) were two additional barriers. Over 75% of women in both countries had no schooling or stopped after primary school, which is an important element when discussing population education and target sites. The majority of women do not consider a breast mass to be a potential health problem, most likely since they do not know what it can represent. This reflects a lack of individual knowledge, a low level of awareness in the population, and a lack of policies supporting programs to ensure health care professionals provide patient education on breast cancer detection and treatment. Cultural beliefs and other behaviors have been identified as barriers to care, such as fear of not being considered desirable by potential marriage partners if a breast mass is present, and fear of losing a breast (mastectomy) if a mass is identified by a physician. In many LMICs, surgery is only available at referral centers. In addition, radiation and adjuvant therapies are often not available, leaving mastectomy as the only option [20]. In cultures where the stigma associated with breast mass and mastectomy is a barrier to care, it is clear that community engagement to lessen the stigma is imperative. Another barrier to care for many women with a breast mass is the lack of access to a trained medical professional. This is part of the larger problem of lack of access to general surgeons. Analysis of data from the administration of SOSAS in Sierra Leone showed a high prevalence of untreated surgical conditions. Specifically, 24.6% of respondents had a condition possibly needing surgical attention, and 25.2% of deaths during the previous year may have been averted with improved access to surgical services [12]. In Rwanda, 38.8–43.6% of the surveyed population reported having at least one operative condition during their lifetime. For the same population, 14.8% reported having an operative condition during the last 12 months, and 6.4% reported having a current operative condition [13]. This burden of surgical disease, which includes women with breast masses, demonstrates the difficulty in meeting the needs of patients with surgical and medical conditions in LICs where there are few physicians and medical staff. The major limitation of this study is the investigators’ reliance on verbal responses from interviewees. No physical examination was done. A respondent’s perception of a breast mass may not be reliable, as it requires trained, experienced medical professionals working within a
906
Ntirenganya et al.
modern health care infrastructure with access to imaging and pathology to differentiate between normal and pathologic breast conditions. This limitation might contribute to an underestimation, as might the fact that breast self‐exams are generally not done in those countries. However, absent physical examination, we can also not exclude overestimation of the prevalence of breast mass.
CONCLUSION This study estimates the prevalence of untreated breast masses in Rwandan and Sierra Leonean women and men by extrapolation from the results of a cross‐sectional, cluster‐based population survey. Many women do not seek medical care when they feel a breast mass, secondary to lack of symptoms, finances, and reliance on traditional healers. In light of current literature that reports patients with breast cancer in developing countries present at later stages and younger ages, and with a very poor prognosis, education about breast cancer and advocacy emphasizing curability should be increased as a strategy to lower the high breast cancer mortality rates in LMICs. To effect change in the culture of ignorance and fear surrounding breast health, community involvement is required.
REFERENCES 1. Beaulieu N, Bloom DE, Bloom LR, Stein RM: Breakaway: The global burden of cancer‐challenges and opportunities. A report from the Economist Intelligence Unit. London, UK, 2009. 2. Knaul FM, Nigenda G, Lozano R, et al.: Breast cancer in Mexico: a pressing priority. Reprod Health Matters 2008;16:113–123. 3. Shulman LN, Willett W, Sievers A, Knaul FM: Breast cancer in developing countries: opportunities for improved survival. J Oncol 2010;2010:595167. 4. Mody GN, Nduaguba A, Ntirenganya F, Riviello R: Characteristics and presentation of patients with breast cancer in Rwanda. Am J Surg 2013;205:409–413. 5. Galukande M, Kiguli‐Malwadde E: Rethinking breast cancer screening strategies in resource‐limited settings. Afr Health Sci 2010;10:89–92. 6. Eniu A, Carlson RW, Aziz Z, et al.: Global Summit Treatment and Allocation of Resources Panel: Breast cancer in limited‐resource countries: treatment and allocation of resources. Breast J 2006;12: S38–S53.
