Skip to main content
Download PDF
Download ePub

Results of a cluster randomized trial testing the Systems Analysis and Improvement Approach to increase cervical cancer screening in family planning clinics in Mombasa County, Kenya

Implementation Science volume 18, Article number: 66 (2023) Cite this article

Abstract

Background

Cervical cancer is the leading cause of cancer death in Kenyan women. Integrating cervical cancer screening into family planning (FP) clinics is a promising strategy to improve health for reproductive-aged women. The objective of this cluster randomized trial was to test the efficacy of an implementation strategy, the Systems Analysis and Improvement Approach (SAIA), as a tool to increase cervical cancer screening in FP clinics in Mombasa County, Kenya.

Methods

Twenty FP clinics in Mombasa County were randomized 1:1 to SAIA versus usual procedures. SAIA has five steps: (1) cascade analysis tool to understand the cascade and identify inefficiencies, (2) sequential process flow mapping to identify bottlenecks, (3) develop and implement workflow modifications (micro-interventions) to address identified bottlenecks, (4) assess the micro-intervention in the cascade analysis tool, and (5) repeat the cycle. Prevalence ratios were calculated using Poisson regression with robust standard errors to compare the proportion of visits where women were screened for cervical cancer in SAIA clinics compared to control clinics.

Results

In the primary intent-to-treat analysis in the last quarter of the trial, 2.5% (37/1507) of visits with eligible FP clients at intervention facilities included cervical cancer screening compared to 3.7% (66/1793) in control clinics (prevalence ratio [PR] 0.67, 95% CI 0.45–1.00). When adjusted for having at least one provider trained to perform cervical cancer screening at baseline, there was no significant difference between screening in intervention clinics compared to control clinics (adjusted PR 1.14, 95% CI 0.74–1.75).

Conclusions

The primary analysis did not show an effect on cervical cancer screening. However, the COVID-19 pandemic and a healthcare worker strike likely impacted SAIA’s implementation with significant disruptions in FP care delivery during the trial. While SAIA’s data-informed decision-making and clinic-derived solutions are likely important, future work should directly study the mechanisms through which SAIA operates and the influence of contextual factors on implementation.

Trial registration

ClinicalTrials.gov, NCT03514459. Registered on April 19, 2018.

Peer Review reports

Introduction

The burden of cervical cancer is particularly high in sub-Saharan Africa, where the high prevalence of human immunodeficiency virus (HIV) infection contributes to cervical cancer incidence and severity [1, 2]. Cervical cancer can be prevented with early screening with visual inspection with acetic acid, cytology, and human papillomavirus (HPV) testing. Early treatment of abnormal cells or dysplastic lesions with cryotherapy, thermal ablation, and loop electrosurgical excision procedures can stop progression to cervical cancer [3,4,5,6]. Unfortunately, cervical cancer screening has been suboptimal in sub-Saharan Africa with screening uptake at 12.9% [7]. In Kenya, cervical cancer is the leading cause of cancer death in women, but a survey in 2014 found that only 14% of Kenyan women had ever been screened [8]. The Kenyan Ministry of Health has published guidelines that task family planning (FP) clinics and HIV clinics with providing cervical cancer screening [9].

Family planning clinics are a logical site to offer screening for reproductive-aged women who are presenting for family planning services, as they are sexually active and at risk for HPV infection, the leading cause of cervical cancer [10]. A survey of 70 FP clinics in Mombasa County conducted in 2017–2018 found that only 54% of clinic managers reported that their facility performed cervical cancer screening [11]. Public facilities and clinics with at least one provider trained to perform cervical cancer screening were more likely to report screening. Clinic managers identified the top three challenges to cervical cancer screening as lack of supplies to screen, lack of training, and clients declining the procedure.

Given the implementation gap between the proven effectiveness of cervical cancer screening and the integration of this evidenced-based intervention into routine care, this cluster randomized trial aimed to test whether a specific implementation strategy, the Systems Analysis and Improvement Approach (SAIA), was effective at increasing cervical cancer screening in FP clinics in Mombasa County, Kenya. SAIA is a multicomponent implementation strategy that uses plan-do-study-act cycles with a cascade analysis tool, sequential process flow mapping, and small tests of change (micro-interventions) to improve care cascades [12,13,14,15]. It was hypothesized that clinics randomized to SAIA would have a higher prevalence of cervical cancer screening compared to control clinics that received no intervention.

Methods

Study setting

This study was conducted in family planning clinics in Mombasa County, Kenya. FP clinics provide contraceptive services to male and female clients and are a source of primary care for reproductive aged individuals. At the start of the study, there were approximately 170 FP clinics in the county, including public and private facilities. All FP clinics receive commodities at no cost from the Mombasa County Department of Health Services (DOHS). Private clinics often charge a convenience fee for FP services including cervical cancer screening. All FP clinics are expected to follow local and national guidelines, which recommend that cervical cancer screening should be performed in FP clinics that are tier 2–tier 5 (dispensaries and private clinics, health centers, sub-county hospitals, county referral hospitals) when resources are available [16, 17]. Care in FP clinics is documented in large paper registry books that capture client demographics, commodities dispensed, and HIV counseling and testing services. In 2008, the FP register was updated to capture cervical cancer screening.

Study design

This was a cluster randomized trial of 20 FP clinics in Mombasa County, Kenya, which was randomized 1:1 to an intervention arm implementing SAIA or a control arm following usual procedures.

