Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=zgha20

Global Health Action

ISSN: 1654-9716 (Print) 1654-9880 (Online) Journal homepage: https://www.tandfonline.com/loi/zgha20

Cervical cancer screening decentralized policyadaptation: an African rural-context-specificsystematic literature review

R. Rahman, M. D. Clark, Z. Collins, F. Traore, E. M. Dioukhane, H. Thiam, Y.Ndiaye, E. L. De Jesus, N. Danfakha, K. E. Peters, T. Komarek, A. M. Linn, P. E.Linn, K. E. Wallner, M. Charles, M. Hasnain, C. E. Peterson & J. A. Dykens

To cite this article: R. Rahman, M. D. Clark, Z. Collins, F. Traore, E. M. Dioukhane, H. Thiam,Y. Ndiaye, E. L. De Jesus, N. Danfakha, K. E. Peters, T. Komarek, A. M. Linn, P. E. Linn, K. E.Wallner, M. Charles, M. Hasnain, C. E. Peterson & J. A. Dykens (2019) Cervical cancer screeningdecentralized policy adaptation: an African rural-context-specific systematic literature review, GlobalHealth Action, 12:1, 1587894, DOI: 10.1080/16549716.2019.1587894

To link to this article: https://doi.org/10.1080/16549716.2019.1587894

© 2019 The Author(s). Published by InformaUK Limited, trading as Taylor & FrancisGroup.

Published online: 02 Apr 2019.

Submit your article to this journal

Article views: 29

View Crossmark data

REVIEW ARTICLE

Cervical cancer screening decentralized policy adaptation: an Africanrural-context-specific systematic literature reviewR. Rahman a, M. D. Clark b,c, Z. Collins d, F. Traoree, E. M. Dioukhanef, H. Thiamg, Y. Ndiayeh,E. L. De Jesus i, N. Danfakhaj, K. E. Peters k, T. Komarekl, A. M. Linn m, P. E. Linn n, K. E. Wallner o,M. Charlesp, M. Hasnainq, C. E. Peterson r and J. A. Dykens s

aUniversity of Toledo College of Medicine and Life Sciences, Toledo, OH, USA; bLibrary of the Health Sciences, University Library,University of Illinois at Chicago, Chicago, IL, USA; cDepartment of Medical Education, College of Medicine, University of Illinois atChicago, Chicago, IL, USA; dDepartment of Family Medicine, Institute for Health Research and Policy, University of Illinois at Chicago(UIC), Chicago, IL, USA; eRegion medical de Kedougou, Bureau de la santé de la reproduction, prevention transmission mere enfant duVIH-SIDA, Kedougou, Senegal; fSenegal Ministry of Health; gRegion medical de Kedougou, Bureau régional de la formation, de lasupervision et de la recherche, Kedougou, Senegal; hDépartement de Recherche, Sénégal Ministère de la Santé et l‘Action Sociale,Dakar, Sénégal; iSchool of Public Health (SPH), University of Illinois at Chicago (UIC), Chicago, IL, USA; jJohn Snow, Inc, Arlington, VA,USA; kSchool of Public Health, Division of Community Health Sciences, Illinois Prevention Research Center, Institute for Health Researchand Policy, University of Illinois at Chicago (UIC), Chicago, IL, USA; lUniformed Services University, Bethesda, MD, USA; mSchool of PublicHealth, Center for Communications Programs, Johns Hopkins University, Baltimore, MD, USA; nLogistics Management Institute, Tysons,VA, USA; oElizabeth Glaser Pediatric AIDS Foundation, Washington, DC, USA; pDepartment of Internal Medicine, University of Illinois atChicago, Chicago, IL, USA; qDepartment of Family Medicine, University of Illinois at Chicago College of Medicine, Chicago, IL, USA;rSchool of Public Health, Division of Epidemiology & Biostatistics, UI Cancer Center, University of Illinois at Chicago (UIC), Chicago, IL,USA; sDepartment of Family Medicine, Center for Global Health, Institute for Health Research and Policy, Cancer Center, University ofIllinois at Chicago (UIC), Chicago, IL, USA

ABSTRACTBackground: Worldwide, nearly 570,000 women are diagnosed with cervical cancereach year, with 85% of new cases in low- and middle-income countries. The African continentis home to 35 of 40 countries with the highest cervical cancer mortality rates. In 2014,a partnership involving a rural region of Senegal, West Africa, was facing cervical cancerscreening service sustainability barriers and began adapting regional-level policy to addressimplementation challenges.Objective: This manuscript reports the findings of a systematic literature review describingthe implementation of decentralized cervical cancer prevention services in Africa, relevant incontext to the Senegal partnership. We report barriers and policy-relevant recommendationsthrough Levesque’s Patient-Centered Access to Healthcare Framework and discuss the impactof this information on the partnership’s approach to shaping Senegal’s regional cervicalcancer screening policy.Methods: The systematic review search strategy comprised two complementary sub-searches. We conducted an initial search identifying 4272 articles, then applied inclusioncriteria, and ultimately 19 studies were included. Data abstraction focused on implementationbarriers categorized with the Levesque framework and by policy relevance.Results: Our findings identified specific demand-side (clients and community) and supply-side (health service-level) barriers to implementation of cervical cancer screening services. Weidentify the most commonly reported demand- and supply-side barriers and summarizesalient policy recommendations discussed within the reviewed literature.Conclusions: Overall, there is a paucity of published literature regarding barriers to and bestpractices in implementation of cervical cancer screening services in rural Africa. Many articlesin this literature review did describe findings with notable policy implications. The Senegalpartnership has consulted this literature when faced with various similar barriers and hasdeveloped two principal initiatives to address contextual challenges. Other initiatives imple-menting cervical cancer visual screening services in decentralized areas may find this con-textual reporting of a literature review helpful as a construct for identifying evidence for thepurpose of guiding ongoing health service policy adaptation.

ARTICLE HISTORYReceived 3 November 2018Accepted 20 February 2019

RESPONSIBLE EDITORJennifer Stewart Williams,Umeå University, Sweden

KEYWORDSImplementation;partnership; global health;visual inspection of thecervix with acetic acid;gynecologic cancer

Background

In 2018, globally 569,847 cervical cancer diagnosesand 311,000 deaths were projected due to this pre-ventable disease [1]. Cervical cancer is the fourthmost common cancer diagnosed among womenworldwide. It has the highest cancer incidence rate

among women in 28 countries and is the most com-mon type of cancer-related mortality among womenin 42 countries, the majority being in sub-SaharanAfrica [1]. While cervical cancer incidence rates aredeclining in high-resource areas, incidence, preva-lence, and mortality rates continue to rise in low-

CONTACT J. A. Dykens jdykens@uic.edu University of Illinois at Chicago, 1919 W. Taylor, Chicago, IL 60612, USA

GLOBAL HEALTH ACTION2019, VOL. 12, 1587894https://doi.org/10.1080/16549716.2019.1587894

© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permitsunrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

and middle-income countries (LMICs) [1].Furthermore, cervical cancer mortality is expectedto increase by 42% to 442,926 deaths in the year2030 [2]. The greatest rise will be in LMICs wherecurrently 85% of incident cervical cancers and 87% ofcervical cancer deaths occur [3,4].

Cervical cancer screening

Various evidence-based cervical cancer screeningtechniques have been developed and tested, and areappropriate for diverse contexts including (a) cytolo-gic screening through Papanicolaou (Pap) smear withfollow-up colposcopy and biopsy to identify earlystage dysplasia and pre-cancers, (b) human papillo-mavirus (HPV) testing through clinician-sampled orself-sampling techniques, and (c) visual inspectionmethods (employing acetic acid and/or Lugol’s solu-tion), which are effective, low-cost approaches appro-priate for low-resource settings. Visual inspectionmethods can complement other screening modalitiesand have been shown to have adequate sensitivity andspecificity to identify later stage pre-cancers. Visualinspection of the cervix with acetic acid (VIA) is themost common screening approach implemented inresource-limited settings in LMICs [5]. VIA is per-formed by applying a vinegar solution to the cervixfollowed by a ‘naked-eye’ visual inspection to identifyprecancerous lesions which are then treated throughfreezing or the loop electrosurgical excision proce-dure (LEEP). The VIA screening test hasa sensitivity of 82.4% (76.3% to 87.3%) anda specificity of 87.4% (77.1% to 93.4%) [6], and hasbeen proven to be safe and cost effective [7–9].However, despite the many screening modalitiesappropriate for low-resource contexts, cervical cancerscreening coverage, globally, remains insufficient.Wide disparities exist between countries, as illu-strated by Austria (effective screening coverageabove 80%) and Ethiopia (less than 1%). This typeof disparity contributes to an average coverage rate of36.9% globally and only 18.5% in less-developedcountries [10]. The contextual consideration of policydevelopment and the adaptation of these policies tochanges over time is critically important for achievingcommunity access to high-quality health servicesglobally [11–15].