Journal of Surgical Oncology
7. Gakwaya A, Kigula‐Mugambe JB, Kavuma A, et al.: Cancer of the breast: 5‐year survival in a tertiary hospital in Uganda. Br J Cancer 2008;99:63–67. 8. William Orme, et al.: Human development index in 2013 report online http://www.undp.org/content/undp/en/home/librarypage/ hdr/human‐development‐report‐2013/ Accessed on March 20, 2014. 9. Jatoi I, Anderson WF, Rao SR, Devesa SS: Breast cancer trends among black and white women in the United States. J Clin Oncol 2005;23:7836–7841. 10. Flannery JT, Sullivan PD: Female breast cancer in Connecticut – incidence, mortality, survival – 1935‐1975. Conn Health Bull 1978;92:4–9. 11. Groen RS, Samai M, Petroze RT, et al.: Pilot testing of population‐ based surgical survey tool in Sierra Leone. World J Surg 2012;36: 771–774. 12. Groen RS, Samai M, Stewart KA, et al.: Untreated surgical conditions in Sierra Leone: a cluster randomised, cross‐sectional, countrywide survey. Lancet 2012;380:1082–1087. 13. Petroze RT, Groen RS, Niyonkuru F, et al.: Estimating operative disease prevalence in a low‐income country: results of a nationwide population survey in Rwanda. Surgery 2013;153: 457–464. 14. Groen RS: SOSAS Surgeons OverSeas Assessment of Surgical Needs, A manual for SOSAS interviewers, Version 3.0, accessed on March 20, 2014. http://www.adamkushnermd.com/files/ INTERVIEWERS_MANUAL_FOR_SOSAS_SL.pdf 15. Obaikol R, Galukande M, Fualal J: Knowledge and practice of breast self examination among female students in a sub Saharan African university. East Central Afr J Surg 2010;15:22–27. 16. CIA world factbook, Index mundi: Rwanda Demographics Profile 2013. https://www.cia.gov/library/publications/the‐world‐factbook/ geos/rw.html Accessed on March 20, 2014. 17. CIA world factbook, Index mundi: Sierra Leone Demographics Profile 2013. https://www.cia.gov/library/publications/the‐world‐ factbook/geos/sl.html Accessed on March 20, 2014. 18. Thomas DB, Gao DL, Ray RM, et al.: Randomized trial of breast‐ self examination in Shanghai: final results. J Natl Cancer Inst 2002; 94:1445–1457. 19. Schwartsmann G: Breast cancer in South America: challenges to improve early detection and medical management of a public health program. J Clin Oncol 2001;19:118S–124S. 20. Adesunkanmi AR, Lawal OO, Adelusola KA, Durosimi MA: The severity, outcome and challenges of breast cancer in Nigeria. Breast 2006;15:399–409.
Prevalence of Breast Masses and Barriers to Care: Results from a Population-Based Survey in Rwanda and Sierra Leone FAUSTIN NTIRENGANYA, MD,1,2 ROBIN T. PETROZE, MD, MPH,3 THAIM B. KAMARA, MD,4 REINOU S. GROEN, MD, MIH,5,6 ADAM L. KUSHNER, MD, MPH,5,7,8 PATRICK KYAMANYWA, MD,2 J. FORREST CALLAND, MD,3 AND T. PETER KINGHAM, MD5,9* 1
Department of Surgery, Kigali University Teaching Hospital, Kigali, Rwanda 2 Faculty of Medicine, National University of Rwanda, Rwanda 3 Department of Surgery, University of Virginia, Charlottesville, Virginia 4 Department of Surgery, College of Medicine and Allied Health Sciences and Connaught Hospital, Freetown, Sierra Leone 5 Surgeons OverSeas, New York, NY 6 Department of Gynecology and Obstetrics, Johns Hopkins Hospital, Baltimore, Maryland 7 Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland 8 Department of Surgery, Columbia University, New York, NY 9 Memorial Sloan Kettering Cancer Center, New York, NY
Background and Objectives: Breast cancer incidence may be increasing in low‐ and middle‐income countries (LMIC). This study estimates the prevalence of breast masses in Rwanda (RW) and Sierra Leone (SL) and identifies barriers to care for women with breast masses. only. Methods: Data were collected from households in RW and SL using Surgeons Overseas Assessment of Surgical Need (SOSAS), a cross‐sectional, randomized, cluster‐based population survey designed to identify surgical conditions. Data regarding breast masses and barriers to care in women with breast masses were analyzed. Results: 3,469 households (1,626 RW; 1,843 SL) were surveyed and 6,820 persons (3,175 RW; 3,645 SL) interviewed. Breast mass prevalence was 3.3% (SL) and 4.6% (RW). Overall, 93.8% of masses were in women, with 49.1% (SL) and 86.1% (RW) in women >30 years. 73.7% (SL) and 92.4% (RW) of women reported no disability; this was their primary reason for not seeking medical attention. Overall, 36.8% of women who reported masses consulted traditional healers only. Conclusions: For women in RW and SL, minimal education, poverty, and reliance on traditional healers are barriers to medical care for breast masses. Public health programs to increase awareness and decrease barriers are necessary to lower breast cancer mortality rates in low‐ and middle‐ income countries (LMIC).