Participants

Family planning clinics were randomly selected from a representative sample of 70 clinics across all sub-counties that were surveyed as part of a landscape analysis [18]. Because this study was conducted in close cooperation with Mombasa County, any public or private FP clinic was eligible to participate in trial if their clinic managers provided assent for participation. Clinics that were expected to close during the study period and the 12 FP clinics that were involved in a previous SAIA trial were not eligible to participate [19]. After approval from the Mombasa County Department of Health, clinic managers were invited to participate in the cluster randomized trial. All clinic staff and managers were invited to participate in SAIA’s implementation and as participants in implementation of SAIA, making these frontline implementers and managers the “study subjects.” Staff from the research team supported clinic staff by facilitating sequential process flow mapping and then abstracting register data and inputting these data into the cascade analysis tool. The research staff also supported the FP clinic staff in idea generation for micro-interventions during monthly cycles and provided training for cervical cancer screening as one of the micro-interventions. The research team had no direct contact with FP clients. De-identified and aggregated data were collected from FP registers. FP clients were not identifiable and were not considered participants in this trial.

Randomization

Randomization was performed by an independent statistician at the Center for AIDS Research Biometrics Core at the University of Washington. The statistician performing the randomization had no other role in implementing the trial. Restricted randomization was performed based on clinic size and whether any cervical cancer screening was being performed during the initial survey. Of the 70 clinics included in an initial survey in 2018, 20 were randomly selected for inclusion and randomized 1:1 to SAIA versus usual procedures. Twelve FP clinics that participated in a previous SAIA trial designed to increase HIV counseling and testing that were randomized to the SAIA intervention arm were not eligible to participate in this current study [19]. Because FP clinic staff participated in SAIA, blinding was not possible.

Experimental implementation strategy versus control

This study evaluated SAIA, an evidenced-based five-step cycle composed of multiple implementation strategies [12,13,14,15]. The first step uses an Excel-based “cascade analysis” tool (CAT) to quantify the numbers of individuals who complete each cascade step and identify priority steps for improvement [20]. Step 2 involves sequential process flow mapping to identify modifiable bottlenecks in the system [13]. Step 3 develops and implements workflow modifications or small tests of change (micro-interventions) to address a bottleneck identified by front-line clinical providers in step 2. Step 4 assesses the impact of the micro-intervention and recalculates the CAT. Step 5 repeats the cycle [19]. SAIA maps to 13 Expert Recommendations for Implementing Change (ERIC) strategies: external facilitation, organization of provider implementation meetings, provision of ongoing consultation, facilitating relay of clinical data to providers, use of audit and feedback of routine data with healthcare teams, modeling and simulation of change, local needs assessment, local consensus discussions that include assessment of readiness and identification of barriers and facilitators, continuous quality improvement, development of a formal implementation blueprint, cyclical tests of change, and purposefully reexamining the implementation process [21].

After FP clinics were randomized, the clinic staff from SAIA facilities were invited for a full-day SAIA training that launched the trial. This training included facility-level flow mapping, discussion of each clinics’ cascade analysis using data from the CAT, and selection of the first micro-intervention. Cascade analysis and micro-intervention development, testing, and implementation were repeated monthly for 18 months from January 2020 to July 2021. Flow mapping was repeated as needed when clinic flow and processes had substantially changed. While the study staff supported the FP clinic staff in developing micro-interventions, the FP clinic staff were the implementers of their micro-interventions. FP clinics randomized to SAIA documented their monthly SAIA cycles on structured implementation plan posters provided by the research team. The posters included the CAT step targeted, barrier identified, and micro-intervention proposed. At the end of each monthly cycle, the clinic and research staff determined whether the micro-intervention was completed fully, partially, or not at all based on the tasks proposed and whether they were completed. The research staff photographed and transcribed these implementation plans to create a record of each SAIA cycle to categorize and track micro-interventions. Members of the research team (MCE and GW) categorized micro-interventions into categories that focused on the individual and health system levels. Where data were missing or questions arose, the staff examined the implementation plan photos to supplement the record. When transcribing the implementation plans, the study staff made notes about contextual factors for consideration when interpreting the data, such as clinic closure, staff on strike, or no FP clients seen during this interval.

The study staff conducted structured interviews with at least one staff member from each FP clinic in the trial to understand clinic and staff characteristics [11]. Data were captured on paper case report forms (CRFs) and entered into a REDCap database [22].

There was minimal interaction between the study staff and clinics randomized to the control arm. The study staff informed these FP clinics about the trial and its objectives, and the staff from control clinics participated in the structured interviews about clinic and staff characteristics. Outcome data from the FP registers was collected in control clinics on a quarterly basis.

Several data-cleaning steps were undertaken. Database entries were examined, and the FP register image was reviewed to determine whether data were available for missing responses. If so, these were added to the CRF and database. In addition, CRFs used to abstract FP register data were reviewed by a second study team member who independently viewed the digital FP register, verifying that the response on the CRF matched the data on the register image. Any discrepancies were resolved by discussion between data abstracters to arrive at a consensus about original register entries. Following data entry, the study staff performed a question-by-question assessment comparing hard copy CRFs to the digital database to identify and correct key-in errors.

Outcomes

The primary outcome of this trial was the proportion of visits with eligible FP clients who were screened for cervical cancer using any approved method in clinics randomized to SAIA versus control clinics. A client was eligible for cervical cancer screening if they were 21–65 years old and were not current on recommended screening when counseled by FP providers.

