key: cord-0761984-in4bxlhd authors: Hutajulu, Susanna Hilda; Prabandari, Yayi Suryo; Bintoro, Bagas Suryo; Wiranata, Juan Adrian; Widiastuti, Mentari; Suryani, Norma Dewi; Saptari, Rorenz Geraldi; Taroeno-Hariadi, Kartika Widayati; Kurnianda, Johan; Purwanto, Ibnu; Hardianti, Mardiah Suci; Allsop, Matthew John title: Delays in the presentation and diagnosis of women with breast cancer in Yogyakarta, Indonesia: A retrospective observational study date: 2022-01-13 journal: PLoS One DOI: 10.1371/journal.pone.0262468 sha: dc07e0ef6f13f95259fb4cfe22ad641c77814ab1 doc_id: 761984 cord_uid: in4bxlhd PURPOSE: To investigate factors associated with delays in presentation and diagnosis of women with confirmed breast cancer (BC). METHODS: A cross-sectional study nested in an ongoing prospective cohort study of breast cancer patients at Dr Sardjito Hospital, Yogyakarta, Indonesia, was employed. Participants (n = 150) from the main study were recruited, with secondary information on demographic, clinical, and tumor variables collected from the study database. A questionnaire was used to gather data on other socioeconomic variables, herbal consumption, number of healthcare visits, knowledge-attitude-practice of BC, and open-ended questions relating to initial presentation. Presentation delay (time between initial symptom and first consultation) was defined as ≥3 months. Diagnosis delay was defined as ≥1 month between presentation and diagnosis confirmation. Impact on disease stage and determinants of both delays were examined. A Kruskal-Wallis test was used to assess the length and distribution of delays by disease stage. A multivariable logistic regression analysis was conducted to explore the association between delays, cancer stage and factors. RESULTS: Sixty-five (43.3%) patients had a ≥3-month presentation delay and 97 (64.7%) had a diagnosis confirmation by ≥1 month. Both presentation and diagnosis delays increased the risk of being diagnosed with cancer stage III-IV (odds ratio/OR 2.21, 95% CI 0.97–5.01, p = 0.059 and OR 3.03, 95% CI 1.28–7.19, p = 0.012). Visit to providers ≤3 times was significantly attributed to a reduced diagnosis delay (OR 0.15, 95% CI 0.06–0.37, p <0.001), while having a family history of cancer was significantly associated with increased diagnosis delay (OR 2.28, 95% CI 1.03–5.04, p = 0.042). The most frequent reasons for delaying presentation were lack of awareness of the cause of symptoms (41.5%), low perceived severity (27.7%) and fear of surgery intervention (26.2%). CONCLUSIONS: Almost half of BC patients in our setting had a delay in presentation and 64.7% experienced a delay in diagnosis. These delays increased the likelihood of presentation with a more advanced stage of disease. Future research is required in Indonesia to explore the feasibility of evidence-based approaches to reducing delays at both levels, including educational interventions to increase awareness of BC symptoms and reducing existing complex and convoluted referral pathways for patients suspected of having cancer. Introduction Breast cancer has the highest cancer incidence in Indonesia and the second highest cause of cancer mortality in females. In 2018, the annual estimated incidence and mortality per 100,000 individuals were 44 and 15.3 per year respectively [1] . In the region of Yogyakarta, Indonesia, breast cancer patients are generally diagnosed at stages III and IV [2, 3] . The 5-year overall survival rate is generally unfavourable. It is 48-50% for the whole disease spectrum [2, 4] and 12% for those with advanced diseases [4] . Delays in breast cancer presentation and diagnosis are likely to be key factors in advancedstage presentation [5] . There are disparities in the length of delays between countries. Reports from high-income countries report median times of 14-60 days [6] [7] [8] , with a presentation delay of >3 months occurring in 17-35% of patients [6, 7, [9] [10] [11] . Reports from low-and middle-income countries report a longer length of delays. A study conducted in the neighbouring country of Malaysia found that the median time to consultation and diagnosis was 2 months and 5.5 months, respectively [12] . In Rwanda, a median time of 5 months for both presentation and diagnosis delays were observed [13] . In Indonesia, previous studies have observed >3-month presentation delay in 36.2% of cases and >1-month diagnosis delay in 25% of cases [14] . Furthermore, a 7-month median time of presentation delay and a 6-month median time of the delay to commencing treatment, relating to the time taken from diagnosis to initial treatment, have been observed [15] . Many sociodemographic factors, clinical factors, and patients' experiences have been reported as influencing presentation delay. Age, residence, distance to a medical facility, marital status, education level, occupation, insurance, health facility visits, visiting traditional medicine practitioners, knowledge of breast cancer, breast self-examination, initial symptom, family history of breast cancer, and comorbidities are factors that have been associated with delays in both presentation and subsequent diagnosis [5, 6, 10, 13, [16] [17] [18] [19] . Reasons for delays derived values for date information to avoid identification of participants. Sharing of the full deidentified dataset is not possible due to restrictions imposed by the ethics committee as most of these contain patient data, albeit de-identified, and it may be possible to determine the identity of participants given the extent of sociodemographic and clinical data available for each participant. Should there be a request for data, this can be sent to the corresponding author (email: susanna. hutajulu@ugm.ac.id). Future researchers