Study of Diagnostic Delay among Symptomatic Breast Cancer Patients in Northern India: A Mixed-Methods Analysis from a Dedicated Breast Cancer Centre

Abstract

Objective:

We conducted this study to understand the factors that contribute to the delay in diagnosis of symptomatic breast cancer patients.

Methodology:

We did a prospective analytical study with both quantitative and qualitative components over 14 months. The total delay in diagnosis of breast cancer from first symptom onset was defined as diagnostic delay. Presentation delay was defined as the time duration between the first symptom and the first visit to a health care provider. Provider delay was defined as the time duration between the presentation to a health care provider and the final diagnosis. Three hundred female breast cancer patients with a diagnostic delay of at least 3 months were interviewed using a pre-tested semi-structured questionnaire.

Results:

We found that more than 50% of patients with delayed presentation were between 30-50 years of age. Painless breast lump was the most common initial symptom. More than 70% of patients presented with locally advanced and metastatic disease. The patient-related delay was more common than provider delay. Breast cancer awareness (p = 0.040) and reasons for delay (p = 0.014) were found to significantly influence the delay. More than 70% of patients reported their symptoms to male members of their families.

Conclusion:

Breast cancer awareness is the single most important determinant influencing diagnostic delay among symptomatic patients.

Introduction

Breast cancer is the most common site-specific cancer and a leading cause of death from cancer in women globally (DeVita et al., 2019). It accounted for 11.7% of new cases and 6.9% deaths of all cancers taken together in the year 2020 as per GLOBOCAN estimates. Breast cancer remains the most common cancer and the most common cause of cancer-related mortality in India with 7.85% of all cases from the country (Sung et al., 2021). The projection for India during the periods between 2020 and 2040 predicts the number of cases to go as high as 3,70,000 from the current 1,80,000 with its approximate relative percentage remaining the same among all the cancers (ICMR, 2016; Sung et al., 2021).

In several developed countries, including the United States, Canada, the United Kingdom, France, and Australia, the fall in incidence during early 2000s was partly attributable to a decline in the use of postmenopausal hormonal treatment after publication of the Women’s Health Initiative trial linking postmenopausal hormone use to increased breast cancer risk (WHI, 2002; Ravdin et al., 2007). Overall, in the Western population there has been a fall in deaths caused by breast cancer as a result of breast screening, early diagnosis and better treatment, although the relative contribution of these factors is yet to be evaluated (Richards et al., 2000).

However, it is still a non-existent entity for majority of Indian population. Healthcare is low on priority and even in major cities screening is an ‘alien’ word for most people. The numerous myths and ignorance that prevail in Indian society about breast cancer have resulted in an unrealistic fear of the disease (Aggarwal et al., 2007). According to various studies, majority of carcinoma breast cases in the West report in stages I and II of disease, whereas in India 45.7% patients report in advanced stages (Kakarala et al., 2010; Leong et al., 2010). Studies suggest that the disease peaks at 40–50 years in Indian women (Chopra et al., 2014). Breast cancer awareness programs are more concentrated in the cities and have still not reached the remote and rural parts of the country (Chopra, 2001). Women often do not present for medical care early enough due to various reasons such as illiteracy, lack of awareness, and financial constraints. Thus, majority of breast cancer patients in India are still treated at locally advanced and metastatic stages (Chopra, 2001; Aggarwal et al., 2008). Breast cancer has been replacing cancer of the cervix as the leading site of cancer in all urban cancer registries in India (ICMR, 2016).

Taken together these facts indicate that most Indian women are diagnosed at an advanced stage of breast cancer during the most productive period of their lives. It has been proven that patients with longer duration of symptoms present with more advanced disease and that a delay in diagnosis between 3-6 months after onset of symptoms reduces 5-years survival by 7% compared to patients diagnosed within 3 months of onset of symptoms (Richards et al., 1999; Montazeri et al., 2003). The time between first symptom and treatment is longer than 3 months in at least a third of all patients, and in about a quarter the time is longer than 6 months (Richards et al., 1999). As organized, mammographic screening programs are not feasible or cost effective in low resource settings, efforts must be made to promote early diagnosis of breast cancer through better awareness along with appropriate and timely treatment (WHO, 2014; Birnbaum et al., 2018; Ngan et al., 2020). A study from sub-Saharan countries reported that more than 25% deaths could be prevented by early diagnosis and timely treatment of breast cancer (McCormack et al., 2020). In a study from the United States, authors reported a significant improvement in breast cancer mortality before the introduction of mammographic screening and adjuvant therapy (Arndt et al., 2002).

Lack of screening program and delay in diagnosis is hindering attempts to improve breast cancer survival in India. Studies over the last 20 years have assessed the delays by both the patients and providers among breast cancer patients, however an in-depth, contextual understanding of the reasons for delay is more important (Arndt et al., 2002). Significant gains can be made by encouraging women to seek help more quickly, while simultaneously improve hospital practices. To reduce the increasing load of mortality due to breast cancer, we need to lay emphasis on early diagnosis and increased use of systemic therapy. Any strategy aimed to shorten the delay in presentation of breast cancer patients requires a comprehensive understanding of the factors that influence the delays.