Decentralized cervical cancer screening

A further challenge to cervical cancer preventionand control is the ongoing disparity of access tocervical cancer services between urban and ruralsettings within countries [16]. Centralized policyoften dictates how decentralized settings are gov-erned and supported financially, through capacitymanagement (personnel placement and trainings).

Policies would ideally reflect the differences in con-text between areas with various levels of ruralityand development. Evidence is necessary to informthe creation and refinement of these policies.However, the 10/90 gap, a global health disparitydefined as less than 10% of global funding forresearch being spent on health issues that afflictmore than 90% of the population, continues toplague the world’s most marginalized communities[17]. This has considerable implications in sub-Saharan Africa and, in particular, for women’shealth issues [18,19]. Hence, there remainsa paucity of literature describing the implementa-tion, strengthening, and sustainment of cervicalcancer prevention and control programs in Africa.Given that decentralized health systems, especiallyin rural, resource-limited settings, have made someprogress toward sustaining accessible screening ser-vices and reducing disparities in cervical cancerrates, it is imperative to gain a better understandingof the context in which cervical cancer preventionand control programs are implemented effectivelyand sustained.

Cervical cancer prevention and control policyrelevance

Local health service practice guidelines and regio-nal health systems policy ideally address the mostcommon implementation barriers within a givencontext in order to optimize impact as well asensure program sustainability. Factors such asacceptability, adoption, appropriateness, and feasi-bility are critical health service implementationcomponents of ensuring sustainable, high-quality,accessible, person-centered health care serviceswithin a community health system [13]. Healthsystems that employ dynamic thinking, systems-as-cause consideration, context analysis, opera-tional thinking, and loop thinking (i.e. viewingcausality as an ongoing process) will make moreimpactful strides in programmatic and policyplanning [20]. Cervical cancer is also a criticalindicator of larger health system challengesincluding poor access to quality primary healthcare services and the lack of culturally competentcommunication – both factors that disproportio-nately affect low-income women [21]. As such,cervical cancer prevention and control, both atthe programmatic level of a single health careservice delivery structure, and at the elevated glo-bal population-based level, requires comprehen-sive systems thinking [20,22]. The assurance ofquality and the long-term sustainment of newlyimplemented cervical cancer screening programsnecessitate considerable attention to systems-levelfactors beyond workforce capacity building.

2 R. RAHMAN ET AL.

Cervical cancer partnership in Senegal

The above considerations have figured prominentlyin the development and advancement of collectiveaction within a partnership formed in Senegal.Global health partnerships take many forms and caninclude various partners including health ministries,academic institutions, and non-governmental organi-zations (NGOs), and have become the dominantorganizational model, globally, to address manyhealth systems challenges globally [23]. In 2010, theKedougou Medical Region in Southeastern Senegal,Peace Corps Senegal, the Institute of Health andDevelopment at Cheikh Anta Diop University,Dakar, Senegal, and the University of Illinois atChicago (UIC) formed a partnership with the pur-pose of improving community access to quality pri-mary health care services. The cervical cancerincidence rate in Senegal (41.1%) places it 15th inthe world in the age-standardized incidence rate ofcervical cancer [24]. Between 2010 and 2013, thepartnership worked to build capacity acrossKedougou to ensure cervical cancer VIA screeningaccess to an estimated 9041 women in the targetedage group (30 to 50 years old, at the time) [25–27].Despite robust community-level education, screeningrates were suboptimal, with less than 5% of thepopulation being screened in some locations [28].A 2014 study concluded that only 38% of women inthe study area were aware of cervical cancer, high-lighting lack of awareness as a major barrier towomen accessing screening services. Despite effortsto target the full age range of at-risk women, themean age was relatively young (35.7) for the totalsample [28]. Furthermore, the highest risk oldercohort of women were the ones least likely to seekscreening services [28]. Based on our findings it wasevident that a standard, non-context-specific, una-dapted approach to raising cervical cancer preventionawareness in this region was inadequate.

In 2014, we undertook a literature review to iden-tify program evaluation or policy-relevant publica-tions with a focus on settings similar to the contextof this rural Senegal region. The intent was both toimprove uptake of cervical cancer screening in thisregion and to inform the future horizontal scale ofthe VIA screening program to neighboring ruralregions by shaping informed, context-specific regio-nal health policy. Concurrently, the Senegal NationalMinistry of Health and Social Action initiated aneffort to develop a National Cancer Control andPrevention Plan that includes cervical cancer screen-ing implementation in the rural zones as an explicitpriority. This systematic literature review reports onresearch describing the implementation of cervicalcancer prevention services at the decentralized level

in Africa. We analyzed the literature to identify cer-vical cancer health services implementation chal-lenges and proposed solutions in very similarcontexts to that of the Kedougou, Senegal, partner-ship. We organized these identified barriers and pol-icy-relevant recommendations according to thePatient-Centered Access to Healthcare Frameworkproposed by Levesque [29]. The aim of this reviewwas to inform ongoing cervical cancer screeninghealth services implementation in Senegal to enhancethe quality of services offered, increase utilizationamong communities, and ensure the sustainabilityof health services. In this report, we describe thefindings of the systematic review and illustrate howthe Senegal partnership utilized this approach toadapt regional policy in developing next-phase initia-tives as a response to ongoing implementation bar-riers [30]. Sub-searches 1 and 2 were merged withineach database (see Appendix A for the PubMedsearch strategy).

Methods

We conducted a systematic review of the electronicliterature published on cervical cancer screening usingvisual inspection methodology in African nationsaccording to the Preferred Reporting Items forSystematic Reviews and Meta-Analyses (PRISMA)guidelines [31]. PubMed (1950 to 11/30/14), EMBASE(1974 to 12/1/14), CINAHL (1937 to 12/1/14), and ISIWeb of Science (1950 to 12/1/14) were searched.

The search strategy comprised two complementarysub-searches, including controlled vocabulary and key-words. Sub-search 1 intersected three concept sets: (a)screening terms (e.g. mass screenings, early detection ofcancer, preventive health services, primary prevention,testing, HPV testing, smears, gynecological examina-tion, obstetric and gynecological diagnostic techniques,health promotions, early interventions); (b) cervix uteriterms (e.g. cervix, cervix uteri); and (c) African nationsand ‘low-resource settings’ terms (e.g. Africa, Algeria,Malawi, low-resource nations, low-resource regions,underdeveloped countries, poorest countries). Sub-search 2 intersected four concept sets: (d) cervix uteriterms (as above); (e) cancer/pre-cancer terms (e.g. can-cers, malignancies, dysplasias, lesions, abnormalities,adenocarcinomas, precursors, warts, tumors); (f) visualinspection terms (e.g. acetic acid, iodine, Lugol’s, visualinspection, visual evaluation); and (g) cervical vertebraterms together with terms to exclude animals (e.g. NOTneck, cervical atlas, root caries, thyroid, rats, cats, dogs).An English-only filter was applied after the results ofsub-searches 1 and 2 were merged.

Selected studies met all inclusion criteria: (a) thearticle reported a clinical study or outlined clinical

GLOBAL HEALTH ACTION 3

guidelines or policy; (b) visual inspection for cervicalcancer screening was focal; (c) the study was set ata decentralized level (regional, district, village)African LMIC. Excluded studies met at least one ofthe following exclusion criteria: (a) the studyincluded mixed population data (e.g. populationsoutside of an African LMIC); (b) the study wasstrictly epidemiologic and without clinical relevance;and (c) the study described screening tool effective-ness without reporting other clinically relevantfindings.

A total of 4272 citations were retrieved fromthe database searches and 3195 citations remainedafter deduplication. The titles and abstracts ofthese citations were screened for relevance byteams of two collaborators, and 57 full-text articleswere subsequently evaluated according to theinclusion criteria. Selection of studies based onapplication of the inclusion criteria was accom-plished by teams of two collaborators by consen-sus, with the principal investigator making thefinal determination in the event of disagreement.Nineteen studies met the inclusion criteria andwere included in the review. The PRISMA flowdiagram in Figure 1 shows the steps taken forarticle selection.