Power and sample size determination

Based on past data [11], it was assumed that FP clinics would see an average of 183 clients per 3-month period. It was also assumed that 8% of visits with eligible FP clients would include cervical cancer screening and that SAIA would increase the proportion of visits with cervical cancer screening of eligible clients to 20%. To calculate a sample size for a cluster RCT, the degree of similarity within a cluster (ρ) must be estimated, with literature often citing ρ≤ 0.25 and rarely exceeding 0.5 [23]. Conservatively, = 0.5 was used for power and sample size calculations, which translates to an intra-class correlation coefficient of approximately 0.05. The cluster size variability of FP clinics was calculated as 0.74. With α = 0.05, 10 clusters per arm were expected to provide 90% power to detect a difference of 12% in the prevalence of cervical cancer screening between SAIA clinics and control clinics.

Statistical analysis

Counts/proportions and median (interquartile range) were calculated for baseline characteristics of SAIA clinics and control clinics. Counts/proportions were also used to describe micro-interventions that were fully, partially, or not implemented. Prevalence ratios were calculated using Poisson regression with robust standard errors to compare the effect of SAIA versus control conditions on the proportion of visits where eligible women were screened for cervical cancer during the last 3 months of the trial. The primary analysis followed the intent-to-treat principle. It was hypothesized that multiple cycles of SAIA might be needed to increase cervical cancer screening when some micro-interventions would not work, and this informed the decision to focus on the last 3 months of the trial. The unexpected COVID-19 pandemic initially restricted research activities as research teams were not permitted to visit the clinics. Also, routine procedures in healthcare clinics were disrupted when COVID-19 cases were high in Mombasa County. For these reasons, the study was extended by an additional 6 months beyond the initial 12 months planned for the trial.

Based on a previous analysis examining cervical cancer screening practices in FP clinics in Mombasa County, having at least one provider trained to perform cervical cancer screening was associated with this procedure being performed [11]. Baseline data highlighted differences between SAIA clinics and control clinics at study entry including the proportion of clinics that had at least one provider trained to perform cervical cancer screening. Best practices for cluster randomized trials often include adjustment for baseline differences in trial arms [24]. To control for these baseline differences, prevalence ratios were adjusted for having at least one provider trained to perform cervical cancer screening.

The national healthcare worker go-slow and strike that occurred in quarters three through five affected clinic operations and likely also impacted the primary results of the trial in quarter 6. To get a better sense of the effect of SAIA during the entire period of the trial, a post hoc secondary analysis was conducted using aggregated results from all 18 months of implementation.

Ethical considerations

This trial is registered at ClinicalTrials.gov (NCT03514459) and was approved by the University of Washington Human Subjects Division and the University of Nairobi-Kenyatta National Hospital Ethics and Research Committee. The Mombasa County DOH endorsed this work as a county-supported public health activity, and clinic managers were asked to provide assent for their clinics’ participation. Clinic staff provided verbal assent to participate in structured interviews. The FP register data were fully de-identified, so consent was not required from individual FP clients.

Results

From the 70 clinics in the cross-sectional survey, 20 were randomly selected for participation in the trial. These clinics were randomized 1:1 to the SAIA and control arms (Fig. 1). Of the clinics randomized to SAIA, one was intermittently closed, but all ten were included in the analysis. Eighty percent (n = 8) of SAIA clinics and 50% (n = 5) of control clinics were public facilities, and 70% (n = 7) of SAIA clinics and 40% (n = 4) of control clinics were in urban locations (Table 1). Clinics randomized to SAIA had a median of five (interquartile range [IQR] 4–7) providers compared to four (2–5) providers in control clinics. Thirty percent (n = 3) of SAIA clinics reported that they performed cervical cancer screening at baseline compared to 60% (n = 6) of control clinics. Of the three SAIA clinics that provided cervical cancer screening, 66% (2/3) charged a convenience fee for screening. Among the six control clinics that provided cervical cancer screening at baseline, 33% (2/6) charged a fee. Clinics randomized to SAIA had a median of zero providers (IQR 0–1) trained to perform cervical cancer screening compared to two providers (IQR 1–3) trained in control clinics.

Fig. 1
figure 1

Flow diagram of FP clinics assessed for eligibility, randomized, participated, and included in final analyses

Full size image
Table 1 Baseline characteristics of family planning clinics randomized to SAIA arm and control arm
Full size table

In the last quarter of the trial, 2.5% (37/1507) of visits with eligible FP clients at SAIA facilities included cervical cancer screening compared to 3.7% (66/1793) in control clinics (prevalence ratio [PR] 0.67, 95% confidence interval [CI] 0.45–1.00) (Table 2). When adjusted for having at least one provider trained to perform cervical cancer screening at baseline, there was no significant difference between screening in SAIA facilities compared to control clinics in the last quarter of the trial (adjusted PR 1.14, 95% CI 0.74–1.75). One SAIA clinic closed in month 2 and remained closed for the duration of the trial. Figure 2 depicts the percentage of eligible FP clients who were screened for cervical cancer in SAIA clinics and control clinics across the six quarters of the trial. The timing of the onset of COVID-19 restrictions, the healthcare worker go-slow, and the healthcare worker strike are shown. The absolute numbers of clients screened and the total number of clients seen in each quarter of the trial are presented.

