Background

Cervical cancer is one of the commonest cancers diagnosed in women and is the fourth leading cause of cancer deaths in women globally [1]. Cervical cancer is the leading cause of cancer-related death in low- and middle-income countries, with the highest regional incidence and mortality occurring in sub-Saharan Africa [1].

In 2020, there was an estimated 604,000 new cases of cervical cancer and 342,000 cervical cancer-related deaths globally, representing age-standardised incidence and mortality rates of 13.3 per 100,000 and 7.3 per 100,000 respectively [1]. The age-standardised incidence rate of cervical cancer in Southern Africa in 2020 was 36.4 per 100,000, three times the global incidence rate, with an age-standardised mortality rate of 20.6 per 100,000 [1].

Malawi is a country with one of the highest incidence and mortality rates due to cervical cancer in the world, registering twice the age-standardised incidence and mortality rates of Southern Africa—age-standardised incidence rate of 67.9 per 100,000 and age-standardised mortality rate of 51.5 per 100,000 [1]. This represents an increasing trend in the occurrence of new cases of cervical cancer in the region [2]. As an AIDS-defining condition, the risk of cervical cancer is significant in the region, owing to the high HIV prevalence irrespective of antiretroviral therapy (ART) uptake [3–5].

Cervical cancer is considered to be one of the most preventable and treatable malignant conditions worldwide, [3] yet efforts to effectively screen and manage this disease fall short. The Malawian screening programme largely relies on the screen-and-treat initiative recommended by the Malawi Ministry of Health (MoH) through Sexual and Reproductive Health Directorate, which utilises visual inspection with acetic acid (VIA) and cryotherapy [6].

The World Health Organisation (WHO) defines cervical cancer elimination as reducing and maintaining the incidence of cervical cancer to less than four in 100,000 women [7]. With this in mind, in May 2018, the WHO issued a call to eliminate cervical cancer as a public health problem, which proposed a new 90-70-90 target for elimination of cervical cancer– 90% HPV vaccination coverage, 70% screening coverage, and 90% of cervical lesions treated by 2030 [7]. The Malawi Cervical Cancer Control Programme (CECAP) outlines a strategic plan to eliminate cervical cancer which proposes that at least 80% of women aged between 25 and 49 years should have been screened for cervical cancer using visual inspection methods for the first time within a 12-month period [8]. This is a higher target that includes a younger age group than the 70% by 35 years prescribed by the WHO in its global strategy of cervical cancer elimination [9].

Despite this strategy, as well as a national roll-out of VIA and screen-and-treat services in Malawi, coverage still remains far below the national target, with a high loss-to-follow-up of previously screened women [6]. By 2015, only 26.5% of women were reported to have screened for cervical cancer, with less than half (43.3%) of those with cryotherapy eligible VIA positivity actually receiving the cryotherapy [6]. An ecological study by Ng’ambi et al in 2023 reported up to 78% cervical cancer screening coverage among HIV positive Malawian women in 2021, an increase of 24% from the 54% reported in 2019 [10]. However, it is important to note that the data utilised in this study were derived from PEPFAR programmatic databases, and thus, the near-target screening uptake observed can be generalised to facilities with PEPFAR support only.

Since the study by Msyamboza et al. in 2016, [6] there has been no updated countrywide estimate of the cervical cancer screening uptake by the general population in Malawi. The Malawi Population-based HIV Impact Assessment (MPHIA) collected data on the self-reported screening practices of a sample of both women living with HIV and HIV negative women representative of the entire country. We thus aimed to estimate the prevalence and spatial distribution of self-reported cervical cancer screening in Malawi from a representative sample of women enumerated in the MPHIA survey, and to understand the factors associated with cervical cancer screening amongst the survey respondents.