Data abstraction and synthesis

In order to elucidate common themes from each text,a data abstraction tool in the form of a DataAbstraction Sheet was created by the authors. Thedata abstraction tool was piloted by two reviewersand edited with feedback. A single, final abstractiontool was used for all included articles. This abstractiontool allowed two separate reviewers to independentlyread each article and abstract pertinent data withoutpersonal interpretation. The group members wereallowed to give feedback on the categories and abstrac-tion tool to help refine the Data Abstraction Sheet andcreate a better understanding of the categories.Abstracted categories include: (a) partnership type,(b) level of studied intervention (demand-side or sup-ply-side), (c) barriers to access, and (d) descriptive(purpose, methodology, major findings). The ‘barriers’category in the abstraction tool followed the LevesquePatient-Centered Access to Health Care Framework,specifying specific barriers on both the demand side(clients and community) and the supply side (healthservices) [29]. Demand-side barriers were subcategor-ized as ‘Perceive’, ‘Pay’, ‘Reach’, ‘Seek’, and ‘Engage’,while supply-side barriers were subcategorized into‘Availability and Accommodation’, ‘Approachability’,

# of records identified through

database searching

(n=4272)

# of additional records identified

through other sources

(n=0)

Identification

# of records after duplicates removed

(n=3195)

Screening

# of records screened

(n=3195)

# of records excluded

(n=3138)

Eligibility

# of full-text articles

assessed for eligibility

(n=57)

# of full-text articles

excluded, with reasons

(n=38)

- Lack of main focus on VIA (n=19)

- Non-decentralized setting (n=15)

- Location not in Africa (n=3)

- Pure Epidemiologic study (n=1)

# of studies included in

qualitative synthesis

(n=19)

Included

Figure 1. PRISMA flow diagram.

4 R. RAHMAN ET AL.

‘Affordability’, ‘Acceptability’, and ‘Appropriateness’.The abstraction categories are shown in more detailin Tables 1 and 2.

Based on the number of mentions, the top fourpolicy-relevant demand- and supply-side barriersfrom the literature review are presented. We alsoidentified and summarize policy recommendationscontained within the selected literature relevant tothe rural African context. All articles referring toa specified barrier or policy recommendation arecited for the category, but only select references areincluded to illustrate specific barriers or policyrecommendations. Other papers within the samplemay have included comments of a similar spirit.

Results

The search identified 19 research papers focused onvisual inspection screening for cervical cancer ata decentralized level in an LMIC. The researchdesigns of the included articles are: descriptiveresearch (e.g. observational, case study, or survey;n = 10, 53%), correlation study (eg. case control,cohort, cross-sectional; n = 5, 26%), and semi-experimental design(eg. field experiment, twin study; n = 4, 21%). Thereare no studies that employed experimental design,

and there are no policy papers included. Two (11%)of the papers were published between 2005 and 2009while 17 (89%) were published between 2010 and2015. The studies took place in eight unique Africancountries: Ghana, Malawi, Mozambique, Nigeria,Tanzania, Uganda, Zambia, and Zimbabwe. Nigeriawas the most studied location, being mentioned infive (26%) different papers, while only three (16%) ofthe papers include multiple study locations. Nine(47%) papers focus on countries classified as lowermiddle income by the World Bank, while 10 (53%)countries are classified as low income [32]. Threemain clinical approaches of ‘see and treat’ were usedin the articles: visual inspection with cryotherapy asa single approach (n = 12, 63%), visual inspectiononly with referral for cryotherapy (n = 1, 5%), andvisual inspection only with no referral (n = 6, 32%).See Table 1 for a description of the studies.

All included articles report projects that weremanaged through global health partnerships invol-ving various partner types. Of these, 11 (58%) of thepartnerships included within-country academicinstitutions, 10 (53%) included international aca-demic institutions, and 13 (68%) had involvmentof local (decentralized) government or formal healthsystems (Table 2).

Our findings identified specific barriers on boththe demand side (clients and community) and thesupply side (health services). Using the LevesquePatient-Centered Access to Health Care Framework,we identified 17 (89%) articles that report demand-side barriers. Most of these articles report barriersacross multiple categories specified in the Levesqueframework: ‘Perceive’ is reported in 13 (76%) articles,‘Pay’ in 12 (71%), ‘Reach’ in 11 (65%), ‘Seek’ in ten(59%), and ‘Engage’ in nine (53%) (Table 3).Seventeen (89%) report supply-side barriers, with 14(82%) articles reporting barriers relevant to‘Availability and Accommodation’. ‘Approachability’is discussed in eight (47%) articles reporting supply-side barriers, ‘Affordability’ in six (35%),‘Acceptability’ in three (18%), and ‘Appropriateness’in three (18%) (Table 4).

Demand-side access barriers and policyrecommendations

● Patient awareness, knowledge or educationdeficiency – 12 articles [33–44].

Barriers: Authors note that a lack of understandingof personal risk or insufficient cervical cancer edu-cation [36] may lead to a decreased sense ofurgency to seek cervical cancer screening.A commonly cited example is a woman’s misun-derstanding that she is only at risk if she displayssigns or symptoms or feels sick [35], but wouldotherwise not need to be screened. A lack of

Table 1. Description of studies (n = 19).n %

Type of studyDescriptive Research 10 53Correlation Study 5 26Semi-Experimental Design 4 21Experimental Design 0 0Policy 0 0Years published2010 - 2015 17 892005 - 2009 2 112000 - 2004 0 0<2000 0 0Location of Study - 8 unique African countriesNigeria 5 26Malawi 2 11Mozambique 2 11Tanzania 2 11Zambia 2 11Ghana 1 5Uganda 1 5Zimbabwe 1 5Study Includes Multiple Countries of Interest (at least oneAfrican country)

3 16

Ghana and Thailand 1 5*Uganda and El Salvador 1 5*Uganda, Peru, and Vietnam 1 5*

Country Income Groups According to World BankClassifications 10 53Low income 9 47Lower middle income 0 0Upper middle income 0 0High incomeClinical ApproachVisual Inspection with Cryotherapy as a Single Approach 12 63Visual Inspection Only - with Cryotherapy Upon Later Visit 1 5Visual Inspection Only - No Explicit Link to Cryotherapy 6 32

* Percentage is reflective of the three articles within the ‘Multiplecountries of interest’ category.

GLOBAL HEALTH ACTION 5

awareness about the presence of cervical cancerpreventative services [35] also decreases screeninguptake. Misperceptions about known services canalso lead to unwarranted fears such as assumptionsthat the procedures used to screen or treat cervicalcancer are uncomfortable or that instrument clean-liness is substandard [44]. Additionally, misunder-standings about screening and treatment may leadto stigma [35], myths [44], and misinformation[41], further impacting screening uptake.

Recommendations: Various methods suggestedfor increasing patient awareness, knowledge, andeducation include but are not limited to: (a) coun-seling sessions that incorporate educational videos[36], (b) health educators emphasizing the benefitsof screening (less pain, potential protectionagainst future cancer, and lower rates of mortality)rather than focusing on the sexual cause of thedisease [34], (c) creating specific curricula targetedat men so they can help motivate and supportwomen to increase screening utilization whileimproving male sentiment toward the screening[42], (d) recruiting peer educators who are attimes more personable and accessible thanTa

ble2.

Partnerships

table.

Mango

ma

2006

Sang

hvi

2008

Peters

2010

Fort

2011

Levine

2011

Mwanaham

untu

2011

Quentin

2011

Audet

2012

Moon

2012

White

2012

Adetokun

bi2013

Chigbu

2013

Paul

2013

Perng

2013

Abiodu

n2014

Chigbu

2014

Chigbu

2014

Maseko

2014

Osing

ada

2014

Total

Partnership

Type

Academ

icinstitu

tion-

International

xx

xx

xx

xx

xx

10

Academ

icinstitu

tion-

National

xx

xx

xx

xx

xx

x11

NGO-International

xx

xx

x5

NGO-Local

xx

x3

Governm

ent/Health

system

-International

xx

x3

Governm

ent/Health

system

-Local

xx

xx

xx

xx

xx

xx

x13

Governm

ent/Health

system

-National

xx

xx

xx

xx

x9

Total:

34

23

45

44

52

41

33

11

11

3

Table 3. Articles reporting demand-side barriers to access(n = 17).

Articles Mentions

n % n = 62

Perceive 13 76Health Llteracy – Patient awareness, knowledge,or education deficiency

12

Beliefs related to health and sickness – Pain ordiscomfort during the procedure

4

Trust and expectations – Distrust of screeningmethodology or efficacy or quality

4

Beliefs related to health and sickness – Side effectson sexual performance

2

Beliefs related to health and sickness – Stereotypedrisk

1

Seek 10 59Personal and social values – Permission requiredfrom family/lacking social support

9

Gender – Embarrassment or anxiety in the clinicalsetting

4

Culture – Belief that cervical cancer is a curse byforeigners/witchcraft

2

Reach 11 65Geographic reach – Access challenges (e.g.multiple visits, distance)

9

Occupational flexibility – Do not have time 5Living environment – Drought 1Pay 12 71Income vs poverty – Inability to pay (significantcost to the patient)

12

Engage 9 47Self-management – Preventive care low priority/not primary reason to visit clinic

6

Caregiver support – Communication challenges(encouragement, teaching mode)

4

*Categories in bold font represent the Levesque framework, categories initalic font represent Levesque framework subcategories, and categorieswithout boldface or italics represent the specific barrier mentioned inthe papers.

**(n) = the number of papers reporting in this category. (%) = thepercentage of 17 papers reporting demand-side barriers. (mentions)= the number of mentions in subcategories (note that one paper mayinclude multiple subcategory mentions).