Quantitative research methods dominate the study of delayed help-seeking among breast cancer patients; focusing mainly on mapping incidences and stages of cancer at diagnosis, establishing the relationship between survival outcomes and duration of delays, and knowledge of breast cancer and demographic attributes (Lin et al., 2015). Seven systematic reviews have been performed between 1999 and 2020, to examine the factors influencing presentation, diagnosis and breast cancer care and of the 144 included studies, only 13 were qualitative and 3 were mixed methods study (Ramirez et al., 1999; Alhurishi et al., 2011; Sharma et al., 2012; Jones et al., 2014; Khan et al., 2015; Espina et al., 2017; Kailemia et al., 2020).

Unlike quantitative research, qualitative research provides in-depth explanations and contextual information crucial to understanding attitudes, decision-making processes, behaviours, concerns, motivations, culture or lifestyles, as well as in generating ideas for new strategies, interventions or theories (Austin and Sutton, 2014). In a study from the UK, authors concluded that delay in presentation of breast cancer patients was not related to socioeconomic status and other demographic variables and instead associated with the beliefs about the breast symptoms and their attribution (Nosarti et al., 2000). Similar findings were reported in a qualitative study among the Iranian women (Khakbazan et al., 2014). These results provide a basis for comprehensive qualitative evaluation of delayed presentation among symptomatic patients. Few qualitative studies evaluating the screening programme (Bener et al., 2002; Lamiyan et al., 2007), social support (Bener et al., 2002), religious and spiritual role (Harandy et al., 2010), living experience with breast cancer (Doumit et al., 2010) have been found but none exploring the factors which influence late presentation or delayed help seeking behaviour among breast cancer patients.

We aimed to evaluate both quantitatively and qualitatively, reasons for diagnostic delay in symptomatic breast cancer patients presenting at Integrated Breast Care Centre (IBCC), All India Institute of Medical Sciences, Rishikesh (India) while also assess the quality and strength of the current evidence related to the factors that may predict delays both by the patients, as well as the primary care physicians.

Materials and Methods

Methodology

We conducted a prospective analytical study at the Integrated Breast Care Centre, All India Institute of Medical Sciences, Rishikesh (India) after obtaining ethical clearance from the Institute’s Ethics Committee. Total of 300 patients who visited the Integrated Breast Care Centre during the study period were included based on the pre-specified selection criteria.

Inclusion criteria

1. Females of any age group willing to participate in the study

2. Patients able to respond in person to the study questionnaire

Exclusion criteria

1. Recurrent breast cancer

2. Second primaries

Data was collected with the help of a pre-tested semi structured questionnaire via personal interviews. Information regarding demographic profile, educational status, marital status, occupation, personal history of breast disease, family history of breast cancer, time lag between noticing symptoms and final diagnosis, awareness regarding breast cancer and its source, prevalence and frequency of self-breast examination (SBE), reasons for delaying presentation after noticing symptom, time taken by doctors to give a final diagnosis and tumor characteristics were collected.

The total delay in diagnosis of breast cancer from first symptom onset was defined as diagnostic delay. Presentation delay was defined as the time duration between first symptom and first visit to a health care provider. Provider delay was defined as the time duration between presentation to a health care provider till the final diagnosis was made (Weller et al., 2012).

Data were coded and recorded in Microsoft Excel Spreadsheet program (2007). SPSS v23 (IBM Corp., USA) statistical package was used for data analysis. Descriptive analysis were elaborated in the form of means/standard deviations and medians/IQRs for continuous variables, and frequencies and percentages for categorical variables. Chi-squared test was used for group comparisons for categorical data. In case the expected frequency was in the contigency tables was found to be <5% for more than 25% of the cells, Fisher’s exact test was used instead. One-Way ANOVA and Kruskal Wallis test were used to compare three or more groups in case of normally or non-normally distributed continuous variables respectively. Linear correlation between two continuous variables was explored using Pearson’s correlation (if data was not normally distributed) and Spearman’s correlation (if data was normally distributed). Statistical significance was kept at p<0.05.

Results

Sociodemographic characteristics

We collected data about the sociodemographic profile (age, marital status, residence, religion, education, occupation, etc.) of our patients. The economic stratification of patients was done using the modified Kuppuswamy scale (Saleem and Jan, 2021). Our findings are summarized in Table 1.

Table 1.

Sociodemographic Characteristics of the Participants

characteristics
Age (Mean ± SD) in years	47.82 ± 11.70
Age (Frequency) in %
20-29 Years	11 (3.7%)
30-39 Years	68 (22.7%)
40-49 Years	94 (31.3%)
50-59 Years	70 (23.3%)
60-69 Years	46 (15.3%)
70-79 Years	10 (3.3%)
80-89 Years	1 (0.3%)
State (Frequency) in %
Uttar Pradesh	166 (55.3%)
Uttarakhand	130 (43.3%)
Haryana	2 (0.7%)
Delhi	1 (0.3%)
Jharkhand	1 (0.3%)
Locality (Frequency) in %
Rural	172 (57.3%)
Urban	128 (42.7%)
Religion (Frequency) in %
Hindu	246 (82.0%)
Muslim	48 (16.0%)
Sikh	6 (2.0%)
Marital Status (Frequency) in %
Unmarried	5 (1.7%)
Married	266 (88.7%)
Widowed	29 (9.7%)
Education (Frequency) in %
None	125 (41.7%)
Primary	73 (24.3%)
Secondary	46 (15.3%)
Graduate and above	56 (18.7%)
Occupation (Frequency) in %
Homemaker	266 (88.7%)
Employed	29 (9.7%)
Retired	5 (1.7%)
Socio-Economic Status (Frequency) in %
Upper	10 (3.3%)
Upper Middle	74 (24.7%)
Lower Middle	98 (32.7%)
Upper Lower	13 (4.3%)
Lower	105 (35.0%)

