Introduction
Cervical cancer, with an incidence rate of 13.3 per 100 000, is the fourth most common cancer among women worldwide.1 In Nepal, the incidence of cervical cancer is 16.4 per 100 000 women, and 11 per 100 000 women lost their lives due to this disease in 2020.2 Elimination of cervical cancer is possible by achieving 80%–100% human papilloma vaccination (HPV) coverage, implementing routine cervical cancer screening and early treatment of abnormalities.3 WHO has launched a global strategy to eliminate cervical cancer by 2030 by reducing the incidence rate to 4 per 100 000 women.4 It adopts a 90-70-90 strategy: this calls for HPV vaccinating at least 90% of girls by the age of 15 years, screening at least 70% of women with high-performance tests by 35–45 years and treating at least 90% of women identified with cervical disease.4
The incidence rate of cervical cancer and its mortality is high in developing countries, including Nepal, because of the lack of proper cancer screening programmes/facilities.5 Therefore, screening services must be available in all health facilities to help rule out cervical cancer in the initial phase. Nepal has institutionalised cervical cancer prevention and screening programmes with the endorsement of cervical cancer screening and prevention (CCSP guideline in 2010), using visual inspection with acetic acid (VIA) as the primary screening method.2 All types of healthcare providers, including auxiliary nurse midwives, nurses, paramedics or physicians, who receive training are certified to provide cervical cancer screening services. Training is provided to local healthcare providers based on the needs of the individual health facilities.6 The guideline adopted the single-visit approach, which refers to screening clients through the VIA method and treating precancerous lesions in the same visit (30 years and above). The guideline was recently updated in 2022, and it includes HPV DNA as the primary screening method. However, the VIA will continue to be a screening method until HPV DNA is fully operational.7
Nepal has set an ambitious target of achieving 90% screening of women of eligible age by 2030.8 However, a recent systematic review showed that less than 1 in 5 (16%) women had used cervical cancer screening services.9 Another nationally representative sample survey conducted in 2019 shows that only 8.2% of women aged 30–49 years have ever been screened for cervical cancer.10 Several factors, such as low level of awareness; limited availability of health services; attitude, practice and competence of health workers; perceived stigma for its association with HPV infection as a sexually transmitted infection;11 and fear of negative evaluation in society have been reported to be the reasons behind the low utilisation of screening services.12 13 In recent years, the cervical cancer screening programme has been gaining priority in the policy and annual work plans. Including screening and vaccination services in the recently endorsed national basic health service package reflects Nepal’s commitment to cervical cancer prevention, early identification and management.14
The Government of Nepal has been expanding basic health services to underserved communities to fulfil its commitment to achieving universal health coverage (UHC).15 Quality of care has been emphasised in UHC initiatives to achieve optimal health outcomes and to improve patient satisfaction.15 16 Although there is no universal definition of quality of care, the Donabedian system’s framework, which conceptualises three dimensions—inputs, process and output—has been the most influential in measuring quality.17 18 Service readiness that measures the provision of inputs is the prerequisite and initial step for improving the quality of care, though service readiness alone doesn’t guarantee the quality of care.19 20 While few existing studies attempt to explore the demand-side constraints of using cervical cancer screening services, there is a lack of nationally representative research that assesses the availability and readiness of these services. Therefore, this study aims to assess the extent to which health facilities are prepared to provide cervical cancer screening services and identify its associated factors. The findings will be helpful for planners and policymakers at all levels to overcome supply side constraints and increase screening coverage. With this in mind, we seek to gather evidence on the available health systems components, such as health infrastructure, institutional procedures, and service readiness, that impact access to and utilisation of cervical cancer screening services at different levels of health facilities. We also hypothesised that management practices, the health facilities’ geographical location and the facilities’ characteristics were associated with screening readiness scores.