6 R. RAHMAN ET AL.

physicians to answer follow-up questions [41], (e)using media such as the radio to build healthliteracy to help increase awareness and encouragewomen to seek out screenings or attend work-shops in the area [33], and (f) implementingmedia-led education to increase recognition ofservices [36].

● Inability to pay – 12 articles [33–38,40–43,45,46]

Barriers: The economic burden on women and theirfamilies (real or perceived) greatly limits screeninguptake. Examples of this burden include but are notlimited to: the perceived cost of screening or treat-ment [33], travel expenses [37], lost wages because ofmissing work [42], and fear of hidden costs [43].

Recommendations: VIA is recommended in cer-tain contexts over other cervical cancer screeningtests, such as the Pap smear, to keep costs for thepatient low; however, not all patients are able to payfor screening and so it is recommended that thehealth care facilities provide discounted or freescreening [35]. One study noted that only 18% ofwomen screened for free would participate if therewas a fee for screening [46]. Furthermore, if morepatients are screened and treated then economies ofscale can potentially lower costs for future patients

[45]. Cost-effectiveness [39] could be increased bycreating models in clinics that offer screening servicesand treatment services for dysplasia in a single visit(screen and treat).

● Geographic reach access challenges – nine arti-cles [35,38,40,42–47]

Barriers: Access to services is a common problem inrural areas with no nearby treatment centers[35,38,40,42–47], or in rural areas where the infra-structure is too deficient to support the keeping andproper care of the equipment required for compre-hensive referral services [47]. If multiple visits arerequired for screening and treatment, responsibilitiesat home may create barriers to screening access orfollow-up [35,38,40,42–47].

Recommendations: Although the papers in theliterature review do not explicitly study the utilizationof mobile clinics for VIA in rural areas in Africa [37],it is recommended to combat problems of access andthis method should be further explored throughresearch [38].

● Permission required from family/lackingsocial support – nine articles [33–36,38,41–44]

Barriers: Lack of social support from a husband [35],friends [41], or community leaders [42] may dissuadewomen from seeking screening services or from fol-lowing up with medical professionals. In addition,cultural or social barriers [41] may cause fear ofexamination when there is a lack of general accep-tance by the community of the benefits of cervicalcancer screening.

Recommendations: Building relationships with orga-nizations and local institutions can help gain rapportwith influential family members [44], peers, andrespected leaders [42] in the community, which maygarner support for life-saving interventions [33].Gaining trust from the local citizens may help solvelogistical issues of social support [34]. Strict standardsof confidential communication enable trust networks toform between medical professionals and patients and isa keystone to success by helping to eliminate worriespatients may have about others’ opinions in their com-munity. The reassurance of confidentiality will increasewomen’s willingness to seek medical help asa preventative measure rather than a last chance at sur-vival [41]. To create a sustainable health service, women’svoices in the service-delivery planning must be priori-tized, to the point where treatment and screening servicesare considered to be an investment in women.

Supply-side barriers and policy recommendations

● Health workforce shortages or low daily pre-sence of service providers – 11 articles [35,37–41,43,44,47–49]

Principal barrier: The number of trained health careproviders (physicians, nurses, midwives, and

Table 4. Articles reporting supply-side barriers to access(n = 17).

Articles Mentions

n % n = 62

Approachability 8 47Outreach – Lack of outreach 8Acceptability 3 18Gender of provider – Female provider preferredbut unavailable

3

Availability and accommodation 14 82Providers – Health workforce shortages or lowdaily presence of service providers

11

Appointments, modes of service provision – Lack offollow-up

8

Facility characteristics – Lack of supplies orelectricity needed for screening

7

Appointments, modes of service provision – Longwait

4

Providers – Burnout and turnover 4Geographic location and context – Distance orspread in rural populations

3

Providers – Lack of administrative oversight ofscreening program

3

Facility characteristics – Functionality ofequipment

2

Affordability 6 35Direct costs of services and related expenses – Costof screening

5

Indirect costs – Clinical space repairs andrenovations

1

Appropriateness 3 18Technical and interpersonal quality – Lack oftraining programs or advancement

3

*Categories in bold font represent the Levesque framework, categories initalic font represent Levesque framework subcategories, and categorieswithout boldface or italics represent the specific barrier mentioned inthe papers.

**(n) = the number of papers reporting in this category. (%) = thepercentage of 17 papers reporting demand-side barriers. (mentions)= the number of mentions in subcategories (note that one paper mayinclude multiple subcategory mentions).

GLOBAL HEALTH ACTION 7

community health workers) is critically insufficientand does not meet patient demand in many rurallocations.

Recommendations: To combat trained service provi-der shortages, a five-day VIA course for health careproviders was effective at teaching skills in Uganda[49]. Furthermore, the study suggested that high-quality, cost-effective standardization in VIA trainingcurricula can help lower mortality, morbidity, and mis-identification in cervical cancer while increasing thenumber of health care providers to help with shortages[49]. Programs should increase correspondence withother clinics in order to produce further training oppor-tunities for local providers [38], while also creatinga network that can share administrative or clinical ideasand resources to create more effective workspaces [48].

● Lack of outreach – eight articles[33,35,38,39,42–44,50]

Barriers: Without an active effort to advertise andinform the community regarding screening serviceavailability and the necessity for prevention, womenmay underutilize [39] the clinical service, especially ifthey interpret a lack of symptoms as a sign of goodhealth [35]. Community health systems must committo leveraging multiple forms of media, such as focusgroups and radio, to spread the message about screen-ing services. However, given the lack of infrastructurein low-resource settings, media intervention may notbe enough to reach or alert all who need care [33].

Recommendations: Mobile units from governmenthospitals can be sent out to provide screening training forhealth care providers at smaller hospitals [38]. Outreachcan be implemented by health promoters and by womenin the community who have received screening [44].Taking a community-engaged approach, health care pro-viders can meet with leaders of traditional institutions toadvocate for screening while asking community leadersto spread awareness about appointments for screeningand treatment [42]. Health care providers should alsorecommend screening whenever women come into theclinic regardless of their chief complaint, because womenwho do not receive this recommendation are 84% lesslikely to be screened [43]. Nurse-midwives should beutilized to encourage screening because their training inreproductive services helps foster a sense of trust andreduces anxiety in female patients [50]. Clinics could alsohave health care providers give reproductive health talksin villages or rural communities to encourage out-reach [44].

● Lack of follow up – eight articles[34,35,37,42,44,46,47]

Barriers: Ensuring follow-up for routine screening ortreatment of abnormal lesions is challenging if appoint-ments are spaced far apart or if patients are routinelykept waiting. Without a reliable record-keeping system,health care providers may lose track of referrals or fol-low-up appointments [35,48]. Moreover, in low-resource

settings, clinicians are often pulled away for trainings orother administrative responsibilities, further challengingtimeliness of patient follow-up [48]. Follow-up examsmay also be delayed if adequate materials and suppliesfor screenings or treatment are not readily available [35].

Recommendations: To increase follow-up, clinicsshould incorporate electronic patient tracking systems,to help eliminate poor record-keeping and repetitivedata entry on paper copies that may be lost [48].

● Lack of supplies or infrastructure needed forscreening – seven articles [35,37,38,42,44,47,48]

Barriers: The lack of supplies was a recurring identifiedbarrier for initial and follow-up screenings. Pooradministrative oversight for stocking supplies playsa critical role in service quality [48]. Other clinicalsupply shortages that were identified include lack ofanesthesia to perform inspection or treatment [47]and lack of gasoline to travel to rural locations [35].Other common health structural challenges impactingcervical cancer services in resource-poor areas includethe lack of reliable electricity and suboptimal measuresfor the prevention of theft or damage to supplies [48].

Recommendations: Addressing supply-side bar-riers is key to growing a sustainable cervical cancerprevention service. This requires robust administrativeoversight and communication with the central healthsystem for mobilization of support and resources.Health care providers will be able to better addresspatient needs if their own barriers to providing serviceare minimized. Ensuring that local staff is monitoringsupplies and re-ordering when necessary is critical. Inaddition, training local technicians to maintain equip-ment for cryotherapy so it does not malfunction andprevent treatment for long periods of time [48] ensuresprevention and rapid resolution of problems.

Discussion

Given that the intent of this literature review was toinform the implementation and strengthening of cervicalcancer screening services in Kedougou, Senegal, and,ultimately, national-level cervical cancer screening policyfocused on decentralized program implementation, wewere forced to reconcile the fact that the literaturelacks policy-specific papers focused on the decentralizedlow-resource settings within low-income countries inAfrica. The absence of this type of knowledge sharing isproblematic given the critical need to develop sustain-able, efficient solutions in these settings, where contextconsiderations are of utmost importance. This literaturereview thus highlights the need for sharing best practices,especially in the policy arena, to benefit decentralizedsettings in LMICs. According to the World Bank, 45countries in Africa are classified as either low incomeor lower middle income [51]. The review identified cer-vical cancer screening implementation reports in decen-tralized settings in only eight African countries. Thus, at

8 R. RAHMAN ET AL.

the time of the search, 37 low- or lower middle-incomeAfrican countries had no reports that fit our inclusioncriteria. Most of the included papers (89%) were pub-lished more recently (2010 to 2015), showing growingattention to this topic. However, the overall lack ofpublished reports in decentralized low-resource settingswithin low-incomeAfrican countries highlights a need toevaluate the implementation of visual inspection cervicalcancer screening implementation in all settings to facil-itate the sharing of best practices.