Clinical characteristics

Painless breast lump was the most common presenting complaint followed by breast pain. However, the nature of breast complaint did not influence diagnostic delay (χ² = 18.070, p = 0.213). (Table 2)

Table 2.

Summary of Symptom at Onset

First Symptom (Frequency) in %
Painless breast lump	233 (77.7%)
Pain	51 (17.0%)
Nipple Discharge	9 (3.0%)
Breast Asymmetry	5 (1.7%)
Axillary Ulcer	1 (0.3%)
Pruritus/Skin changes	1 (0.3%)

Tumor characteristics

More than 70% patients presented with locally advanced and advanced breast cancer. 98.6% patients had an invasive breast cancer at the time of presentation. A large number of patients presented at cT4b disease (36.0%). Thirty-four patients (11.3%) had metastatic disease at the time of presentation. T4b stage is associated with skin changes of breast cancer (ulceration, satellite nodules and Peau d’orange appearance) (Kalli et al., 2018). However, we did not find any association between the cT (χ² = 27.190, p = 0.296) or anatomical tumor stage (χ² = 17.875, p = 0.657) and delayed presentation.

Delay in diagnosis of breast cancer

Diagnostic delay was more common in women without any formal education (41.7%) compared to those who received some form of education. However, at least 15% patients who were graduate and above also delayed help seeking. Statistical analysis failed to show any association between literacy level and diagnostic delay (χ² = 9.546, p = 0.388) (Table 3).

Table 3.

Summary of Delay in Diagnosis of Breast Cancer

Delay	Mean ± SD	Median (IQR)
Diagnostic Delay (Months)	9.63 ± 8.62	6.33 (4.96-12.33)
Presentation Delay (Months)	7.98 ± 8.68	5.00 (4.00-8.00)
Provider Delay (Months)	1.64 ± 3.25	0.50 (0.33-1.00)

Reasons for delay

More than 50% patients sought help late due to an ‘apparently’ asymptomatic breast lump not associated with any other complaint. At least 20% patients took other form of treatments (Ayurvedic/Homeopathic) before reaching our centre. Covid-19 imposed lockdown was responsible for the delay in 4.8% patients (Table 4).

Table 4.

Summary of the Reasons for Delay in Diagnosis of Breast Cancer

Reasons for Delay
Reasons for Presentation Delay (Frequency) in %
Asymptomatic breast lump	149 (55.0%)
Took Ayurvedic medicines	37 (13.7%)
Took Homeopathic medicines	30 (11.1%)
COVID-19 lockdown	13 (4.8%)
Lack of family support	11 (4.1%)
Misinformed by unqualified practitioners (‘quacks’)	10 (3.7%)
Fear of diagnosis of cancer	6 (2.2%)
Poor access to health services	6 (2.2%)
Financial issues	4 (1.5%)
Lack of knowledge	3 (1.1%)
Engaged in family activities	1 (0.4%)
Pregnancy	1 (0.4%)
Reasons for Provider Delay (Frequency) in %
Misdiagnosed by primary physician	47 (65.3%)
Delay in investigations	13 (18.1%)
Delayed referral to higher centre	12 (16.7%)

Knowledge and practices

Only 39.0% patients had some awareness about signs or symptoms of breast cancer. Family and friends were the most common source of their awareness. Health workers/professionals contributed just 4.3% to breast cancer awareness. Self-breast examination was practiced by only 7% patients. More than 80% patients were not aware of the cause of breast cancer. Others attributed it to a range of causes (Table 5).

Table 5.

Summary of Knowledge and Practices among the Patients

characteristics
Awareness about breast cancer (Frequency) in %
Yes	117 (39.0%)
No	183 (61.0%)
Source of Awareness (Frequency) in %
Family/Friends	78 (66.6%)
TV/Newspaper	26 (22.2%)
Health-Workers/Professionals	13 (11.2%)
Self-Breast Examination (Frequency) in %
Yes	21 (7.0%)
No	279 (93.0%)
Perceived Causes (Frequency) in %
Not Aware	250 (83.3%)
Lack of Breastfeeding	9 (3.0%)
Trauma to Breast	8 (2.7%)
Hormonal Factors	6 (2.0%)
Genetic Factors	4 (1.3%)
Lifestyle Changes	4 (1.3%)
Old Age	3 (1.0%)
Tobacco Smoking	3 (1.0%)
Environmental Pollution	2 (0.7%)
Fated	2 (0.7%)
Dietary Factors	1 (0.3%)
Family History	1 (0.3%)
Food Habits	1 (0.3%)
Infection	1 (0.3%)
Medications	1 (0.3%)
Menopause Induced	1 (0.3%)
Oral Contraceptive Pills	1 (0.3%)
Oral Tobacco Use	1 (0.3%)
Post-Hysterectomy	1 (0.3%)
First Reported To (Frequency) in %
Husband	179 (59.7%)
Daughter	42 (14.0%)
Son	30 (10.0%)
Daughter-in-law	21 (7.0%)
Mother	8 (2.7%)
Sister	8 (2.7%)
Sister-in-law	5 (1.7%)
Mother-in-law	4 (1.3%)
Brother	1 (0.3%)
Cousin	1 (0.3%)
Friend	1 (0.3%)