can contact the institutional ethics committee (email: mhrec_fmugm@ugm.ac.id) at Universitas Gadjah Mada, Indonesia with data access queries as well. include lighter symptoms, fear of informing other people, negative attitudes toward medical health professionals and fear of treatment [13, 20] . In a local publication, initially consulting with non-medical practitioners for breast-related complaints and consuming non-medical treatments have also been associated with diagnosis and treatment delays [14] . Observed by qualitative studies, a lack of awareness and knowledge of cancer, cancer beliefs, treatment beliefs, financial problems, emotional burden, paternalistic style of communication, and unmet information needs are related to psychosocial and cultural reasons for patient delay [15, 21, 22] . To date, research focusing on delays in presentation and diagnosis for patients with breast cancer [15, 22, 23] in Indonesia have included qualitative studies [15, 23, 24] , including a study undertaken in Yogyakarta, the region in which this study is focused [21] . There is very limited research quantifying factors affecting delays in Indonesia [14, 15] , with none exploring factors in Yogyakarta. The Special Region of Yogyakarta, with a current population of 3,882,288 [24] , has the highest frequency of cancer in the country [25] , with breast cancer being the commonest malignancy [26] . The objective of this study is to quantitatively investigate the factors associated with presentation and diagnosis delays, the relationship of delays to stage at presentation and reasons for patient delay within local breast cancer cases. This cross-sectional nested study recruited 150 Indonesian female breast cancer patients registered in a prospective ongoing main study. The main study aims to analyse the risk of side effects from chemotherapy and determine their effect on the survival and quality of life in 250 breast cancer patients. The registered participants are patients visiting and receiving their first chemotherapy treatment in the Hematology and Medical Oncology Division, "Tulip"/Integrated Cancer Clinic, Dr. Sardjito General Hospital, Yogyakarta, Indonesia, from 2018-2022. Women aged �18 years with histopathologically confirmed breast cancer without terminal condition and severe congestive heart failure have been recruited. Cases received chemotherapy as neoadjuvant treatment (before surgery), adjuvant treatment (after surgery), or palliative treatment with or without surgery. Study subjects were contacted and approached to participate in the cross-sectional observation study with consent before recruitment. From the main study database, we collected secondary information on demographic variables (age, education, and residence), comorbidity, family history of cancer, date of the first symptom, date of first medical visit, nutritional status (determined by body mass index/BMI), first symptom of breast cancer, and cancer staging upon diagnosis. To acquire determinants that were unavailable initially in the study database, we developed a questionnaire with 27 questions in the initiation phase. Questions that were listed in the questionnaire were adapted from multiple sources, including the Indonesian Family Life Survey Wave 5 [27] for living arrangement, socioeconomic status and accessibility to the first medical facility visit and Breast Module of the Cancer Awareness Measure (Breast-CAM) for knowledge of breast cancer risk factors and habit of breast self-exam [28] . We also adapted variables from existing studies to develop questionnaire items [13, 17, 29, 30] . Questionnaire development started with the identification of variables that will be measured and the items from existing questions that related to the selected variables. Items were further selected before the translation process. Three independent translators performed forward-backward translation to produce a questionnaire with semantic equivalence to the original versions in English. Assessment of validity included face validity and questionnaire finalization (see S1 Fig) . Face validation was conducted through pilot testing the questionnaire with six lay members of the public, convenience sampled through contacts known to the research team. Three women were primary care facility patients who came for a routine appointment, two women were primary care facility administrative officers, and one woman was a household assistant for a research team member. We conducted a discussion with the participants and gained helpful suggestions for determining which questions required refinement. Through this process, from an initial set of 27 questions, one question about the type of facilities visited to check their symptoms was removed due to redundancy. We also added nine questions to accommodate issues that had not been covered in the initial set of questions. These questions were about patients' address based on their ID card, whether patients still lived at the same address, date of initial symptom, date of first medical contact, distance from their place of residence to the site of first medical contact, and four questions about herbal medicine consumption (the type, dose, and frequency of consumption). While the date of the first breast cancer symptom was already provided in the database, we asked it again in the questionnaire to confirm the information. Five questions were re-worded to improve readability and clarity, and open-ended questions were included at the end of the questionnaire to elicit participants' reasons for presentation delay. The process of face validity resulted in a total set of 35 questions. From September 2020 to February 2021, a face to face interview was conducted with the