Despite no papers from the review having an explicitpolicy focus, the included articles did describe findingswith notable policy implications, in particular cervicalcancer screening implementation barriers. We identifiedan equal number of articles (n = 17, 89%) describingsupply-side barriers to those describing demand-sidebarriers. However, when evaluating the number of bar-rier mentions, we identified nearly twice as manydemand-side barrier mentions (82) as compared to sup-ply-side mentions (45) across all papers. This findingreinforces the perspective that the successful implemen-tation of a visual inspection cervical cancer screeningprogram requires careful consideration of the local con-text into which the health service is being implementedas well as the need for robust engagement of individualsand the community as a whole. There is a current lack ofimplementation science projects reporting best practicesand the barriers to effective cervical cancer preventionand control programs in these community health sys-tems. These are some of the most marginalized commu-nities in the world. Research capacity may be lackingeven where the development of new programs isongoing. Given this disparity, the development andresearch communities should align more robustly tobuild services and evidence where they are most needed.

Using the Levesque framework, we identified aneven spread of reported demand-side barriers acrossthe access categories. Supply-side barrier reports,however, were concentrated within three specificareas: the most reports were from Availability andAccommodation, followed by Approachability, and,finally, Affordability. Community health systems inrural Africa may be able to strengthen their under-performing services or apply new principles toexpand current services to other locations witha focus on the policy-relevant recommendations ofthis review. These salient points address the mostcommon problems regarding the implementation ofcervical cancer visual inspection screening serviceand sustainability in similar contexts.

Policy relevance of findings for the Senegalpartnership

The literature lacks policy-specific papers focused on thedecentralized low-resource settings in Africa. Reflectingthe findings of this literature, since 2014, the Senegal

partnership has encountered many barriers. We haveused the analysis of this literature review to developand contextualize further steps in our implementationplan. As a result, we have developed two key initiatives toaddress many of these contextual challenges. The intent,in upholding the principles of theDynamic SustainabilityFramework [52], is to strengthen the local health systemthrough establishing inherent, iterative processes ofresponsiveness to changing contexts. In so doing, wehope to facilitate appropriate and timely adaptation ofthe local health service policy and, ultimately, ensure thesustainability of the screening program. For this reason,we are reporting the policy relevance of the findings ofthis literature review specifically on the two key strategicimplementation initiatives to illustrate how the findingsof this review have been applied within the specific con-text of our described partnership.

Initiative 1: addressing determinants of uptakeand behavioral intervention

One of the partnership’s most significant challengeshas been increasing screening uptake among thehighest risk women [28]. The estimated participationrate for cervical cancer screening in Senegal (whereincidence peaks between the ages of 45 and 54) isvery low, nationwide – 1.9% for women ages 40 to 49and 0% for women over the age of 50 – because of thelack of accessible screening services [53]. Given thatin Senegal, HPV positivity is increased in older-agewomen (a 2002 study of 1639 women showed that8.6% of women ages 55–59 are high-risk HPV posi-tive) [54], screening in this age range is critical.

Our first principal initiative is, therefore, centeredaround a robust investigation of the determinants ofcervical cancer screening uptake and sustained utilizationin this region. The information from this review guidedthe development and refinement of a context-specificpeer education behavioral intervention to improvescreening uptake. To inform the participatory develop-ment of this program, we have assessed barriers andfacilitators of screening at multiple levels: individuals(women aged 30 to 59), households (family orprincipal social unit of at-risk women), and the commu-nity (immediate village or neighborhood with commonamenities of at-risk women). We hypothesize that a peereducation program (implemented through Care Groups[55,56]) that adapts to changing contexts over time and istargeted at a multi-level audience will result in early,widespread uptake and sustained use of the VIA cervicalcancer screening program. This hypothesis has yet to befully tested. We are also evaluating the intervention’simpact on reducing stigma surrounding cervical cancer.Study findings continue to inform programmatic plan-ning in the Kedougou Region, and the peer educationcurriculum we have developed may serve as a templatefor maximizing early impact of new cervical cancer

GLOBAL HEALTH ACTION 9

screening services implemented in other areas of ruralSenegal. The long-term goal with these data is to informnational-level policy to guide the implementation ofcervical cancer screening programs in other ruralSenegal regions (total rural population of 8.8 million[57]) with no current access to cervical cancer screeningservices.

Initiative 2: addressing workforce attrition andsustained midwifery training

Another major barrier that has shaped the strategy forensuring the long-term sustainability is the considerableattrition rate among midwives in the region. BecauseKedougou is very underdeveloped, with limited infra-structure, health care personnel routinely leave after twoto five years to work in areas closer to the capital city ofSenegal. Of the original 63 health care workers trained bythe end of 2013, only 19 remained in the region at theend of 2015, an attrition rate of 70%. By the end of 2017,an additional 24 midwives were transferred out of theregion. Because midwives do not currently learn VIAscreening skills through their formal training, all newmidwives posted into the region must receive this in-service training. There are often delays, leaving accessi-bility of the screening service often unreliable.

To respond to this considerable challenge, thepartnership’s second principal initiative is to collaboratewith the midwifery training center supported by theSenegal Ministry of Health and Social Action in theneighboring region of Tambacounda. With multiplecomponents of the intervention being informed by thisliterature review, we will establish VIA screening andcryotherapy procedural skills as part of the standard pre-service training curriculum. This will also ensure reliableaccess to high-quality training formidwives in Kedougouand other surrounding regions. All midwives will also betaught a cervical cancer screening health services qualityimprovement process based on EngenderHealth’sCOPE® approach [25–27] and to develop basic skills inimplementation research. Through an ongoing qualityimprovement process at the primary health care facilitylevel and developing research capacity in the region, wehope that health service policy will be informed by thelocal context and will be adaptable as future challengesarise.

Limitations

This literature review is focused on a specific topic (visualinspection cervical cancer screening) within a specificcontext (policy relevance in rural decentralized regionsof Africa). Furthermore, given that partner types, con-texts, and objectives differ considerably across globalhealth partnerships, we have attempted to aggregate thereported barriers across all papers into a policy narrativethat is most relevant to the Senegal–Peace Corps–UIC

partnership’s specific purposes. Therefore, the general-izability of these findings is limited. In addition, theknowledge gained from this literature review was appliedto the partnership strategic planning at the time. We arepublishing this information some time after the occur-rence of this exercise to illustrate how this knowledgehelped inform the direction of our own partnership. Assuch, more recently published literature is not containedwithin this review. However, the implementation scienceimplications within our partnership and context holdsignificant value and, therefore, this shortcoming doesnot lessen the value of the primary lessons gainedthrough this review and application of key findings.This report may provide value, lessons learned, andbest practices for others confronting questions ofongoing barriers and ultimate programmatic sustainabil-ity for implementation of cervical cancer screeninghealth services.

Conclusions

Cervical cancer prevention requires a multi-prongedapproach including both primary and secondary pre-vention through accessible vaccination and screeningprograms [3]. Given that access to vaccination inmany LMICs remains limited by considerable sys-tems-level constraints, comprehensive cervical cancerprevention and control in low-resource settingsrequires a coordinated focus on building accessibleand low-cost screening programs. To achieve thisgoal, advancements in implementation research inthis area and a keen understanding of how to adaptevidence-based solutions to local contexts are critical.The affordable screening technique using visualinspection has demonstrated sound evidence of effec-tiveness for decades. Nonetheless, adoption of thiscervical cancer screening technique into local con-texts has been limited, resulting in continued absenceor limited access in many low-resource settings glob-ally. Through this review, the local context-informedreflection and the careful consideration of thereported lessons learned in similar external contextshave been critical in shaping our partnership’s inter-ventions aimed at addressing challenges to sustain-ability. This contextual literature review has,therefore, proven informative in shaping cervical can-cer screening policy and strategic planning inSenegal. As additional cervical cancer screening tech-nologies are developed in the coming years, these newapproaches may contribute to the refinement of clin-ical algorithms guiding cervical cancer preventionand control programs. The assurance of quality andthe long-term sustainment of newly implementedcervical cancer screening programs necessitate con-siderable attention to systems-level factors beyondthe development of technologies and building healthworkforce capacity. The establishment of high-

10 R. RAHMAN ET AL.

quality, culturally appropriate, effective cervical can-cer screening programs requires adequate supplychains, reliable referral systems, registries, and track-ing systems that ensure timely follow-up, qualityassurance provision, robust person- and family-centered outreach programs, and governance systemsthat facilitate responsiveness to capacity challengessuch as workforce attrition, among many other func-tional requirements.

