Duration of intervals in the care seeking pathway for lung cancer in Bangladesh: A journey from symptoms triggering consultation to receipt of treatment.

Timeliness in seeking care is critical for lung cancer patients' survival and better prognosis. The care seeking trajectory of patients with lung cancer in Bangladesh has not been explored, despite the differences in health systems and structures compared to high income countries. This study investigated the symptoms triggering healthcare seeking, preferred healthcare providers (including informal healthcare providers such as pharmacy retailers, village doctors, and "traditional healers"), and the duration of intervals in the lung cancer care pathway of patients in Bangladesh. A cross-sectional study was conducted in three tertiary care hospitals in Bangladesh among diagnosed lung cancer patients through face-to-face interview and medical record review. Time intervals from onset of symptom and care seeking events were calculated and compared between those who sought initial care from different providers using Wilcoxon rank sum tests. Among 418 study participants, the majority (90%) of whom were males, with a mean age of 57 ±9.86 years, cough and chest pain were the most common (23%) combination of symptoms triggering healthcare seeking. About two-thirds of the total respondents (60%) went to informal healthcare providers as their first point of contact. Living in rural areas, lower levels of education and lower income were associated with seeking care from such providers. The median duration between onset of symptom to confirmation of diagnosis was 121 days, between confirmation of diagnosis and initiation of treatment was 22 days, and between onset of symptom and initiation of treatment was 151 days. Pre-diagnosis durations were longer for those who had sought initial care from an informal provider (p<0.05). Time to first contact with a health provider was shorter in this study compared to other developed and developing countries but utilizing informal healthcare providers caused delays in diagnosis and initiation of treatment. Encouraging people to seek care from a formal healthcare provider may reduce the overall duration of the care seeking pathway.

Introduction

Lung cancer is an aggressive disease with poor survival. As the leading cause of cancer deaths, it has a higher incidence and mortality rate than any other cancers globally [1]. According to the World Cancer Report 2020, there were 2.1 million new cases and 1.8 million deaths in 2018 due to lung cancer. Early stage diagnosis of lung cancer may allow surgical resection, the most effective treatment, with the five-year survival rate being as high as 50–70% [2-4] if diagnosed at Stage I. However, around 75% of people present at an advanced stage of their disease [5], where the overall survival is only 10–20% [1] and 1–5% if diagnosed at Stage IV [2-4].

The importance of timely diagnosis and treatment of lung cancer, and the need for indicators to measure timeliness of care are increasingly discussed in the literature [6, 7]. Guidelines for lung cancer management include recommended timeframes for the intervals from one timepoint to another in the disease care pathway [8]. Clinical guidelines in some western countries recommend start of treatment (surgery, chemotherapy or radiotherapy) within two to eight weeks of diagnosis [9-12]. However, these guidelines are developed based on the strength of resourceful health systems of High Income Countries (HICs) [8] and are hard, if not impossible to implement effectively in Low and Middle Income Countries (LMICs) [13]. There is no existing guideline for lung cancer management in most LMICs, including South Asian countries [14].

Despite consensus around best practice for diagnosis and treatment expressed in guidelines, time from the onset of symptoms to diagnosis, on which prognosis is highly dependent, appears to vary widely in practice, as does time to start treatment. Most studies have been conducted in western countries and a major focus of existing studies is about the waiting time between diagnosis and treatment in line with the clinical guidelines. There is a high degree of variability in the patient journey prior to diagnosis according to the existing literature; however, little is known about LMICs. A study in India [15] found similar results from onset of symptom to initiation of treatment in comparison with those conducted in the western countries [16], but there have been no studies conducted in Bangladesh. Bangladesh is one of the most densely populated LMICs with a population of 163 million people. There were over 250,000 premature deaths from noncommunicable diseases in 2018, 26% of which were due to cancer (108,000) [17]. Lung cancer is the most common cancer diagnosis in males in Bangladesh and is the commonest type of cancer-related death [18]. Most cases of lung cancer are diagnosed at an advanced stage when the options for treatment are limited [19].

The health system of Bangladesh is structured quite differently from those in western countries which have informed current understanding of the lung cancer patient journey. There is no structured referral system from primary care to specialist care and there are no restrictions on patients seeking care from specialists without having an appropriate referral from primary care providers. Referral pathways are fragmented and there is no mechanism to know whether consumers have taken up referrals [20]. Doctors with Bachelor of Medicine and Surgery (MBBS) degrees are considered to be general practitioners in Bangladesh. Secondary and tertiary healthcare service is provided in hospitals, which can be public or private services run by non-government organisations and not-for-profit organizations. General physicians and specialist doctors can be accessed publicly or privately in those settings. There are 37 specialized hospitals (15 public and 22 private) providing regular cancer treatment in Bangladesh, and an estimated 1.7 radiation oncologists per 10,000 cancer patients [17].

In LMICs, informal healthcare providers who lack recognised training comprise a significant portion of the health system, operating generally as individuals rather than in institutions at a primary care level [21]. Pharmacies, “traditional healers” (ayurvedic, homeopathic, unanie/kabiraji—herbal medicine providers) and village doctors are the most common informal healthcare providers in Bangladesh [22]. Pharmacies, which might be more accurately described as retail medicine shops or drug sellers, are often the first point of contact for people in Bangladesh, particularly those from lower socioeconomic status [23]. These pharmacies provide symptomatic treatment and advice on diseases and health conditions, and dispense medication without a prescription, despite this being prohibited under the drug licence law [24].

There is an overall lack of evidence in Bangladesh and across the subcontinent regarding timeliness of seeking care of lung cancer patients. The aim of this study was to measure the intervals between timepoints in the care-seeking pathway of patients with lung cancer in Bangladesh. Sub-questions were (1) do durations differ according to the symptom(s) that motivated patients to seek healthcare [triggering symptom(s)], (2) do durations differ according to which healthcare providers the patients chose to seek initial healthcare from.

Materials and methods

Study design and population

A cross-sectional study was conducted in three study sites in Dhaka, Bangladesh. The sites were purposely selected considering variations in patient profiles, and included one public hospital, one university teaching hospital and one low cost philanthropic cancer hospital. The National Institute of Cancer Research and Hospital (NICRH) is a 300-bed tertiary level public hospital dedicated only for cancer care and provides the maximum load of cancer management in the country. Bangabandhu Sheikh Mujib Medical University (BSMMU) is a postgraduate medical institute with a limited 36-bed inpatient capacity oncology department which accepts both public and private referrals. Ahsania Mission Cancer & General Hospital (AMCGH) is the largest non-profit hospital in the country, currently having a 250-bed inpatient capacity dedicated for cancer care. Patients from across the country receive care from these three hospitals.