J. Surg. Oncol. 2014;110:903–906. ß 2014 Wiley Periodicals, Inc.
KEY WORDS: prevalence; barriers to care; breast mass; developing countries
INTRODUCTION The prevalence of breast cancer in developing countries is not known. However, the incidence may be rising sharply due to lifestyle changes, reproductive factors, and increased life expectancy [1]. In addition to Westernization of lifestyle, the seeming increase in incidence may be related to past underdiagnosis, a situation that is now only beginning to be alleviated through early efforts that heighten public awareness of breast cancer as a health issue. Breast cancer is a leading cause of cancer‐ related mortality among women in developing countries as well as in the developed world [2]. Case fatality rates are, however, higher in low‐ and middle‐income countries (LMICs) than in high‐income countries (HICs). This disparity in outcomes may be due to a lack of awareness of the benefits of early detection; a scarcity of adequate facilities for screening, diagnosis, and treatment; poor access to primary health care; and shortages in skilled personnel [3]. In Sub‐Saharan Africa, studies on breast cancer epidemiology or outcomes are small and retrospective hospital‐based reports. These studies report an increase in breast cancer incidence, along with late‐ stage presentation, poor prognosis, and low survival rates [4–7]. In Rwanda (East Africa) and Sierra Leone (West Africa), health policy makers are working to improve access to primary health care as well as to modern medical technologies such as mammography, CT, MRI, advanced surgical procedures, and adjuvant therapies [8]. Data indicate that breast cancer survival rates improved in the United States in the early 1970s, a period that pre‐dates widespread
ß 2014 Wiley Periodicals, Inc.
use of the technologies mentioned above. Authors who analyzed data from prior to 1975 attributed this improvement in survival rates to more effective breast education programs, increased breast cancer awareness, increased detection of tumors palpable with self or clinical examination, and better diagnostics [3,9,10]. Thus, while it is important to increase access to advanced technologies in the developing world, these patients’ survival rates may be more immediately improved by intensifying the awareness of breast cancer and promoting the benefits of early treatment. The objective of this study was to estimate the prevalence of untreated breast masses in the general populations of Rwanda and Sierra Leone and, further, to identify and compare barriers to medical treatment for women with breast masses in these two low‐income countries (LICs).
Grant sponsor: Surgeons OverSeas, a non‐profit organization. Conflict of Interest: The authors have no conflicts of interest to declare. *Correspondence to: T. Peter Kingham, MD Hepatopancreatobiliary Service, Memorial Sloan Kettering Cancer Center 1275 York Avenue, New York, NY 10065. Fax: þ212‐794‐5842. E‐mail: [email protected] Received 8 April 2014; Accepted 18 June 2014 DOI 10.1002/jso.23726 Published online 2 August 2014 in Wiley Online Library (wileyonlinelibrary.com).
904
Ntirenganya et al.