Methods
Data source
Our study is based on information obtained from Nepal’s second comprehensive cross-sectional facility survey in 2021, the Nepal Health Facility Survey (NHFS). The NHFS aimed to evaluate the capacity and preparedness of health facilities in Nepal to deliver essential health services. The survey used four validated data collection tools: Facility Inventory Questionnaires, Health Provider Questionnaires, Exit Interview Questionnaires, and Observation Protocols for antenatal care, family planning, and service for sick children. The data regarding the availability and readiness of health facility services were provided by the most knowledgeable person present at the facility on the day of the survey. Our analysis focused on variables from the Facility Inventory and Health Provider Questionnaire adapted from the Demographic and Health Survey (DHS) programme’s worldwide Service Provision Assessment Survey. The details about the survey methodology can be found elsewhere.21
Sample and sampling procedures
We used a nationally representative sample of 1633 health facilities selected through an equal probability systematic sampling method with sample allocation. Stratification was achieved by separating health facilities by facility type within each province. The sample allocation incorporated power allocation among provinces to ensure comparable survey precision. All government hospitals were unequivocally included in the sample due to their relatively low number and significant role in the healthcare system. Likewise, considering their small numbers, all non-government hospitals with at least one bed and those in the Karnali and Sudurpashchim provinces were also included with certainty. Among the non-government hospitals sampled, 54% were included in the sample with certainty, while the remaining 46% were randomly selected.
Furthermore, all primary healthcare centres (PHCCs) and stand-alone HIV testing and counselling centres (HTCs) were included in the sample with certainty. We removed 7 duplicate samples of health facilities and 50 that either declined to participate, were closed on interview days or were unreachable. Thus, NHFS-2021 evaluated 1576 health facilities.
For this study, we concentrated solely on the 338 health facilities (unweighted sample) that provided screening services for cervical cancer. Applying the sampling weights (please see the statistical analysis section below) led to a total analytical sample of 183 health facilities (weighted sample). The details of the sampling procedure are outlined elsewhere.21
Measurement of variables
Outcome variable
The readiness of healthcare facilities to provide cervical cancer screening services was defined as the facility’s capability and willingness to provide cervical cancer screening services. This was determined by observation and interviews on the availability and functionality of key items in three domains—trained staff and guidelines, equipment, and diagnostics—which align with the WHO’s Service Availability and Readiness Assessment (SARA) manual (online supplemental table S1). Trained staff was defined as healthcare providers who received structured inservice training on VIA or single visit approach within the 24 months before the survey. The availability of guidelines was determined by the presence of any relevant guidelines that included information on cervical cancer on the day of the visit. The facility must have had a functioning speculum available on the day of the visit to meet the equipment requirement and the screening tests, including VIA or pap smear tests.
Supplemental material
The outcome variable in our study was the readiness score of cervical cancer services, which is calculated based on the availability of tracer items in each domain outlined in online supplemental table S1. We dichotomised each item to generate the service readiness scores, assigning a code of 1 if it was available and 0 if it was not. Subsequently, we calculated individual domain scores by summing up the indicators within each domain, dividing that sum by the number of indicators in the domain, multiplying the result by 100, and then dividing by the total number of domains in the index. This process yielded the readiness score for each domain.
To illustrate, we combined the scores of the three domains to establish an overall readiness score for cervical cancer services. To ensure a standardised scale of 0%–100%, each of the three domains contributed equally to the total score, that is, a third (33.3%) of the total score. The trained staff and guideline domain had two tracer items; hence, its score was further divided by the number of items (33.3/2 = 16.67). The detailed breakdown of this approach is provided in online supplemental table S1, as recommended in the literature.22 23
Exposure variables
We analysed a range of exposure or independent variables that could potentially be associated with the readiness of healthcare facilities to offer cervical cancer screening services.
Facility type: We divided facility type into four categories: public hospitals, private hospitals, PHCCs, and basic healthcare centres (BHCCs), which comprise health posts, urban health clinics, and community health units.
Location of health facilities: Based on their location, we divided facilities into two categories: urban and rural.
Management structure and practices: The managing authority of the facility was divided into public or private categories. Non-governmental organisation was included in the private category. Quality assurance activities were based on the presence of an official record indicating any quality assurance measures such as reports, meeting minutes, checklists for supervision, mortality reviews, or audits of records. These activities were classified as either conducted or not conducted. Staff management meetings referred to regular meetings focused on managing the staff and were categorised as No or Yes. A management meeting with the health facility management committee was defined as a meeting held with committee members at least once every 6 months, with documented evidence of a recent meeting. The presence of a system to gather client opinions was classified as either absent or present. External supervision within the past 4 months was categorised as either absent or present. The term ‘routine user fee’ indicated whether the facility implemented any regular fees or charges for the services offered to clients, which could include charges for health cards, client registration and other services. Based on the provision of user fees, we grouped facilities into two categories, that is, Yes or No. Facilities with routine user fees or client service charges, including health card charges and client registration, were classified as ‘Yes’.