Table 2 Prevalence ratios of visits with eligible family planning clients where cervical cancer screening occurred in SAIA clinics compared to control clinics
Full size table
Fig. 2
figure 2

Percent of visits with eligible FP clients with cervical cancer screening in SAIA clinics compared to control clinics with associated SAIA cycles and SAIA interventions placed

Full size image

In a post hoc secondary analysis including 18 months of the trial, 4.2% (289/6940) of visits with eligible FP clients at SAIA clinics compared to 2.3% (198/8740) of visits in control clinics included cervical cancer screening (PR 1.84, 95% CI 1.54–2.20). This association was stronger after adjustment for baseline provider training for cervical cancer screening (aPR 2.27, 95% CI 1.86–2.77). A post hoc power calculation showed that the secondary analysis had > 99% power to detect this difference in prevalence.

Clinics randomized to SAIA had SAIA micro-interventions placed in 129 of 180 possible months. The micro-interventions were categorized into client-level and clinic-level interventions (Table 3). More than half of micro-interventions involved sensitization of clients (56%, 72), and interventions to mobilize clients occurred in 3% (4) of SAIA cycles. Clinic-level micro-interventions included addressing lack of commodities (26%, 34), training in cervical cancer screening (8%, 10), restructuring of clinic flow (2%, 2), improving documentation (2%, 2), making renovations to facilitate cervical cancer screening (2%, 2), and addressing the cost of cervical cancer screening for clients (1%, 1). Out of the 129 micro-interventions proposed, 44% (57) were implemented fully, 31% (40) were implemented partially, and 25% (32) were not implemented at all. There were no adverse events from participating in this trial.

Table 3 Categorization of micro-interventions in SAIA clinics
Full size table

Discussion

In the primary intent-to-treat analysis, SAIA was not associated with a significant difference in visits where women were screened for cervical cancer in SAIA versus control clinics in the last quarter of the trial. Because the COVID-19 pandemic and healthcare worker strike impacted routine clinic operations in SAIA and control clinics, the pre-specified primary analysis may not be the best reflection of SAIA’s impact in this trial. When results were aggregated across 18 months, there is preliminary evidence that SAIA was associated with a nearly twofold increase in the proportion of visits with cervical cancer screening compared to controls, offering preliminary evidence.

Integrating cervical cancer screening into FP clinics is a promising implementation strategy, particularly in low-and middle-income countries (LMICs), which can lower the burden on clients for additional visits and could increase the efficiency of resources being used [25]. A study conducted in Kenya highlighted that women prefer breast and cervical cancer screening to be integrated into routine care [26]. This is also supported by a framework for cervical cancer elimination in LMICs that supports integrating care for cervical cancer screening into antenatal care, HIV clinics, and FP clinics [27].

While the integration of services is one promising implementation strategy for increasing cervical cancer screening, other implementation strategies have also been evaluated. A scoping review in 2020 mapped different implementation strategies to improve cervical cancer screening in sub-Saharan Africa [28]. Of the 19 studies included in the review, more than half tested education-based interventions targeted to potential cervical cancer screening clients. In general, these interventions were not very effective. Studies testing interventions utilizing community health educators and peer educators had more success. Innovative service delivery models including community health worker outreach and enhanced counseling were also associated with increases in cervical cancer screening. A different systematic review also concluded that while education-based strategies were common, these were not the most effective strategies [29]. This review stressed the need for new implementation strategies. The different context-specific micro-interventions that FP clinics implemented during the trial could be categorized as adapting service delivery within the FP clinic, education of providers and clients, and improved counseling to clients by healthcare providers. These newer implementation strategies may explain some of SAIA’s impact on cervical cancer screening.

To date, SAIA has been tested in other care cascades including preventing-mother-to-child transmission (PMTCT) in Kenya, Mozambique, and Cote d’Ivoire, HIV counseling and testing in FP clinics in Kenya, mental health in Mozambique, and the pediatric HIV cascade in Kenya. When used to improve the PMTCT cascade, SAIA was associated with improvements in antiretroviral coverage and the proportion of HIV-exposed infants screened for HIV in pre-specified sub-group analyses, though these associations were not statistically significant. HIV testing in antenatal clinics was not different in clinics implementing SAIA compared to usual procedures [12]. In FP clinics in Mombasa County, Kenya, SAIA was associated with significantly more HIV counseling and testing compared to usual procedures in a cluster randomized trial of 24 clinics [19]. These results were maintained after transitioning SAIA’s facilitation to Mombasa County Department of Health staff [30]. In the mental healthcare cascade in Mozambique, SAIA was associated with higher odds of on-time follow-up for appointments, adherence to medications, and functional improvement but was not associated with initial mental health diagnosis or medication selection, enrollment in follow-up care, or return to care after initial mental health consultation [31]. When tested to improve the pediatric and adolescent HIV care cascade, SAIA was significantly associated with an increase in ordered viral load tests but was not associated with more HIV testing, linkage to care, or viral load suppression in a pre-post analysis [32]. In these prior implementations of SAIA, results have ranged from improvements in one step of a care cascade to significant improvements in the entire cascade. While endpoint results have been mixed, variability may be explained by the range of barriers and bottlenecks present in different steps in these varied cascades across different countries.