METHODS Data were collected and aggregated from respondents with self‐ reported breast masses, as identified by the Surgeons Overseas Assessment of Surgical Need (SOSAS) survey in the two Sub‐ Saharan African Nations [11]. SOSAS is a cross‐sectional, cluster‐based population survey designed to identify untreated surgical conditions and provide estimations of prevalence. It was conducted in Rwanda in October 2011 and in Sierra Leone in January 2012, and was administered via electronic tablets [12,13]. SOSAS version 2.0 was used in Rwanda and version 3.0 [14] was used in Sierra Leone, with almost no changes between the two versions and very few local modifications. The survey instrument contains specific questions regarding breast pathology and barriers to seeking medical treatment for breast masses. The sample size was calculated using Cochran’s formula where estimated prevalence was previously established in a pilot study of the same survey in Sierra Leone [11]. Clusters were randomly selected in two stages with a probability proportional to population size after stratification by urban and rural populations. Medical and nursing students were recruited and trained in questionnaire content, informed consent, interview techniques, sampling strategies, and electronic tablet use [14]. These surveys were done in collaboration with local physicians and surgeons. Permission was obtained from the respective national health ministries, and ethical approval was obtained from local ethical and scientific review committees as well as from the institutional review boards of the Royal Dutch Tropical Institute in the Netherlands and the University of Virginia in the United States. Written informed consent was obtained from all survey respondents in both Rwanda and Sierra Leone [12,13], and respondents did not receive a stipend. In the selected village or cluster, separate households in each cluster were randomly assigned for investigation. After the household representative was identified, he or she was asked questions from sections A,B,C, and D of the SOSAS questionnaire. After the interview with the household representative, two other household members were selected using either a random number calculator, or random number table. Both household members were interviewed with questions from sections E,F,G,H,I,J, and K, and the last section (L) for women only [14]. Collected data were screened on site by the field supervisors and feedback was provided to enumerators on the same day. Final, full assessment was done during the study period after data collection to identify inconsistencies and missing items [12]. For the present study, we used data only from questionnaire sections A,B,C,D, and L from the two countries. Data were retrieved, aggregated, and analyzed using Microsoft Excel version 12 and descriptive statistics.
TABLE I. Demographic Data for Women With Breast Masses
Median age (range) Age group 30 31–60 60 Highest level of education none primary school secondary school tertiary school Literate Illiterate Employment none home maker domestic helper farmer Self‐employed/small business Menstrual cycle last year No Yes Pregnant Breast feeding
Sierra Leone (N ¼ 57)
Rwanda (N ¼ 79)
31 (0.5–80)
43 (1–80)
29 (50.9%) 24 (42.1%) 4 (7.0%)
11 (13.9%) 48 (60.8%) 20 (25.3%)
38 3 14 2 42 15
33 (41.8%) 44 (55.7%) 2 (2.5%) 0 39 (49.4%) 40 (50.6%)
(66.7%) (5.3%) (24.6%) (3.5%) (73.7%) (26.3%) 10 7 2 26 12
9 0 0 70 0
14 (24.6%) 43 (75.4%) 5 (8.8%) 9 (15.8%)
45 (57%) 34 (43%) NA 21 (26.6%)
NA ¼ not available.
12 months. 73.7% of Sierra Leonean and 92.4% of Rwandan women respondents reporting breast mass stated that no disability was associated with these masses. This absence of disability was the reason reported most commonly for not seeking medical treatment. Another common reason for not seeking medical care was lack of money, as reported by 35.1% of women in Sierra Leone and 11.4% of women in Rwanda. Additionally, some women reported that they did not consult a medical practitioner because of a lack of trust in the health care system, long distance required to reach the provider, or stigma associated with having a breast problem. For respondents in both countries combined, 36.8% of women with breast masses consulted traditional healers instead of going to health centers, and over 75% of women had no schooling or stopped after primary school. Because SOSAS is a cross‐sectional, cluster‐based population survey, the results may be extrapolated to the general populations of Rwanda and Sierra Leone as estimates, and with limitations.
DISCUSSION RESULTS For SOSAS in Sierra Leone, we analyzed data from 1,843 (98%) of the 1,875 households targeted and from 3,645 respondents [12]. For Rwanda, we analyzed data from 1,626 (99%) of the 1642 households targeted and from 3,175 respondents [13]. Data from a total of 6,820 subjects were analyzed [12,13]. In Sierra Leone, the breakdown of the respondents was 46% men and 54% women; in Rwanda it was 43% men and 57% women. Combined, the ratio of women to men was 1.25. The prevalence of breast mass reported was 3.3% (2.9% women and 0.4% men) and 4.6% (4.4% women and 0.2% men) in Sierra Leone and Rwanda, respectively. Combined, this prevalence is close to 4%. The majority (93.8%) of these masses were reported in women. Overall, 70.6% of breast masses were reported in women aged more than 30 years (Table I). However, we found that in Sierra Leone, 49.1% of women with breast masses were over 30 years of age, compared to 86.1% in Rwanda. All women reporting breast masses described a single lump (Table II). Of these women, 88.6% and 66.7% in Rwanda and Sierra Leone, respectively, reported that the mass was present for more than Journal of Surgical Oncology
In this study, the number of women complaining of breast masses is unconfirmed by physical exam and there is no information about TABLE II. Characteristics of Breast Masses (Women Only)
Number of masses One More than one Mass discovered 12 months Unknown Present currently Not present currently NA ¼ not available.