participants. Trained researchers performed the interview at the ambulatory clinic for 77 participants and, due to the pandemic situation, through phone calls for 74 participants. The questionnaire was only administered once patient informed consent had been obtained. The joint ethics committee from the Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada/Dr Sardjito General Hospital, Yogyakarta has approved the main ongoing prospective study (reference number: KE/FK/0417/EC/2018) and provided specific ethical approval for the study reported in this manuscript (amendment from the ongoing study with reference number KE/FK/0444/EC/2020). Presentation delay was defined as the number of months between breast symptom onset and the patient's first presentation to a medical professional (doctor, nurse, or midwife). Diagnosis delay was used to define the number of months between the first visit to a medical professional and the date of the first pathology report confirming a diagnosis of breast cancer. When respondents could not remember the date of when the symptoms first appeared or the date when they first visited a health facility, we asked for a time span in months. After providing the range of months, we sought to direct the patient to remember the distance between those dates from important dates/events such as family/respondent's birthdays, or religious holidays, to further narrow the range and improve the recall closer to the exact date. When provided with a single month, we tried to ask about the exact day or its distance with other important events in the month, to further narrow the date into a single exact day. We also set breast surgery/procedure dates that were recorded in clinical notes as a benchmark, because it was considered by most patients as an important event. Participants were asked the date of first symptom and first medical visit, based on surgery date as a benchmark. When respondents were unable to provide a date for when their symptoms began or the first provider visit, they were asked to provide a month or month range and year. If they provided a month, the date was estimated as the 15th of that month; if they provided a month range, the estimated date was the midpoint between the 15th of those months. If patients were only able to provide a year, the estimated date was June 30th of that year [13] . An agreement was decided when there was conflicting information of the date of first symptoms in the existing database and that was collected from the interview. In such a case, we relied on the self-report data in the database. We used a delay of �3 months to define presentation delay based on substantial evidence that such delays are associated with lower survival. We used �1 month to define diagnosis delay because one month is indicated as an adequate time for the physician to take appropriate action and shorter delays are not clinically significant [20] . The key independent variables included various parameters that were obtained from the existing study database and the interview. Sociodemographic information included age at diagnosis, residence (urban or rural, referred to the regional statistical bureau), distance to the first health facility visited (<3 km or �3 km), living arrangement (living alone, with spouse only, with other than a spouse, or with spouse and others), educational attainment (3 times and family history of cancer. Our findings highlighted those women who presented late often presented with an advanced stage of the disease. This is consistent with extended data for various presentation and diagnosis times (see S2 Table) . Moreover, the risk of lower BMI was also observed (see S3 Table) , implying an increased cancer burden and reduced survival potential. The �3-month patient delay rate of 43.3% in the study population is higher than those observed in most high-income countries such as United States (17%) [7] , Europe (17.3-20%) [6, 11] or Asia (29-35%) [9, 10] . It is comparable to reports from Asian countries including Malaysia and Iran (42.5-43.3%) [12, 29] and lower than other low-and middle-income countries like Pakistan (84%) [19] and Kenya (73.08%) [34] . A high proportion (64.7%) of �1 month diagnosis delay in our local breast cancer cohort is also higher than those previously documented (15.5-38%) [8, 11] . Factors that influence both delays in presentation and diagnosis have been extensively explored across international publications. However, their impact and mitigation strategies are yet to be described and evaluated in Indonesian settings. In our local setting, lower-income is potentially associated with a presentation delay, supporting findings in other countries [10, 19, 29] . The introduction of national universal health coverage in 2014 sought to increase access to healthcare to the population [35] . However, various out-of-pocket expenses remain, not covered by universal health coverage, such as transportation and logistics linked to presentation, diagnosis and subsequent treatments. Despite a large proportion of medical-related costs covered, there is still a financial burden placed on families with lower economic status. In addition, patients from lower socio-economic groups may have less health awareness and suboptimal family support [10] . Frequent (>3 times) medical visits after first initial consultation was the only significant predictor of delay in receiving a diagnosis in the present study. Although we explored patientlevel factors, we did not specifically explore elements of the health service surrounding provider delays. The main factors for system-oriented delay include failure of medical practitioners to act on clinical findings, maltreatment of symptoms as benign breast disease, falsenegative