In this present study when SAIA was tested to improve cervical cancer screening, a number of unique factors contributed to the lack of significant effect of the primary analysis. This study occurred concomitantly with the go-slow and strike and the COVID-19 pandemic, both of which unexpectedly and profoundly disrupted healthcare delivery across Mombasa. One SAIA clinic closed in month 2 of the trial. In other clinics, the pandemic restrictions and healthcare worker strike disrupted operations. This may explain why less than half of the proposed micro-interventions were implemented fully. Cervical cancer screening declined sharply in both SAIA and control clinics during the strike, highlighting the impact of healthcare disruptions on necessary preventative care. The primary analysis was scheduled for the end of the trial period, to allow time for SAIA’s effect to have the greatest impact over the 18-month study period, but this coincided with the end of the months-long healthcare worker go-slow and strike. Our experience with this trial and previously with HIV counseling and testing in FP clinics highlighted that SAIA’s effects were observed soon after the intervention and generally persisted over time with the exception of the major contextual impacts detailed previously. In this context, we propose that future SAIA trials should consider including data from the full implementation period in their primary analyses.

This study had several strengths. First, the cluster randomized trial design rigorously tested SAIA in the cervical cancer screening care cascade. Second, this research included a collection of details about specific micro-interventions that clinics proposed and implemented. Granular data on the micro-interventions selected, how well they were implemented, and the effect on cervical cancer screening provided a better understanding of how SAIA worked. Finally, the quality control procedures employed in this study helped to optimize the capture and quality of the programmatic data examined to identify study outcomes.

A few limitations are important to note. These data present the prevalence of visits where cervical cancer screening was performed rather than the proportion of eligible women who were screened. To measure the proportion of eligible women screened, there would need to be a robust system for following individual FP clients over time. This was not possible with the paper registers in use at the time of the study. Despite this limitation, a post hoc secondary analysis of data for 18 months of implementation was able to detect an overall effect of the SAIA implementation strategy on the proportion of visits at which cervical cancer screening was performed. This secondary analysis had > 99% power in a post hoc power calculation. Next, despite randomization, the relatively small number of clinics in the trial contributed to baseline differences between the SAIA arm and the control arms, particularly in provider training for cervical cancer screening. This limitation was mitigated by performing analyses that adjusted for baseline differences [24]. Lastly, this implementation study used routinely collected program data, and clinicians with busy workloads may not have time to document fully in the FP registers. As a result, it can be difficult to know the extent to which improved rates of cervical cancer screening were related to better documentation versus actual increases in the proportion of visits at which screening was performed. In this context, it is notable that few of the micro-interventions in the SAIA arm of the trial focused on improving documentation.

Conclusion

In conclusion, this cluster randomized trial tested SAIA as a tool to improve cervical cancer screening. Although the primary intent-to-treat analysis did not show an effect, data from 18 months of the randomized trial provide preliminary evidence that SAIA could be a useful strategy to improve rates of cervical cancer screening in FP clinics. While SAIA’s data-informed decision-making and clinic-derived solutions that incorporate a variety of implementation strategies may be important, future work should directly evaluate the mechanisms through which SAIA operates and the influence of contextual factors on implementation.

Availability of data and materials

This study was conducted with approval from the Kenyatta National Hospital—University of Nairobi Ethics and Research Committee (KNH-UON ERC), which requires that we release data from Kenyan studies (including de-identified data) only after they have provided their written approval for additional analyses. As such, data for this study will be available from the authors upon request, with written approval for the proposed analysis from the KNH/UON ERC. Their application forms and guidelines can be accessed at https://erc.uonbi.ac.ke/. Please contact the principal of the KNH-UON ERC at principal-cae@uonbi.ac.ke.

References

  1. Denny LA, Franceschi S, de Sanjose S, Heard I, Moscicki AB, Palefsky J. Human papillomavirus, human immunodeficiency virus and immunosuppression. Vaccine. 2012;30(Suppl 5):F168–74. https://doi.org/10.1016/j.vaccine.2012.06.045. PubMed PMID: 23199960.

    Article  CAS  PubMed  Google Scholar 

  2. UNAIDS. The GAP Report. Geneva: Joint United National Programme on HIV/AIDS (UNAIDS); 2014.

    Google Scholar 

  3. International Agency for Research on Cancer. Cervical cancer estimated incidence, mortality and prevalence worldwide in 2012. Geneva: World Health Organization; 2012.