We included variables based on existing literature, which indicates their influence on service availability and readiness.19 23 24 However, we excluded managing authority (public and private) from the regression analysis since managing authority demonstrated collinearity with facility types.
Statistical analysis
In the descriptive analyses, categorical variables were summarised using frequency, counts and proportions. We presented the percentages of facilities offering cervical cancer screening services along with individual items related to readiness, accompanied by 95% CIs for the proportions. We used the facility weights available in the NHFS data set to ensure that the sampled facilities accurately represented the population. Since the 2021 NHFS sample was stratified, sampling weights were determined based on the sampling probabilities unique to each stratum. The details of sampling procedures and the sampling weights can be seen in the final 2021 NHFS report published elsewhere.21 Further, Stata’s ’svy’ commands in V.17.0 (StataCorp, College Station, Texas, USA) were used to account for the complex sampling design. The readiness score was non-normally distributed and skewed to the left, as shown in online supplemental figure S1. A box plot where the median and third quartiles are equal suggests that most observations are concentrated in the lower half of the distribution. The Breusch-Pagan test statistic for heteroskedasticity revealed a significant difference from 0 (p<0.001) using ordinary least squares (online supplemental table S2). Additionally, we examined the residuals graphically using weighted least squares regression, which still showed non-normality and heteroskedasticity. Thus, to account for the non-normality of residuals, we used quantile regression models for this study at 0.25, 0.5 and 0.75 conditional quantile points.25–27 We conducted model validation and assessed the assumptions of the quantile regression model through residual plots. We examined the linearity of the association between dependent and independent variables, checked for statistical independence of the residuals, ensured constant variance of the errors, and evaluated the normality of the error distribution using graphical methods. Further, we used a postestimation command for quantile regression to test the equality of slope,28 and the slopes were significantly different (p=0.0061).
Supplemental material
Patient and public involvement
None.
Results
Of the 1565 facilities surveyed in Nepal (excluding stand-alone HTCs), 183 (12%) provided cervical cancer screening services. Of the health facilities providing cervical cancer screening services, around 39% were BHCCs, and 38% were private hospitals. Publicly managed authority accounted for 62% of the facilities. By province, less than 5% were in Karnali province and the highest concentration (36%) was in Bagmati province. Most of the facilities (77%) were situated in urban areas. About a third (34%) of the facilities had conducted quality assurance, and almost 70% had regular staff management meetings and external supervision in the last 4 months. More than half (52%) of the facilities offering cervical cancer screening services held management meetings, while 80% had a mechanism for determining client feedback. Additionally, around a fourth (24%) of the facilities providing cervical cancer screening services did not impose routine user fees (table 1).
Cervical cancer screening service readiness (n=183)
Availability of supportive items for cervical cancer screening
Our results show that cervical cancer management services were available in only 22% of the facilities offering cervical cancer screening services (online supplemental table S3). Additionally, while 74% of facilities had the VIA test available, only 26% had pap smear tests, and even fewer had access to loop electrosurgical excision procedure and cryotherapy (9%). Similarly, only 12% of facilities that offer cervical cancer screening services had thermocoagulation, and only 15% had access to colposcopy.
Around a fourth (23%) of the facilities had any guidelines that have content on cervical cancer screening and/or management, and 10% had RH clinical protocols, both of which provide information on cervical cancer screening services. Of the facilities that provided cervical cancer screening services, 13% had at least one VIA-trained staff member, while 3% had at least one staff member trained in the single-visit approach (online supplemental table S3).
Health facility readiness to provide cervical cancer screening services
Among the facilities offering cervical cancer screening services, 24% had at least one healthcare provider trained in VIA or the single-visit approach within the past 24 months. However, only 2% had cervical cancer screening guidelines, such as RH clinical protocol available (data not shown in the figure).
Among the facilities offering cervical cancer screening services, 22% were prepared regarding staff and guidelines, 76% were prepared with equipment and 80% had diagnostic tests available (figure 1). The overall readiness of cervical cancer screening services was 59.1%, with an SD of 23.8% and a median of 66.7%.
Readiness of cervical cancer screening services (overall) and the readiness across three domains (trained staff and guidelines, equipment and diagnostics).
Cervical cancer screening service readiness by facility-related variables
The findings revealed that public hospitals had the highest readiness at 67%, followed by PHCCs at 60%, BHCCs at 59% and private hospitals at 56% (table 1).
