Exploring primary care level provider interpretation and management of potential breast and cervical cancer signs and symptoms in South Africa
Jennifer Moodley1,2,3,a, Jane Harries1,b, Suzanne Emilie Scott4,c, Amos Deogratius Mwaka5,d, Smiji Saji6,e and Fiona Mary Walter6,7,f
1Women’s Health Research Unit, School of Public Health and Family Medicine, Faculty of Health Sciences, University of Cape Town, Observatory, Cape Town 7925, South Africa
2Cancer Research Initiative, Faculty of Health Sciences, University of Cape Town, Observatory, Cape Town 7925, South Africa
3South African Medical Research Council Gynaecology Cancer Research Centre, Faculty of Health Sciences, University of Cape Town, Observatory, Cape Town 7925, South Africa
4Centre for Oral, Clinical and Translational Sciences, Faculty of Dentistry, Oral and Craniofacial Sciences, King’s College London, London SE1 9RT, UK
5Department of Medicine, School of Medicine, College of Health Sciences, Makerere University, Upper Mulago Jill Road, PO Box 7072, Kampala 256, Uganda
6The Primary Care Unit, Department of Public Health and Primary Care, University of Cambridge, Cambridge CB1 8RN, UK
7Wolfson Institute of Population Health, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London E1 4NS, UK
ahttps://orcid.org/0000-0002-9398-5202
bhttps://orcid.org/0000-0001-7359-8419
chttps://orcid.org/0000-0001-5536-9612
dhttps://orcid.org/0000-0001-7952-2327
ehttps://orcid.org/0000-0003-0002-6326
fhttps://orcid.org/0000-0002-7191-6476
Abstract
Objectives: Women with potential breast and cervical cancer symptoms in South Africa (SA) usually self-present to primary health care (PHC) clinics. The aim of this study was to explore PHC provider interpretation and management of potential breast and cervical cancer signs and symptoms.
Methods: In-depth interviews with PHC providers incorporating vignettes were conducted between April and May 2019 in two sites in SA. Four vignettes (two breast and two cervical) were developed by the research team to capture aspects of provider symptom interpretation, reasoning, actions and challenges. The content of the vignettes was informed by a preceding community-based survey and qualitative interviews with symptomatic women. Interviews were audio recorded, transcribed verbatim and analysed using a thematic analysis approach.
Results: Twenty-four PHC providers were interviewed (12 urban, 12 rural; median age: 43 years). Four main themes relating to clinical assessment and reasoning; referral and feedback challenges; awareness of breast and cervical cancer policy guidelines and training and education needs emerged. Vignette-prompted questions relating to presenting symptoms, and possible accompanying symptoms and signs, demonstrated comprehensive proposed history taking and clinical assessment by PHC providers. Cancer was considered as a potential diagnosis by the majority of PHC providers. PHC providers also considered the possibility of infectious causes for both breast and cervical vignettes indicating they would ask questions around human immunodeficiency virus status, use of anti-retroviral therapy, and, for those with cervical symptoms, would need to rule out a sexually transmitted infection. Sexual assault was considered in assessing the cervical symptom scenarios. Providers raised issues around cumbersome booking systems and lack of feedback from referral centres. The need for provider and patient education and training to improve timely diagnosis of breast and cervical cancer was raised. Most providers were not aware of current breast or cervical cancer policy guidelines.
Conclusion: Clinical assessment at PHC level is complex and influenced by local health issues. Providing context-relevant training and support for PHC providers, and improving referral and feedback systems, could assist timely diagnosis of women with symptomatic breast and cervical cancer.
Keywords: primary health care providers, cancer symptom interpretation
Correspondence to: Jennifer Moodley.
Email: Jennifer.moodley@uct.ac.za
Published: 30/09/2021
Received: 25/06/2021
Copyright: © the authors; licensee ecancermedicalscience. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Introduction
Breast and cervical cancer are leading causes of cancer morbidity and mortality in South Africa (SA), with breast cancer being the most commonly diagnosed cancer and cervical cancer the leading cause of cancer deaths among women [1, 2]. Both cancers have been recognised as priority diseases by the South African National Department of Health, with recent policy guidelines emphasising the importance of early diagnosis of symptomatic disease as it enables more opportunities for curative treatment and improved outcomes [3, 4]. This emphasis resonates with the World Health Organization’s recent positioning statement that recognition of symptomatic cancer at an early stage is one of the most effective public health measures in cancer control [5].
SA has a tiered public sector health care system (primary, secondary and tertiary levels). Typically, women with potential breast and cervical cancer symptoms will self-present to primary health care (PHC) clinics and will be initially assessed by a PHC provider. The majority (80%) of first contact care and assessment is by PHC nurse providers [6]. PHC providers attend to patients with a range of symptoms and undifferentiated health problems on a daily basis and need to prioritise among the many potential differential diagnoses. Whilst potential cancer symptoms are common, cancer is relatively rare. Research in high-income countries has shown that the vast majority of patients that present with possible cancer symptoms do not end up with a cancer diagnosis [7–11]. Local studies have shown that the presence of co-morbidities can complicate referral and cancer diagnosis [12, 13]. In one South African study, women with co-morbidities had a significantly longer diagnostic interval for breast cancer compared to those without co-morbidities [13]. Studies conducted elsewhere in Africa have shown that PHC providers have limited breast and cervical cancer knowledge. A study in Nigeria showed that 58% of PHC providers had fair or poor breast cancer knowledge [14]. Another Nigerian study reported that the majority of PHC workers interviewed showed poor understanding of cervical cancer disease, symptoms and risk factors [15]. In a needs assessment conducted in Rwanda, 60% of general practitioners based at district level hospitals identified a need for more information about various areas of cancer management, including diagnostic skills to detect cancers earlier [16]. PHC providers have a challenging but pivotal role in appraising women’s symptoms, assessing the likelihood of the symptoms being due to cancer and deciding whether to refer patients to secondary level facilities for further diagnostic tests and assessments. A timely and accurate initial appraisal of potential symptoms at the PHC level is a critical step in the cancer diagnostic pathway, yet little is known about factors influencing this assessment in low- and middle-income settings, including in SA. This qualitative sub-study formed part of a larger study aimed at better understanding pathways to care for women with possible breast and cervical cancer symptoms. We used both quantitative and qualitative methods to meet the overall study aim [17–20].