Methods

Results

Characteristics of survey participants

Table 1 describes the baseline characteristics of the survey participants included in the analysis. The median age of the participants was 30.0 (IQR 21.0, 42.6), with 47.3% (2,650,403/5,604,578) being within the target age group (25 to 49 years) for cervical cancer screening. The proportion of women who were HIV positive was 10.5% (587,488/5,604,578), of whom 88.8% were on ART (521,529/587,488). The majority of women (82.0%, 4,587,368/5,604,578) reported to have ever been pregnant before, and were mostly married or living together with a partner (88.1%, 3,246,174/5,604,578).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Weighted summaries of the characteristics of survey participants.

https://doi.org/10.1371/journal.pone.0309053.t001

The women surveyed were mostly from rural areas (82.6%, 4,629,194/5,604,578), with more than two thirds of the population coming from the Central West (20.8%), the Southeast (20.6%) and the Southwest (20.5%) zones. The North, Central East, Lilongwe City and Blantyre City only accounted for 11.3%, 15.8%, 5.9% and 5.2% respectively.

The commonest education level was primary school level (64.0%, 3,582,010/5,604,578), with most women reporting to be unemployed (64.6%, 1,248,132/1,932,701). Survey respondents were largely from the middle to highest wealth quintiles (20.7%, 20.8% and 20.4% for the middle, fourth and highest quintiles respectively).

Participants reporting having undergone cervical cancer screening were significantly older than those who had never undergone cervical cancer screening (median age of 36.0 [28.0, 45.0] versus 28.0 [20.0, 41.0], p <0.001). The target age group (25 to 49 years) outlined in the National Cervical Cancer Control Strategy was most represented in those who reported to have screened than in those who had never screened (632,099/925,235 [68%] versus 2,018,304/4,679,343 [43%], p <0.001). Women who reported having undergone cervical cancer screening had a higher proportion of HIV positivity (24.0% versus 7.8%, p <0.001) than in those who had never been screened and had a higher proportion that had ever been pregnant (96.2% versus 79.2%, p <0.001).

Prevalence and geographical distribution of self-reported cervical cancer screening

The overall prevalence of self-reported cervical cancer screening was 16.5% (95% confidence interval (CI) of 15.5 to 18.0%). Among women living with HIV, the prevalence was 37.8% (95% CI 34.8–40.9), compared to 14.0% (95% CI 13.0–15.0) in HIV-negative women.

Fig 1A illustrates the geographical distribution of cervical cancer screening in the country. The highest prevalence of screening was reported in Blantyre City (25.9%, 95% CI 22.9–29.0), followed by the Southwest (24.1%, 95% CI 21.3–26.9) and Lilongwe City (19.6%, 95% CI 18.0–21.3). The Central East zone had the lowest prevalence of self-reported cervical cancer screening in the country (9.4%, 95% CI 7.5–11.2). In all the zones, the majority of the respondents were who reported having undergone screening were from urban regions (22.7% [95% CI 21.4–24.0] versus 15.2% [95% CI 14.0–16.8])–Fig 1C.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Geographical distribution of self-reported cervical cancer screening in Malawi.

(A) Overall distribution of cervical cancer screening by region. (B) Regional distribution of cervical cancer screening stratified by HIV status. (C) Prevalence of screening by residential type.

https://doi.org/10.1371/journal.pone.0309053.g001

For women living with HIV, the prevalence of self-reported screening is highest in Blantyre City (47.6%, 95% CI 40.8–54.3), followed by Lilongwe City (45.8%, 95% CI 35.1–56.5) and the Southwest zone (44.4, 95% CI 38.7–50.0). The lowest reported screening prevalence is observed in the Central East (24.6%, 95% CI 17.0–32.2)–Fig 1B. The screening prevalence distribution for HIV negative women was similar to the overall picture, with the highest prevalence of screening being reported in Blantyre City (21.5%, 95% CI 17.6–25.3), followed by the Southwest (20.1%, 95% CI 17.6–22.6) and Lilongwe City (16.3%, 95% CI 14.7–18.0), and the Central East zone had the lowest prevalence of self-reported cervical cancer screening in the country (8.6%, 95% CI 6.8–10.3).

Fig 2 summarises the prevalence of cervical cancer screening by granulated age group and HIV status. The highest screening prevalence (28.8%) was observed between the ages of 40 to 44 years, followed closely by ages 45 to 49 years at 28.2%. in all age groups, women living with HIV had higher screening prevalence than HIV negative women.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Prevalence of self-reported cervical cancer screening.