No significant differences in readiness were observed among the location categories, with 59% in urban areas compared with 58% in rural areas. The readiness for cervical cancer screening services was higher among facilities that conducted regular/monthly staff management meetings, had a system to determine client opinions, had a routine user fee for services, and performed quality assurance. The readiness was higher in facilities without supervision in the last 4 months, with only marginal differences. The readiness for cervical cancer screening services varied across provinces, with the highest readiness observed in Sudurpashchim province (69%) and the lowest in Karnali (44%) and Madhesh (48%) provinces (table 1).
Factors associated with cervical cancer screening service readiness
Table 2 presents the coefficients derived from quantile regression analyses, which were performed at various points along the distribution of cervical cancer screening readiness (0.25, 0.50 0.75) to assess the correlation between facility-related characteristics and service readiness.
Association between characteristics of health facilities and cervical cancer screening service readiness, using quantile regression (n=183)
Quantile regression estimates indicated a significant and heterogeneous positive association between provinces (specifically Gandaki and Sudurpashchim provinces compared with Koshi province) and service readiness. Facilities in Gandaki province showed a significantly higher readiness score at all three quantiles than those in Koshi province. Facilities in Bagmati and Sudurpashchim provinces had a higher readiness score at 0.50 quantile. On the other hand, facilities in Karnali province had a significantly lower readiness score compared with facilities in Koshi province at 0.75 quintile.
Similarly, facility types and staff management meetings also showed a significant association with readiness score in the 0.75 quantile. BHCCs had a lower readiness score than private hospitals, and facilities that reported a staff management meeting reported having a higher readiness score than facilities not reporting staff management meetings.
Discussion
This study aimed to assess the availability of cervical cancer screening services and the readiness among health facilities offering cervical cancer screening in Nepal. Among 1576 health facilities in Nepal (including public and private), 183 (12%) facilities have cervical cancer screening services available. The overall readiness of the health facility was 59%, with the highest readiness with diagnostic tests (80%) and the lowest with staff and guidelines (22%). The availability of cervical cancer services and tests was consistently less (in less than 50% of the facilities), except for the VIA test (74%).
The current analysis found that just over 1 in 10 health facilities (12%) in Nepal perform cervical cancer screening. In Ethiopia, 30% of healthcare facilities provide cervical cancer screening services,29 three times higher than our findings. Similarly, cervical cancer screening availability and readiness in Tanzania and Libya were 8% and 6%, respectively, lower than in Nepal, indicating that cervical cancer screening services’ availability and readiness were low in other developing countries too.30 Women in many low-income countries face supply side barriers to cervical cancer screening, such as unavailability of services and long waiting times. Additionally, financial constraints, lack of information regarding the benefits of screening, lack of family/partner permission, stigma, misconception, fear of pain and fear of positive results are some of the demand-side barriers.31 32 To mitigate these challenges, the WHO emphasises expanding facilities providing cervical cancer screening services and conducting awareness programmes to enhance the utilisation of screening services.
Although there is no existing national cervical cancer control programme in Nepal, the government has started various initiatives with the collaboration of several international organisations for its prevention and management, and authorities in Nepal have shared the target of achieving 90% coverage by 2030.8 Nonetheless, the current initiatives are insufficient to meet the WHO’s target for cervical cancer elimination, which aims for 90% of women aged 30–49 years to be screened by 2030 and 47% by 2022. According to the 2022 Nepal DHS, only 11% of women in this age group had been screened.33
Current investigation showed that nearly three quarters of the facilities offering cervical cancer screening (74%) have VIA technology. The CCSP guideline 2010 has recommended VIA as the primary method of screening. VIA was recommended by the Health Technology Assessment Report of the Department of Health Research and WHO, which is the most efficient and cost-effective screening technique for low-income and middle-income countries.34 VIA can be done with minimal resources, provides instant results, and can be provided by primary healthcare workers with intensive training and supervision.35 However, the recently endorsed global strategy has recommended HPV DNA as the primary screening method and VIA as the triage test.4 In line with the recommendation, Nepal has recently revised the CCSP guideline in which HPV DNA is included as the primary screening method, but also plans to continue VIA as a screening method until HPV DNA is fully operational. Nepal will continue VIA as the primary screening method or triage test;7 thus, it would be a promising approach to strengthen the readiness of VIA services.