Inpatients aged ≥18 years with a known diagnosis of lung cancer, who were able to understand the Participant Information and Consent Form and could nominate a family member or carer, if needed, to provide information on their behalf were eligible to participate in the study. Potential participants who were inpatients on the data collection date were identified by doctors or nurses at the study sites from the inpatient files as per the inclusion criteria, after which the investigators approached admitted consecutively eligible participants for written informed consent. Eligible participants were given information about the voluntary nature of their involvement in the study, details of the study and contact details of the investigators. The participant information was read to those participants who were illiterate. If the patient was feeling unwell and did not wish to provide answers themselves, patients were asked if they wanted to nominate caregivers to provide data on their behalf, and with consent from patients, caregivers were interviewed separately. Participants provided a signature as a sign of their consent or thumb impression for those who were illiterate. There was no compensation provided for participating in the study.

Questionnaire and data collection procedure

A questionnaire was developed, informed by relevant literature about delays in cancer care [25, 26], national surveys and knowledge of the local health system. Questions and response options were designed to be appropriate to the country context and study objectives. The questionnaire was developed in English and then translated into Bengali with the whole process going through ‘translation and back translation’ involving two of the investigators and two public health researchers, all of whom were bilingual. The Bengali version of the questionnaire was pilot-tested to ascertain consistency, appropriateness of language, and the best sequence of questions. The questionnaire was pilot-tested with eight lung cancer patients to ensure the questions were understandable and interpreted in a similar way by the participants. Based on the pilot test, several questions were rephrased for better understanding. Participants in the pilot testing were excluded from the final sample for data analysis. The questionnaire collected data under four sections 1) Socio-demographics such as age, sex, area of residence, education, marital status, type of family, and monthly household income, 2) History of illness (comorbidities) 3) Symptoms information including date of onset of symptoms, first symptoms, symptoms that triggered healthcare seeking, 4) Help-seeking practice including choice of healthcare provider for first contact, time to travel to healthcare provider, whether additional healthcare providers were consulted before diagnosis, date of diagnosis, time to travel to diagnostic facility, date of referral for treatment, additional healthcare provider consulted after diagnosis before starting treatment, and date of initiation of treatment. Participants’ area of residence was categorised as urban or rural based on the presence of municipal services in the public administrative area.

A total of 418 questionnaires were completed between 10 October 2019 and 13 February 2020, of which 318 were with the patients and 100 were with the patients’ nominated caregivers. Of 424 eligible patients, only six refused to participate, with 285 participants from NICRH, 27 from BSMMU and 106 from AMCGH were included in the study. Our intention was to calculate the mean and median for the total lung cancer population admitted to the three hospitals. A sample size of 418 patients enabled a margin of error of less than 1 day in the calculation of the shortest duration indicator (an average of 4 days from diagnosis to referral for treatment) and less than 14 days in calculating the longest duration indicator (an average of 186 days for the entire journey) with a 95% confidence level [27]. The structured questionnaire was administered face-to-face by three data collectors, who were medical doctors, and none was involved in providing treatment to the participants. Two were trained to conduct the survey by the first author who also participated in data collection. Information on different timepoints in the healthcare seeking pathway was collected from hospital clinical files, and from the patient-held clinical files to minimise recall bias. Permission was obtained from the hospital authority and informed consent was sought from the patients to examine both types of clinical files. After obtaining informed consent, a participant’s interview was done at the bedside if privacy could be maintained and, in the case of the caregiver, the interview was conducted in the duty doctor’s room in private. In addition to asking about different timepoints of care seeking in the disease cycle as part of the questionnaire, data were extracted from patient-held clinical files available at the time of interview. In Bangladesh, patients keep a hard-copy record of their previous treatment documents which includes previous prescriptions, investigation reports and discharge summary of previous hospital admission. They generally have this with them when receiving care in a hospital. This process served as cross-checking to minimise recall bias. Where there was a mismatch of information provided by the participant and that extracted from hospital documents, the dates derived from the hospital were accepted. To ensure data quality, the first author monitored administration of the questionnaire throughout the data collection period. Five percent of the collected data were reviewed to check the accuracy and completeness of the data. No discrepancies were found in the data.

Measurement of healthcare seeking intervals

Date variables

(i) date of first symptom, (ii) date of first contact with any healthcare provider, (iii) date of diagnosis, (iv) date of referral for treatment and (v) date of treatment initiation.

Date of first symptoms was reported by the respondents and clinical files were also checked to confirm where possible. Date of first contact with a healthcare provider included the date to visit any provider. The date of diagnosis is the date of histopathological or radiological confirmation of lung cancer. Date of referral was the date when the patient was referred by a formal healthcare provider to a tertiary healthcare facility to initiate treatment e.g., chemotherapy, radiotherapy or surgery. There was no referral from informal health care providers to tertiary healthcare facilities in our study.

Time intervals

The duration of time intervals was calculated in days from one time point to another. The time interval was recorded as ‘0’ days if the starting and end events occurred on the same day. Seven time intervals were calculated which were as follows:

Date of onset of symptom(s) to the date of the first contact with any healthcare provider

Date of the first contact with any healthcare provider to the date of the confirmed diagnosis

Date of the confirmed diagnosis to the date of the referral for treatment

Date of the referral for treatment to the date of the initiation of treatment

Date of the onset of symptom(s) to the date of the confirmation of diagnosis

Date of the confirmation of diagnosis to the date of the initiation of treatment

Date of onset of symptom(s) to the date of the initiation of treatment

Data analysis

Descriptive analyses were performed using frequency, mean and standard deviation (SD), median and interquartile range (IQR) depending on the distribution of the data. Comparison of sociodemographic characteristics and type of healthcare providers for first contact (formal and informal healthcare provider) were performed using chi-square analysis; post-hoc comparisons were adjusted using the Bonferroni approach [28]; Wilcoxon Rank Sum tests were done to compare intervals according to the type of healthcare providers first contacted. The data were analysed using SPSS (version 25) and significance was set at p-value <0.05.

Ethics

This study obtained ethical approval from the University Human Ethics Committee of La Trobe University, Melbourne, Australia (HEC19154); National Institute of Cancer Research & Hospital, Dhaka, Bangladesh (NICRH/Ethics/2019/504); and Ahsania Mission Cancer and General Hospital, Dhaka Bangladesh (DAM/AMCGH/1900-2019-2063).