In this sub-study, we aimed to provide an in-depth, more nuanced understanding of PHC providers experience and clinical practices interpreting and managing potential breast and cervical cancer signs and symptoms.
Methods
Study design, population and setting
Qualitative research was deemed important and suitable to our research objectives. We used a qualitative in-depth interview (IDI) study design augmented by vignettes. The study was conducted between April and May 2019 at one urban and one rural site in SA. Details on the study site have been described previously [20, 21]. The research co-ordinator approached health facility managers in both research sites, explained the overall purpose of the study and requested a list of PHC providers who would encounter women with symptoms in their daily practice. We obtained information on professional status, level of experience, age and gender so that we could purposively recruit a mix of participants in both rural and urban sites. We then approached participants, invited them to participate in the study and set up a suitable time to be interviewed. Written informed consent was obtained from providers prior to the interview process. Consent (written) was also obtained to record all interviews. Confidentiality and anonymity were ensured, and participants were assured that responses would not affect their employment in any way.
Data collection
Interviews were conducted by female research assistants trained in qualitative research methods, in a private room in the health care facility and were 40–60 minutes duration. All interviews were audio recorded and conducted in isiXhosa or English. Interviews were transcribed verbatim and translated into English by an independent transcriber.
Four vignettes were developed by the research team: two on breast symptoms and two on cervical symptoms. Vignettes are brief stories or scenarios that describe hypothetical situations to which a participant responds [22, 23]. Our vignettes were developed to capture aspects of provider symptom interpretation; reasoning, judgements and actions related to investigations and management; referral and feedback processes and consideration of infective or alternative diagnoses. The content of the vignettes was informed by findings from the preceding community-based cross-sectional survey, qualitative IDIs with symptomatic women and from drawing on the literature [18, 20]. Briefly, the vignettes described a 50-year-old woman with diabetes mellitus and a breast lump (scenario 1); a 50-year-old woman with diabetes mellitus and breast skin changes (scenario 2); a 35-year-old woman with human immunodeficiency virus (HIV), a vaginal discharge and post coital bleeding (scenario 3) and a 60-year-old woman with post-menopausal vaginal bleeding (scenario 4) (Appendix A). The interview guide and vignettes were piloted, and changes were made to improve flow and clarity of the vignettes. The vignettes were presented to participants in stages, with prompts to explore provider opinions and reasoning. Each provider was presented with two scenarios (one breast and one cervical) and we ensured an equal distribution of the two scenarios between both rural and urban sites. Study participants were all presented the scenarios in the same order – one breast and one cervical scenario. At the end of the scenario, providers were also asked about policy and guidelines related to the management of possible breast and cervical cancer in SA.
Interviewers recorded reflective fieldnotes after each interview which provided added context when reviewing the transcripts. Socio-demographic information was collected prior to the interview and included age, gender, professional status and years of experience including number of years working at the health care facility.
Data analysis
We adapted aspects of the framework method to enhance thematic analysis by following the different stages of transcription, familiarisation with the interview, coding, developing and applying an analytical framework, charting data into the framework matrix and interpreting the data [24]. Initial categories for analysing data were drawn from the interview guide and vignettes, and then themes and patterns were identified after reviewing the data. The transcripts were coded by JH using the qualitative software package NVivo 12 Pro which facilitated the sorting and management of the data. SS sorted and coded the clinical data, capturing the key elements of responses related to history taking, clinical examination and investigations in a matrix framework (see Appendix B). The clinical coding was reviewed by JM and FMW. A code book with 30 codes, 19 a priori codes and 11 emergent codes was developed by JH based on the research objectives and interview guide and then shared with the research team and cross checked for coder variation. Coding discrepancies were resolved through discussion and consensus. Memos were recorded alongside the coding process and were useful in exploring relationships of links across categories, or reflections about a particular phenomenon. The data were then reviewed for major trends, crosscutting themes and issues for further exploration. Given the common themes occurring across the interviews, we concluded that data saturation had been achieved as little new information related to the main research questions emerged from the data.
To address dependability, we established an audit trail by developing a detailed track record of the data collection process, coding was undertaken by more than one person and coding was checked amongst the study team for coder variation. Coding discrepancies were resolved through discussion and consensus. For confirmability, all interviewers kept a reflective journal and we held regular investigator and researcher team meetings to discuss research progress. Further, data collection used a range of different interviewing techniques, vignettes and more structured questions with study respondents. Transferability was addressed through purposeful sampling to represent a range of respondents in both urban and rural sites. The findings of the study have been reported following the Consolidated Criteria for Reporting Qualitative Research [25].
Ethical considerations
Ethical approval was obtained from the Human Research Ethics Committee, University of Cape Town (859/2017). Permission was also obtained from the local health departments to undertake research in the health care facilities. All study participants provided written consent to be interviewed and recorded.
Results
A total of 24 PHC providers were interviewed (12 urban and 12 rural). The participant median age was 43 years (inter-quartile range: 27–59 years). The majority were female (20/24). Fifteen of the participants were professional nurses, five clinical nurse practitioners, one community service nurse and three were operations managers. Results are presented according to the main themes which included clinical assessment and reasoning; referral and feedback challenges; awareness of breast and cervical cancer policy guidelines and training and education needs.
Clinical assessment and reasoning
Assessments for cancer
PHC provider accounts of the history that would be taken were comprehensive, with detailed questions around the presenting symptoms and on possible accompanying symptoms linked to the vignette scenario (see Appendix B). Similarly, descriptions of clinical examinations were detailed, related to the presenting symptom and included a full clinical breast examination (scenarios 1 and 2), vaginal examination with a speculum (scenarios 3 and 4) and for both a more general ‘head-to-toe’ assessment.
For breast symptoms, history taking included questions around physical symptoms such as pain, size and texture of lump, and duration of symptoms. In addition to physical symptoms, providers also asked about familial history of breast cancer. Providers suggested similar symptom history taking questions for both breast symptom scenarios.
The first question is I’m going to ask the woman, when did she first notice the lump, is the lump growing or not growing; is the lump painful or not painful… And I am going to ask her if she is taking her medication well and ask her about the history of family and cancer – any cancer in the family (Scenario 1 rural)
Providers also mentioned the importance of asking the patient about nipple and breast skin changes and similarly noted any changes during their clinical examination. A provider noted that skin changes, especially colour and skin texture and appearance (peau d’ orange) could be indicative of breast cancer. Checking for lymph nodes under the axilla was also deemed important to rule out breast cancer.