  4. Sudenga SL, Rositch AF, Otieno WA, Smith JS. Knowledge, attitudes, practices, and perceived risk of cervical cancer among Kenyan women: brief report. Int J Gynecol Cancer. 2013;23(5):895–9. https://doi.org/10.1097/IGC.0b013e31828e425c. PubMedPMID:23694983;PMCID:PMC3662490.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Fitzmaurice C, Dicker D, Pain A, Hamavid H, Moradi-Lakeh M, MacIntyre MF, Allen C, Hansen G, Woodbrook R, Wolfe C, Hamadeh RR, Moore A, Werdecker A, Gessner BD, Te Ao B, McMahon B, Karimkhani C, Yu C, Cooke GS, Schwebel DC, Carpenter DO, Pereira DM, Nash D, Kazi DS, De Leo D, Plass D, Ukwaja KN, Thurston GD, Yun Jin K, Simard EP, Mills E, Park EK, Catala-Lopez F, deVeber G, Gotay C, Khan G, Hosgood HD 3rd, Santos IS, Leasher JL, Singh J, Leigh J, Jonas JB, Sanabria J, Beardsley J, Jacobsen KH, Takahashi K, Franklin RC, Ronfani L, Montico M, Naldi L, Tonelli M, Geleijnse J, Petzold M, Shrime MG, Younis M, Yonemoto N, Breitborde N, Yip P, Pourmalek F, Lotufo PA, Esteghamati A, Hankey GJ, Ali R, Lunevicius R, Malekzadeh R, Dellavalle R, Weintraub R, Lucas R, Hay R, Rojas-Rueda D, Westerman R, Sepanlou SG, Nolte S, Patten S, Weichenthal S, Abera SF, Fereshtehnejad SM, Shiue I, Driscoll T, Vasankari T, Alsharif U, Rahimi-Movaghar V, Vlassov VV, Marcenes WS, Mekonnen W, Melaku YA, Yano Y, Artaman A, Campos I, MacLachlan J, Mueller U, Kim D, Trillini M, Eshrati B, Williams HC, Shibuya K, Dandona R, Murthy K, Cowie B, Amare AT, Antonio CA, Castaneda-Orjuela C, van Gool CH, Violante F, Oh IH, Deribe K, Soreide K, Knibbs L, Kereselidze M, Green M, Cardenas R, Roy N, Tillmann T, Li Y, Krueger H, Monasta L, Dey S, Sheikhbahaei S, Hafezi-Nejad N, Kumar GA, Sreeramareddy CT, Dandona L, Wang H, Vollset SE, Mokdad A, Salomon JA, Lozano R, Vos T, Forouzanfar M, Lopez A, Murray C, Naghavi M, Global Burden of Disease Cancer C. The global burden of cancer 2013. JAMA Oncol. 2015;1(4):505–27. https://doi.org/10.1001/jamaoncol.2015.0735. PubMed PMID: 26181261; PMCID: PMC4500822.

    Article  PubMed  Google Scholar 

  6. Control of cancer of the cervix uteri. A WHO meeting. Bull World Health Organ. 1986;64(4):607–18 PubMed PMID: 3490930; PMCID: PMC2490893.

    Google Scholar 

  7. Yimer NB, Mohammed MA, Solomon K, Tadese M, Grutzmacher S, Meikena HK, Alemnew B, Sharew NT, Habtewold TD. Cervical cancer screening uptake in sub-Saharan Africa: a systematic review and meta-analysis. Public Health. 2021;195:105–11. https://doi.org/10.1016/j.puhe.2021.04.014. Epub 20210531. PubMed PMID: 34082174.

    Article  CAS  PubMed  Google Scholar 

  8. Kenya National Bureau of Statistics, Ministry of Health/Kenya, National AIDS Control Council/Kenya, Kenya Medical Research Institute, National Council for Population and Development/Kenya, and ICF International. Kenya Demographic and Health Survey 2014. Rockville: Kenya National Bureau of Statistics, Ministry of Health/Kenya, National AIDS Control Council/Kenya, Kenya Medical Research Institute, National Council for Population and Development/Kenya, and ICF International; 2015.

  9. Ministry of Public Health and Sanitation, Ministry of Medical Services. National Guidelines for Prevention and Management of Cervical, Breast and Prostate Cancer. Nairobi: Ministry of Health Division of Reproductive Health; 2012.

  10. Cohen PA, Jhingran A, Oaknin A, Denny L. Cervical cancer. Lancet. 2019;393(10167):169–82. https://doi.org/10.1016/S0140-6736(18)32470-X. PubMed PMID: 30638582.

    Article  PubMed  Google Scholar 

  11. Eastment MC, Wanje G, Richardson BA, Mwaringa E, Patta S, Sherr K, Barnabas RV, Mandaliya K, Jaoko W, McClelland RS. A cross-sectional study of the prevalence, barriers, and facilitators of cervical cancer screening in family planning clinics in Mombasa County, Kenya. BMC Health Serv Res. 2022;22(1):1577. https://doi.org/10.1186/s12913-022-08984-2. Epub 20221223. PubMed PMID: 36564740; PMCID: PMC9789596.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Rustagi AS, Gimbel S, Nduati R, Cuembelo MeF, Wasserheit JN, Farquhar C, Gloyd S, Sherr K; Team wiftSS. Implementation and Operational research: impact of a systems engineering intervention on PMTCT service delivery in Côte d’Ivoire, Kenya, Mozambique: a cluster randomized trial. J Acquir Immune Defic Syndr. 2016;72(3):e68-76. https://doi.org/10.1097/QAI.0000000000001023. PubMed PMID: 27082507; PMCID: PMC4911259.

  13. Wagner AD, Crocker J, Liu S, Cherutich P, Gimbel S, Fernandes Q, Mugambi M, Ásbjörnsdóttir K, Masyuko S, Wagenaar BH, Nduati R, Sherr K. Making smarter decisions faster: systems engineering to improve the global public health response to HIV. Curr HIV/AIDS Rep. 2019;16(4):279–91. https://doi.org/10.1007/s11904-019-00449-2. PubMedPMID:31197648;PMCID:PMC6635031.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Sherr K, Gimbel S, Rustagi A, Nduati R, Cuembelo F, Farquhar C, Wasserheit J, Gloyd S, With input from the SST. Systems analysis and improvement to optimize pMTCT (SAIA): a cluster randomized trial. Implement Sci. 2014;9:55. https://doi.org/10.1186/1748-5908-9-55. PubMed PMID: 24885976; PMCID: PMC4019370.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Gimbel S, Voss J, Mercer MA, Zierler B, Gloyd S, CoutinhoMde J, Floriano F, CuembeloMde F, Einberg J, Sherr K. The prevention of mother-to-child transmission of HIV cascade analysis tool: supporting health managers to improve facility-level service delivery. BMC Res Notes. 2014;7:743. https://doi.org/10.1186/1756-0500-7-743. PubMedPMID:25335783;PMCID:PMC4216351.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Ministry of Health, Kenya. Kenya National Cancer Screening Guidelines. Nairobi: Ministry of Health; 2018.

