Human immunodeficiency virus (HIV)-positive women have a six-fold increase in risk for developing cervical cancer compared to women who are HIV-negative.¹ In southern Africa, 63.8% of women with cervical cancer are living with HIV, while globally, this figure is only 5.8%.¹ It is estimated that up to 91% of all cases of women living with HIV and cervical cancer live in southern (64%) and eastern Africa (27%).¹

According to the Cancer Report by Statistics SA, between 2008 and 2019² the highest incidence of cervical cancer was found among black African women, a finding that was attributed to the high burden of HIV and human papillomavirus (HPV) infections in this population.³ This co-occurrence is problematic, as HIV exacerbates HPV. Human immunodeficiency virus contributes to the depletion of pro-inflammatory and anti-inflammatory pathways in HIV-positive women’s cervical mucosa, meaning they are less likely to clear an HPV infection. The latter increases their risk for cervical cancer.⁴ However, when detected early through screening programmes, cervical cancer is curable.

Unfortunately, socio-economic inequalities make screening less accessible to lower-income patients in sub-Saharan Africa.^5,6,7

The national guidelines for cervical screening in South Africa⁸ recommend that, upon an HIV-positive diagnosis, all women should receive a PAP smear. Thereafter, this screening should be repeated every 3 years if cytology is negative. However, research conducted in KwaZulu-Natal reported that only 39.2 % of HIV-positive women had ever received a PAP smear.⁹ A similar notion was found in Limpopo province; however, this was a qualitative study and more research is needed.¹⁰ The reasons for failure to adhere to the national guidelines are unclear. International Agency for Research on Cancer concluded that there is sufficient evidence that cervical cancer screening can reduce cervical cancer mortality by 80% or more among screened women,¹¹ making it crucial to understand inadequate cervical cancer screening practices. Therefore, our study aimed to investigate the factors contributing to a delayed diagnosis of cervical cancer in HIV-positive women at a district hospital in Limpopo province, South Africa.

A qualitative study was conducted at Mokopane Hospital. This hospital is situated near Mokopane town and is the referral centre for the Waterberg health district of Limpopo province in South Africa, serving a population of more than one million, predominantly rural Sepedi-speaking people.²

All 20 patients on the database in the gynaecology unit at Mokopane hospital who were HIV-positive and had cervical cancer stage 3 or stage 4 were approached to participate in the study. However, 2 patients were too sick to participate in the interviews, 5 patients died before the interview date because of complications of their disease, and 3 were unavailable for the interviews. The remaining 10 selected patients took part in the research.

Free attitude, non-directive interviews were conducted between June and August 2023. The initial exploratory question was: ‘What do you think are the factors that contributed to you being diagnosed late with cervical cancer?’ Further probing was carried out with three open-ended questions:

Interviews were performedin a private room, where no interruptions were expected. An experienced research assistant conducted the interviews with the researcher present, taking field notes. The interviews were conducted in the language that participants were comfortable in, either English or Sepedi. The average duration of the interviews was 20 min. All interviews were audio-recorded. Demographic information, including the symptoms before their diagnosis of cervical cancer, was collected as an introduction to the interviews.

All the recorded interviews were transcribed verbatim by the research assistant, and the transcriptions were checked by the researcher for correctness. The transcribed interviews were then translated into English by a language expert and rechecked by the researcher. The data were analysed using interpretative content analysis. Familiarisation with the content of the interviews started during the interview, when the researcher was taking field notes. The researcher and the research assistant further familiarised themselves with the data by reading the transcripts and listening to the audiotapes several times. Key ideas and recurrent themes were noticed as they emerged. The thematic content analysis was guided by the five stages of thematic analysis, namely familiarisation, theme identification, indexing, charting, mapping and interpretation.¹² The data were explored for common themes by creating a list of codes after the familiarisation process. These codes were then organised into categories. The final identified themes were then confirmed by using the cut-and-paste manual qualitative analysis method.¹¹ During the interpretation stage, themes were arranged into a schematic model. The model represents the factors influencing late presentation with cervical cancer at the research site and demonstrates the interrelatedness between the themes.

Active attempts were made to ensure the trustworthiness of the study. Credibility, transferability, dependability and confirmability were enhanced with the following steps: prolonged engagement with the data, checking the transcription and emerging themes with the participants, triangulating the findings between the researcher and research assistant, and review by the research supervisor.

All 10 participants were Sepedi-speaking black South Africans from Mokopane, with ages ranging from 39-50-years-old. Most of the participants had secondary education (n = 6), were unemployed (n = 7) and single (n = 7) (Table 1). One participant was in a cohabiting relationship, and 2 participants were married. Participants’ main presenting symptoms were menorrhagia, post-coital bleeding and pain (Table 1).

From the interviews, we identified four main themes that contributed to the delay in diagnosis of cervical cancer in these 10 HIV-positive women, namely: (1) patients’ perceptions and knowledge of the cervical cancer screening process; (2) their fears and beliefs; (3) health systems limitations; and (4) healthcare worker knowledge, attitudes and practices.

Figure 1 represents the themes and their interrelatedness diagrammatically.

Theme 1 encompasses inadequate knowledge about the early warning signs of cervical cancer and the perception that PAP-smears are diagnostic rather than for screening as major barriers to cervical cancer screening. As a participant explained: ‘I thought it was normal because every month I bleed I thought my monthly cycle was changing, […] but I did not realize it was it [cancer]’ (Participant 8).