(A) Prevalence of self-reported cervical cancer screening by age. (B) Prevalence of self-reported cervical cancer screening by HIV status.

https://doi.org/10.1371/journal.pone.0309053.g002

Factors associated with self-reported cervical cancer screening

Table 2 details the relationships between self-reported cervical cancer screening and the independent variables. Women aged 25 to 49 years had almost three times the odds of reporting having screened for cervical cancer than younger women (adjusted odds ratio (aOR) of 2.86, 95% CI 2.15–3.80)). Women aged 50 and older were 2.90 times (95% CI 2.06–4.10) more likely to report screening than women aged less than 25 years. Women living with HIV were 2.83 times more likely to report screening than HIV negative women (95% CI 2.29–3.50). Women with a higher education level and with the highest wealth quintile were more likely to report having screened compared to the baselines (adjusted odds ratio (aOR) 2.74 [95% CI 1.67–4.52] and aOR 2.86 [95% CI 2.01–4.08] respectively). There was no difference in reported cervical cancer screening between women who reported accessing modern contraceptive methods and those who don’t (aOR 1.03, 95% CI 0.86–1.24, p = 0.716).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Univariable and multivariable logistics regression describing the factors associated with screening for cervical cancer.

https://doi.org/10.1371/journal.pone.0309053.t002

Discussion

This study provides an updated nationwide overview of the uptake of cervical cancer screening in Malawi, and describes the spatial distribution and individual-level characteristics associated with cervical cancer screening. We found a low prevalence of reported screening, with screening being more prevalent in the HIV population and in urban settlements. The study also found the prevalence of cervical cancer screening to be lowest in Central East zone of the country, followed by the Central West zone.

The prevalence of cervical cancer screening by self-report in this study was 16.5% (95% CI 15.5–18.0%). It is difficult to make direct comparisons with previous studies due to the lack of comprehensive information on the coverage and uptake of cervical cancer screening in Malawi. The MPHIA 2016 report only reported the prevalence of self-reported cervical cancer screening for the ages of 30 years to 49 years in HIV positive women, which was estimated to be 18.7% [12]. A study by Msyamboza et al. using data from the CECAP database examined the coverage of cervical cancer screening in women aged 30 to 45 years only, which they estimated to be 26.5% in 2015; an increase from the 9.3% reported in 2011 [6]. The prevalence of screening in this age group in our population shows a prevalence of 26.1%, denoting no real change in the coverage of cervical cancer screening.

Cervical cancer screening was more common in women living with HIV than in HIV-negative women. The prevalence of screening in women living with HIV in this study was more than twice the estimated prevalence in HIV negative women. We postulate that this is due to the deliberate efforts to deliver integrated services to people living with HIV by PEPFAR implementing partners such as Elizabeth Glaser Paediatric AIDS Foundation (EGPAF), Johns Hopkins Program for International Education in Gynaecology and Obstetrics (JHPIEGO), Partners in Hope (PIH), and Lighthouse Trust, and other partners such as Médecins Sans Frontières (MSF), including deliberate screening for tuberculosis, enhanced surveillance of sexually transmitted diseases, and scheduled screening for cervical cancer in women [13]. Women who present for refills of their antiretrovirals are routinely referred to the cervical cancer screening clinics in health facilities. Such enhanced referral mechanisms are lacking in most disease and health programs, with postnatal clinics and family planning clinics being the only platform through which HIV negative women receive recommendations for cervical cancer screening, resulting in the strong association between cervical cancer screening and history of pregnancy and access to modern contraceptive methods identified in this study.

The regional HIV prevalence and the activities of partners in the country also inform the distribution of cervical cancer screening uptake in the country. Currently, the MSF, and through PEPFAR funding, EGPAF and AIDS Health Foundation (AHF), have activities in the south of Malawi. These supplement the routine screening facilities set up by the Malawi MoH and the Lighthouse Trust in central hospitals. The central region is served by Baylor and PIH. The PEPFAR funding targets districts with the highest HIV prevalence. As such, districts in the Central and Northern regions, where HIV prevalence is lowest, [11] receive the least partner support for cervical cancer screening.