Sierra Leone
Rwanda
57 (100%) 0
79 (100%) 0
6 12 38 1 30 27
1 7 70 1
(10.5%) (21.0%) (66.7%) (1.8%) (52.6%) (47.4%)
(1.3%) (8.9%) (88.6%) (1.3%) NA NA
Breast masses in Rwanda and Sierra Leone whether the masses are benign or malignant. Most of the women did not undergo medical evaluations. For that reason, the term “breast mass” here applies equally to all breast lesions, i.e., fibroadenomas, cysts, normal “lumpy” breast tissue, and cancer. The breakdown of benign versus malignant disease in settings such as Rwanda and Sierra Leone is not known. In both countries, there was effectively little or no breast cancer care. There was little capacity to perform biopsy, process pathology, perform mastectomy and axillary node dissection, or to administer chemotherapy, hormone therapy, or radiotherapy. These circumstances can be an additional reason for little motivation on the part of women to seek medical help. The pattern of care for women with breast masses, however, may be slowly changing, at least in Rwanda, where there are some unpublished data suggesting that women may be seeking care earlier. Although these data are preliminary and not published, the fact of the studies’ existence emphasizes that cancer care in LMICs is slowly being recognized as a major priority for global health efforts. In patients with breast cancer, Gakwaya et al. in Uganda reported a 5‐year survival rate of 56%, which is lower than the rates for both South African blacks (64%) and North American whites (82–88%). In the same study, only 23% of patients presented in early stages (0‐IIB); 26% presented with distant metastasis, and 75% underwent mastectomy and axillary clearance [7]. It is difficult to plan optimally effective oncologic public health initiatives to improve these outcomes because prospectively obtained baseline data on the prevalence of cancers is lacking. SOSAS was developed to provide policy makers with prospective baseline data to support the position that surgical conditions constitute a substantial and important part of global public health. Given the availability of only small, retrospective, hospital‐based reports, the prevalence in LMICs of breast cancer, specifically, is not known. Considering breast alone, our study revealed that 3.3% and 4.6% of the survey populations in Sierra Leone and Rwanda, respectively, reported masses in the breast that require surgical evaluation. Of note is that Obaikol et al. reported a similar rate (4.8%) of incidental finding of breast mass, with data from a limited sample of 320 female university students in Kampala, Uganda [15]. The Sub‐Saharan African population is relatively young compared to HIC populations with breast cancer. In addition, the incidence of breast cancer in LMICs is expected to increase due to changes in lifestyle, reproductive patterns, and life expectancy, as Westernization increases. In Sierra Leone, the majority (50.9%) of breast masses were found in women aged less than or equal to 30 years; however, an additional 42.1% of breast masses were found in women aged greater than 30 and less than 60. All of these masses require evaluation by a trained medical professional to determine the need for surgical evaluation. Of the women reporting breast masses in Sierra Leone, 66.7% stated that the mass was present for more than 12 months, and 73.7% stated that the mass was not disabling. With the young age, the length of time the masses have been present, and the low percentage that are disabling, we suspect that many of these masses are benign. It is, however, also possible that many of these masses are malignant, as breast cancer develops at a younger age in LMICs. Thus, there is a large population of women with breast masses in Sierra Leone and Rwanda that require, at the least, examination by a trained medical professional. By July 2013, Rwanda had an estimated population of 12,012,589, growth rate of 2.7%, and male‐to‐female ratio of 0.99. The fact that this is a household study is reflected in the relative oversampling of women by our surveys in both Sierra Leone, and Rwanda. Rwanda has a young population: 61.4% is aged below 25 years and only 6.4% is above 55 years [16,17]. In this study, 4.6% of the surveyed Rwandan population reported a breast mass. An extrapolation to the general Rwandan population would equate to 552,579 individuals, 528,554 women and 24,025 men, with breast masses needing surgical evaluation. A similar extrapolation to the general population of Sierra Leone, using the corresponding survey result (3.3%), estimates that 185,219 individuals have breast masses that should be evaluated. Admittedly, Journal of Surgical Oncology
905