Results

The demographic characteristics of the respondents from each site are described in Table 1. The mean age of the sample was 57 years (SD ±9.86 with a range of 25–82 years). The majority of patients were over 55 years old (68%), with less than 10% aged < 45 years. Participants were predominantly from rural areas (88%) and the majority of participants were illiterate (64%). The mean number of years of education was 4 years (SD ±5 and range 0–17 years). The median monthly household income was BDT 15500 (USD 182). The lowest quartile had an income of BDT 10000 (US$ 118) and below, whereas the upper quartile had an income of BDT 100000 (US$ 1176) and above. Significant income differences were found across the three study sites (p<0.001). We did not explore stage of lung cancer at the time of diagnosis because the information is generally not available in patient files in Bangladesh; similarly, the data on comorbidities were not sufficient to analyse.

Table 1

Characteristics of the participants (n = 418).

	NICRH (n, %)	BSMMU (n, %)	AMCGH (n, %)	Total (n, %)
Age Group (in years)
<35	7 (2.5)	0	1 (0.9)	8 (1.9)
35–44	20 (7)	4 (14.8)	5 (4.7)	29 (6.9)
45–54	65 (22.8)	2 (7.4)	29 (27.4)	96 (23)
55–64	120 (42.1)	12 (44.4)	45 (42.5)	177 (42.3)
≥65	73 (25.6)	9 (33.3)	26 (24.5)	108 (25.8)
Sex
Male	260 (91.2)	23 (85.2)	92 (86.8)	375 (89.7)
Female	25 (8.8)	4 (14.8)	14 (13.2)	43 (10.3)
Marital Status
Married	266 (93.3)	25 (92.6)	97 (91.5)	388 (92.8)
Not Married	19 (6.7)	2 (7.4)	9 (8.5)	30 (7.2)
Area of Residence
Urban	34 (11.9)	2 (7.4)	14 (13.2)	50 (12)
Rural	251 (88.1)	25 (92.6)	92 (86.8)	368 (88)
Family Structure
Nuclear family	120 (42.1)	10 (37.0)	34 (32.1)	164 (39.2)
Joint/extended family	165 (57.9)	17 (63.0)	72 (67.9)	254 (60.8)
Highest Education Level
Illiterate & below primary education	196 (68.8)	21 (77.8)	52 (51.1)	269 (64.4)
Primary	48 (16.8)	3 (11.1)	13 (12.3)	64 (15.3)
Secondary	24 (8.4)	1 (3.7)	14 (13.2)	39 (9.3)
Higher Secondary	7 (2.5)	2 (7.4)	9 (8.5)	18 (4.3)
Bachelor’s degree & above qualification	10 (3.5)	0	18 (17)	28 (6.7)
Monthly Household Income
BDT ≤15000 (≤US$176)	190 (66.7)	13 (48.1)	6 (5.7)	209 (50)
BDT 15001–50000 (US$176–588)	80 (28.1)	13 (48.1)	33 (31.1)	126 (30.1)
BDT 50001–100000 (US$588–1176)	13 (4.6)	0	38 (35.8)	51 (12.2)
BDT ≥100001 (≥US$1176)	2 (0.7)	1 (3.7)	29 (27.4)	32 (7.7)
Travel time to first healthcare provider
<10 min	91 (31.9)	8 (29.6)	43 (40.6)	142 (34)
10 to 30 min	125 (43.9)	11 (40.7)	24 (22.6)	160 (38.3)
31 to 60 min	43 (15.1)	4 (14.8)	25 (23.6)	72 (17.2)
61 min or more	26 (9.1)	4 (14.8)	14 (13.2)	44 (10.5)

*1 US$ = 85 BDT (Bangladeshi Taka), NICRH—National Institute of Cancer Research and Hospital, BSMMU—Bangabandhu Sheikh Mujib Medical University, AMCGH—Ahsania Mission Cancer and General Hospital

Symptoms triggering healthcare seeking

Table 2 shows the first symptom(s) experienced and the symptoms that led to respondents seeking healthcare. Half of the participants (226, 54%) mentioned just one triggering symptom. However, 46% of the participants reported a combination of symptoms triggering healthcare seeking. At least two symptoms triggered healthcare-seeking in 34% of participants, while 11% of participants mentioned three or four symptoms as their triggering factor.

Table 2

First symptoms and triggering symptoms to seek healthcare provider (n = 418).

	Symptoms first noticed		Triggering symptoms
	Frequency*	%	Frequency*	%
Cough	369	88.28	275	65.79
Chest pain	236	56.5	162	38.76
Shortness of breath	108	25.84	68	16.27
Haemoptysis	103	24.64	62	14.83
Lack of appetite	43	10.29	19	4.55
Hoarseness of voice	40	9.57	19	4.55
Weight loss	10	2.39	2	0.48
Fatigue	6	1.44	5	1.20
Persistent or recurrent infections	2	0.48	0	0
Other uncommon symptoms	65	15.6	51	12.20

*as multiple triggering symptoms were reported by a single respondent, the total number is more than 418. % are calculated as a proportion of the total sample.

Table 3 shows the differences in timeframe in relation to the combination of triggering symptoms at diagnosis. Participants who reported cough and chest pain experienced longer duration to get diagnosed compared with others (p<0.002). However, some of those patients had other symptoms in addition to cough and chest pain.

Table 3

Association with triggering symptoms and duration of intervals (N = 418).

	Frequency (%)*	Onset of symptoms to diagnosis (Median, Range)	p-value
Only Cough	95 (22.72)	123 (1069)	0.31
Cough and chest pain	91 (21.77)	157 (1426)	0.002
Cough and shortness of breath	54 (12.92)	139 (496)	0.08
Cough and haemoptysis	37 (8.85)	131 (671)	0.31
Chest pain and shortness of breath	22 (5.26)	146 (467)	0.18

*Multiple triggering symptom reported.

Healthcare providers

The majority of participants (60%) had contact with an informal healthcare provider (pharmacy 43%, village doctor and traditional healer 17%) as their first point of contact, whilst 22% went to a General Practitioner (GP) and the remainder to other formal healthcare providers (Table 4). Only 8% of the participants sought care from a single provider before undergoing confirmation of a lung cancer diagnosis. The majority (92%) went to multiple healthcare providers before getting confirmation of their lung cancer diagnosis. Two-fifths consulted two additional healthcare providers (42%), one quarter consulted one additional healthcare provider and 81 (19%) respondents consulted three additional healthcare providers pre-diagnosis. After first contact, most of the additional pre-diagnosis care seeking was with public healthcare providers (76%) and specialists (48%) (Table 4).

Table 4

Contact with healthcare providers at different timepoints.