or misinterpretation of mammogram, and false-negative results of fine-needle aspiration cytology [9, 10, 12, 16, 36] . These factors may be targets for inclusion in future investigation and improvement initiatives. In the research literature, delays in diagnosis may be reduced through the provision of more efficient training programmes for members of the medical profession. Furthermore, our findings indicate a need for simpler and more efficient referral systems to access centres providing cancer care. As a response to long referral and waiting times in cancer care, the Swedish government launched a national policy called "standardized cancer patient pathways" [37] . This policy assigned all phases, from the first suspicion of cancer until the point of receiving treatment, with an appointed maximum time-scale, based on the optimal time for the patients and variation between diagnoses. Similar policies have been introduced across countries including Denmark and Norway [38, 39] . Furthermore, a family history of cancer is also associated with an increased risk of diagnosis delay in our local cohort, supporting other reports [40, 41] . Fear of treatment and its side effects that may have been witnessed in other family members could be a reason for delaying diagnosis and requires further exploration. Most patients tend to ignore symptoms, are not alarmed when they initially appear and do not limit daily activity. This may lead individuals to believe it is not harmful and thus delay seeking medical help. As much as 41.5% of women discredited the symptom due to the absence of pain and 27.7% thought it was not serious and thus did not require any medical attention. Delays in help-seeking behaviour have a major effect on a patient's prognosis and survival, yet, are potentially preventable. Recent studies reported a low awareness level of breast cancer risk factors, barriers, attitudes and breast cancer screening among Indonesian women. This may reflect an inadequate or a lack of breast cancer awareness in the country [42, 43] . Additional education programs aiming to increase awareness and public education has been recommended [43] to improve awareness of breast cancer signs and symptoms. There is an urgent need to develop communication and education strategies regarding breast cancer symptoms and early detection for Indonesia, such as specifically motivating early detection practices and breast self-examination [44] . Various methods could be explored, including mass media, various forms of literature, and programmes. One example is the 'Be Cancer Alert Campaign', using mass media for raising breast and colorectal cancer awareness conducted in Malaysia [45] . In one systematic review, a breast cancer awareness campaign was found to increase the initiation of breast self-examination behaviours and increase the attendance of breast cancer screening [46] . There is scope to explore the feasibility and wider evaluation of similar campaigns in Indonesia. For example, a policy of breast cancer self-examination was introduced by the Indonesian Ministry of Health in 2015 but has yet to be comprehensively evaluated. Our findings reemphasize the need for cancer prevention programs to focus on making women aware of any symptoms in the breast, especially in those with lower economic status. Others indicated that women living in urban areas have a poorer level of knowledge of breast cancer risk factors compared to those living in more rural areas. This group may also serve as a target for future awareness programs [43] . During this study, reluctance to go to attend a health facility during the COVID-19 pandemic was a reason for delayed presentation in one case. Among 65 patients who delayed their first consultation, 10 initially presented to the health facility at the time of the COVID-19 pandemic in Indonesia. When considering the whole panel (n = 150), 2 cases out of 45 women stated the same reason. There was an indication that this became a reason for common delays during the study period. Its effect on the overall health problem in Indonesia still needs to be evaluated. We gathered detailed data from participants in an existing cohort study, using a validated questionnaire that was used by trained researchers in ambulatory clinics. Inclusion from the existing cohort study may introduce some limitations. The study mostly recruited women with good ECOG index, without terminal disease, and who were chemotherapy-naive. Some patients have died because of a terminal disease that developed after diagnosis and treatment administration. Patients with very poor clinical performance and heavily pretreated cases were not included due to the exclusion criteria of the main study. Furthermore, the vast majority of patients underwent staging in the hospital site in which the research team is based (type A hospital) so that there were time intervals between diagnostic confirmation in the district hospitals (type B hospitals) and stage establishment. Some cases possibly have already developed a more advanced stage during the elapsing time. However, a 1-month interval is indicated as representative staging time after being diagnosed in the previous health care [47] . Those who did not experience diagnosis delay in our cohort had a median referral time interval from type B hospitals to ours of 54 days. Many interviews carried out in this study were completed months after participant recruitment to the main study. The first case was recruited in May 2018 while interviews began in September 2020. As a consequence, there is a 1-29 month interval between the first intake in the main study and interviews for this study. Although strategies were in place to mitigate recall bias, this