You also ask about the changes. As you already told me there is changes on the breast, but you check if the change is in the nipple or on the skin. you also ask the client, is there any lump on the breast …So we talk about the lymph nodes, is there any lymph nodes on the armpit. So those are the things we first want to know and you check all those before you do your examinations... you can check those changes because if you query cancer there can be a change on the skin or sometimes it becomes orange or any discolouration on the skin or on the nipple.…(Scenario 2 urban)
Some providers suggested total body examinations from ‘head to toe’ for breast changes, possibly to rule out other underlying conditions (anaemia, jaundice cyanosis, scoliosis).
Examine the client from head to toe. She must take off all the clothes so that we can look and examine her… for the nurse to check the head and the eyes and the mouth, all the parts around the head because as a nurse you’re going to look for other problems except the one she comes with like maybe she’s got a problem with anaemia …you check if this rash is septic or is just a rash or is reddish and you just do the massage on the breast… to see if maybe there’s no lumps around the breast ..maybe this rash coming from inside the breast – (Scenario 2 rural)
For cervical symptoms, the scenarios were often linked to other possible causes of vaginal bleeding such as hormonal contraception, possible miscarriage or pregnancy, retained products of conception post abortion and possible sexually transmitted infection (STI) and sexual violence (see later Ruling out other infectious diseases and Querying sexual violence).
It is important because we want to check if this vaginal discharge is caused by a sexually transmitted infection, or is it thrush which is not sexually transmitted, it can be any kind of discharge so that’s what we want to exclude by asking that question with the symptoms that she’s presenting with it can be either between the sexually transmitted infections and signs of cancer, miscarriage or even after an abortion, then maybe there was products of conception that were left inside the uterus, So that can give her those signs that she’s presenting with, so asking those questions help me decide, this patient could likely be suffering from this or that (Scenario 3 urban)
Almost all providers indicated that they would perform a Pap smear to rule out cervical cancer (for Scenario 4) as post-menopausal bleeding could be indicative of cervical cancer. However, some noted that they would not do a Pap smear if there was too much bleeding with a large lesion (as per scenario).
You ask when the bleeding started because this client is not supposed to bleed again, that is the first thing… you suspect as the nurse that there is something wrong ... And you are not even going to do anything, you are not going to do a vaginal examination or anything, that is not what you are going to do because you suspect that there is cervical cancer … menopause long ago there’s nothing you’re going to do, you are going to refer this person to the doctor immediately; because you can’t do a smear on a bleeding client (Scenario 4 rural)
Some providers recounted how they would examine the patient from ‘head to toe’ looking for generalised signs of bleeding, followed by a vaginal examination including for the presence of genital warts.
First of all you need to do a physical examination on the whole body … I want to check if everything is normal from head to toe, to check if she’s not pale, there are no sores on the body and no rash before I do the vaginal examination (Scenario 3 rural)
Considering HIV
Questions around HIV status and use of anti-retroviral therapy (ART) were an important part of history taking and influenced clinical reasoning and guidance.
She is HIV positive. First of all I am going to ask if she is taking her treatment [ART] well. I am going to ask the client if she is using a condom and concerning her complaint I am going to ask all that, how many partners she has, when the bleeding started and then I am going to ask if she’s had a Pap smear before and then if the answer is no I’m going to continue with counselling that every woman that is on ART should do a Pap smear yearly (Scenario 3 rural)Some providers (both rural and urban) noted that an HIV status could further compound the risk of cervical cancer.
She’s HIV positive, the risk is much higher now for that cervical cancer as well, so that is why we need, to be much more vigilant with her (Scenario 3 urban)
…if a client is HIV positive you should do a Pap smear from 30 years of age, ….We want to exclude cervical cancer (Scenario 3 rural)
For women with breast symptoms, a history of use of HIV medication (efavirenz) was considered important as it is a possible cause of breast lipodystrophy.
The medication can help you to diagnose this client … For example, I’m working on ART, so if the client maybe comes with an enlarged breast, so that is where I know this client is on efavirenz and complaining about the enlargement of both breasts. So, it can be caused because of the efavirenz. So that is why it is very important to also check what kind of medication the client is on (Scenario 2 urban)
Ruling out other infectious diseases
For the scenarios related to possible cervical cancer symptoms, ruling out a STI was deemed important. A provider described how she would establish whether symptoms (vaginal discharge, bleeding) could be attributed to an STI due to unprotected sex or retained products of conception, versus a non-infective condition such as thrush or a miscarriage or whether signs of possible cervical cancer.
A smelly, yellow vaginal discharge in an HIV positive woman was viewed by some providers as an indicator of an STI.
Like I said the yellow discharge, obviously that’s clear that it’s an STI, you will ask if it’s smelling, then you know how to treat what type of infection it is (Scenario 3 urban)When was she diagnosed with HIV, has she been taking treatment or not, … How many sexual partners does she have, is she using a condom and when did she notice this unusual discharge and what else is there, does she have low abdominal pains, itchiness or any other problems and for how long has she been having this discharge and when did the bleeding start? (Scenario 3 urban)
In relation to breast skin and colour changes, a provider noted that skin changes might be due to cellulitis, a breast abscess or mastitis and needed to be ruled out prior to assuming symptoms might be attributed to breast cancer.
And then, I think also the colour of the lump or the size of the lump, I think is also important because also defining it as to what might be the conditions. Because there might be cellulitis, there may be breast cancer, all those things, they may manifest in the same manner you see…, we need to establish as to whether or not she has skin problems generally or not the skin problem can also affect the breast like breast abscess or mastitis (Scenario 1 rural)
Querying sexual violence
Rape or sexual assault was considered a possible causative factor for the vaginal bleeding symptom described in Scenario 4 but did not arise in the history taking for Scenario 3. Often this questioning was related to the high levels of crime in SA.