Nearly 60% patients reported their complaint to their husbands, followed by daughter, son, daughter-in-law, mother, sister, sister-in-law, mother-in-law, brother, cousin and friend (Table 5).

Association between diagnostic delay and other variables

We did a subgroup analysis of all the variables to identify those which influenced the presentation of breast cancer patients (Table 6). We found that awareness of breast cancer and reasons for presentation delay were significantly associated with diagnostic delay (p < 0.05).

Table 6.

Association between Diagnostic Delay and Other Variables

Parameters	Diagnostic Delay												p- Value
	3-6 Months (n = 133)			6-12 Months (n = 86)				12-24 Months (n = 58)			>24 Months (n = 23)
I. Sociodemographic characteristics
Age (Mean ± SD) in years	48.46 ± 11.26			47.19 ± 11.39				46.05 ± 11.40			50.91 ± 15.52		0.3121
Age (Frequency) in %													0.1982
20-29 Years	3 (27.3%)			3 (27.3%)				3 (27.3%)			2 (18.2%)
30-39 Years	28 (41.2%)			22 (32.4%)				15 (22.1%)			3 (4.4%)
40-49 Years	43 (45.7%)			27 (28.7%)				17 (18.1%)			7 (7.4%)
50-59 Years	33 (47.1%)			19 (27.1%)				15 (21.4%)			3 (4.3%)
60-69 Years	21 (45.7%)			14 (30.4%)				6 (13.0%)			5 (10.9%)
70-79 Years	5 (50.0%)			1 (10.0%)				2 (20.0%)			2 (20.0%)
80-89 Years	0 (0.0%)			0 (0.0%)				0 (0.0%)			1 (100.0%)
State (Frequency) in %													0.8283
Uttar Pradesh	74 (44.6%)			49 (29.5%)				32 (19.3%)			11 (6.6%)
Uttarakhand	56 (43.1%)			37 (28.5%)				26 (20.0%)			11 (8.5%)
Haryana	1 (50.0%)			0 (0.0%)				0 (0.0%)			1 (50.0%)
Delhi	1 (100.0%)			0 (0.0%)				0 (0.0%)			0 (0.0%)
Jharkhand	1 (100.0%)			0 (0.0%)				0 (0.0%)			0 (0.0%)
Locality (Frequency) in %													0.6852
Rural	73 (42.4%)			54 (31.4%)				32 (18.6%)			13 (7.6%)
Urban	60 (46.9%)			32 (25.0%)				26 (20.3%)			10 (7.8%)
Religion (Frequency) in %													0.3513
Hindu	109 (44.3%)			66 (26.8%)				49 (19.9%)			22 (8.9%)
Muslim	21 (43.8%)			19 (39.6%)				7 (14.6%)			1 (2.1%)
Sikh	3 (50.0%)			1 (16.7%)				2 (33.3%)			0 (0.0%)
Marital Status (Frequency) in %													0.1243
Unmarried	1 (20.0%)			0 (0.0%)				2 (40.0%)			2 (40.0%)
Married	118 (44.4%)			78 (29.3%)				52 (19.5%)			18 (6.8%)
Widowed	14 (48.3%)			8 (27.6%)				4 (13.8%)			3 (10.3%)
Education (Frequency) in %													0.3882
None	45 (36.0%)			43 (34.4%)				27 (21.6%)			10 (8.0%)
Primary	36 (49.3%)			21 (28.8%)				10 (13.7%)			6 (8.2%)
Secondary	21 (45.7%)			11 (23.9%)				11 (23.9%)			3 (6.5%)
Graduate and above	31 (55.4%)			11 (19.6%)				10 (17.9%)			4 (7.1%)
Occupation (Frequency) in %													0.1663
Homemaker	117 (44.0%)			77 (28.9%)				54 (20.3%)			18 (6.8%)
Employed	14 (48.3%)			9 (31.0%)				3 (10.3%)			3 (10.3%)
Retired	2 (40.0%)			0 (0.0%)				1 (20.0%)			2 (40.0%)
Socio-Economic Status (Frequency) in %													0.3622
Upper	4 (40.0%)			2 (20.0%)				3 (30.0%)			1 (10.0%)
Upper Middle	36 (48.6%)			17 (23.0%)				16 (21.6%)			5 (6.8%)
Lower Middle	52 (53.1%)			25 (25.5%)				12 (12.2%)			9 (9.2%)
Upper Lower	5 (38.5%)			5 (38.5%)				2 (15.4%)			1 (7.7%)
Lower	36 (34.3%)			37 (35.2%)				25 (23.8%)			7 (6.7%)
II. Clinical characteristics
First Symptom (Frequency) in %													0.2133
Painless breast lump	97 (41.6%)			72 (30.9%)				44 (18.9%)			20 (8.6%)
Pain	25 (49.0%)			12 (23.5%)				12 (23.5%)			2 (3.9%)
Nipple discharge	7 (77.8%)			0 (0.0%)				1 (11.1%)			1 (11.1%)
Parameters			Diagnostic Delay												p- Value