may still have influenced participant responses. Many breast cancer patients in our local setting delayed seeking advice for symptoms later diagnosed as breast cancer symptoms. A high proportion of women experienced a delay in diagnosis. Delays significantly increased the risk of presentation with advanced disease and its association with high mortality probability. Frequent medical visits before diagnosis and family history of cancer were significant determinants of diagnosis delays. Feasibility testing of approaches to promoting community education to promote breast cancer awareness and training for health care professionals is required to explore strategies for potentially minimising delays and mortality from breast cancer in our local region and other settings across Indonesia. Steps for developing the study questionnaire. Questionnaire development started from identifying variables that will be measured and items that were obtained from existing questions related to the selected variables. It is followed by item selection and translation. The last steps included validity assessment or face validity and questionnaire finalization. (TIF) S1 Global Cancer Observatory: Indonesia Population Fact Sheet Age at diagnosis predicted survival outcome of female patients with breast cancer at a tertiary hospital in Yogyakarta, Indonesia Pathological profiles and clinical management challenges of breast cancer emerging in young women in Indonesia: a hospital-based study. BMC Women's Health Analisis ketahanan hidup 5 tahun pada penderita kanker payudara di rumah sakit kanker Dharmais. Universitas Indonesia Delays in diagnosis and treatment of breast cancer: a multinational analysis Factors affecting breast cancer treatment delay in Turkey: a study from Turkish Federation of Breast Diseases Societies Breast cancer presentation and diagnostic delays in young women Patient and Care Delays of Breast Cancer in China Barriers to early presentation of self-discovered breast cancer in Singapore and Malaysia: a qualitative multicentre study Delayed presentation of symptomatic breast cancers in Hong Kong: experience in a public cancer centre Reasons for delay in treatment of breast cancer detected due to breast self-examination in women from the Lubelskie region Diagnosis delay of breast cancer and its associated factors in Malaysian women Delays in Breast Cancer Presentation and Diagnosis at Two Rural Cancer Referral Centers in Rwanda. Oncologist Profil cancer delay pada kasus kanker payudara di RS Onkologi Surabaya Psychosocial and cultural reasons for delay in seeking help and nonadherence to treatment in Indonesian women with breast cancer: a qualitative study Diagnosis delay in Libyan female breast cancer The effect of age on delay in diagnosis and stage of breast cancer Reasons for patient's delay in diagnosis of breast carcinoma in Pakistan Patient Self Delay Among Women With Breast Cancer Understanding why women delay in seeking help for breast cancer symptoms Exploring possible causes for delays seeking medical treatment among Indonesian women with breast cancer Determinants of early breast cancer presentation: a qualitative exploration among female survivors in Indonesia. Psychol Health Factors associated with late detection of breast cancer: application of health belief model theory Jogja Cancer Registry. Laporan data registrasi kanker berbasis populasi 2017 (report of populationbased cancer registry in 2017). 2020. Available from Community-facility Survey Questionnaire for the Indonesia Family Life Survey, Wave 5. Rand Cancer Research UK. Breast module of the Cancer Awareness Measure (Breast-CAM): Toolkit (version 2) Patient delay in women presenting with advanced breast cancer: an Iranian study. Public Health Factors associated with delayed diagnosis of breast cancer in northeast Thailand Factors that modify breast cancer risk in women The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM Using thematic analysis in psychology Delayed presentation of breast cancer patients Universal health coverage in Indonesia: concept, progress, and challenges Breast cancer early detection and diagnostic capacity in Uganda. Cancer Implementing standardized cancer patient pathways (CPPs)-a qualitative study exploring the perspectives of health care professionals Cancer patient pathways in Denmark as a joint effort between bureaucrats, health professionals and politicians-a national Danish project. Health Policy Implementering av pakkeforløp for kreft. Nasjonal plan for implementering av pakkeforløp for kreft Determinants of delay in diagnosis and end stage at presentation among breast cancer patients in Iran: a multicenter study. Sci Rep Socioeconomic Factors, Health Behavior, and Late-Stage Diagnosis of Breast Cancer: Considering the Impact of Delay in Diagnosis Determinants of cancer screening awareness and participation among Indonesian women Awareness Level about Breast Cancer Risk Factors, Barriers, Attitude and Breast Cancer Screening among Indonesian Women Can Breast Self-examination and Clinical Breast Examination Along With Increasing Breast Awareness Facilitate Earlier Detection of Breast Cancer in Populations With Advanced Stages at Diagnosis?. Clin breast cancer The 'Be Cancer Alert Campaign': protocol to evaluate a mass media campaign to raise awareness about breast and colorectal cancer in Malaysia The impact of breast cancer awareness interventions on breast screening uptake among women in the United Kingdom: A systematic review Delay in breast cancer: implications for stage at diagnosis and survival. Front Public Health We gratefully thanked Irfan Haris, Riani Witaningrum, Yufi Kartika Astari, and Betrix Rifana Kusumaning Indah for technical assistance.