The first question I will ask, because our country has no law; has she been raped…
Is there any trauma, how long and when did this bleeding start, because of her age she shouldn’t bleed…. Because all these old mothers are being raped nowadays because there is no law here in South Africa. Maybe someone broke into her house at night and then raped her and that caused the bleeding or maybe the trauma by being abused by the thugs or robbers in her house (Scenario 4 urban)
Is there any history of assault? Because maybe if in cases of assault maybe, bleeding can occur in cases of rape (Scenario 4 rural)
Referral and feedback challenges
Providers raised a range of health systems issues in the process of diagnosis and treatment of women with potential breast or cervical cancer symptoms. These ranged from cumbersome booking systems to the way test results were received, tracked and followed up.
Most providers indicated that doctors played a key role in terms of patient management and referral. In situations where there were no facility, doctors available at primary care clinics providers would refer directly to a higher-level facility by making a telephonic appointment and/or providing the patient with a referral letter to take to the referral centre.
The referral process in all the facilities is that it has to go via the doctor before they can accept a client it has to go via a doctor in the primary health care facility first and then to a tertiary hospital (Scenario 1 urban)
Depending on where the health care facility was located, investigation or test results were received in different ways. In rural facilities, results were received via a courier, whereas in urban facilities with better infrastructure, providers were able to view results online. Informing patients of their test results was cumbersome and inefficient for both patients and providers. Whilst providers were able to eventually receive results, the system of contacting patients was complicated, relying on telephone contact, using community workers to access patients, or relying on patients returning to receive their results or, in some cases, relying on the referral doctor.
A provider explained how lines of communication and feedback were dependent on an ad hoc system and not a formal feedback process:
The only thing is the curiosity from the nurse to keep on asking the doctor, but sometimes we always get the same answer, ‘No, I’m still waiting, that patient is still in hospital,’ you know, yes that is the problem. The only solution that can help us is for us to have direct referral system of these cases (Scenario 2 urban)
Providers also voiced concerns related to lack of feedback, often relying on community workers to follow-up and in some instances finding out from others that a patient had died:
As our nurses here we need a back-referral system and this was discussed for a long time about the referral system; but nothing is happening, and you have to know what happened to the client that you referred, and not hear from other people that the patient has passed away… You need to know how the doctor managed the patient, we have an interest in that but it’s not happening. I call, we phone, to give results, and if she doesn’t have the phone, we have the community workers they do the follow up (Scenario 4 rural)
We want to know what is happening with the client that we referred. If the referral system could improve – the back-referral system,… As I said, the back-referral system is bad, they don’t come back to us and tell us what was done on the client, there’s no back-referral system.[feedback] (Scenario 1 rural)
Interviewer : is the nurse in this situation likely to get feedback from the referral centres? Respondent: only for the colposcopy, that’s when you are able to get feedback. But for the other cases, unless you are lucky enough to, for the patient to come back and tell you what happened (Scenario 4 urban)
Awareness of breast and cervical cancer policy guidelines
Most providers interviewed were not aware of the existing breast or cervical cancer policy guidelines or their content nor had the opportunity to review the policies. A few providers mentioned that policy and guidelines might be available at their facility.
‘no ….I can’t remember now … I don’t have the guidelines with me, but they are in a file, but I know what you are supposed to do. . . . I don’t want to lie, … I don’t remember’, (Scenarios 1 and 3 urban)
Because this [breast cancer] is a well-known condition, so I believe that there are policies for it but myself I’ve never come across those policies (Scenario 2 rural)
For those few providers who had some knowledge of the cervical cancer screening policy, it was mostly related to screening guidelines (age and time intervals) including for HIV positive women; however, these providers were also unsure of where to locate the guidelines in their facility should they need to refer to it.
There are guidelines because we do, this Pap smear that we do at 30 – Pap smear, 40 – Pap smear, 50 – pap smear, those are the guidelines…3 years every 3 years if HIV positive (Scenario 3 urban)
Some providers whilst being familiar with the SA cervical and breast cancer guidelines did express an interest in engaging with the policy and guidelines.
A provider located in the rural site requested information on current cancer policies to increase her knowledge about how to manage the types of scenarios described (reflections rural)
Training and education needs
Providers raised the need for education and training to improve timely access, diagnosis and treatment spontaneously. Regular educational workshops on breast cancer detection and diagnosis were deemed important to improve timely diagnosis and treatment of breast cancer.
Yes, for me I think there should be more workshops for us Clinical Nurses to attend because most of the people if you can see the stats they are dying because of breast cancer, meaning which we are misdiagnosing here in the clinics. If we can be sure of what we are doing when we assess cancer and we assess the patients here first, we can detect it at an early stage because hospitals are full of breast cancer patients. And you ask yourself how and when the clinics missed these people; because when they reach us it’s too late, we should detect then when it starts here at the clinic. If we can have regular workshops, even twice a year, that’s fine (Scenario 2 urban)
Many providers spoke about their role in patient education including teaching patients about breast awareness, breast self-examination and the importance of regular Pap smears and allaying fears about the procedure, as such education could facilitate early detection of female cancers.
It is important for us to diagnose or for quick diagnosis or referral, we are depending on the patient because she’s the one who examined herself, her breast at home. So, we teach them how to see the weight of the two breasts, if they are equal and if she sees that there’s a difference, they are not equal, maybe there’s something that is not normal (Scenario 1 rural)
Some patients do not want to do a Pap smear because they have a fear of the unknown that the Pap smear is painful this … but I’m going to do clear health education that the patient may see the importance of doing a Pap smear (Scenario 1 urban)
Discussion
This is the first study in SA exploring PHC provider interpretation and management of potential breast and cervical cancer symptoms. In responding to the scenarios, our study showed that PHC providers are likely to take a detailed history and perform a comprehensive examination. Cancer was considered as a potential diagnosis, but providers also reflected on the local context as they worked through a differential diagnosis. The burden of disease in SA includes infectious diseases, maternal and child health and injuries and violence, as well as cancer and other non-communicable diseases [26]. The country has a very high prevalence of HIV and STIs, with a particularly heavy burden of disease among women [27, 28]: in 2019, HIV prevalence among women aged 15–49 years was 25% [27], approximately 1.1. million new cases of STIs are treated annually [28] and sexual violence remains a major problem [29, 30]. Given this disease profile, it is not surprising that infectious diseases and sexual violence remained foremost in PHC providers’ minds.