Healthcare providers	First point of contact	Additional	Additional
		pre-diagnosis	pre-treatment
	Frequency	Frequency	Frequency
	N = 418 (%)	N = 383* (%)	N = 96* (%)
Public Healthcare	28 (6.7)	319 (76.3)	34 (8.1)
Private and NGO Healthcare	27 (6.5)	119 (28.5)	47 (11.2)
GPs (MBBS)	91 (21.8)	123 (29.4)	15 (3.6)
Specialist Doctors	22 (5.3)	200 (47.9)	35 (8.4)
Pharmacy (over the counter)	179 (42.8)	7 (1.7)	0
Village Doctor & Traditional healer	71 (17)	17 (4.1)	5 (1.2)

* Multiple responses were allowed.

Once a diagnosis of lung cancer was confirmed by a formal healthcare provider, almost one-quarter of participants consulted at least one additional healthcare provider before starting treatment. In our study, all of the participants were referred by their GPs to the tertiary hospitals (study sites) to initiate treatment. Private and NGO healthcare providers were consulted by 47 (11% of 418) participants, 35 (8% of 418) consulted additional specialist doctors and 34 (8% of 418) consulted in public healthcare facilities. Notably, consulting with traditional healers and village doctors was less frequent compared to first point of contact (17%), at additional pre-diagnosis consultation (4%) and additional pre-treatment consultation (1%) timepoints. There is also a large drop in seeking additional care from pharmacy at pre-diagnosis and pre-treatment stage (Table 4).

The type of health care provider who was the first point of contact did not differ by age, sex, marital status or family structure of the participants. However, patients from rural areas (61%), illiterate patients (67%) and patients from the lowest income groups (69%) were more likely to choose an informal healthcare provider as the first point of contact (p<0.05). Education level was associated significantly with type of healthcare provider: illiterate patients were more likely to seek informal care compared with those with bachelor and above education (p<0.001, statistically significant after Bonferroni correction). Similarly household income was associated significantly with type of healthcare provider; participants with household income of BDT 100,000 (p<0.002, significant after Bonferroni correction) (Table 5).

Table 5

First healthcare contact according to sociodemographic characteristics (n = 418).

Patient characteristics	Formal healthcare provider	Informal healthcare provider	p-value
	(n, %)
		(n, %)
Age Group (in years)
<35	3 (37.5)	5 (62.5)	0.40
35–44	11 (37.9)	18 (62.1)
45–54	31 (32.3)	65 (67.7)
55–64	79 (44.6)	98 (55.4)
≥65	44 (40.7)	64 (59.3)
Sex
Male	152 (40.5)	223 (59.5)	0.40
Female	16 (37.2)	27 (62.8)
Marital Status
Married	156 (40.2)	232 (59.8)	0.57
Not Married	12 (40.0)	18 (60.0)
Family Structure
Nuclear family	70 (42.7)	94 (57.3)	0.23
Joint/extended family	98 (38.6)	156 (61.4)
Area of Residence
Urban	27 (54.0)	23 (46.0)	0.03
Rural	141 (38.3)	227 (61.7)
Highest Education Level
Illiterate & below primary education	89 (33.1)	180 (66.9)	<0.001
Primary education	28 (43.8)	36 (56.3)
Secondary & higher secondary education	31 (54.4)	26 (45.6)
Bachelor’s & above qualification	20 (71.4)	8 (28.6)
Monthly Household Income
BDT ≤15000 (≤US$176)	65 (31.1)	144 (68.9)	<0.001
BDT 15001–50000 (US$176–588)	54 (42.9)	72 (57.1)
BDT 50001–100000 (US$588–1176)	28 (54.9)	23 (45.1)
BDT ≥100001 (≥US$1176)	21 (65.6)	11 (34.4)

* p-value distribution for Chi-square tests of distribution of participants between seeking care from formal and informal healthcare by participant characteristics.

Duration of intervals in the care seeking pathway

First contact with informal care (in comparison with formal care) was associated with a shorter duration between onset of symptoms and first contact (p<0.05), but longer durations from first contact to diagnosis, from onset of symptoms to diagnosis, and from onset of symptoms to initiation of treatment (p<0.05) (Table 6).

Table 6

Duration of intervals (days) in care seeking with first point of contact*.

	Formal healthcare provider	Informal healthcare provider	p-value
	(N = 168, 40%)	(N = 250, 60%)
	Median (Range)	Median (Range)
Duration between onset of symptoms and first contact with provider	13 (68)	10 (358)	0.03
Duration between first contact with provider and diagnosis	89 (723)	121 (1434)	<0.001
Duration between onset of symptoms and diagnosis	104 (1434)	136 (1081)	<0.001
Total duration from onset of symptoms to initiation of treatment	131 (1460)	171 (1095)	<0.001

*p-value for Wilcoxon Rank Sum test. Not all the intervals are related with the first contact; the most relevant intervals were examined.

The entire care pathway was segmented in seven intervals. The median time between onset of symptom and initiation of treatment was 151 days. As a segment, the longest duration in the care seeking pathway was the duration from symptom to diagnosis (median 121 days) and the second longest duration was from first contact with any healthcare provider to diagnosis (median 107 days). The shortest duration was observed between diagnosis and referral for treatment (median 3 days) and onset of symptom to first contact with any healthcare provider (median 10 days) (Fig 1).

Fig 1

Timepoints and intervals in the care seeking pathway for lung cancer.

Discussion

This study is the first to investigate the care trajectory and duration of intervals in the care seeking pathway for lung cancer in Bangladesh. The focus of this study was on healthcare seeking and symptoms triggered healthcare seeking. In our study, cough was the most frequent first symptom reported by respondents (88%) (Table 2); however, only 23% of the respondents mentioned cough as their only symptom (Table 3) triggering them to seek healthcare or take remedial action. These findings may indicate that not everyone considered a cough as an alarming symptom and potential sign of lung cancer. Apart from the obvious symptom presentation, lung cancer symptoms can be complex in patients with comorbidities [29] and can also be absent from any clear symptom presentation [30]. A history of chronic lung comorbidity and smoking is associated with delayed presentation to a healthcare provider [31]. Patients with lung cancer may also present with a variety of symptoms which are not unique to that condition [32]. The presence of comorbidities with similar symptoms can make the underlying cancer even more difficult to detect [29]. Sometimes, symptoms are misattributed to smoking or to comorbidities such as chronic obstructive pulmonary disease (COPD) or cardiac diseases, resulting in delayed care-seeking [33]. Moreover, stigma [34-36] and fear [34, 37] are associated with delayed care seeking behaviour for lung cancer among smokers which can lead to advanced stage presentation at diagnosis [38].