			3-6 Months (n = 133)			6-12 Months (n = 86)			12-24 Months (n = 58)			>24 Months (n = 23)
II. Clinical characteristics
First Symptom (Frequency) in %															0.2133
Breast asymmetry			4 (80.0%)			1 (20.0%)			0 (0.0%)			0 (0.0%)
Axillary ulcer			0 (0.0%)			0 (0.0%)			1 (100.0%)			0 (0.0%)
Pruritus/Skin changes			0 (0.0%)			1 (100.0%)			0 (0.0%)			0 (0.0%)
III. Tumor characteristics
Type of Breast Cancer (Frequency) in %															0.3603
Ductal carcinoma in-situ (DCIS)			1 (50.0%)			1 (50.0%)			0 (0.0%)			0 (0.0%)
DCIS with Paget's Disease			0 (0.0%)			1 (50.0%)			0 (0.0%)			1 (50.0%)
Invasive Carcinoma, NOS			125 (45.1%)			79 (28.5%)			53 (19.1%)			20 (7.2%)
Invasive Lobular Carcinoma			4 (44.4%)			2 (22.2%)			2 (22.2%)			1 (11.1%)
Invasive Metaplastic Carcinoma			2 (66.7%)			1 (33.3%)			0 (0.0%)			0 (0.0%)
Invasive Medullary Carcinoma, NOS			0 (0.0%)			1 (50.0%)			1 (50.0%)			0 (0.0%)
Invasive Mucinous Carcinoma			1 (50.0%)			1 (50.0%)			0 (0.0%)			0 (0.0%)
Invasive Adenosquamous Carcinoma			0 (0.0%)			0 (0.0%)			1 (100.0%)			0 (0.0%)
Invasive Apocrine Carcinoma			0 (0.0%)			0 (0.0%)			1 (100.0%)			0 (0.0%)
Invasive Colloid Carcinoma			0 (0.0%)			0 (0.0%)			0 (0.0%)			1 (100.0%)
T Stage (Frequency) in %															0.2962
Tis			1 (33.3%)			2 (66.7%)			0 (0.0%)			0 (0.0%)
T1			5 (83.3%)			0 (0.0%)			1 (16.7%)			0 (0.0%)
T2			34 (47.9%)			19 (26.8%)			15 (21.1%)			3 (4.2%)
T3			43 (49.4%)			29 (33.3%)			11 (12.6%)			4 (4.6%)
T4a			7 (46.7%)			5 (33.3%)			1 (6.7%)			2 (13.3%)
T4b			40 (37.0%)			28 (25.9%)			27 (25.0%)			13 (12.0%)
T4c			2 (25.0%)			3 (37.5%)			2 (25.0%)			1 (12.5%)
T4d			0 (0.0%)			0 (0.0%)			1 (100.0%)			0 (0.0%)
N Stage (Frequency) in %															0.5582
N0			42 (50.0%)			22 (26.2%)			13 (15.5%)			7 (8.3%)
N1			51 (46.8%)			31 (28.4%)			21 (19.3%)			6 (5.5%)
N2a			22 (37.3%)			20 (33.9%)			11 (18.6%)			6 (10.2%)
N2b			0 (0.0%)			0 (0.0%)			1 (100.0%)			0 (0.0%)
N3a			8 (34.8%)			5 (21.7%)			8 (34.8%)			2 (8.7%)
N3b			0 (0.0%)			2 (50.0%)			1 (25.0%)			1 (25.0%)
N3c			10 (50.0%)			6 (30.0%)			3 (15.0%)			1 (5.0%)
M Stage (Frequency) in %															0.7932
M0			120 (45.1%)			74 (27.8%)			52 (19.5%)			20 (7.5%)
M1			13 (38.2%)			12 (35.3%)			6 (17.6%)			3 (8.8%)
Anatomical Stage (Frequency) in %															0.6572
Stage 0			1 (33.3%)			2 (66.7%)			0 (0.0%)			0 (0.0%)
Stage 1			4 (80.0%)			0 (0.0%)			1 (20.0%)			0 (0.0%)
Stage 2a			16 (48.5%)			7 (21.2%)			8 (24.2%)			2 (6.1%)
Stage 2b			24 (50.0%)			15 (31.2%)			8 (16.7%)			1 (2.1%)
Stage 3a			27 (46.6%)			18 (31.0%)			9 (15.5%)			4 (6.9%)
Stage 3b			36 (40.0%)			25 (27.8%)			17 (18.9%)			12 (13.3%)
Stage 3c			16 (38.1%)			11 (26.2%)			12 (28.6%)			3 (7.1%)
Stage 4			9 (42.9%)			8 (38.1%)			3 (14.3%)			1 (4.8%)
Parameters			Diagnostic Delay												p- Value
			3-6 Months (n = 133)			6-12 Months (n = 86)			12-24 Months (n = 58)			>24 Months (n = 23)
Stage (Frequency) in %															0.4762
Early			44 (50.0%)			24 (27.3%)			17 (19.3%)			3 (3.4%)
Locally advanced			80 (41.9%)			54 (28.3%)			38 (19.9%)			19 (9.9%)
Advanced			9 (42.9%)			8 (38.1%)			3 (14.3%)			1 (4.8%)
IV. Reasons for delay
Reasons for Presentation Delay*** (Frequency) in %															0.0142
Asymptomatic breast lump			80 (53.7%)			41 (27.5%)			19 (12.8%)			9 (6.0%)
Took Ayurvedic medicines			9 (24.3%)			9 (24.3%)			12 (32.4%)			7 (18.9%)
Took Homeopathic medicines			6 (20.0%)			11 (36.7%)			9 (30.0%)			4 (13.3%)