We are hopeful that others attempting to imple-ment or strengthen cervical cancer visual screeningservices in decentralized areas of Africa will alsorecognize the findings of this review to be helpful asa construct for identifying evidence for potential bar-riers, recommendations for overcoming them, andinforming policy for tailoring their approaches foroptimal cervical cancer screening and control pro-gram development and sustainability. This is impera-tive for addressing a preventable cancer which hasdeadly consequences for women, families, and socie-ties when left unscreened and untreated.

Acknowledgments

The authors would like to acknowledge and thank thefollowing individuals and institutions:

Mariama Touré Diarra, Dr. Abib Ndiaye, Dr. AbdoulAziz Kassé, Dr. Manuel Pina, Dr. Cheikh Senghor,Dr. Evrard Kabou, Landing Sagna, Mamadou Diaw,Chris Hedrick, Vanessa Dickey, MaureenCunningham, Cheryl Faye, Adji Thiaw, Pape Camara,Imane Sene, Dr. John Hickner, Dr. Stevan Weine,Dr. Rithvik Balakrishnan, Tina Schuh, Dr. Paul Rotert,Dr. Tracy Irwin, Dr. Marielle Goyette, Chris Coox,Nicole Aspros, Sarah Mollenkopf, Aaron Persing, TessKomarek, Elle De Jesus, Laurie Ohlstein, EmilyJohnson, Arielle Kempinsky, Lesa Young, CarmenDibaya, Maria Castrillon, Zola Collins, Ethan Quinn,Gracey McGrory, Aaron Macoubray, Sherry Vazhayil,Ashley Prettyman, Hans-Martin Ishida, Brendan Gray,Elizabeth Costello, Emma Murphy, Ellen Hendrix,Cason Kirby, and Catherine Lind, and the Universityof Illinois at Chicago Department of Family Medicineand the Center for Global Health. We are grateful forfunding through the Centers for Disease Control andPrevention and the UIC Department of FamilyMedicine that enabled the partnership to respond toidentified barriers through the development ofa quality improvement program.

Author contributions

All authors contributed to the conception and design of thestudy (RR, MC, FT, EMD, YN, KEP, ELD, MH, JAD) or tothe acquisition, analysis, and interpretation of data (ZRC,HT, ND, TK, AML, PEL, KEW, MC, CEP), and drafted themanuscript (RR, MC, JAD) or revised it critically for con-tent (ZRC, FT, EMD, HT, YN, ELD, ND, KEP, TK, AML,PEL, KEW, MC, MH, CEP). All authors read and approvedthe final manuscript.

Disclosure statement

No potential conflict of interest was reported by theauthors.

Ethics and consent

This paper does not require ethical clearance or consent asit is based entirely on material from secondary publishedsources and the authors’ own insights and experience.

Funding information

This work was supported and funded by the UICDepartment of Family Medicine and the US Centers forDisease Control and Prevention, Prevention ResearchCenters Program [Cooperative agreement:#3U48DP005010-02S2]. The findings and conclusions inthis report are those of the authors and do not necessarilyrepresent the official position of the Centers for DiseaseControl and Prevention.

Paper context

In 2014, a partnership involving a rural region of Senegal wasfacing cervical cancer screening service sustainability barriersand began adapting regional-level policy to address imple-mentation challenges. This literature review empowered thepartnership to develop a strategic plan to address majorstructural challenges including workforce retention and cli-ent uptake, among others. This article reports the literaturereview as well as a narrative describing the evolution of theservice implementation and policy adaptation since 2014.

ORCID

R. Rahman http://orcid.org/0000-0002-0122-1900M. D. Clark http://orcid.org/0000-0003-4015-9596Z. Collins http://orcid.org/0000-0002-6254-8165E. L. De Jesus http://orcid.org/0000-0001-7719-231XK. E. Peters http://orcid.org/0000-0002-8334-1606A. M. Linn http://orcid.org/0000-0001-9720-2378P. E. Linn http://orcid.org/0000-0003-2143-0045K. E. Wallner http://orcid.org/0000-0003-1633-4737C. E. Peterson http://orcid.org/0000-0002-9868-9576J. A. Dykens http://orcid.org/0000-0002-4194-8725

References

[1] Bray F, Ferlay J, Soerjomataram I, et al. Global cancerstatistics 2018: GLOBOCAN estimates of incidenceand mortality worldwide for 36 cancers in 185countries. CA Cancer J Clin. 2018. [Internet].DOI:10.3322/caac.21492

[2] Organization WH, Others. Projections of mortalityand causes of death, 2015 and 2030. Geneva,Switzerland: The World Health Organization; 2015.

[3] Randall TC, Ghebre R. Challenges in prevention andcare delivery for women with cervical cancer inSub-Saharan Africa. Front Oncol. 2016;6:160.

[4] Chuang LT, Temin S, Camacho R, et al. Management andcare of women with invasive cervical cancer: Americansociety of clinical oncology resource-stratified clinicalpractice guideline. J Glob Oncol. 2016;2:311–340.

GLOBAL HEALTH ACTION 11

[5] Sankaranarayanan R, Nessa A, Esmy PO, et al. Visualinspection methods for cervical cancer prevention.Best Pract Res Clin Obstet Gynaecol. 2012;26:221–232.

[6] Fokom-Domgue J, Combescure C, Fokom-Defo V,et al. Performance of alternative strategies for primarycervical cancer screening in sub-Saharan Africa: sys-tematic review and meta-analysis of diagnostic testaccuracy studies. BMJ. 2015;351:h3084.

[7] Blumenthal PD, Gaffikin L, Deganus S, et al. Cervicalcancer prevention: safety, acceptability, and feasibility ofa single-visit approach in Accra, Ghana. Am J ObstetGynecol. 2007;407:407.e1–e8,discussion 407.e8–e9.

[8] Goldie SJ, Gaffikin L, Goldhaber-Fiebert JD, et al.Cost-effectiveness of cervical-cancer screening in fivedeveloping countries. N Engl J Med.2005;353:2158–2168.

[9] Goldie SJ, Kuhn L, Denny L, et al. Policy analysis ofcervical cancer screening strategies in low-resourcesettings: clinical benefits and cost-effectiveness.JAMA. 2001;285:3107–3115.

[10] Gakidou E, Nordhagen S, Obermeyer Z. Coverage ofcervical cancer screening in 57 countries: low averagelevels and large inequalities. PLoS Med. 2008;5:e132.

[11] Mitchell SA, Chambers DA. Leveraging implementa-tion science to improve cancer care delivery andpatient outcomes. J Oncol Pract. 2017;13:523–529.

[12] Tabak RG, Khoong EC, Chambers DA, et al. Bridgingresearch and practice: models for dissemination andimplementation research. Am J Prev Med.2012;43:337–350.

[13] Peters DH, Tran NT, Adam T, et al. Implementationresearch in health: a practical guide. Geneva,Switzerland: World Health Organization; 2013.

[14] Peters DH, Bhuiya A, Ghaffar A. Engaging stake-holders in implementation research: lessons from thefuture health systems research programme experience.Health Res Policy Syst. 2017;15:104.

[15] Brownson RC, Colditz GA, Proctor EK.Dissemination and implementation research in health:translating science to practice. New York, NY: OxfordUniversity Press; 2017.

[16] Denny L, Quinn M, Sankaranarayanan R. Chapter 8:screening for cervical cancer in developing countries.Vaccine. 2006;24:S3/71–77.

[17] Commission on health research for development.Health research: essential link to equity in develop-ment. New York, NY: Oxford University Press; 1990.

[18] Kilama WL. The 10/90 gap in sub-Saharan Africa:resolving inequities in health research. Acta Trop.2009;112:S8–S15.

[19] Doyal L. Gender and the 10/90 gap in health research.Bull World Health Organ. 2004;82:162–162.

[20] De Savigny D, Adam T, Alliance for Health Policy andSystems Research, et al. Systems thinking for healthsystems strengthening. Geneva: Alliance for HealthPolicy and Systems Research: World HealthOrganization; 2009.

[21] Freeman HP, Chu KC. Determinants of cancer dispa-rities: barriers to cancer screening, diagnosis, andtreatment. Surg Oncol Clin N Am. 2005;14:10–1016.

[22] Suba EJ, Murphy SK, Donnelly AD, et al. Systemsanalysis of real-world obstacles to successful cervicalcancer prevention in developing countries. AmJ Public Health. 2006;96:480–487.

[23] Bill, Foundation MG, McKinsey. Global health part-nerships: assessing country consequences. FlorhamPark, NJ: McKinsey & Company Florham Park; 2005.