Historically, smoking rates [39] and lung cancer prevalence [40] are low among women in Bangladesh, which may explain the predominance of males in this study. The mean total duration from onset of symptoms to initiation of treatment for our study was 186.4 ± 140.4 days with a median of 151 days (13–1460 days). This is longer than the duration reported in a study conducted in India, a neighbouring country, where the median duration was 126 days [41]. The total duration in our study, from onset of symptom(s) to initiation of treatment is also longer in comparison to similar studies conducted in western countries such as USA (median of 52 days) [42], Greece (median of 81 days) [43], and Turkey (median 57 days) [16].

Some of the intervals appeared to be shorter than those observed in studies from other countries. The mean duration between onset of symptoms and first care seeking was 15.2 ± 24.4 (median 10) days in this study. This is shorter than time periods reported from symptom onset to contact with a medical doctor in other countries, including 71 days in Nepal [44], 94 days in India [45], and 35 days or 49.9 ± 96.9 days in Turkey [16, 46], however, these studies did not include informal providers as first contact. In this study, informal healthcare providers were the first choice for seeking healthcare reported by most respondents (60%), hence this duration is shorter than other studies mentioned above. Studies conducted in LMICs have reported that, closer vicinity, easier access, and affordability are the main reasons that patients seek care from informal healthcare providers compared to formal healthcare providers [21].

The period from onset of symptoms to confirmed diagnosis was longer than in other countries where similar studies were conducted. In our study the mean duration between onset of symptoms and diagnosis was 151 ± 132 days with a median of 121 days. This is longer compared to studies conducted in India (median of 107 days) [41], Greece (median of 52 days) [43], Turkey (median of 49 days) [16], and Canada (median of 90 days) [47]. As noted above, this difference may reflect the inclusion of informal care providers as the first healthcare provider. This is supported by the finding that the mean duration in our study from first contact with any healthcare provider to diagnosis was longer than other studies (135.8 ± 126.7 days). In Nepal, this step took 50 days on average [44] where the first contact was a medically trained provider. This interval involves the patient reaching qualified doctors and the hospital system, and suggests that the sooner the patient gets diagnosed, the sooner the treatment initiates. In order to ensure early diagnosis, early engagement with a medically trained rather than informal healthcare provider is paramount.

It is noteworthy to mention that the delay occurred in the early phase of the disease care pathway, mostly in the pre-diagnosis phase, and after diagnosis the intervals became shorter compared to pre-diagnosis intervals. The mean interval from diagnosis to initiation of treatment was 35.3 ± 42.9 days with a median of 22 days. Although it was not within the acceptable maximum timeframe for diagnosis to initiation of treatment described in guidelines in Australia, Denmark and Sweden (14 days), it was within an acceptable range for guidelines in the USA (42 days) and the UK (48 days). Further, the duration (diagnosis to initiation of treatment) in this study was shorter than the same interval reported in the USA (median of 27 days [48] and mean of 24.4 ± 54.9 days) and was within the range for medians of 6–45 days reported by Jacobsen et al [49].

The pattern in the duration of intervals between key timepoints in the care seeking pathway for lung cancer in this study suggests that the reason for the duration from onset of symptoms to initiation of treatment was long as care seeking mostly started from informal healthcare in Bangladesh. There were significant differences in timepoints according to whether first contact was with informal healthcare providers compared with formal. While the median duration from onset of symptoms to first contact was significantly shorter for those who used informal providers, intervals between onset of symptoms and diagnosis and treatment were significantly longer for people who sought help from informal providers. Moreover, those who sought care from informal providers were more likely to come from rural areas, had fewer years of education and a lower income. Similar findings were reported in other studies conducted in Bangladesh [50, 51] and India [52]. In addition, traditional healing practice is very widely used in rural Bangladesh, with 75–80% of the rural population dependent on it in the absence of access to formal healthcare [51]. However, only 17% of study participants in this study reported consulting village doctors and traditional healers for management of lung cancer (Table 4). Reliance on traditional healers poses a threat for timely initiation of proper medical care for patients with cancer which is reported in studies conducted in Bangladesh [53], India [54] and Malaysia [55]. So, while time to seek and receive this care is short, it is unlikely to be effective for someone with lung cancer.

In LMICs like Bangladesh, resource shortage and access barriers are likely to prevent cancer patients from getting timely diagnosis although those who are diagnosed may be treated within the timeline recommended by the clinical guidelines developed in the HICs. This highlights the importance of addressing challenges in the care pathway prior to diagnosis in the LMICs. Unfortunately, there is a lack of contextualised clinical guidelines tailored to the needs of the LMICs. We suspect that there is a possibility that a significant number of people with respiratory diseases, particularly lung cancer, are undiagnosed in Bangladesh’s largely rural and disadvantaged population. Exploration of issues of unidentified and unmet need is beyond the scope of this paper. Investment in basic primary health care and use of community health workers could help uncover the huge burden of chronic respiratory disease which is often undiagnosed and unmanaged [56]. While setting up facilities for radiotherapy is a costly option at sub-district levels, investment can be considered for providing incentives to the oncologists who would be interested to work in those settings. Efforts should be given to ‘fast-tracking’ access to formal healthcare providers for any suspected case of lung cancer for early diagnosis and management. Informal healthcare providers should be trained to recognise lung cancer symptoms and to refer suspected patients early to formal healthcare providers; introducing a reward system could encourage adoption of such practice. Further research will be needed to explore factors associated with longer and shorter durations in the health seeking pathway and their effects on lung cancer survival in Bangladesh.

This study is not devoid of limitations. The date of onset of symptom(s) was collected from patients’ or care givers’ recollection and could have recall bias, and date of first presentation to a provider is often difficult to ascertain from records in the presence of longstanding comorbidities. This study used data from patients in three urban hospitals in Bangladesh, therefore, it is a limited sample study, so findings cannot represent the experiences of people receiving a diagnosis and treatment for lung cancer in the entire country. However, the included hospitals were the three key referral hospitals for cancer management in Bangladesh and the sample included patients from all over the country and from all levels of socioeconomic status. As this study was conducted in hospital settings there is a possibility of sampling bias; it did not capture the experience of anyone who did not receive treatment from a hospital. A robust study using data from other cancer hospitals in different cities and a country-specific validated tool could increase and ensure generalisability.