Local studies have shown that women with breast cancer symptoms may visit PHC facilities several times before referral and diagnosis [12, 13]. A particular challenge facing all PHC providers is the overlap between cancer symptoms and symptoms arising from common local health conditions. Abnormal vaginal bleeding, vaginal discharge and lower abdominal pain, all known symptoms of cervical cancer, can also occur in women with STIs and those who have experienced sexual violence. In addition, STIs may occur concurrently with cervical cancer, further complicating symptom interpretation and diagnosis. There are also infections and other conditions with symptoms similar to breast cancer. For example, symptoms such as skin induration, puckering and ulceration can also occur in women with a breast abscess, and with lactational and non-lactational breast infections. Lipodystrophy, a side-effect of some ART, results in a redistribution of body fat with loss over limbs and gain in various sites including the breast [31]. Symptoms of breast cancer in patients on ART may thus be misattributed to lipodystrophy and could contribute to a delay in timely diagnosis. Our study shows that PHC providers do consider cancer as a potential differential diagnosis but the extent to which symptom overlap contributes to delayed diagnosis of breast and cervical cancer at a primary care level in SA requires further investigation.
Within the SA public health care system, once a PHC nurse suspects that a patient may have cancer, referral for further assessment either to a PHC based doctor or to a secondary level service is required. Appropriate well-functioning referral and feedback systems between primary and secondary level providers are thus critical to timely cancer diagnosis. Well organised feedback systems not only facilitate timely diagnosis but can also serve as a means of continuing education for PHC providers. The challenges of cumbersome appointment, referral and feedback systems have been raised previously [32, 33], and our study underscores the urgent need to comprehensively address this. In our study, very few PHC providers were aware of the national breast and cervical cancer policies, highlighting the importance of having mechanisms to ensure providers are aware of policy developments and how these can support them in clinical decision making and management. Health care providers should also use these policies as a guide to inform and support their clinical decision making in recognising and treating breast and cervical cancer.
The COVID-19 pandemic has resulted in an increased use of online communication within the health sector providing new opportunities for remote clinical assessments and support [34]. In the future, greater use of online communication systems could improve timely diagnosis of symptomatic cancer and address the training needs identified by the PHC providers in our study. However, any online support and training strategy has technological and financial implications and requires stable Internet services.
In the UK, various clinical decision support instruments have been developed to help primary care providers identify patients requiring further investigation for possible cancer [35–37]. These include electronic cancer Risk Assessment Tools (RATs) that are developed using data from UK cohort studies and then validated on a separate patient cohort [35–37] as well as artificial intelligence technology-driven symptom management tools [38]. Primary care providers in the UK reported that the RATs were helpful [11], and there is currently a large pragmatic primary care trial underway, investigating their utility and effectiveness. As SA develops electronic medical records and databases at multiple levels of the health system, development of locally relevant RATs becomes a possibility and could facilitate timely diagnosis of symptomatic cancer. SA breast and cervical cancer RATs will need to make provision for the high prevalence of infectious diseases and sexual violence. Developing of a local symptom management and referral e-Tool could further support PHC providers. Where possible, this e-Tool could link with electronic booking systems at referral, with the aim of streamlining the referral process.
Our study is based in one rural and one urban site in SA and insights gained are likely to be transferrable to other such settings in SA. Further, our study provides an in-depth and nuanced understanding of PHC provider’s appraisal and management of possible breast and cervical cancer symptoms – a strength of qualitative research. We chose to use vignettes in the form of clinical scenarios in our study as it allows for provider views and actions to be explored in a less threatening way thereby avoiding potential social desirability bias. Furthermore, the specificity of vignettes allows for the exploration of contextual influences on judgements [22, 23].
Conclusion
Clinical assessment at PHC level is complex and influenced by local health issues. Providing context-relevant training and support for PHC providers and improving referral and feedback systems could assist timely diagnosis of women with symptomatic breast and cervical cancer.
List of abbreviations
ART, Anti-retroviral therapy; HIV, Human immunodeficiency virus; PHC, Primary health care; SA, South Africa; STI, Sexually transmitted infection; RAT, Risk assessment tools; UK, United Kingdom.
Conflicts of interest
The authors declare that they have no conflicts of interest.
Funding statement
Research reported in this article was jointly supported by the Cancer Association of South Africa, the University of Cape Town and the SA Medical Research Council with funds received from the SA National Department of Health, GlaxoSmithKline Africa Non-Communicable Disease Open Lab (via a supporting grant Project Number: 023), the UK Medical Research Council (via the Newton Fund). Authors retained control of the final content of the publication. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.
Acknowledgments
The authors thank the study participants for sharing their knowledge and experiences; Dr Jennifer Githaiga for her commitment, enthusiasm and support throughout the project; the project advisory committee members, field staff and community liaison managers for their support in data collection; and Dr Chukwudi Nnaji for assistance with formatting. FMW is Director and SES is co-investigator of the multi-institutional CanTest Collaborative, which is funded by Cancer Research UK (C8640/A23385).
Authors’ contributions
JM, FMW, SES, JH and ADM conceptualised and designed the study. JH oversaw data collection. JH coded the transcripts data. SS sorted and coded the clinical data. JM and FMW reviewed the clinical coding. JM prepared the first draft, incorporated revisions and prepared the final draft. All the authors reviewed drafts and approved the final manuscript.