Theme 2 alluded to how personal beliefs and fears further delay cervical cancer screening. These included being emotionally not ready, procrastinating, as well as fears and embarrassment about doing the procedure. One participant explained:

Another explained how cultural beliefs also influence them:

Theme 3 describes health system limitations that contribute to the delayed diagnosis. Poor follow-up arrangements, inadequate equipment, inconclusive PAP smear results, decentralisation of HIV clinics, and unpleasant experiences at facilities are all barriers to effective cervical cancer screening. As a participant mentioned:

Theme 4 emphasises how health workers’ inadequate knowledge and low index of suspicion influence their screening practices, resulting in delayed diagnoses. The unwillingness of healthcare workers further bars patients from getting PAP smears as required. This is highlighted by the quotation: ‘I came back many times, but they would give pills without doing anything. [it] felt like they are not giving me attention’ (Participant 4).

Our study highlights several factors contributing to the delay in the diagnosis of cervical cancer. Firstly, there is a misunderstanding that PAP smears are a diagnostic test rather than a screening tool necessary for the early identification of cervical cancer. Secondly, personal issues such as fear, anxiety, shyness, embarrassment, procrastination and a belief in traditional healers are barriers to screening practices. Thirdly, health system challenges such as unavailability of diagnostic consumables, broken equipment, competing health programmes, inconclusive results and inadequate treatment facilities exacerbate this. Fourthly, a low index of suspicion and apparent unwillingness of health workers to do cervical screening further delays the process.

The wrong information and perceptions patients had about cervical cancer screening led to delayed diagnosis of cervical cancer in HIV-positive women (theme 1). We found that, while women know about cervical cancer, they believed PAP smears were used to diagnose the condition rather than for early detection, so most did the test only when they were already unwell. Our study highlights those deficiencies in terms of patients’ understanding of the procedure itself led to delayed diagnosis. Some women poorly recognised the symptoms of cervical cancer and thought it was because of their HIV status. Similar factors have been identified elsewhere on the continent. Studies from Ethiopia, Tanzania, Botswana and KwaZulu-Natal confirm that a lack of knowledge about cervical cancer risk factors, early symptoms, and the frequency of screening, as well as the notion that PAP smears are diagnostic, all play a role in delayed diagnoses of cervical cancer.^13,14,15,16

An interesting but concerning finding was that only one participant, with tertiary education, searched for her own information on the internet when she experienced unpleasant symptoms (see Table 2: theme 1.4). She was then able to seek appropriate care, unlike other participants who were less informed. For younger and educated participants, searching for medical information online has become an intrinsic part of modern medicine.¹⁷ It is therefore concerning that only one out of 10 participants in this study was empowered enough to do so.

Several barriers to internet searching of health information in Africa have previously been identified, including a lack of appropriate devices, inadequate internet access, limited skills and understanding.¹⁸ This is a phenomenon that needs more research in our setting.

Our participants highlighted several personal barriers to going for PAP smears (theme 2), including not being ready or being afraid to do the test, procrastination, being embarrassed and shy to undress in front of healthcare workers. Researchers in Botswana and Kenya also found that a lack of privacy and perceived violation of privacy were barriers to early screening for cervical cancer.^12,18 As in our study, patients in KwaZulu-Natal frequently go to traditional healers before consulting at health facilities, leading to delays in diagnosing cervical cancer.¹⁴

Healthcare system deficiencies, such as poor follow-up arrangements and limited resources to do PAP smears, were mentioned as barriers to the diagnosis of cervical cancer (theme 3). These limitations in the healthcare system have been mentioned in the National cervical cancer screening guidelines as well as in studies carried outin Malawi and Botswana.^8,13,20 Negative experiences at health facilities influence participants’ health-seeking behaviour and their confidence in the health system to help them (theme 3). The negative consequences of unpleasant experiences at health facilities were also described in studies performed in Botswana, Malawi and Uganda.^7,13,20 We found that inconclusive results, repeating the tests without adequate explanation, uncertainty expressed by healthcare workers and unpleasant treatment interventions frustrated participants (themes 3.3, 3.5 and 4.1).

Some of our participants felt that when HIV services were decentralised and moved from the hospital to primary care clinics, it led to a reduction in PAP smear screening for women living with HIV (theme 3.4). Studies from Nigeria and Tanzania found the opposite, suggesting that decentralisation allows for more screening.^6,16,21 Adepuju et al. suggested in the Nigerian study that so-called ‘more important’ services, such as maternal health, compete negatively with cervical cancer screening.²¹

Healthcare worker attitudes and their competence were viewed as some of the main reasons for the delayed diagnosis of cervical cancer (theme 4). Our findings indicate that some participants perceived healthcare workers as being unwilling or not interested in doing cervical cancer screening (theme 4.3). Three South African qualitative studies among nurses and patients and a KAP study among Saudi Arabian nurses, echoed this reluctance among health workers, who claimed cultural taboos and limited resources as barriers to cervical cancer screening.^5,10,14,22 These findings are concerning, considering that primary healthcare clinics are the source of information for most patients about appropriate and recommended screening practices.

Four major factors appear to contribute to delays in cervical cancer diagnoses in HIV-positive women in a rural hospital in Limpopo. These included that patients perceived PAP smears to be diagnostic rather than for screening purposes, and that they had limited knowledge about the frequency of PAP smear screening. In addition, patients’ fears and beliefs contributed to delays. Healthcare system limitations also played a role. Finally, health workers’ limited knowledge, low index of suspicion and unwillingness to do PAP smears prevented patients from receiving a timely diagnosis.

Health education on cervical cancer prevention strategies, adequate resources, monitoring of screening practices and the motivation of health workers, coupled with a patient-centred approach, are needed to ensure adequate screening for cervical cancer in HIV-positive women. Further research on empowering patients from rural settings to search for their own health information may be crucial to combat the high burden of cervical cancer in HIV-positive women in southern Africa.