Conclusion

This paper is the first to describe the lung cancer patient journey in Bangladesh. The study suggests that the patterns in the average duration between timepoints in healthcare seeking from onset of symptoms to initiation of treatment in lung cancer care in Bangladesh are different from western and other developing countries, being shorter at some points and longer at others. The findings suggest that the first point of care for most patients was through informal care providers, possibly due to their convenience and ease of access. However, quicker access to informal care may not necessarily translate into quicker provision of lung cancer treatment. In this study, we found that it can lead to significant delays in diagnosis and initiation of treatment. To ensure timely lung cancer diagnosis and care across the whole patient journey, there is a need to better understand the reasons for the intervals between care points observed in this study. This may help to develop health system specific and clinically evidence-based benchmarks for Bangladesh.

20 Apr 2021

PONE-D-20-39263

Duration of intervals in the care seeking pathway for lung cancer in Bangladesh: a journey from symptoms triggering consultation to receipt of treatment

PLOS ONE

Dear Dr. Ansar,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jun 04 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Muhammed Elhadi, MBBCh

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide further details on sample size and power calculations.

3. In statistical methods, please refer to any post-hoc corrections to correct for multiple comparisons during your statistical analyses. If these were not performed please justify the reasons. Please refer to our statistical reporting guidelines for assistance (https://journals.plos.org/plosone/s/submission-guidelines.#loc-statistical-reporting).

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Partly

Reviewer #6: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

Reviewer #6: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: No

Reviewer #5: Yes

Reviewer #6: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #4: Yes

Reviewer #5: Yes

Reviewer #6: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this study, Ansar and colleagues evaluate the impact of resource and referral utilization in Bangladesh in duration of wait times between symptom development and treatment of lung cancer. Patients were recruited from three tertiary care hospitals in Dhaka, Bangladesh. They find that rural location, low income, lower educational level and primary self-referral to non-medical providers (such as pharmacists and traditional healers) at the time of symptom onset were all associated with delays in time from symptom onset to diagnosis. Overall, the manuscript is well-written with minor spelling errors, although the introduction is lengthy. Limitations including being descriptive nature, the smaller patient size and limited applicability to analysis (given that it only studies this in Bangladesh, a single low-to-middle income country). These limitations are offset by the specificity of information regarding area of residence, education, and education. Additionally, there is a need for more clear information regarding access to appropriate care in middle to low income countries. However, concerns remain regarding stage at diagnosis, how the time from symptoms to first provider are acquired and concerns about possible bias in attribution of patient behaviors which would need to be addressed.

Critiques:

The introduction is too long, reads like a review article and includes information that could be summarized or is already provided in the discussion. Alternatively, some of this could be briefly summarized in a table under methods.

In p. 6, line 119, the authors refer to “traditional healthcare providers?” I believe the author later refers to these as people who practice holistic medicine (referred to in some places as “traditional healer” or “village doctor”)? If so, the authors should use a consistent label for this group, and possibly place this term in quotes or clearly describe these providers.

p. 10, Time Intervals methods: It is unclear how “estimations” were performed or how much estimation was required (rather than a clear calculation of duration time intervals). Although this may be needed for duration from symptoms to seeking care, this is concerning for recall bias (particularly in other intervals) if based only on patient or family member recollections.

Methods: Definitions of how urban and rural are defined are not clearly stated.

Table 5, p-value for marital status should be 0.98 (decimal omitted)

Table 6: ranges in duration are typically provided in addition to medians. This is addressed for some of these in the discussion, but should be included in the results section rather than the discussion.

The authors do not report the stage of lung cancer at the time of diagnosis. Most lung cancers diagnosed in symptomatic patients are at an advanced stage, but it would be of important to know at what stage(s) these patients were diagnosed in order to compare that to other studies, which may vary in stage at diagnosis (which has been associated with variances in time from diagnosis to treatment, for instance, and for which variations in these durations can variably impact outcomes).

Other spelling errors: p. 18, Line 391, should be “finding.” P. 19, line 395 should be “healers.” P. 19, line 402: should read “symptoms.”

Some of the subjective writing in the discussion and conclusion needs explanation. For instance, in the conclusion, the authors note that “informal care was convenient and easy to access, which…” Were these quotes from patient interviews (qualitative research component), or the writers’ beliefs? If so, they should be presented as hypotheses or potential causes rather than facts. For instance, distrust of established healthcare services and fear of diagnosis have been identified as barriers to care/delays in care in some qualitative research studies.

Avoiding attribution of cause will help with the authors’ later point that “there is a need to better understand the reasons for the intervals…”

The figure needs to be more crisp (did not show up well/was blurry on the provided pdf formatting).

Reviewer #2: The study methodology was sound enough and the article was well written in palatable sentences that can meet the publication criteria in this well reputed journal, PLOS One. The analysis was good as well.

Reviewer #3: This is a very interesting study that is well designed and well written. I have the following comments for the authors to consider that may improve the manuscript.

1. I would be interested in knowing the relationship of the authors (all Australian) with the people they worked with in Bangladesh. I found it somewhat surprising that no Bengali collaborator was included on this manuscript and would consider adding one if possible. This does not necessarily need to be included in the manuscript, although I would like a comment in the response.

2. For those who are not familiar with the term “informal healthcare providers”, please provide explanation or example in the abstract.

3. The wording in the sentence “Although only one fifth the size of its neighbour 92 Pakistan in geographical area, it has just on 75% the population of that country.” Is odd on the bolded section. Perhaps this should read “geographical area, the population is only slightly smaller (75%) that Pakistan’s population”

4. I really appreciated the detailed contest on the Bengali lung cancer statisticis and the health system.

5. I am a little confused about how “traditional healthcare providers” are informal healthcare providers. To me this term indicated traditional hospital-based. I wonder if this is mean to include more of the village doctors? Can you define or clarify this term. I see this is defined later… I might take out the word healthcare or at least move the examples up. Later this is called a traditional healer (which may be a good term)

6. The introduction is very long. It could be shortened by removing some of the non-specific information about lung cancer since there is ample in the discussion of lung cancer in Bangladesh

7. It would be helpful to know how often the providers did or didn’t refer. Was this all consecutive patients? It would be helpful to be able to know a little about potential selection bias. Did all providers refer?

8. Was there any patient compensation?

9. Please comment on the predominance of men in the study.

10. There is a missing ( in line 279.

11. This would be a little easier to read with tables at the end of the manuscript.

12. The discussion is also quite long. It would be nice to consolidate some of the pharmacy discussion, as this is included with references in separate paragraphs.

13. In limitations, please commend on how this sample compares to all people getting a diagnosis. I suspect this is a limited sample, which should be included as a limitation.