COVID-19 lockdown			7 (53.8%)			5 (38.5%)			1 (7.7%)			0 (0.0%)
Lack of family support			7 (63.6%)			3 (27.3%)			0 (0.0%)			1 (9.1%)
Misinformed by unqualified			3 (30.0%)			2 (20.0%)			5 (50.0%)			0 (0.0%)
practitioners (‘quacks’)
Fear of diagnosis of cancer			2 (33.3%)			1 (16.7%)			2 (33.3%)			1 (16.7%)
Poor access to health services			4 (66.7%)			2 (33.3%)			0 (0.0%)			0 (0.0%)
Financial issues			0 (0.0%)			2 (50.0%)			2 (50.0%)			0 (0.0%)
Lack of knowledge			1 (33.3%)			1 (33.3%)			0 (0.0%)			1 (33.3%)
Engaged in family activities			1 (100.0%)			0 (0.0%)			0 (0.0%)			0 (0.0%)
Pregnancy			1 (100.0%)			0 (0.0%)			0 (0.0%)			0 (0.0%)
Reasons for Provider Delay (Frequency) in %															0.4153
Misdiagnosed by primary physician			11 (23.9%)			16 (34.8%)			18 (39.1%)			2 (2.2%)
Delay in investigations			4 (30.8%)			7 (53.8%)			2 (15.4%)			0 (0.0%)
Delayed referral to higher centre			3 (25.0%)			7 (58.3%)			2 (16.7%)			0 (0.0%)
V. Knowledge and practices
Awareness (Yes)***			59 (50.4%)			26 (22.2%)			19 (16.2%)			13 (11.1%)			0.0402
Source of Awareness (Frequency) in %															0.2302
Family/Friends			40 (50.6%)			21 (26.6%)			11 (13.9%)			7 (8.9%)
TV/Newspaper			14 (53.8%)			3 (11.5%)			5 (19.2%)			4 (15.4%)
Health-Workers/Professionals			5 (38.5%)			3 (23.1%)			3 (23.1%)			2 (15.4%)
SBE (Present)			14 (66.7%)			3 (14.3%)			3 (14.3%)			1 (4.8%)			0.2293
Perceived Causes (Frequency) in %
Not aware			113 (45.2%)			75 (30.0%)			47 (18.8%)			15 (6.0%)
Lack of breastfeeding			4 (44.4%)			1 (11.1%)			3 (33.3%)			1 (11.1%)
Trauma to breast			3 (37.5%)			1 (12.5%)			3 (37.5%)			1 (12.5%)
Hormonal factors			1 (16.7%)			2 (33.3%)			1 (16.7%)			2 (33.3%)
Genetic factors			3 (75.0%)			0 (0.0%)			0 (0.0%)			1 (25.0%)
Lifestyle changes			1 (25.0%)			1 (25.0%)			1 (25.0%)			1 (25.0%)
Old age			0 (0.0%)			1 (33.3%)			1 (33.3%)			1 (33.3%)
Tobacco smoking			2 (66.7%)			0 (0.0%)			0 (0.0%)			1 (33.3%)
Environmental pollution			1 (50.0%)			1 (50.0%)			0 (0.0%)			0 (0.0%)
Fated			0 (0.0%)			2 (100.0%)			0 (0.0%)			0 (0.0%)
Dietary factors			0 (0.0%)			0 (0.0%)			1 (100.0%)			0 (0.0%)
Family history			0 (0.0%)			1 (100.0%)			0 (0.0%)			0 (0.0%)
Food habits			1 (100.0%)			0 (0.0%)			0 (0.0%)			0 (0.0%)
Infection			0 (0.0%)			1 (100.0%)			0 (0.0%)			0 (0.0%)
Medications			1 (100.0%)			0 (0.0%)			0 (0.0%)			0 (0.0%)
Menopause induced			1 (100.0%)			0 (0.0%)			0 (0.0%)			0 (0.0%)
Parameters		Diagnostic Delay												p- Value
		3-6 Months (n = 133)			6-12 Months (n = 86)		12-24 Months (n = 58)			>24 Months (n = 23)
Perceived Causes (Frequency) in %
Oral Contraceptive Pills		1 (100.0%)			0 (0.0%)		0 (0.0%)			0 (0.0%)
Oral tobacco use		1 (100.0%)			0 (0.0%)		0 (0.0%)			0 (0.0%)
Post-Hysterectomy		0 (0.0%)			0 (0.0%)		1 (100.0%)			0 (0.0%)
First Reported To (Frequency) in %														0.1712
Husband		78 (43.6%)			57 (31.8%)		34 (19.0%)			10 (5.6%)
Daughter		19 (45.2%)			10 (23.8%)		11 (26.2%)			2 (4.8%)
Son		16 (53.3%)			9 (30.0%)		2 (6.7%)			3 (10.0%)
Daughter-in-law		7 (33.3%)			6 (28.6%)		4 (19.0%)			4 (19.0%)
Mother		2 (25.0%)			1 (12.5%)		3 (37.5%)			2 (25.0%)
Sister		4 (50.0%)			1 (12.5%)		3 (37.5%)			0 (0.0%)
Sister-in-law		1 (20.0%)			1 (20.0%)		1 (20.0%)			2 (40.0%)
Mother-in-law		4 (100.0%)			0 (0.0%)		0 (0.0%)			0 (0.0%)
Brother		1 (100.0%)			0 (0.0%)		0 (0.0%)			0 (0.0%)
Cousin		0 (0.0%)			1 (100.0%)		0 (0.0%)			0 (0.0%)
Friend		1 (100.0%)			0 (0.0%)		0 (0.0%)			0 (0.0%)

