Available via license: CC0
Content may be subject to copyright.
1
1Title: Perceptions of barriers and facilitators for cervical cancer screening from women and
2healthcare workers in Ghana: Applying the Dynamic Sustainability Framework
3Short title: Barriers and facilitators for cervical cancer screening in Ghana.
4 Adwoa Bemah Boamah Mensah1*, Thomas Okpoti Konney2, Ernest Adankwah3, 4, John Amuasi4,
55, Madalyn Nones6, Joshua Okyere1,7, Kwame Ofori Boadu8, Felicia Maame Efua Eduah9, Serena
6 Xiong10, J. Robin Moon11, Beth Virnig12, Shalini Kulasingam6
71Department of Nursing, School of Nursing and Midwifery, Kwame Nkrumah University of
8 Science and Technology, Kumasi, Ghana
92School of Medical Sciences, Kwame Nkrumah University of Science and Technology, Kumasi,
10 Ghana
11 3Department of Medical Diagnostics, Faculty of Allied Health Sciences, Kwame Nkrumah
12 University of Science and Technology, Kumasi, Ghana
13 4Global Health Department, School of Public Health. Kwame Nkrumah University of Science and
14 Technology, Kumasi, Ghana
15 5Kumasi Center for Collaborative Research in Tropical Medicine, UPO PMB, KNUST- Kumasi-
16 Ghana
17 6Division of Epidemiology and Community Health, University of Minnesota, Minneapolis,
18 Minnesota, USA.
19 7Department of Population and Health, University of Cape Coast, University Post Office, Cape
20 Coast, Ghana.
21 8Ghana Health Service, Kumasi South Regional Hospital, Ashanti Region, Ghana
22 9Ghana Health Service, Suntreso Government Hospital, Post Office Box 14775, Kumasi- Ashanti
23 Region, Ghana.
24 10Division of Public Health Sciences, Washington University in St. Louis, St. Louis, MO
25 11Department of Health Policy and Management, City University of New York, Graduate School
26 of Public Health and Health Policy, USA
27 12College of Public Health and Health Professions, University of Florida, USA.
28 Emails
29 ABBM: bbemahc2000@gmail.com (Corresponding author)
30 TOK: tom.konney@yahoo.com
31 EA: ernestadankwah@yahoo.com
32 JA: amuas001@umn.edu
33 MN: nones002@umn.edu
34 JO: joshuaokyere54@gmail.com
35 KOB: kwame.boadu@ghs.gov.gh
36 FMEE: fmeeduah@gmail.com
37 SX: xiongs@wustl.edu
38 JRM: Robin.Moon@sph.cuny.edu
39 BV: bvirnig@phhp.ufl.edu
40 SK: kulas016@umn.edu
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.
2
41 Abstract
42 Cervical cancer screening has reduced cervical cancer-related mortality by over 70% in countries
43 that have achieved high coverage. However, there are significant geographic disparities in access
44 to screening. In Ghana, although cervical cancer is the second most common cancer in women,
45 there is no national-level cervical cancer screening program, and only 2 to 4% of eligible Ghanaian
46 women have ever been screened for cervical cancer. This study used an exploratory, sequential
47 mixed-methods approach to examine barriers and facilitators to cervical cancer screening from
48 women and healthcare workers perspectives, guided by the Dynamic Sustainability Framework.