these numbers could be viewed as an overestimation for malignant breast pathology since breast masses frequently can be benign, but the numbers do support the possibility of significant unmet health care needs for women with breast masses. The substantial difference in age distribution between the two countries is due to the Rwandan genocide in 1994. The major loss of life during that period means, that there is a large number of women in their late 20s that will be entering the age groups where breast cancer is increasingly common. For that reason Rwanda, in particular, must prepare for a rapid rise in breast cancer presentation due to the young genocide survivors coming of age. Rwanda and Sierra Leone are among the most improved performers in the 2013 United Nations human development index [8] as health policy makers are working to improve access to primary health care as well as to modern technologies such as mammography, CT, MRI, advanced surgical procedures, and adjuvant therapies. For breast cancer diagnosis alone, a modern health care infrastructure is required for proper evaluation of breast lumps. In addition to the availability of a trained practitioner to perform a physical examination, equipment and training are required for ultrasound and/or mammogram, biopsy, and pathology. Also required is the availability of a trained pathologist to interpret the biopsy. Nevertheless, because implementing improvements in access to primary health care and modern technologies will take years, breast cancer survival rates may be more immediately improved by intensifying the awareness of breast cancer and the stressing that early diagnosis can lead to curative treatment [18,19]. Several barriers limiting access to care were identified in this study. The primary reason for not seeking medical treatment was that the masses were not painful. Financial hardship and consultation with a traditional healer (instead of a medical professional) were two additional barriers. Over 75% of women in both countries had no schooling or stopped after primary school, which is an important element when discussing population education and target sites. The majority of women do not consider a breast mass to be a potential health problem, most likely since they do not know what it can represent. This reflects a lack of individual knowledge, a low level of awareness in the population, and a lack of policies supporting programs to ensure health care professionals provide patient education on breast cancer detection and treatment. Cultural beliefs and other behaviors have been identified as barriers to care, such as fear of not being considered desirable by potential marriage partners if a breast mass is present, and fear of losing a breast (mastectomy) if a mass is identified by a physician. In many LMICs, surgery is only available at referral centers. In addition, radiation and adjuvant therapies are often not available, leaving mastectomy as the only option [20]. In cultures where the stigma associated with breast mass and mastectomy is a barrier to care, it is clear that community engagement to lessen the stigma is imperative. Another barrier to care for many women with a breast mass is the lack of access to a trained medical professional. This is part of the larger problem of lack of access to general surgeons. Analysis of data from the administration of SOSAS in Sierra Leone showed a high prevalence of untreated surgical conditions. Specifically, 24.6% of respondents had a condition possibly needing surgical attention, and 25.2% of deaths during the previous year may have been averted with improved access to surgical services [12]. In Rwanda, 38.8–43.6% of the surveyed population reported having at least one operative condition during their lifetime. For the same population, 14.8% reported having an operative condition during the last 12 months, and 6.4% reported having a current operative condition [13]. This burden of surgical disease, which includes women with breast masses, demonstrates the difficulty in meeting the needs of patients with surgical and medical conditions in LICs where there are few physicians and medical staff. The major limitation of this study is the investigators’ reliance on verbal responses from interviewees. No physical examination was done. A respondent’s perception of a breast mass may not be reliable, as it requires trained, experienced medical professionals working within a
906
Ntirenganya et al.
modern health care infrastructure with access to imaging and pathology to differentiate between normal and pathologic breast conditions. This limitation might contribute to an underestimation, as might the fact that breast self‐exams are generally not done in those countries. However, absent physical examination, we can also not exclude overestimation of the prevalence of breast mass.