Regional inequalities in healthcare access across Malawi also exist, with wealth-based disparities in access to maternal healthcare highest in the central region, followed by the north, and lowest in the south, [14] potentially contributing to the observed disparities between the regions. Additionally, the Southern region has a higher number of facilities providing maternal health care services that the Central and Northern regions, [15] though district-level granular data on the number of facilities providing cervical cancer screening services is lacking. Other factors that affect health access may have also contributed to the observed effects between the regions, including geographical terrain, road infrastructure, poverty levels, and community and cultural dynamics [14, 16, 17]. Thus interventions to improve uptake of cervical cancer screening need to incorporate the complex interplay between all the factors potentially contributing to inequities in care access, taking into considerations of the domains of the comprehensive health access framework [18]. Further, cervical cancer screening uptake was found to be higher in urban regions than in rural areas. Differences in healthcare uptake between rural and urban areas can be attributed to several factors. Rural areas are more likely to be impoverished, are at higher risk of social vulnerabilities, and have greater challenges overcoming geographical barriers [17, 19]. Transportation challenges can further hinder screening uptake. A study by Varela et al. demonstrated that transportation was a significant barrier to accessing healthcare in Malawi, and transportation as a challenge was more prevalent for rural residents [20]. Additionally, urban residents tend to have higher socioeconomic status, better health literacy, and more access to educational campaigns about cervical cancer. Moreover, urban areas often benefit more from government and NGO health initiatives due to logistical ease.

Cervical cancer screening was strongly associated with higher education level and higher wealth quintile. Gerstl et al. described both education and income as factors associated with screening for cervical cancer in two southern districts of Malawi [21]. This is also corroborated by a study by Gowokani Chirwa who investigated socioeconomic disparities in cervical cancer uptake using the MPHIA 2015 to 2016 data [22]. It is likely that women who have attained a higher education level and are in a higher income bracket are more aware of cervical cancer and its preventative measures, and allows for overcoming facility barriers such as distance to health facilities, access to health facilities with adequate human, laboratory and clinical resources to conduct screening, and patient navigation challenges [23, 24].

A key facet to preventing cervical cancer is HPV vaccinations. The National Cervical Cancer Control Strategy outlines a target HPV vaccine coverage in girls aged 9–13 years who have of 90% in its strategic framework [8]. The document also detailed challenges to the programme’s targets after reviewing the 2004 strategic document, which included the shortfalls in scale-up implementation of the HPV vaccination programme, after the promising results of a pilot vaccine implementation programme in two districts [11, 25]. To our knowledge, up to date studies that evaluate the HPV vaccination programme in Malawi are lacking, suggesting a knowledge gap that needs to be addressed. Dorji et al., in their meta-analysis, described the uptake of the HPV vaccine in low-income countries including Malawi. They noted that despite some countries reporting high vaccine uptake during reported studies, there is a drastic drop when funding for the particular campaign expires [26]. By the time the MPHIA study was being conducted, the Malawi vaccine programme was only a year old, thus making it difficult to obtain reliable responses about vaccine uptake in the surveyed population. Nonetheless, there remains a large unmet need for vaccination in the target age group in the country, as anecdotal reports outline frequent stockouts of the vaccine, hindering routine provision.

This paper has limitations. Firstly, measurement of the outcome is based on self-reported information, thus there may be a likelihood of recall bias as well as social desirability bias led by societal expectations which may potentially have inflated the prevalence estimate. The reliability of the reported outcome was also vulnerable to effects of response fatigue, especially considering that the MPHIA survey was a relatively lengthy survey. Additionally, due to the self-reported nature of the screening data, we did not have reliable data on treatment and outcomes of screening and treatment, and hence did not report these. Secondly, we could only use the variables that were available in the MPHIA dataset due to the secondary nature of the study. There could be potential confounding in the associations that has not been explored. Despite these, the study recruited a large nationally representative sample of the women in Malawi, allowing for broad and precise generalisations on the uptake of cervical cancer screening in the country.

In conclusion, the uptake of cervical cancer screening in Malawi remains low and is primarily driven by the efforts to address the HIV/AIDS epidemic. Screening efforts are hence concentrated on high HIV-burden, urban regions. Despite this, the uptake of screening remains far below the national target of 80%. There is a need to identify funding support and implementation approaches to potentiate screening coverage in both HIV positive and HIV negative women in the country. Further studies of a longitudinal nature are required in order to describe the complete care cascade, from the proportion of abnormalities identified on screening, to the numbers of women treated for pre-cancerous cervical lesions, and the short- and long-term outcomes of the screen-and-treat services for cervical cancer in Malawi.

Supporting information