References
1. Bray F, Ferlay J, and Soerjomataram I, et al (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries CA Cancer J Clin 68(6) 394–424 https://doi.org/10.3322/caac.21492 PMID: 30207593
2. South African National Cancer Registry (2017) Cancer in South Africa 2017 (Johannesburg: NCR) [https://www.nicd.ac.za/centres/national-cancer-registry/] Date accessed: 3/05/21
3. South African National Department of Health, South Africa (2017) Breast cancer: prevention and control Policy (Pretoria: NDoH)
4. South African National Department of Health (2017) Cervical cancer prevention and control policy (Pretoria: NDOH)
5. World Health Organization (WHO) (2020) WHO report on cancer: setting priorities, investing wisely and providing care for all (Geneva: World Health Organization) Licence: CC BY-NC-SA 3.0 IGO
6. Mash R, Ogunbanjo G, and Naidoo SS, et al (2015) The contribution of family physicians to district health services : a national position paper for South Africa : forum S Afr Fam Pract 57(3) 54–61
7. Koo MM, Hamilton W, and Walter FM, et al (2018) Symptom signatures and diagnostic timeliness in cancer patients: a review of current evidence Neoplasia 20(2) 165–174 https://doi.org/10.1016/j.neo.2017.11.005
8. Koo MM, Unger-Saldaña K, and Mwaka AD, et al (2021) Conceptual framework to guide early diagnosis programs for symptomatic cancer as part of global cancer control J Glob Oncol 7 35–45 https://doi.org/10.1200/GO.20.00310
9. Rubin G, Berendsen A, and Crawford SM, et al (2015) The expanding role of primary care in cancer control Lancet Oncol 16(12) 1231–1272 https://doi.org/10.1016/S1470-2045(15)00205-3 PMID: 26431866
10. Nekhlyudov L, and Latosinsky S (2010) The interface of primary and oncology specialty care: from symptoms to diagnosis J Natl Cancer Inst Monogr 2010(40) 11–17 https://doi.org/10.1093/jncimonographs/lgq001 PMID: 20386049 PMCID: 3482946
11. Hamilton W (2010) Cancer diagnosis in primary care Br J Gen Pract 60(571) 121–128 https://doi.org/10.3399/bjgp10X483175 PMID: 20132704 PMCID: 2814263
12. Moodley J, Cairncross L, and Naiker T, et al (2016) Understanding pathways to breast cancer diagnosis among women in the Western Cape Province, South Africa: a qualitative study BMJ Open 6(1) e009905 https://doi.org/10.1136/bmjopen-2015-009905 PMID: 26729392 PMCID: 4716174
13. Moodley J, Cairncross L, and Naiker T, et al (2018) From symptom discovery to treatment - women’s pathways to breast cancer care: a cross-sectional study BMC Cancer 18(1) 312 https://doi.org/10.1186/s12885-018-4219-7 PMID: 29562894 PMCID: 5863383
14. Pruitt LCC, Odedina S, and Anetor I, et al (2020) Breast cancer knowledge assessment of health workers in Ibadan, Southwest Nigeria JCO Glob Oncol 6 387–394 https://doi.org/10.1200/JGO.19.00260 PMID: 32125900 PMCID: 7126761
15. Onyenwenyi AOC and McHunu GG (2019) Primary health care workers’ understanding and skills related to cervical cancer prevention in Sango PHC centre in south-western Nigeria: a qualitative study Prim Health Care Res Dev 20 e93 https://doi.org/10.1017/S1463423619000215
16. Martin AN, Kaneza KM, and Kulkarni A, et al (2019) Cancer control at the district hospital level in Sub-Saharan Africa: an educational and resource needs assessment of general practitioners J Glob Oncol 5 1–8
17. Moodley J, Constant D, and Mwaka AD, et al (2021) Anticipated help seeking behaviour and barriers to seeking care for possible breast and cervical cancer symptoms in Uganda and South Africa Ecancermedicalscience 15 1171 https://doi.org/10.3332/ecancer.2021.1171 PMID: 33680085 PMCID: 7929770
18. Harries J, Scott SE, and Walter FM, et al (2020) Women’s appraisal, interpretation and help-seeking for possible symptoms of breast and cervical cancer in South Africa: a qualitative study BMC Women’s Health 20(1) 251 https://doi.org/10.1186/s12905-020-01120-4 PMID: 33187501 PMCID: 7666481
19. Mwaka AD, Walter FM, and Scott S, et al (2021) Symptom appraisal, help-seeking and perceived barriers to healthcare seeking in Uganda: an exploratory study among women with potential symptoms of breast and cervical cancer BMJ Open 11(2) e041365 https://doi.org/10.1136/bmjopen-2020-041365 PMID: 33550241 PMCID: 7925866
20. Moodley J, Constant D, and Mwaka AD, et al (2020) Mapping awareness of breast and cervical cancer risk factors, symptoms and lay beliefs in Uganda and South Africa PLoS One 15(10) e0240788 https://doi.org/10.1371/journal.pone.0240788 PMID: 33091035 PMCID: 7580973
21. Moodley J, Scott SE, and Mwaka AD, et al (2019) Development and validation of the African Women Awareness of CANcer (AWACAN) tool for breast and cervical cancer PLoS One 14(8) e0220545 10.1371/journal.pone.0220545>https://doi.org/10.1371/journal.pone.0220545 PMID: 31386684 PMCID: 6684059
22. Finch J (1987) The vignette technique in survey research Sociology 21(1) 105–114 https://doi.org/10.1177/0038038587021001008
23. Hughes R and Huby M (2012) The construction and interpretation of vignettes in social research Soc Work Soc Sci Rev 11(1) 36–51
24. Gale NK, Heath G, and Cameron E, et al (2013) Using the framework method for the analysis of qualitative data in multi-disciplinary health research BMC Med Res Methodol 13 117 https://doi.org/10.1186/1471-2288-13-117 PMID: 24047204 PMCID: 3848812
25. Tong A, Sainsbury P, and Craig J (2007) Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups Int J Qual Health Care 19(6) 349–357 https://doi.org/10.1093/intqhc/mzm042 PMID: 17872937
26. Pillay-van Wyk V, Msemburi W, and Laubscher R, et al (2016) Mortality trends and differentials in South Africa from 1997 to 2012: second national burden of disease study Lancet Glob Health 2014(9) e642–e653 https://doi.org/10.1016/S2214-109X(16)30113-9
27. UNAIDS Country factsheet (Switzerland: UNAIDS) [https://www.unaids.org/en/regionscountries/countries/southafrica] Date accessed 3/05/21
28. Let our actions count South Africa’s national strategic plan for HIV, TB and STIs: 2017–2022 [https://sanac.org.za//wp-content/uploads/2017/06/NSP_FullDocument_FINAL.pdf] Date accessed 3/05/21
29. Steele SJ, Abrahams N, and Duncan K, et al (2019) The epidemiology of rape and sexual violence in the platinum mining district of Rustenburg, South Africa: prevalence, and factors associated with sexual violence PLoS One 14(7) e0216449 https://doi.org/10.1371/journal.pone.0216449 PMID: 31365536 PMCID: 6668777
30. Dunkle KL, Jewkes RK, and Brown HC, et al (2004) Gender-based violence, relationship power, and risk of HIV infection in women attending antenatal clinics in South Africa Lancet 363(9419) 1415–1421 https://doi.org/10.1016/S0140-6736(04)16098-4 PMID: 15121402
31. Zinn RJ, Serrurier C, and Takuva S, et al (2013) HIV-associated lipodystrophy in South Africa: the impact on the patient and the impact on the plastic surgeon J Plast Reconstr Aesthet Surg 66(6) 839–844 https://doi.org/10.1016/j.bjps.2013.02.032 PMID: 23545227