Multivariable quantile regression results indicated lower readiness in BHCCs compared with private hospitals and Karnali province compared with Koshi province at 0.75 quintile. These findings suggest that concerted efforts are needed to enhance service readiness in lower-level health facilities serving most rural populations and in Karnali province. In pursuing UHC, it is imperative that the Government of Nepal and its development partners urgently act to improve the availability of cervical cancer screening services throughout the country, focusing on BHCCs and Karnali province to enhance readiness.
Furthermore, we found that facilities that reported regular staff management meetings had a higher readiness score compared with health facilities not reporting such meetings, at 0.75 quintile. This finding is in line with our earlier work on antenatal and non-communicable disease readiness in Nepal, suggesting that facilities that hold regular meetings are able to review, reflect on, and act on the shortcomings and improve service readiness.36
Programme and policy implications
Choosing an evidence-based technique for screening is one aspect of the screening programme. For the smooth operation of cervical cancer screening services, a functional health system is necessary where trained staff and guidelines, necessary equipment, and diagnostics are readily available. The study shows the limited availability of cervical cancer screening services with critical gaps in human resource capacity and the availability of guidelines. As the country is gearing its efforts towards the elimination of cervical cancer as a public health problem, with only eight training clinical sites providing VIA training and a concentration of these sites in one province, there’s a clear need for more widespread distribution of training facilities across the country. The provision of training to health workers on cervical cancer screening during preservice training and inservice training could effectively improve the availability of trained manpower.37 The current situation also calls for strengthening the training system by adopting new approaches such as training service providers in camp settings, a hybrid method of distance learning and inperson clinical exposure. Increasing the pool of trainers and expanding training sites nationwide is a potential strategy for improving the number of trained manpower. Hence, collaborating with academic institutes and private hospitals to increase the pool of trainers can improve the number of trained manpower. A systematic review showed that regular staff assessment, feedback and provider incentives improved cervical cancer screening uptake.38 These strategies might also improve service readiness through increased staff availability and retention, especially in lower-level health facilities (PHCCs and BHCCs).
Furthermore, regular production, the provision of guidelines with user-friendly instructions, and the provision of training and supportive supervision can increase the availability and use of guidelines and job aids for cervical cancer screening.39 Likewise, the availability of guidelines and equipment can be improved by incorporating the readiness of cervical cancer screening services in minimum service standards (MSS). MSS is a quality improvement initiative that involves periodic assessment of the readiness and performance of health facilities and the development of improvement plans based on the gaps identified.40 MSS is regularly conducted in public hospitals and is being scaled up at the PHCC level and BHCC level.
Health-seeking behaviour among women also plays a great role in determining screening participation. Evidence suggests that implementing culturally sensitive group education by mobilising community health workers may effectively increase screening participation among the Asian population.41 Additionally, the use of invitation letters for screening,41 health interventions2 and the use of multimedia42 to provide education may be some of the alternative approaches shown to be successful elsewhere.
The critical gaps identified by this study have implications from the supply and demand sides of cervical cancer screening services. First, the availability of health facilities with sufficient equipment and staff must be ensured by both policy and health facility levels. Local governments must collaborate with provincial and federal governments and local health facilities and adopt strategies to address the gaps. An important strategy should be staff capacity-building through continuous training and availability of resources (including equipment, supplies and guidelines) to the providers for the administration of services from policy and health facility levels.43 Second, the integration of cervical cancer screening services with chronic disease management has promoted health facility readiness by implementing a team-based approach.44 To ensure the implementation and adherence to service guidelines, a standard operating procedure needs to be developed by health facilities to achieve the set target of providing services.45 Additionally, continuous monitoring of readiness and availability of quality services and the actual delivery of the services is pertinent.43 Assessment of change readiness (education, motivation, opportunities) of clinical staff and strategies to improve their readiness are important for implementing change.46
Strengths and limitations
To the best of our knowledge, this research is the first attempt to assess the readiness of cervical cancer screening services in Nepal by using data from the NHFS. The NHFS is a comprehensive nationwide survey of health facilities in Nepal, and this particular year, it incorporated the service readiness component for cervical cancer services for the first time. The current study focused on analysing this aspect of the survey. Service readiness, which served as the outcome measure, was assessed based on indicators recommended by the WHO’s SARA framework.
Nevertheless, it is crucial to recognise this study’s limitations. Although the survey achieved national representation across Nepal, the findings may not be universally applicable to other low-income and middle-income countries. Additionally, the NHFS is a cross-sectional survey, making it challenging to establish a temporal relationship between service readiness and covariates.
This post was originally published on https://bmjopen.bmj.com