  17. Ministry of Health DoFH, Family planning program. National Family Planning Guidelines for Service Providers. 6th ed. Nairobi, Kenya: RMHSU; 2018.

    Google Scholar 

  18. Eastment MC, Wanje G, Richardson BA, Nassir F, Mwaringa E, Barnabas RV, Sherr K, Mandaliya K, Jaoko W, McClelland RS. Performance of family planning clinics in conducting recommended HIV counseling and testing in Mombasa County, Kenya: a cross-sectional study. BMC Health Serv Res. 2019;19(1):665. https://doi.org/10.1186/s12913-019-4519-x. Epub 2019/09/14. PubMed PMID: 31521157; PMCID: PMC6744633.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Eastment MC, Wanje G, Richardson BA, Mwaringa E, Sherr K, Barnabas RV, Perla M, Mandaliya K, Jaoko W, McClelland RS. Results of a cluster randomized trial testing the systems analysis and improvement approach to increase HIV testing in family planning clinics. AIDS. 2022;36(2):225–35. https://doi.org/10.1097/QAD.0000000000003099. PubMedPMID:34628439;PMCID:PMC8702477.

    Article  PubMed  Google Scholar 

  20. Wagner AD, Gimbel S, Ásbjörnsdóttir KH, Cherutich P, Coutinho J, Crocker J, Cruz E, Cuembelo F, Cumbe V, Eastment M, Einberg J, Floriano F, Gaitho D, Guthrie BL, John-Stewart G, Kral AH, Lambdin BH, Liu S, Maina M, Manaca N, Matsuzaki M, Mattox L, Mburu N, McClelland RS, Micek MA, Mocumbi AO, Muanido A, Nduati R, Njuguna IN, Oluoch G, Oyiengo LB, Ronen K, Soi C, Wagenaar BH, Wanje G, Wenger LD, Sherr K. Cascade analysis: an adaptable implementation strategy across HIV and non-HIV delivery platforms. J Acquir Immune Defic Syndr. 2019;82(Suppl 3):S322–31. https://doi.org/10.1097/QAI.0000000000002220. PubMedPMID:31764270;PMCID:PMC6880809.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Gimbel S, Ásbjörnsdóttir K, Banek K, et al. The Systems Analysis and Improvement Approach: specifying core components of an implementation strategy to optimize care cascades in public health. Implement Sci Commun. 2023;4:15. https://doi.org/10.1186/s43058-023-00390-x.

  22. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap) - a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–81.

    Article  PubMed  Google Scholar 

  23. Hayes RJ, Bennett S. Simple sample size calculation for cluster-randomized trials. Int J Epidemiol. 1999;28(2):319–26 PubMed PMID: 10342698.

    Article  CAS  PubMed  Google Scholar 

  24. Nixon RM, Thompson SG. Baseline adjustments for binary data in repeated cross-sectional cluster randomized trials. Stat Med. 2003;22(17):2673–92. https://doi.org/10.1002/sim.1483. PubMed PMID: 12939779.

    Article  CAS  PubMed  Google Scholar 

  25. Denny L, de Sanjose S, Mutebi M, Anderson BO, Kim J, Jeronimo J, Herrero R, Yeates K, Ginsburg O, Sankaranarayanan R. Interventions to close the divide for women with breast and cervical cancer between low-income and middle-income countries and high-income countries. Lancet. 2017;389(10071):861–70. https://doi.org/10.1016/S0140-6736(16)31795-0. Epub 20161101. PubMed PMID: 27814963.

    Article  PubMed  Google Scholar 

  26. Gakunga R, Ali Z, Kinyanjui A, Jones M, Muinga E, Musyoki D, Igobwa M, Atieno M, Subramanian S. Preferences for breast and cervical cancer screening among women and men in Kenya: key considerations for designing implementation strategies to increase screening uptake. J Cancer Educ. 2023. https://doi.org/10.1007/s13187-023-02274-z. Epub 20230220. PubMed PMID: 36808563.