[24] Ferlay J, Soerjomataram I, Dikshit R, et al. Cancerincidence and mortality worldwide: sources, methodsand major patterns in GLOBOCAN 2012.Int J Cancer. 2015;136:E359–E386.

[25] Huezo C, Diaz S. Quality of care in family planning:clients’ rights and providers’ needs. Adv Contracept.1993;2:129–139.

[26] EngenderHealth. COPE handbook: a process forimproving quality in health services ® revised edition.New York, NY: EngenderHealth; 2003.

[27] EngenderHealth. COPE for cervical cancer preventionservices: a toolbook to accompany the COPE hand-book ® engenderhealth’s quality improvement series.EngenderHealth. [Internet]. 2004. Available from:http://screening.iarc.fr/doc/cope_cxca_toolbook.pdf

[28] Dykens JA, Linn AM, Irwin T, et al. Implementingvisual cervical cancer screening in Senegal: across-sectional study of risk factors and prevalencehighlighting service utilization barriers.Int J Womens Health. 2017;9:59–67.

[29] Levesque J-F, Harris MF, Russell G. Patient-centredaccess to health care: conceptualising access at theinterface of health systems and populations.Int J Equity Health. 2013;12:18.

[30] Chambers DA, Norton WE. The adaptome: advancingthe science of intervention adaptation. Am J PrevMed. 2016;51:S124–S131.

[31] Moher D, Liberati A, Tetzlaff J, et al. Preferred report-ing items for systematic reviews and meta-analyses:the PRISMA statement. Ann Int Med. 2009;151:264–269, W64.

[32] World bank country and lending groups [Internet].World Bank Data Help Desk. [cited 2017 Jul 18].Available from: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups

[33] Perng P, Perng W, Ngoma T, et al. Promoters of andbarriers to cervical cancer screening in a rural settingin Tanzania. Int J Gynaecol Obstet. 2013;123:221–225.

[34] Audet CM, Silva Matos C, Blevins M, et al.Acceptability of cervical cancer screening in ruralMozambique. Health Educ Res. 2012;27:544–551.

[35] Mangoma JF, Chirenje MZ, Chimbari MJ, et al. Anassessment of rural women’s knowledge, constraints andperceptions on cervical cancer screening: the case of twodistricts in Zimbabwe. Afr J Reprod Health.2006;10:91–103.

[36] Abiodun OA, Olu-Abiodun OO, Sotunsa JO, et al.Impact of health education intervention on knowledgeand perception of cervical cancer and cervical screen-ing uptake among adult women in rural communitiesin Nigeria. BMC Public Health. 2014;14:814.

[37] Adetokunbo AO. The understanding and perceptionof service providers about the community-based cer-vical screening in Nigeria. Ann Trop Med PublicHealth. 2013;6:637.

[38] Peters LM, Soliman AS, Bukori P, et al. Evidence for theneed of educational programs for cervical screening inrural Tanzania. J Cancer Educ. 2010;25:153–159.

[39] Sanghvi H, Limpaphayom KK, Plotkin M, et al.Cervical cancer screening using visual inspectionwith acetic acid: operational experiences fromGhana and Thailand. Reprod Health Matters.2008;16:67–77.

[40] Maseko FC, Chirwa ML, Muula AS. Client satisfactionwith cervical cancer screening in Malawi. BMC HealthServ Res. 2014;14:420.

12 R. RAHMAN ET AL.

[41] White HL, Mulambia C, Sinkala M, et al. Motivationsand experiences of women who accessed “see andtreat” cervical cancer prevention services in Zambia.J Psychosomatic Obstet Gynecol. 2012;33:91–98.

[42] Chigbu CO, Onyebuchi AK, Ajah LO, et al.Motivations and preferences of rural Nigerianwomen undergoing cervical cancer screening viavisual inspection with acetic acid. Int J GynaecolObstet. 2013;120:262–265.

[43] Fort VK, Makin MS, Siegler AJ, Ault K, Rochat R.Barriers to cervical cancer screening in Mulanje,Malawi: a qualitative study. Patient Prefer Adherence[Internet]. 2011 Mar 14;5:125–31. Available from:http://dx.doi.org/10.2147/PPA.S1731.7

[44] Paul P, Winkler JL, Bartolini RM, et al. Screen-and-treat approach to cervical cancer prevention usingvisual inspection with acetic acid and cryotherapy:experiences, perceptions, and beliefs from demonstra-tion projects in Peru, Uganda, and Vietnam.Oncologist. 2013;18 Suppl:1278–1284.

[45] Quentin W, Adu-Sarkodie Y, Terris-Prestholt F, et al.Costs of cervical cancer screening and treatment usingvisual inspection with acetic acid (VIA) and cryother-apy in Ghana: the importance of scale. Trop MedInt Health. 2011;16:379–389.

[46] Chigbu CO, Onyebuchi AK. See-and-treat manage-ment of high-grade squamous intraepithelial lesionsin a resource-constrained African setting.Int J Gynaecol Obstet. 2014;124:204–206.

[47] Chigbu CO, Onyebuchi AK. Use of a portable dia-thermy machine for LEEP without colposcopy duringsee-and-treat management of VIA-positive cervicallesions in resource-poor settings. Int J GynecolObstetrics. 2014;125:99–102.

[48] Moon TD, Silva-Matos C, Cordoso A, et al.Implementation of cervical cancer screening usingvisual inspection with acetic acid in ruralMozambique: successes and challenges using HIVcare and treatment programme investments inZambézia Province. J Int AIDS Soc. 2012;15:17406.

[49] Levine LD, Chudnoff SG, Taylor K, et al. A 5-day educa-tional program for teaching cervical cancer screeningusing visual inspection with acetic acid in low-resourcesettings. Int J Gynaecol Obstet. 2011;115:171–174.

[50] Osingada CP, Ninsiima G, Chalo RN, et al.Determinants of uptake of cervical cancer screeningservices at a no-cost reproductive health clinic mana-ged by nurse-midwives. Cancer Nurs. 2014;38:177–184.

[51] World Bank Country and Lending Groups [Internet].World bank data help desk. [cited 2017 Jul 18].Available from: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups

[52] Chambers DA, Glasgow RE, Stange KC. The dynamicsustainability framework: addressing the paradox of sus-tainment amid ongoing change. Implement Sci.2013;8:117.

[53] de Sanjosé S. BLB-RLAGSBMMGDMJBF. human papil-lomavirus and related diseases report: senegal [Internet].©ICO/IARC Information Centre on HPV and Cancer(HPV Information Centre). 2017. Available from: http://www.hpvcentre.net/statistics/reports/SEN.pdf

[54] Xi LF, Touré P, Critchlow CW, et al. Prevalence ofspecific types of human papillomavirus and cervicalsquamous intraepithelial lesions in consecutive, pre-viously unscreened, West-African women over 35years of age. Int J Cancer. 2003;103:803–809.

[55] Perry H, Morrow M, Davis T, et al. Care groups – aneffective community- based delivery strategy forimproving reproductive, maternal, neonatal andchild health in high-mortality. Resource-ConstrainedSettings A Guide for Policy Makers and Donors;[Internet]. 2014. Available from: https://www.fsnnetwork.org/sites/default/files/resource_uploads/care_group_policy_guide_final_8_2014_0.pdf

[56] Perry HB, Zulliger R, Rogers MM. Community healthworkers in low-, middle-, and high-income countries:an overview of their history, recent evolution, andcurrent effectiveness. Annu Rev Public Health.2014;35:399–421.

[57] 2018World Population Data Sheet [Internet]. [cited 2018Oct 18]. Available from: http://www.worldpopdata.org/

Appendix: PubMed Search Conducted 11/30/2014

Three duplicates were removed from 2698 citations aftermoving them from PubMed to RefWorks - bringing thetotal citations to 2695.Note, duplication of this search at a later date is moreclosely approximated using two date filter to compensatefor the ongoing post-search expansion of PubMed: (“1950/01/01„[Date - Create] : “2014/11/30„[Date - Create])AND (“1950/01/01„[Date - Publication] : “2014/11/30„[Date - Publication])

High yield search

1/

((“Uterine Cervical Neoplasms„[Mesh] OR “Uterine CervicalDysplasia„[Mesh] OR “Cervical Intraepithelial Neoplasia„[Mesh] OR ((Cervical OR cervix OR cervixes OR cervicesOR exocervix OR exocervical OR exocervices OR endocervixOR endocervices OR endocervical OR ectocervix OR ectocer-vices OR ectocervical OR “Cervix Uteri„[Mesh] OR cervicova-ginal) AND (cancer[tw] OR cancers[tw] OR malignancy[tw]OR malignancies[tw] OR dysplasia OR dysplasias OR neopla-sia OR neoplasias OR neoplasm OR neoplasms OR neoplasticOR lesion OR lesions OR abnormal OR abnormalities ORadenocarcinoma OR adenocarcinomas OR precancerous ORprecursor ORwarts OR tumor[tw] OR tumors[tw] OR tumour[tw] OR tumours[tw]) OR HPV OR human papillomavirusOR “Papillomavirus Infections"[Mesh]))) AND ((((("AceticAcid"[Mesh] OR “acetic acid”[tw] OR iodine OR Lugol's ORLugol) AND ((VIA AND visual) OR (VILI AND visual) OR“visual inspection” OR “visual screening” OR “visual observa-tion” OR “visual assessment” OR “visual identification” OR“visual recognition” OR “visual detection” OR “visual evalua-tion” OR “naked eye” OR “naked-eye”)))))