49 Two convenience samples of 215 women and 17 healthcare personnel were recruited for this study.
50 All participants were from one of three selected clinics (Ejisu Government Hospital, Kumasi South
51 Hospital, and the Suntreso Government Hospital) in the Ashanti region of Ghana. Descriptive
52 analyses were used to group the data by practice setting and ecological system. Statistical
53 differences in means and proportions were used to evaluate women’s barriers to cervical cancer
54 screening. Quantitative findings from the women’s survey informed qualitative, in-depth
55 interviews with the healthcare workers and analyzed using an inductive thematic analysis. The
56 median age of women and healthcare workers was 37.0 years and 38.0 years respectively. Most
57 women (n=194, 90.2%) reported never having been screened. Women who had not been screened
58 were more likely to have no college or university education. Ecologic factors identified were lack
59 of knowledge about available services, distance to a clinic and requiring a spouse’s permission
60 prior to scheduling. Practice setting barriers included long clinic wait times and culturally sensitive
61 issue. The quantitative and qualitative data were integrated in the data collection stage, results, and
62 subsequent discussion. These findings highlight the need for non-clinician-based culturally
63 sensitive tool options for screening such as self-collected HPV tests to increase screening
64 participation in Ghana.
65
66 Keywords: Cervical cancer, Screening, Barriers, Facilitators, Women, Healthcare worker,
67 Dynamic Sustainability Framework, Mixed-methods, Ghana.
68
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
3
69 Introduction
70 Cervical cancer is a leading cause of cancer death among women in low-and-middle-income
71 countries (LMIC) primarily due to a lack of access to screening (1). Ghana has a population of
72 10.6 million women (or approximately 63.2% of the female population) aged 15 years and older
73 who are at risk of developing cervical cancer (2). Additionally, cervical cancer is the second most
74 frequently diagnosed cancer in Ghanaian women with an age-standardized incidence rate of 27.4
75 per 100,000 women and an age-standardized mortality rate of 17.8 per 100,000 women. In contrast,
76 despite a similar population size, Spain has age standardized cervical cancer incidence and
77 mortality rates of 8.2 and 1.7 per 100,000 women, respectively, due to the widespread availability
78 of screening and treatment (3).
79
80 Ghana faces many barriers in its effort to address the high incidence of cervical cancer. Currently,
81 there is no national human papillomavirus (HPV) vaccination campaign to protect women from
82 contracting HPV, the main cause of cervical cancer (4). As a result, both vaccination and screening
83 rates remain low (2). The majority of women present to clinics with advanced stage cervical cancer
84 (5). To address this, Ghana’s National Reproductive Health Policy was revised in 2014 to integrate
85 cervical cancer screening into existing reproductive health programs such as family planning and
86 sexually transmitted infections management services (6,7). The policy includes recommendations
87 for screening using Visual Inspection with Acetic Acid (VIA) and Papanicolaou tests for women
88 ages 25–45 years and cryotherapy for the treatment of precancerous cervical lesions. Despite this
89 policy, which has been in place for nine years, only 2.4 to 4% of eligible Ghanaian women are
90 screened for cervical cancer annually (8–11). Interviews with women in Ghana who have been
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
4
91 diagnosed with late stage cervical cancer confirm that barriers to preventative care include high
92 costs, a lack of knowledge, and lack of access to screening facilities (14).
93
94 Until recently, the World Health Organization (WHO) has recommended cervical cytology and
95 VIA for cervical cancer screening in LMICs. Of the two, VIA has been more readily adopted
96 across sub-Saharan Africa (SSA) due to the availability and low cost of acetic acid for visualizing
97 abnormal regions, the low cost to train personnel to conduct VIA-based screening, and the ability
98 to carry out immediate treatment if abnormal regions of the cervix are identified (referred to as
99 ‘see and treat’). However, limitations of this approach include a low sensitivity, low
100 reproducibility, and the need for trained personnel to conduct pelvic exams (15–18). Although
101 cytology has an improved sensitivity compared to VIA, it also involves a pelvic exam by trained
102 clinic staff for specimen collection, is more expensive than VIA, and requires training to interpret
103 laboratory results (16). As a result, the WHO updated their cervical cancer screening guidelines in
104 2022 to primarily recommend HPV testing in conjunction with self-collected samples for women
105 ages 30 - 50 years (19).