  27. Shin MB, Liu G, Mugo N, Garcia PJ, Rao DW, Bayer CJ, Eckert LO, Pinder LF, Wasserheit JN, Barnabas RV. A framework for cervical cancer elimination in low-and-middle-income countries: a scoping review and roadmap for interventions and research priorities. Front Public Health. 2021;9:670032. https://doi.org/10.3389/fpubh.2021.670032. Epub 20210701. PubMed PMID: 34277540; PMCID: PMC8281011.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Lott BE, Trejo MJ, Baum C, McClelland DJ, Adsul P, Madhivanan P, Carvajal S, Ernst K, Ehiri J. Interventions to increase uptake of cervical screening in sub-Saharan Africa: a scoping review using the integrated behavioral model. BMC Public Health. 2020;20(1):654. https://doi.org/10.1186/s12889-020-08777-4. Epub 20200511. PubMed PMID: 32393218; PMCID: PMC7216595.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Johnson LG, Armstrong A, Joyce CM, Teitelman AM, Buttenheim AM. Implementation strategies to improve cervical cancer prevention in sub-Saharan Africa: a systematic review. Implement Sci. 2018;13(1):28. https://doi.org/10.1186/s13012-018-0718-9. Epub 20180209. PubMed PMID: 29426344; PMCID: PMC5807829.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Long JE, Eastment MC, Wanje G, Richardson BA, Mwaringa E, Mohamed MA, Sherr K, Barnabas RV, Mandaliya K, Jaoko W, McClelland RS. Assessing the sustainability of the Systems Analysis and Improvement Approach to increase HIV testing in family planning clinics in Mombasa, Kenya: results of a cluster randomized trial. Implement Sci. 2022;17(1):70. https://doi.org/10.1186/s13012-022-01242-3. Epub 20221004. PubMed PMID: 36195890; PMCID: PMC9530422.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Fabian KE, Muanido A, Cumbe VFJ, Manaca N, Hicks L, Weiner BJ, Sherr K, Wagenaar BH. Optimizing treatment cascades for mental healthcare in Mozambique: preliminary effectiveness of the Systems Analysis and Improvement Approach for Mental Health (SAIA-MH). Health Policy Plan. 2021;35(10):1354–63. https://doi.org/10.1093/heapol/czaa114. PubMed PMID: 33221835.

    Article  PubMed  Google Scholar 

  32. Wagner AD, Augusto O, Njuguna IN, Gaitho D, Mburu N, Oluoch G, Carimo N, Mwaura P, Cherutich P, Oyiengo L, Gimbel S, John-Stewart GC, Nduati R, Sherr K. Systems analysis and improvement approach to optimize the pediatric and adolescent HIV cascade (SAIA-PEDS): a pilot study. Implement Sci Commun. 2022;3(1):49. https://doi.org/10.1186/s43058-022-00272-8. Epub 20220510. PubMed PMID: 35538591; PMCID: PMC9087970.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

This research was supported by a grant from the National Institutes of Health/National Cancer Institute K08-CA228761. RSM received support for mentoring from K24-HD88229. Research reported in this publication was supported by the University of Washington/Fred Hutch Center for AIDS Research (CFAR), an NIH-funded program under award number P30-AI027757 which is supported by the following NIH Institutes and Centers: NIAID, NCI, NIMH, NIDA, NICHD, NHLBI, NIA, NIGMS, and NIDDK. CFAR supported randomization through the biometrics core and has provided infrastructure support for the Mombasa research site. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author information

Authors and Affiliations

  1. Departments of Medicine, University of Washington, Seattle, WA, 98104, USA

    McKenna C. Eastment, Ruanne V. Barnabas & R. Scott Mcclelland

  2. Global Health, University of Washington, Seattle, WA, USA

    George Wanje, Barbra A. Richardson, Kenneth Sherr, Ruanne V. Barnabas, Kishorchandra Mandaliya & R. Scott Mcclelland

  3. Biostatistics, University of Washington, Seattle, WA, USA

    Barbra A. Richardson

  4. Vaccine and Infectious Disease Division, Fred Hutchinson Cancer Research Center, Seattle, WA, USA

    Barbra A. Richardson

  5. Mombasa County Department of Health, Mombasa, Kenya

    Emily Mwaringa & Shem Patta

  6. Epidemiology, University of Washington, Seattle, WA, USA

    Ruanne V. Barnabas & R. Scott Mcclelland

  7. Department of Medical Microbiology and Immunology, University of Nairobi, Nairobi, Kenya

    Walter Jaoko & R. Scott Mcclelland

Authors
  1. McKenna C. Eastment
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. George Wanje
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Barbra A. Richardson
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Emily Mwaringa
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Shem Patta
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Kenneth Sherr
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Ruanne V. Barnabas
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Kishorchandra Mandaliya
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Walter Jaoko
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. R. Scott Mcclelland
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

MCE has been the primary author of this manuscript with significant editing by RSM. GW, BAR, EM, SP, KS, RB, KM, WJ, and RSM were involved in the project design, study implementation, and data review. All authors have reviewed the manuscript and approved this submission.

Corresponding author

Correspondence to McKenna C. Eastment.

Ethics declarations

Ethics approval and consent to participate

This trial was approved by the University of Washington Human Subjects Division and the University of Nairobi-Kenyatta National Hospital Ethics and Research Committee. The Mombasa County DOH endorsed this work as a county-supported public health activity, and clinic managers were asked to provide assent for their clinics’ participation. The clinic staff provided verbal assent to participate in structured interviews. The FP register data were fully de-identified, so consent was not required from individual FP clients.

Consent for publication

Not applicable.

Competing interests

RSM has received STI testing supplies, provided to the University of Washington, from Hologic Corporation. RVB declares support from the US National Institutes of Health, the Bill and Melinda Gates Foundation, and the World Health Organization. Regeneron Pharmaceuticals covered abstract and manuscript writing costs outside the submitted work. RVB serves on a Gilead Sciences DMC, for which she receives an honorarium. The other authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Eastment, M.C., Wanje, G., Richardson, B.A. et al. Results of a cluster randomized trial testing the Systems Analysis and Improvement Approach to increase cervical cancer screening in family planning clinics in Mombasa County, Kenya. Implementation Sci 18, 66 (2023). https://doi.org/10.1186/s13012-023-01322-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13012-023-01322-y

Implementation Science

ISSN: 1748-5908