Low yield search (multi-faceted)

2/

“Uterine Cervical Neoplasms"[Mesh] OR “Uterine CervicalDysplasia"[Mesh] OR “Cervical IntraepithelialNeoplasia"[Mesh] OR ((Cervical OR cervix OR cervixes ORcervices OR exocervix OR exocervical OR exocervices ORendocervix OR endocervices OR endocervical OR ectocervixOR ectocervices OR ectocervical OR “Cervix Uteri"[Mesh] ORcervicovaginal) AND (cancer[tw] OR cancers[tw] OR malig-nancy[tw] OR malignancies[tw] OR dysplasia OR dysplasiasOR neoplasia OR neoplasias OR neoplasm OR neoplasms OR

GLOBAL HEALTH ACTION 13

neoplastic OR lesion OR lesions OR abnormal OR abnormal-ities OR adenocarcinomaOR adenocarcinomas OR precancer-ous OR precursor ORwarts OR tumor[tw] OR tumors[tw] ORtumour[tw] OR tumours[tw]) OR HPV OR human papillo-mavirus OR “Papillomavirus Infections"[Mesh])

3/

test[tw] OR testing[tw] OR tests[tw] OR tested[tw] OR screen[tw] OR screening[tw] OR screenings[tw] OR screened[tw]OR prevention[tw] OR preventions[tw] OR preventive[tw]OR preventing[tw] OR prevent[tw] OR prevents[tw] ORcampaign[tw] OR campaigns[tw] OR campaigning[tw] OR“health promotion"[tw] OR “health promotions"[tw] OR“early detection"[tw] OR “early diagnosis"[tw] OR “earlydiagnoses"[tw] OR “early medical intervention"[tw] OR“early medical interventions"[tw] OR “early intervention"[tw]OR “early interventions"[tw] OR vaginal smear*[tw] OR cer-vical smear*[tw]OR “Mass Screening"[Mesh] OR “PreventiveHealth Services"[Mesh] OR “Preventive Medicine"[Mesh] OR“Primary Prevention"[Mesh] OR “Early Detection ofCancer"[Mesh] OR “Early Diagnosis"[Mesh] OR “EarlyMedical Intervention"[Mesh] OR “prevention andcontrol"[Subheading] OR “Health Promotion"[Mesh] OR“Gynecological Examination"[Mesh] OR “DiagnosticTechniques, Obstetrical and Gynecological"[Mesh]

4/

("Africa"[Mesh] OR Africa OR African OR Sahara OR Sub-Saharan OR subSaharan OR “Africa South of theSahara"[Mesh] OR “Southern Africa” OR “Africa,Central"[Mesh] OR “Africa, Eastern"[Mesh] OR “EasternAfrica” OR “Africa, Southern"[Mesh] OR “Africa,Western"[Mesh] OR “Western Africa” OR “Western Sahara”OR “Africa, Northern"[Mesh] OR “North Africa” OR“Northern Africa” OR “Sahrawi Arab Democratic Republic”OR Sahrawi OR Algeria OR Algerian OR Egypt OR EgyptianOR Libya OR Libyan OR Morocco OR Moroccan OR TunisiaOR Tunisian OR Cameroon OR Cameroonian OR “CentralAfrican Republic” OR Chad OR Chadian OR Congo OR“Congo Brazzaville” OR “Democratic Republic of the Congo”OR Congolese OR “Equatorial Guinea” OR “EquatorialGuinean” OR Gabon OR Gabonese OR Burundi ORBurundian OR Djibouti OR Djiboutian OR Eritrea OREritrean OR Ethiopia OR Ethiopian OR Kenya OR KenyanOR Rwanda OR Rwandian OR Rwandan OR Somalia ORSomalian OR Somali OR Somaliland OR Sudan OR SudaneseOR “South Sudan” OR Tanzania OR Tanzanian OR UgandaOR Ugandan OR Angola OR Angolan OR Botswana ORBotswanan OR Lesotho OR Basotho OR Malawi ORMalawian OR Mozambique OR mozambican OR NamibiaOR Namibian OR “South Africa” OR “South Africa"[Mesh]OR “South African” OR Swaziland OR Swazi OR Zambia ORZambian OR Zimbabwe OR Zimbabwean OR Benin ORBeninian OR ”Burkina Faso” OR “Cape Verde” OR “CapeVerdian” OR “Cote d'Ivoire” OR Gambia OR Gambian ORGhana OR Ghanaian OR Guinea OR Guinean OR “Guinea-Bissau” OR “Guinea-Conakry” OR Liberia OR Liberian ORMali OR Malian OR Mauritius OR Mauritania ORMauritanian OR Niger OR Nigerien OR Nigeria OR NigerianOR Senegal OR Senegalese OR “Sierra Leone” OR “SierraLeonean” OR Togo OR Togolese OR Afr OR “DevelopingCountries"[Mesh] OR “medical missions"[tw] OR “officialmedical missions"[tw] OR “Medical Missions, Official"[Mesh]

OR ngo[tw] OR ngos[tw] OR “non-governmental organiza-tions”[tw] OR “emerging countries”[tw] OR “emergingnations”[tw] OR “emerging economies”[tw] OR “emergingeconomy”[tw] OR “low resource countries”[tw] OR “lowresource nations”[tw] OR “low resource regions”[tw] OR“low resource region”[tw] OR “low resource setting”[tw] OR“low resource settings”[tw] OR “underdeveloped coun-tries"[tw] OR “underdeveloped country”[tw] OR “poorestcountries”[tw] OR “poorest nations”[tw] OR “impoverishednations”[tw] OR “impoverished nation”[tw] OR “developingnation”[tw] OR “developing nations”[tw] OR “developingcountries”[tw]OR “developing nation”[tw] OR “less developednation"[tw] OR “less developed nation”[tw] OR “less devel-oped country”[tw] OR “less developed countries”[tw] OR“least developed nations"[tw] OR “least developed coun-try”[tw] OR “least developed countries”[tw] OR “third-world”[tw] OR multi-country[tw] OR multi-national[tw] OR“emerging economy countries"[tw])

5/

("Neck"[Mesh] OR neck OR “head and neck” OR “CervicalVertebrae"[Mesh]OR “cervical vertebrae”OR “cervical plexus”OR “cervical atlas” OR “neck pain” OR “EsophagealDiseases"[Mesh] OR “Spinal Diseases"[Mesh] OR “SpinalCord Diseases"[Mesh] OR transcervical OR lumbosacral ORlumbar OR thorax OR thoracic OR larynx OR “disc”[tw] ORdiscs OR arthroplasty OR osteoporotic OR discectomy ORmyelopathy OR “Mycobacterium Infections"[Mesh] OR tuber-culin OR tuberculosis OR tuberculomas OR tuberculoma OR“root caries” OR “dental caries” OR dentin OR “tooth cervix”OR dentition OR “thyroid diseases” OR “ThyroidDiseases"[Mesh] OR thyroid[ti] OR parathyroid OR “thyroidcancers” OR “thyroid cancer” OR “thyroid neoplasms” OR“thyroid malignancies” OR “thyroid nodules” OR “ThyroidNeoplasms"[Mesh] OR thyroidectomy OR “ectopic pregnan-cies”[ti] OR “ectopic pregnancy”[ti] OR preterm birth*[ti] ORpreterm deliver*[ti] OR premature rupture*[ti] OR fetal[ti] ORcaesarean[ti] OR “Abortion, Induced"[Majr] OR abortion*[ti]OR “cervical cerclage” OR “cervical ripening” OR “cervicalincompetence” OR fistula OR “Obstetric LaborComplications"[Majr] OR “obstetric complications”[ti] OR“labor complications”[ti] OR “African-American”[tw] ORlions OR cows OR bovine OR bovines OR cattle OR heifersOR chimpanzees OR pigs OR “guinea-pigs” OR mice OR ratsOR cats OR dogs OR rabbits OR zebrafish OR hamsters)

6/

#2 AND #3 AND #4

7/

#6 NOT #5 (Completes the Low Yield retrieval set)

Combine the Low Yield and High Yield searches withEnglish filter and date limits

8/

#7 OR #1 (Order is important)

9/

#8 AND ("1950/01/01"[Date - Create] : “2014/11/30"[Date -Create]) AND ("1950/01/01"[Date - Publication] : “2014/11/30"[Date - Publication]) AND English[lang]

14 R. RAHMAN ET AL.