106
107 HPV-based screening has a significantly higher sensitivity compared to VIA and cytology for the
108 detection of high-grade cervical lesions (20,21). HPV testing can be performed with self-collected
109 samples, which allows patients’ privacy and, importantly, reduces reliance on clinics with trained
110 personnel. While this approach to cervical cancer screening is appealing, its implementation in
111 countries in SSA is not straightforward. Hurdles include determining when and where to offer
112 screening, how to best instruct women on self-collection methods, which collection device and
113 HPV test to use, and how to triage HPV-positive women. Deciding how to address these hurdles
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
5
114 requires an understanding of the unique barriers faced by a population operating within a specific
115 healthcare setting in a given country system. The Dynamic Sustainability Framework provides a
116 structure for re-examining interventions that have already been implemented, such as cervical
117 cancer screening in Ghana, and identifying key domains and constructs both within the practice
118 setting itself as well as the wider ecological system that can be used to improve patient outcomes
119 (22). To accomplish this, it’s important to understand barriers to optimizing patient outcomes from
120 the perspective of those utilizing healthcare services as well as those providing these services. This
121 study used the general DSF framework in combination with a mixed-methods sequential approach
122 to identify potential barriers to cervical cancer screening from the perspectives of both women and
123 healthcare workers.
124 Materials and Methods
125 Study design
126
127 Approval for this study was provided by the Committee on Human Research, Publication and
128 Ethics, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
129 (CHRPE/AP/043/22) and the Cancer Center Protocol Review Committee, University of Minnesota
130 (CPRC# 2022LS045). We used aspects of the Dynamic Sustainability Framework to better
131 understand the factors that affect the uptake of the current cervical screening practices currently
132 implemented in Ghana. In particular, we utilized two domains, “Practice settings” and “Ecological
133 system”, from the DSF (Fig 1) to provide a general framework for our data collection and analysis
134 (22). In addition, we adopted an exploratory, sequential mixed-methods study design to obtain
135 detailed information on the constructs for each of these domains. In particular, we used an initial,
136 quantitative phase to obtain information on “other practice setting constructs” and “population
137 characteristics” in the Ecological System domain. For this phase, we focused on Ghanaian women,
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
6
138 overall and stratified by key characteristics. Analysis of the quantitative data informed the second,
139 qualitative phase of the study. This second phase was designed to obtain in-depth information to
140 better understand practice setting constraints such as staffing. In particular, we conducted key
141 informant interviews with clinicians and laboratory workers who were directly involved in cervical
142 cancer screening and testing at the same facilities as those used for recruitment of women eligible
143 for screening for the quantitative data collection (Phase I). Our overall goal was to obtain
144 information from each of these domains that could help us better understand why cervical cancer
145 screening as an intervention may not be working well and how it can potentially be improved.
146
147 Study setting
148 Individuals were selected from three healthcare facilities: the Ejisu Government Hospital, Kumasi
149 South Hospital, and the Suntreso Government Hospital. Kumasi South and Suntreso Government
150 hospitals are urban facilities located in the city of Kumasi, Ghana’s second largest city, which has
151 a total population of 6,630,000 individuals (23). Ejisu Government Hospital is peri-urban and is
152 located in the municipality of Ejisu, which has a smaller population of approximately 181,000
153 individuals. Together, the selected facilities serve individuals from three metropolitan assemblies
154 in the Ashanti region: Ejisu, Asokwa, and Kumasi. These areas were chosen because participation
155 in screening is low and women served by health facilities in this region either have to wait for
156 outreach services or travel to the few health facilities offering cervical cancer screening (24).