Reviewer #4: PONE-D-20-39263

Duration of intervals in the care seeking pathway for lung cancer in Bangladesh: a journey from symptoms triggering consultation to receipt of treatment.

This is a cross-sectional study of 418 patients, majority male, seeking care for lung cancer in Bangladesh. Participants were interviewed in-person and underwent medical record review. Overall, this is extremely well written, logical, and succinct with only a few minor grammatical and typo errors. The majority of participants visited informal healthcare providers first rather than formal healthcare providers (60%). This is incredibly important information as to where people whether in rural or urban settings are getting their healthcare. The authors should be commended in researching a population that is often overlooked due to therapeutic nihilism and that so many were illiterate (64%!). This can make research incredibly difficult, but they took every precaution including informed consent and have documented this incredibly important information. Well done!

Minor comments:

1. Why do the authors think the majority of the population is male? Are women not seeking care? Do they not smoke? Are they not seeking care at all? Mentioning this in the discussion section would be helpful.

2. Line 92: sentence is awkward, would revise. “it has just on 75% the population of that country”. Would suggest: “Bangladesh is geographically much smaller but has 75% of the population of Pakistan.” Or something similar.

3. Line 191 & 208: typo in “minimise” should be: “minimize” but I defer to the editor

4. Line 257: would say “less than or < 45 years” rather than “under 45 years”

5. Line 257: rural areas. The area needs an (s).

6. Line 282: end sentence after cancer diagnosis and remove “of lung cancer”. This is redundant.

7. Table 5: please explain if the p-value is for the distribution of the covariates within the column rather than comparing the urban percentages across the row between formal and information. This can be denoted as a footnote.

8. Line 391: “finding” is misspelled.

9. Line 406: Recognize rather than “recognize” but I defer to the editor

Reviewer #5: This is an interesting paper in an area where there’s little evidence.

I note all of the guidelines quoted in the introduction section. It would be good to see any kind of guidance designed for lower income settings, as it is difficult to apply the standards of very wealthy countries to poorer settings (perhaps something from IARC instead?).

I am not sure how relevant some of the other references are either, for example reference 13 on changes in presenting symptoms – these are very UK specific and are not really applicable to LMIC settings. I would be inclined to restructure the introduction de-emphasise the western country examples – I know there is some LMIC material, but I think there is too much from developed countries.

Reference 21 should be updated, there must be some more recent information on the numbers of hospitals providing cancer treatment in the country.

It would be good to know which literature informed the questionnaire; was it adapted from any existing instrument? Was there any formal validation process?

It might be good also to describe the provision of primary care in Bangladesh – for example, whilst general practitioners are referred to in the paper, this can mean something quite different in LMIC’s. Typically not many medical doctors are classified as general practitioners.

The study draws some interesting comparisons between diagnostic intervals in Bangladesh and other countries – but there are a few methodological issues.

• Date of first symptom can be very difficult to define with lung cancer. It often occurs on the background of chronic respiratory symptoms and the change can be very subtle eg in the nature of the cough or some additional chest pain. Sometimes there is little change at all in symptoms which may have been present for 20 years or more. This should be highlighted in the discussion section along with comments on the validity of the self-reports used

• Date of first presentation can also be difficult to define in the context of multiple or long lasting symptoms and there is often a discrepancy between what patients say and primary care records. It may be helpful to refer to some of the methodological diagnostic intervals work – eg the papers on appraisal intervals by Fiona Walter or the Aarhus statement on diagnostic interval studies. The International Cancer Benchmarking Partnership has included lung cancer in its work and that may be worth referring to as well.

Some other points:

• There should be more emphasis on the huge amount of chronic respiratory disease which goes undiagnosed in Bangladesh. While I appreciate the authors acknowledge this limitation, there should be greater recognition that much lung cancer remains undiagnosed, particularly in poor, rural and remote populations in South Asia.

• I note the recommendation for awareness raising campaigns. I am not sure how meaningful such recommendations would be in the context of the study. As partly highlighted in the paper, there are enormous financial, cultural and social barriers to presenting with lung cancer symptoms in LMIC’s.

• Some authors describe the interval between first noticing a symptom and reporting it, even reporting it to a family member; diagnostic journeys can be rather convoluted and certainly not linear. There is no mention in the paper of stigma, yet this has been recognised as an important barrier to presenting with symptoms that may indicate cancer in LMIC’s.

• It would be good to see some recommendations which are a little bit more tailored to the local financially constrained environment. For example, investment in basic primary health care and use of community health workers can help uncover the huge burden of chronic respiratory disease which is often undiagnosed and unmanaged. Any programme targeting lung cancer would need to be undertaken within this broader context of chronic respiratory disease. Treatments for lung cancer are expensive as are diagnostic investigations such as CT scanning and bronchoscopy. It would be better to ground any recommendations for improved diagnosis of lung cancer in the reality of the current capacity in Bangladesh. What kind of investment might achieve the most in settings where treatments such as chemotherapy, radiotherapy and thoracic surgery are severely limited?

The paper is, nevertheless, of interest because we have very little information to date on cancer diagnostic intervals in South Asian countries and this would be an important addition. I think if it could be better grounded in the international diagnostic interval literature, and if more attention could be paid to local health system and individual factors which might limit any improvements, the paper would be considerably improved.

Reviewer #6: This is an interesting paper on the care seeking pathway for lung cancer in Bangladesh, a subject that had not been studied yet. It shows that the time between the first contact with a healthcare provider and the diagnosis of lung cancer is quite long. This seems to be related to the fact that many patients contact an informal healthcare provider, mainly a pharmacy, before they contact a formally trained healthcare provider. This finding is important as timely diagnosis is crucial in lung cancer. Creating awareness of lung cancer symptoms and encouraging people to seek care from a formal healthcare provider may reduce the overall duration of the care seeking pathway.

My main comments:

Abstract:

As the care seeking trajectory of lung cancer has been studied extensively, please explain why a study Bangladesh is needed.

Please specify “other countries” in the conclusion, are these developing countries or western countries?

Introduction:

The introduction is quite lengthy. Some suggestions to shorten it:

- Please shorten the description of the recommended timeframes in guidelines. Maybe just describe the range or mean for the different time intervals.

- The part about presenting symptoms seems to be a bit out of place and could be left out, as the main presenting symptoms of lung cancer are well-known. Unless you have a hypothesis why presenting symptoms would be different in Bangladesh? Then you should also compare them to the literature.

- I found it very interesting to read about the health system in Bangladesh, it definitely helps to put the results in perspective, so this should remain part of the paper. However, details not directly relevant to the research question, such as e.g. the population of Pakistan, could be left out. Besides, the number of hospitals is based on an article from 2013, maybe more recent data are available?