***Significant at p<0.05, ¹, One-Way ANOVA; ², Chi-Squared Test; ³, Fisher's Exact Test; ⁴, Kruskal Wallis Test

There was a significant difference between the various groups in terms of distribution of diagnostic delay (χ² = 8.303, p = 0.040). The source of awareness was not significantly associated with the diagnostic delay (χ² = 11.709, p = 0.230). The reasons for delay in presentation was also significantly associated with diagnostic delay (χ² = 53.320, p = 0.014).

Discussion

Majority of patients in our study who delayed presentation were between 30 – 50 years of age. Three studies done previously concluded that younger women were more likely to have delayed presentation than older women (Facione and Facione, 2006; Li et al., 2012; Friedman et al., 2006). However, we did not find any significant difference between any age group and diagnostic delay (χ² = 22.802, p = 0.198).

We witnessed a huge influx of patients from western Uttar Pradesh and Uttarakhand. A detailed analysis of the public health care infrastructure of the draining districts of Uttar Pradesh reveals adequate public health facilities in these regions (NRHM, 2013; Sarwal et al., 2021). Therefore, the delay in diagnosis of breast cancer in patients coming from these regions cannot be attributed to lack of healthcare access. Affordability is also not an issue as majority of the facilities are public funded with highly subsidized services (Banerjee, 2020). Patients from Uttarakhand on the other hand face unique challenges due to geographical barriers imposed by hilly terrain. However, these concerns were not matched by our analysis as geographical location was not found to influence delayed presentation of the patients (χ² = 8.625, p = 0.828) (Table 1).

We did not find formal education to influence diagnostic delay of symptomatic patients (χ² = 9.546, p = 0.388). In a study from the United States, authors found that at least 66% women with delay were college educated (Facione and Facione, 2006). In another study comparing cancer prevention knowledge and behavior among Blacks and Whites in the United States, authors found education to be a significant predictor of the outcome (Jepson et al., 1991). Similar findings were reported by others in studies done exclusively among the breast cancer patients (Coates et al., 1992; Hunter et al., 1993; Lam et al., 2009; Khakbazan et al, 2014). However, few other studies did not find any association between the education level and delayed diagnosis (Vernon et al., 1985; Mor et al., 1990).

In a study from North America, authors found that religious beliefs could delay the presentation of symptomatic breast cancer patients in women who believe that ‘cure is in the hand of God’ (Mitchell et al., 2002). A meta-analysis evaluating the effect of religion, spirituality and physical health in cancer patients found that religion and spirituality were associated with better patient reported physical health (Jim et al., 2015). In a study among African American women, authors found an association between delay in diagnosis of breast cancer and disclosing the breast symptoms to God (Gullatte et al., 2010). In another study from North America, authors found that religious beliefs positively influenced presentation time among whites but delayed presentation among African American women (Moorman et al., 2019). We did not find any association between the religion of patients and delay in diagnosis of breast cancer (χ² = 6.780, p = 0.351). However, we believe that general religious beliefs and spiritual inclinations more strongly influence patient presentation rather than religion itself.

We did not find any association between socioeconomic status of the patients and delayed diagnosis of breast cancer (χ² = 13.096, p = 0.362). In a study evaluating the effect of ethnicity, socioeconomic status and diagnostic delay on the survival of breast cancer patients, authors concluded that socioeconomic status affects the survival of patients when considered separately but not when all the variables were included (Vernon et al., 1985). However, some other studies found an association between lower family incomes with delayed help-seeking behavior (Jepson et al., 1991; Friedman et al., 2006; Li et al., 2012; Khakbazan et al., 2014).

The nature of breast complaint did not influence the delay in diagnosis of symptomatic patients (χ² = 18.070, p = 0.213). However, others reported lump-symptoms as more prompting to seek help rather than non-lump symptoms (O’Mahony et al., 2013; Khakbazan et al., 2014). Patients with painful breast lumps also seek help earlier, as painless symptoms are generally attributed to physiological processes such as pregnancy (Lam et al., 2009; Khakbazan et al., 2014).