157 Recruitment took place from 1st August 2022 and ended on 30th November 2022.
158
159
160 Fig 1. Dynamic sustainability framework for cervical cancer
161 screening in Ghana
162
163 Ecological System
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
7
164 Quantitative Data Collection (Survey)
165 The quantitative portion of the study examined other practice setting and population characteristics
166 that affect women’s abilities to obtain cervical cancer screening. A survey was adapted based on
167 a published study that evaluated the knowledge about and attitudes towards cervical cancer and
168 cervical cancer screening in a population of 900 women living in Uganda (25). The questionnaire
169 was drafted in English, translated into Twi and back-translated into English to check for potential
170 errors. The survey was divided into seven sections. The first section asked for demographic
171 information including the participant’s age, education level, marital status, occupation, average
172 income, number of dependents, and length of living in the current district. The second and third
173 sections asked about the participant’s knowledge of cervical cancer, cervical cancer screening and
174 vaccination. This section included questions regarding symptoms of cervical cancer, sources of
175 information for patient knowledge about cervical cancer and when and how often individuals
176 should be vaccinated and/or screened for cervical cancer. The fourth section was a knowledge
177 scale on the risk factors of cervical cancer in which participants were asked to respond ‘Yes,’ ‘No’
178 or ‘Don’t know’ (25). The fifth section was a belief scale that participants were asked to rank on a
179 five-point Likert scale with ‘1’ indicating strong disagreement and ‘5’ indicating strong agreement
180 with the statement (25). The sixth section asked about other practice setting characteristics, such
181 as wait times and distance to clinic, that affect participation in cervical cancer screening and the
182 final section asked about the woman’s sexual history and history of pregnancies. The questionnaire
183 was programmed in REDCap (Research Electronic Data Capture) to allow real time data collection
184 and uploading, and to also help with quality control and allow for efficient data processing (26,27).
185 The cross-sectional survey was conducted at the three selected healthcare facilities as noted above.
186 Eligible individuals were women 30 years of age and older who could undergo cervical cancer
187 screening, lived in the catchment area of one of the three hospitals, could provide informed consent
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
8
188 and could comprehend the survey questions either verbally or in written form. A convenience
189 sampling approach was used to enroll women; women presenting at one of the three facilities were
190 approached by a research assistant regarding possible interest in the study. If a woman was
191 interested, she was invited to a private room where the research assistant provided further
192 information and assessed participant’s eligibility. If the woman was eligible and willing to
193 participate after disclosure of the study details, the assistant asked for consent and administered
194 the questionnaire. All women received a compensation of GHC 30.00 for their participation in the
195 study.
196
197 Analysis of quantitative data
198 All analyses were performed using R Statistical Software (v4.1.3; R Core Team 2021) (28). Data
199 cleaning included evaluating missingness and non-response rates for various questions. All
200 variables used for analysis had less than 1% of missing responses. Descriptive statistics were used
201 to characterize population characteristics and evaluate other practice setting constraints that affect
202 participation in cervical cancer screening. Continuous variables were assessed for median and IQR
203 while frequencies were assessed for all categorical variables. Additionally, other practice setting
204 constraints were stratified by education level (some college or university versus no college or
205 university), marital status (married or living together versus single, divorced or widowed) and rural
206 versus urban. Categorical variables were created based on prior studies examining socio-
207 demographic variables that predict Pap-smear uptake in Ghanaian women (29). For marital status,
208 women who were married or living with a partner were compared to women who were single,
209 divorced, widowed, or separated. For educational status, women who had some college or
210 university education were compared to women with secondary school education or less. Women
211 living in rural areas were also compared to those living in urban areas, which has previously been
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
9
212 shown to influence uptake of cervical cancer screening (30). After stratifying by demographic
213 variables, prevalence ratios were used to assess differences in response frequency by demographic
214 category, and two-sided t-tests were used to analyze a difference in means and differences in
215 proportions for continuous variables. All statistical tests were deemed significant at the α=0.05
216 level.
217
218 From quantitative to qualitative methods
219 For the second phase, a semi-structured interview guide was developed that incorporated findings
220 that emerged as important and were related to the Practice Setting during the first phase. Based on
221 women’s reporting of limited accessibility to clinics, needing permission to schedule a doctor
222 appointment, and clinic wait times of one hour or greater, interview questions explored the views
223 of healthcare workers on health system hurdles faced by women accessing preventive care.