CONCLUSION This study estimates the prevalence of untreated breast masses in Rwandan and Sierra Leonean women and men by extrapolation from the results of a cross‐sectional, cluster‐based population survey. Many women do not seek medical care when they feel a breast mass, secondary to lack of symptoms, finances, and reliance on traditional healers. In light of current literature that reports patients with breast cancer in developing countries present at later stages and younger ages, and with a very poor prognosis, education about breast cancer and advocacy emphasizing curability should be increased as a strategy to lower the high breast cancer mortality rates in LMICs. To effect change in the culture of ignorance and fear surrounding breast health, community involvement is required.
REFERENCES 1. Beaulieu N, Bloom DE, Bloom LR, Stein RM: Breakaway: The global burden of cancer‐challenges and opportunities. A report from the Economist Intelligence Unit. London, UK, 2009. 2. Knaul FM, Nigenda G, Lozano R, et al.: Breast cancer in Mexico: a pressing priority. Reprod Health Matters 2008;16:113–123. 3. Shulman LN, Willett W, Sievers A, Knaul FM: Breast cancer in developing countries: opportunities for improved survival. J Oncol 2010;2010:595167. 4. Mody GN, Nduaguba A, Ntirenganya F, Riviello R: Characteristics and presentation of patients with breast cancer in Rwanda. Am J Surg 2013;205:409–413. 5. Galukande M, Kiguli‐Malwadde E: Rethinking breast cancer screening strategies in resource‐limited settings. Afr Health Sci 2010;10:89–92. 6. Eniu A, Carlson RW, Aziz Z, et al.: Global Summit Treatment and Allocation of Resources Panel: Breast cancer in limited‐resource countries: treatment and allocation of resources. Breast J 2006;12: S38–S53.
Journal of Surgical Oncology
7. Gakwaya A, Kigula‐Mugambe JB, Kavuma A, et al.: Cancer of the breast: 5‐year survival in a tertiary hospital in Uganda. Br J Cancer 2008;99:63–67. 8. William Orme, et al.: Human development index in 2013 report online http://www.undp.org/content/undp/en/home/librarypage/ hdr/human‐development‐report‐2013/ Accessed on March 20, 2014. 9. Jatoi I, Anderson WF, Rao SR, Devesa SS: Breast cancer trends among black and white women in the United States. J Clin Oncol 2005;23:7836–7841. 10. Flannery JT, Sullivan PD: Female breast cancer in Connecticut – incidence, mortality, survival – 1935‐1975. Conn Health Bull 1978;92:4–9. 11. Groen RS, Samai M, Petroze RT, et al.: Pilot testing of population‐ based surgical survey tool in Sierra Leone. World J Surg 2012;36: 771–774. 12. Groen RS, Samai M, Stewart KA, et al.: Untreated surgical conditions in Sierra Leone: a cluster randomised, cross‐sectional, countrywide survey. Lancet 2012;380:1082–1087. 13. Petroze RT, Groen RS, Niyonkuru F, et al.: Estimating operative disease prevalence in a low‐income country: results of a nationwide population survey in Rwanda. Surgery 2013;153: 457–464. 14. Groen RS: SOSAS Surgeons OverSeas Assessment of Surgical Needs, A manual for SOSAS interviewers, Version 3.0, accessed on March 20, 2014. http://www.adamkushnermd.com/files/ INTERVIEWERS_MANUAL_FOR_SOSAS_SL.pdf 15. Obaikol R, Galukande M, Fualal J: Knowledge and practice of breast self examination among female students in a sub Saharan African university. East Central Afr J Surg 2010;15:22–27. 16. CIA world factbook, Index mundi: Rwanda Demographics Profile 2013. https://www.cia.gov/library/publications/the‐world‐factbook/ geos/rw.html Accessed on March 20, 2014. 17. CIA world factbook, Index mundi: Sierra Leone Demographics Profile 2013. https://www.cia.gov/library/publications/the‐world‐ factbook/geos/sl.html Accessed on March 20, 2014. 18. Thomas DB, Gao DL, Ray RM, et al.: Randomized trial of breast‐ self examination in Shanghai: final results. J Natl Cancer Inst 2002; 94:1445–1457. 19. Schwartsmann G: Breast cancer in South America: challenges to improve early detection and medical management of a public health program. J Clin Oncol 2001;19:118S–124S. 20. Adesunkanmi AR, Lawal OO, Adelusola KA, Durosimi MA: The severity, outcome and challenges of breast cancer in Nigeria. Breast 2006;15:399–409.