Breast lump was the most common symptom in our patients followed by breast pain. Lack of awareness and knowledge about the breast symptoms, and interpretation of the symptoms as harmless and temporary and insufficient to seek care were primarily responsible for the delay in diagnosis among our patients. Our findings are consistent with studies conducted elsewhere (Nosarti et al., 2000; O’Mahony and Hegarty, 2009; Li et al., 2012). In many instances the breast symptoms were reassured by the family members who were first informed by the patient. This is vital concept that needs to be addressed in the awareness campaign and can cause significant improvement in early presentation of breast cancer patients.

Significant delay was caused among our patients due to preference of indigenous medicines for breast cancer treatment (p = 0.014). The sale and availability of any non-evidence-based treatment that claims to ‘cure’ malignancies should therefore be tightly regulated. The Covid-19 pandemic was also responsible for significant delay of symptomatic patients. Even though most centers continued to offer services to cancer patients, the imposition of curfews, lack of public transport possible could have led to the delay. Nearly 4% patients first visited unqualified practitioners (‘quacks’). Most of them revealed that easy access, low consulting costs were the among the main reasons for seeking their advice. Guaranteed treatment and cure offered by these practitioners is another factor that brings patients to them.

A significant proportion of patients were informed that their symptoms were benign and were reassured. Many of these patients were made to undergo a series of investigations before finally referred to a higher dedicated centre. This delay ranged from 2 days to 3 months and was particularly higher in high volume public institutions. This delay calls for better training of primary care physicians to better identify breast related signs and symptoms.

Nearly 40% of our patients had some awareness about breast cancer. The source of awareness, however did not influence the delay in diagnosis (χ² = 11.709, p = 0.230). In a study from Malaysia, authors found that public awareness program reduced the proportion of women presenting with stage III and stage IV disease from 77 to 37% (χ²= 17.0, p < 0.0001) between 1993 and 1998 respectively (Devi et al., 2007). A cross-sectional survey from Ghana evaluated the impact of breast cancer awareness program and concluded that those who attended the program were more likely to perform self-breast examination and have higher knowledge scores. Whether, it improves early breast cancer detection however is not clear (Mena et al., 2014). A randomized control trial involving an educational intervention to improve attitudes, behavioural intentions and early-stage diagnosis concluded that ‘fear appeals’ led to improvements in attitude (p = 0.01) and intentions (p = 0.001) but no effect was seen in early-stage diagnosis (p = 0.78) (Zonouzy et al., 2018).

Clinical and self-breast examination are well known to facilitate early diagnosis of breast cancer especially when mammographic screening is not available or feasible (Holleb, 1981; Wu and Lee, 2018). Based on these findings, breast self-examination was included in international cancer control programs particularly in the low and middle-income countries (Sullivan et al., 2015). However, self-breast examination was not found to influence the delay in diagnosis of breast cancer among our patients (χ² = 4.716, p = 0.229). In a study from Kashmir, India, authors found breast cancer awareness and breast self-examination in 7.35% women. They also concluded that breast cancer awareness was linked to the general education of the women (Malik et al., 2020). As majority of our patients did not have any formal education/training or exposure about how to practice breast self-examination, the practice itself might have been rendered ineffective. More than 80% of our patients were not aware of the cause of breast cancer. Others attributed it to lack of breastfeeding, breast trauma, hormonal factors and genetic factors. However, patients’ perception was not found to influence the delay in presentation breast cancer (χ² = 52.206, p = 0.544).

There was no association between the person breast symptoms were reported to by the patient and diagnostic delay χ² = 37.214, p = 0.171). Nearly 70% of our patients reported their breast complaints to male members their family (husband/son). This allows us to effectively engage adult males in female breast cancer awareness programs. Easy access and outreach to men in the society can prove especially effective in early presentation of breast cancer patients.

In conclusion, breast cancer is a fast-increasing public health problem with significant burden on health care expenditure. We did a mixed methods study, with a large sample size available for qualitative analysis. We assessed a number of factors ranging from sociodemographic to clinical variables, tumour characteristics and knowledge and practices pertaining to breast cancer. Breast cancer awareness was found to be the most effective method of ensuring early-stage presentation of symptomatic breast cancer patients. We conclude that breast awareness is not synonymous with educational status. More effort must be made to improve awareness of women towards breast symptoms and early reporting. As more than 70% patients report their complaints to male members of their families, we suggest a gender-neutral approach to breast cancer awareness programs.

Author Contribution Statement

Dr. Ankit Rai and Dr. Bina Ravi conceptualized the study; Dr. Ankit Rai, Dr. Bina Ravi, Dr. Pradeep Aggarwal prepared the study protocol; Dr. Ankit Rai, Dr. Bina Ravi, Dr. Prateek Sharda did the data collection; Dr. Ankit Rai, Dr. Pradeep Aggarwal did the data interpretation and analysis; Dr. Ankit Rai, Dr. Bina Ravi, Dr. Pradeep Aggarwal, Dr. Prateek Sharda prepared the final manuscript.