224
225 Practice Setting
226 Qualitative section (Interviews)
227 Understanding barriers to cervical cancer screening from a health system perspective
228
229 To better understand how healthcare settings, affect cervical cancer screening, we interviewed
230 healthcare workers from each of the three hospitals (the Ejisu Government Hospital, Kumasi South
231 Hospital, and the Suntreso Government Hospital). Healthcare workers were defined as
232 nurses/midwives, doctors or laboratory personnel who were involved in cervical cancer screening
233 (nurses/midwives and doctors) or testing (laboratory personnel). Individuals were eligible for the
234 study if they were 21 years of age or older, involved in cervical cancer screening, could
235 comprehend the questionnaire verbally, and could provide informed consent. Possible participants
236 were identified by administrators at each selected facility. Research assistants contacted each of
237 the possible participants to review the study details and eligibility criteria. Eligible participants
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
10
238 decided on the date, time and venue for the interview. All of the interviews were conducted in a
239 private office within the health facility to guarantee participant comfort and privacy. Eligible
240 individuals interested in participating were asked for consent prior to administering the interview.
241 Total sample size was 17 and this was based on the principle of theoretical saturation, which was
242 determined from the interviews (31). We determined we had reached data saturation when
243 additional interviews were redundant to previously collected data (31). By the 15th interview, there
244 was no new information. We carried out two more interviews to confirm that we had reached data
245 saturation. In total, 17 interviews were conducted. The healthcare workers’ questionnaire was split
246 into two sections with the first section asking for demographic information such as age, gender,
247 education level and length of employment at the current location. The second section asked about
248 current cervical cancer screening procedures at their respective clinic and potential constraints and
249 facilitators to accessing screening by women. This section used semi-structured interview
250 questions and data was collected through face-to-face in-depth interviews that were audio
251 recorded. On average, the interviews lasted 40 minutes.
252
253 Qualitative data analysis
254 Qualitative data analysis followed an inductive thematic analysis framework. The process of
255 analysis began with verbatim transcription of both the English (n=4) and local language (Twi,
256 n=13) audio-recorded data. For the interviews conducted in Twi, two expert translators were used
257 in a back-back translation process. A random selection of audio data was evaluated by a bilingual
258 research assistant to ensure accuracy in interviewing, transcribing and translation. Transcripts were
259 imported into QSR NVivo-12 Plus for data management and coding. The ‘nodes’ function in QSR
260 NVivo-12 was used for preliminary inductive coding (32). Intercoder agreement was 95% for
261 codes on questions regarding barriers to cervical cancer screening and 90% agreement was
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
11
262 observed on questions about factors that would facilitate a woman to seek cervical cancer
263 screening. To enhance trustworthiness of the data, strategies such as prolonged engagement, peer
264 debriefing, member checking (n=2), audit trail, reflexivity, detailed and appropriate descriptions
265 of the methodological processes and context were employed (33) The inductive coding results
266 were categorized according to emerging patterns and major themes around barriers to cervical
267 cancer screening and testing. Key quotes were extracted to reflect the various themes to support
268 the barriers.
269
270 Integration of the mixed-methods data
271 Following the structure of an exploratory, sequential mixed-methods design, the quantitative and
272 qualitative data were integrated in the data collection stage, results and subsequent discussion.
273 Themes and key quotes from the in-depth interviews were used to provide additional insight into
274 health system barriers women might face in accessing cervical cancer screening.
275 Ethics Approval and Consent to Participate
276 Approval for this study was provided by Committee on Human Research, Publication and Ethics,
277 Kwame Nkrumah University of Science and Technology (CHRPE/AP/043/22) and the Cancer
278 Center Protocol Review Committee, University of Minnesota (CPRC# 2022LS045). The
279 participants received written information about the study. Formal consent was obtained by writing
280 and participation was voluntary and anonymous.