32. Moodley J, Constant D, and Botha MH, et al (2019) Exploring the feasibility of using mobile phones to improve the management of clients with cervical cancer precursor lesions BMC Women’s Health 19(1) 2 https://doi.org/10.1186/s12905-018-0702-1 PMID: 30616579 PMCID: 6322222
33. Momberg M, Botha MH, and Van der Merwe FH, et al (2017) Women’s experiences with cervical cancer screening in a colposcopy referral clinic in Cape Town, South Africa: a qualitative analysis BMJ Open 7(2) e013914 https://doi.org/10.1136/bmjopen-2016-013914 PMID: 28213603 PMCID: 5318645
34. Nnaji CA and Moodley J (2021) Impact of the COVID-19 pandemic on cancer diagnosis, treatment and research in African health systems: a review of current evidence and contextual perspectives Ecancermedicalscience 15 1170 https://doi.org/10.3332/ecancer.2021.1170 PMID: 33680084 PMCID: 7929764
35. Hamilton W, Green T, and Martins T, et al (2013) Evaluation of risk assessment tools for suspected cancer in general practice: a cohort study Br J Gen Pract 63(606) e30–e36 https://doi.org/10.3399/bjgp13X660751 PMID: 23336455 PMCID: 3529290
36. Hippisley-Cox J and Coupland C. (2013) Symptoms and risk factors to identify women with suspected cancer in primary care: derivation and validation of an algorithm Br J Gen Pract 63(606) e11–e21 https://doi.org/10.3399/bjgp13X660733 PMID: 23336450 PMCID: 3529288
37. Hippisley-Cox J and Coupland C (2015) Development and validation of risk prediction algorithms to estimate future risk of common cancers in men and women: prospective cohort study BMJ Open 5(3) e007825 https://doi.org/10.1136/bmjopen-2015-007825 PMID: 25783428 PMCID: 4368998
38. Cthesign [https://cthesigns.co.uk] Date accessed 3/05/21
Appendix A. Vignettes.
Vignettes and IDI guide
Prior to description of scenario and prior to the interview remember:
• Greeting and introduction
• Provide background and information on study
• Complete informed consent before proceeding
• Complete socio-demographic data collection sheet
• Each participant to be presented with one breast and one cervical cancer scenario
Interviews to be audio-recorded.
Introduction to scenario
I am interviewing PHC providers about management of certain symptoms experienced by women. I am going to describe different scenarios for you. In each scenario, a female patient is seen by a PHC provider at a clinic. I would like you to help me complete the scenario by describing what the primary care provider in the scenario will ask or do at each stage. There are no correct or incorrect answers.
With your permission, I would like to record our conversation today. The information we discuss today will remain confidential and will not be shared with anyone outside of our immediate research team.
Once participant has consented to record interview say: I am going to begin recording the interview now. (State the participant ID, date, location interviewer name and scenario numbers at the start of the recording)
Scenario 1
A 50-year-old woman with a history of diabetes noticed a lump on her right breast. She comes to this clinic, (insert name of clinic), to have the lump on her right breast checked. At the clinic, she is seen by the Nurse, who must assess and manage the patient.
Stage 1 – History
• After the Nurse introduces herself and makes the woman comfortable, can you tell me what questions would the nurse ask the patient to help with the assessment of this patient?
For each question provided, interviewer to probe
◦ Could you explain to me why it is important for the nurse to ask that question?
◦ Could you explain how does it help her to manage the patient?
Prompt
• Is there any other question/s the nurse is likely to ask the patient? Interviewer to keep prompting until there are no further questions.
Stage 2 – Clinical examination. We are now going to talk about clinical examinations
I am continuing with the scenario. The nurse asks the woman if she can examine her.
• How would the nurse go about doing the clinical examination?
• Can you tell me what the nurse will be looking out for in the clinical examination?
• For each answer ask
◦ Why do you think that is important?
◦ How do you think it helps the nurse know what to do?
Prompt:
• Is there anything else the nurse will look for in the examination? Interviewer to continue prompting participant until there are no further response/things to look out for.
Stage 3: Investigation and referral
The nurse finds out that the patient has had a breast lump for 1 month. On examination, she finds a lump in the right breast and nothing else of significance.
• Can you tell me, how the nurse, who is based at (insert name of clinic), will manage the patient?
• Probes for each response
◦ Why will she do this? i.e. why was that decision made?
• Are there any investigations/tests that the nurse could do or order?
If yes
◦ Could you indicate what test the nurse would order? Why would one do this test?
◦ Who would do test?Where will the test be done?
◦ How does one order this test?
◦ How long does it take for results to be available? – What is the average timeframe for results to become available?How will the results be available? – telephone, electronic, post, return to clinic to get results, etc.
Stage 4: Referral
I am going back to the scenario:
The nurse is concerned that the patient might have breast cancer and decides to refer her for further assessment:
• Can you explain how the referral process works? Including who or where the nurse will refer the patient to. Could you discuss any challenges with referring patients like this? What are the challenges?
• Is there anything that makes it easier to refer patients like this patient? If yes, could you describe what makes it easier to refer patients like this patient?Will the nurse get to know the outcome of the referral? If no, why not and if yes, how do you think the nurse will get to know the outcome of the referral?
• Is there anything else you want to add to this scenario? Could you discuss
We have come to the end of this scenario.
Scenario 2
A 50-year-old woman with a history of diabetes is worried about changes on the skin of her right breast. She comes to this clinic, (insert name of clinic), to have this checked. At the clinic she is seen by a nurse, who must assess and manage her.
Stage 1 – History
• After the nurse introduces herself and makes the patient comfortable, can you tell me what questions she will ask the patient to help with her assessment?
For each question provided, interviewer to probe
◦ Why is it important for the nurse to ask that question?
◦ How does it help her to manage the patient?
Prompt
• Is there any other question the nurse is likely to ask the patient? Interviewer to probe until there are no further questions.