- As to the research questions, I am missing the ‘why’ for the first two. Why is it important to look at the symptoms that motivated patients to seek care and which healthcare providers they visited? What problem could it solve? I would say your main question is the timeline of the care seeking pathway. A subquestion would then be whether some symptoms lead to a longer care seeking pathway as raising awareness about these symptoms could possibly shorten this time. Another subquestion would be for which health care providers the timeline is longer and then find ways to improve this.

Methods:

- Page 8, line 189: A part of questionnaire is about “history of illness”, what exactly does this mean? comorbidities?

- Page 10, line 218: Please further describe how the different time-points were determined. E.g:.

-date of first symptoms: was this as reported by the patient in the questionnaire? Based on the medical file?

- date of first contact with a health care provider: also including traditional health care providers and pharmacy?

- date of diagnosis of lung cancer: either histological or radiological?

- date of referral for treatment: by which health care providers and to whom, which treatments included? I assume surgery/radiotherapy/chemotherapy?

- date of treatment initiation, which treatments included?

- Page 10, line 222: Instead of the long text explaining the different time intervals, I would rather suggest a graph.

Results:

- Please use subheadings to structure the results.

- Page 11, line 253: Very impressive inclusion rate! Do you know the reason for refusal in the 6 patients?

- Table 1: What is the reason not to test for differences between locations in patient characteristics?

- Table 3: This table could be left out, as it does not add very much to the whole story. Rather, it could be interesting to add whether the timeframes differ by symptoms at diagnosis.

- Page 14, line 295: It is mentioned that the additional health care professionals visited at pre-diagnosis / pre-treatment were much less often traditional healers/village doctors, but there is also a large drop in pharmacy. This should be mentioned.

- Page 14, line 298: Rather describe something like: The type of health care provider who was the first point of contact did not differ by age, sex, marital status or family structure. However, patients from rural areas, illiterate patients and patients from the lowest income groups were more likely to choose an informal healthcare provider as the first point of contact.

Table 6: I assume the numbers in the table are days? Please add the number of patients in the header of the two columns?

A lot of different time intervals are discussed. I would prefer to only discuss in the text the 4 time intervals between the 5 time points and not the other overlapping time intervals. Just mentioning them in the figure or table seems enough to me

Discussion:

Were informal care providers also included in the other studies that you cite? This could explain the shorter duration between first onset of symptoms and first contact with a healthcare provider.

The comparison with existing literature is quite extensive and could be shortened, while the strengths and weaknesses of the study are hardly discussed and could be given more attention.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Shahjada Selim

Reviewer #3: No

Reviewer #4: No

Reviewer #5: No

Reviewer #6: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: PONE-D-20-39263_reviewer (2).pdf

Click here for additional data file.

6 Jul 2021

PONE-D-20-39263 Response to reviewers’ comments provided as an attachment which contains all the responses in a table.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

3 Aug 2021

PONE-D-20-39263R1

Duration of intervals in the care seeking pathway for lung cancer in Bangladesh: a journey from symptoms triggering consultation to receipt of treatment

PLOS ONE

Dear Dr. Ansar,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Sep 17 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Muhammed Elhadi, MBBCh

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

Reviewer #5: All comments have been addressed

Reviewer #6: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #5: (No Response)

Reviewer #6: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #5: (No Response)

Reviewer #6: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #5: (No Response)

Reviewer #6: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

Reviewer #5: (No Response)

Reviewer #6: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors present a study of provider and referral utilization in Bangladesh and the impact on wait times between symptoms to diagnosis and ultimately treatment. The authors' have adequately responded to the reviewers' critiques, leading to improvements in the manuscript clarity and impact. Minor concerns remain as detailed below, one of which includes a concern about the provided Figure 1:

p.13, line 292- In Table 1, it appears the upper quartile had a monthly income of BDT >/= 100,001, but the text states this upper quartile was had monthly income BDT 50,000. Could the authors please correct or clarify?

p. 23, line 507: The authors cannot directly suggest that informal care was “convenient and easy to access,” as this study did not evaluate reasons for self-referral to informal care providers. At most, the authors can highlight that most patients’ first point of care is through informal care providers, possibly due to convenience and ease of access (as previously described). The authors can also point out that patients were seen more quickly after initiation of symptoms when the first point was care was an informal care provider.

P23, line 511: Do the authors want to highlight that informal care led to significant delays in lung cancer diagnosis and initiation of treatment (not just initiation of treatment alone as it is currently written)?

Figure 1: Unclear why data for “Onset of symptom to diagnosis” is included twice in this figure. Should the bottom value be “Onset of symptom to Treatment”?

Reviewer #2: Well written article of a good study. In future other studies may include many more centre to describe more details of it.

Reviewer #3: Thank you for addressing comments, manuscript is improved. I appreciated the improved flow and detail within the manuscript.

Reviewer #5: (No Response)

Reviewer #6: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Shahjada Selim

Reviewer #3: No

Reviewer #5: No

Reviewer #6: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: PONE-D-20-39263_reviewer (2).pdf

Click here for additional data file.

5 Aug 2021

A rebuttal letter responding to each point raised by the academic editor and reviewer(s) uploaded (labelled as 'Response to Reviewers').

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

31 Aug 2021

Duration of intervals in the care seeking pathway for lung cancer in Bangladesh: a journey from symptoms triggering consultation to receipt of treatment

PONE-D-20-39263R2

Dear Dr. Ansar,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Muhammed Elhadi, MBBCh

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: All comments have been addressed

Reviewer #5: (No Response)

Reviewer #6: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: (No Response)

Reviewer #3: Yes

Reviewer #5: (No Response)

Reviewer #6: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: (No Response)

Reviewer #3: Yes

Reviewer #5: (No Response)

Reviewer #6: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: (No Response)

Reviewer #3: Yes

Reviewer #5: (No Response)

Reviewer #6: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: (No Response)

Reviewer #3: Yes

Reviewer #5: (No Response)

Reviewer #6: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #3: Nice work., the manuscript is improved. The authors responded appropriately to the comments that were provided.

Reviewer #5: (No Response)

Reviewer #6: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: No

Reviewer #5: Yes: Professor David Weller

Reviewer #6: No

2 Sep 2021

PONE-D-20-39263R2

Duration of intervals in the care seeking pathway for lung cancer in Bangladesh: a journey from symptoms triggering consultation to receipt of treatment

Dear Dr. Ansar:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Muhammed Elhadi

Academic Editor

PLOS ONE