Stage 2 – Clinical examination
I am continuing with the scenario. The nurse asks the woman if she can examine her.
• How would the nurse go about doing the clinical examination?
• What could the nurse be looking out for in the clinical examination?
For each answer ask?
◦ Why is that important?
◦ How does it help the nurse to know what to do?
Prompt:
• Is there anything else that the nurse could look for in the examination? Interviewer to keep asking this until there are no further response/things to look out for.
Stage 3: Investigation and referral
The nurse finds out that the patient had noticed the skin changes on her right breast 1 month ago. On examination, the nurse notices the following change in skin on right breast (show picture of breast skin orange peel appearance). She finds nothing else of significance.
• How will the nurse, who is based at (insert name of clinic), manage the patient?
Probe – for each response
◦ Why will she do this? i.e. why was that decision made?
• Are there any investigations/tests that the nurse could do or order?
If yes
◦ What test?
◦ Who would do test?
◦ Where will the test be done?
◦ How does one order this test?
◦ When will the result be available?
◦ How will the results be available? – telephone, electronic, post, return to clinic, etc.
Stage 4: Referral
In this scenario, the nurse decides to refer the patient for further assessment:
• Can you describe how this referral process works?Who will she refer to?
• Any challenges with referring patients like this? What are the challenges?
• Is there anything that makes it easier to refer patients like this patient?Is the nurse likely to get feedback from the referral centre?
• Will the nurse get to know the outcome of the referral?
• Is there an appointment system for making referrals?
• Is there anything else you want to add to this scenario?
Scenario 3
A 35-year-old woman who is HIV positive comes to the clinic (insert name of clinic) because she has had a yellow vaginal discharge and vaginal bleeding after sexual intercourse for the past month. The woman is seen by the nurse.
Stage 1 – History
• After the nurse introduces herself and makes the patient comfortable, what questions is she likely to ask the woman to help with the assessment?
For each question provided, interviewer to probe
◦ Why do you think it is important for her to ask that question?
◦ How does it help her to manage the patient?
Prompt
• Is there any other question that the nurse is likely to ask the patient? Interviewer to continue probing until there are no further questions.
Stage 2 – Clinical examination
I am continuing with the scenario. The nurse asks the patient if she can examine her.
• How would the nurse go about doing a clinical examination?
• What could she be looking out for in the clinical examination?
For each answer ask?
◦ Why do you think this is important?
◦ How does it help her know what to do?
Prompt:
• Is there anything else that the nurse will look for in the examination? Interviewer to keep asking this until there are no further response/things to look out for.
Stage 3: Investigation
The nurse does a speculum examination and notices a large lesion blocking her view of the cervix. There is blood oozing from the lesion.
• Could you describe how the nurse, who is based at (insert name of clinic), would manage the patient?
Probe
• Why will she do this? i.e. why was that decision made?
• Are there any investigations/tests that she could do or order?
If yes
◦ What kinds of tests might be ordered?
◦ Who will do the test?Where will the test be done?
◦ When would the result be available?
◦ How will it be available? Telephone, electronic, post, etc.
Stage 4: Referral
I am continuing with the scenario. The nurse is concerned that the patient might have cervical cancer and decides to refer her for further management.
• Could you describe how the referral process works?
• Is there some sort of appointment system?
• Who will the nurse refer the patient to?
• How far is the referral centre?
• Are there any challenges with referring patients like this?
If yes
◦ What are the challenges?
◦ Is there anything that makes it easier to refer patients like this patient? If yes, could you describe what these might be?
• Is the nurse likely to get feedback from the referral centre? If yes, how long would this process take?
• Will the nurse get to know the outcome of the referral?
• Is there anything else you want to add to this scenario?
Scenario 4
A 60-year-old woman comes to the clinic (insert name of clinic) because she has vaginal bleeding. She is worried because her periods stopped more than 10 years ago so she cannot understand why she has started bleeding again. The patient is seen by the Nurse.
Stage 1 – History
• After the nurse introduces herself and makes the patient comfortable, what questions will she ask to help with the assessment?
For each question provided, interviewer to probe
◦ Why is it important for her to ask that question?
◦ How does it help her to manage the patient?
Prompt
• Is there any other question the nurse is likely to ask the patient? Interviewer to continue probing this until there are no further questions.
Stage 2 – Clinical examination
I am continuing with the scenario. The nurse asks the patient if she can examine her.
• How would the nurse go about doing the clinical examination?
• What will the nurse be looking out for in the clinical examination?
For each answer ask?
◦ Why is that important?
◦ How does it help her know what to do?
Prompt:
• Is there anything else the nurse will look for in the examination? Interviewer to keep asking this until there are no further response/things to look out for.
Stage 3: Investigation
The nurse cannot see anything of significance on clinical examination.
• How will the nurse, who is based at (insert name of clinic), manage the patient?
Probe
• Why will she do this? i.e. why was that decision made?
• Are there any investigation/tests that she would do or order?
• If yesWhat test?
◦ Who will do the test?
◦ Where will the test be done?
◦ When would the result be available?
◦ How will it be available? Telephone, electronic, post, etc.
Stage 4: Referral
Coming back to the scenario. The nurse does a Pap smear and the result comes back as invasive carcinoma of the cervix.
• How will the nurse inform the patient of the result?
• How will the patient be managed?
• Where will she be referred to for further management?
• How does the referral process work?
• Any there any challenges with referring patients like this?If Yes – What are the challenges?
◦ Is there anything that makes it easier to refer patients like this patient?Is the nurse likely to get feedback from the referral centre?
• Will the nurse get to know the outcome of the referral?
• Is there anything else you want to add to this scenario?
Post-scenario questions
The following questions are to be asked at the end of the interview after the participant has completed two scenarios
1. Are you aware of any policies or guidelines for the management of patients with suspected breast cancer in SA?
If Yes
• Could you tell me the name of the policy/guideline? And date of guideline?
• Can you briefly describe what the policy/guideline says?
2. Are you aware of any policies or guidelines for the management of patients with suspected cervical cancer in SA?
If Yes
• Could you tell me the name of the policy/guideline? Could you tell me the date of the guideline?
• Can you briefly describe what the policy/guideline says?
Thank participant and close interview.
Appendix B. Matrix framework