281 Results
282 Population characteristics for the women’s sample are presented in Table 1a and characteristics
283 for the healthcare worker’s sample are presented in Table 1b. The median age of women was 37.0
284 (IQR: 32.0 - 45.5). Overall, women had varying levels of educational attainment. The majority of
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
12
285 women were married (69.3%, n=149) and self-employed (62.8%, n=135). The median number of
286 dependents was four. The median age of healthcare workers was 38.0 (IQR: 36.5 - 43.5). The
287 majority of healthcare workers identified as female (78.9%, n=15) and had received a university
288 degree (63.2%, n=12). The median length of time that healthcare workers had worked within their
289 profession was 12.0 years (IQR: 10.0 - 15.5).
290
291 Table 1a: Demographics for women participants
Participant Characteristic (n=215)
Age*
37.0 (32.0–45.5)
Education, n (%)
No primary school
12 (5.6)
Primary school
22 (10.2)
Junior high school
78 (36.3)
Senior high school
53 (24.7)
Some college or university
50 (23.2)
Marital Status, n (%)
Single
23 (10.7)
Married
149 (69.3)
Widowed
18 (8.4)
Divorced/Separated
12 (5.6)
Living with a partner
13 (6.0)
Occupation, n (%)
Agriculture/Farming
12 (5.6)
Self-employed
135 (62.8)
Housewife
12 (5.6)
Civil/Government/Private Sector Employee
40 (18.6)
Other
16 (7.4)
Dependent
Number of Dependents*
4.0 (3.0–5.5)
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
13
Length Living in District, n (%)
All of my life
19 (8.8)
Less than 1 year
22 (10.2)
Between 1 and 5 years
74 (34.4)
More than 5 years
100 (46.5)
292 *Age and number of dependents are reported as median (IQR)
293
294 Table 1b: Demographics of healthcare workers
Participant Characteristic
All healthcare workers (n=19)
Age*
38.0 (36.5–43.5)
Gender, n (%)
Female
15 (78.9)
Male
4 (21.1)
Education, n (%)
Some college
7 (36.8)
A college/university degree
12 (63.2)
Employment
Length of employment in years*
12.0 (10.0–15.5)
295 *Age and length of employment are presented in median (IQR)
296
297 Quantitative data describing other practice setting issues that women face when accessing cervical
298 cancer screening are presented in Table 2. The median distance that women reported traveling to
299 get to the closest healthcare facility was 5.5 kilometers (IQR: 3.1 - 10.1) and the median travel
300 time to the clinic was 25.0 minutes (IQR: 15.0 - 30.0). One hundred and twenty-four women
301 (57.7%) reported that they had to wait longer than one hour before being seen at the clinic. The
302 majority of women took public transportation to get to the healthcare facility (58.1% or n=125).
303 Aside from public transportation, women usually took either their own vehicle or a ridesharing
304 vehicle (31.6% of n=68). Seventy-three women (34.0%) responded that they needed permission
305 prior to scheduling a doctor’s appointment. One hundred and thirty-five women (62.3%) reported
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
14
306 that they knew of healthcare facilities that provided cervical cancer screening; however, most
307 women (90.2% or n=194) reported that they had never been screened for cervical cancer.
308
309 Table 2: Other practice setting constraints for accessing cervical cancer screening for all
310 women (n=215)
Factor
All Women
How far is it, in kilometers, to the local healthcare facility/ clinic from your home?*
5.5 (3.1 - 10.1)
How much time, in minutes, does it take you to get to the clinic?*
25.0 (15.0 - 30.0)
Wait longer than 60 minutes to be seen at the clinic n (%)
124 (57.7)
Primary mode of transportation, n (%)
Car (either own or rideshare)
68 (31.6)
Bicycle
1 (.004)
Walking
18 (8.4)
Public Transportation
125 (58.1)
Needs permission to schedule a healthcare visit n (%)
73 (34.0)
Knows of clinics/hospitals that offer cervical cancer screening n (%)
80 (37.2)
Has never been screened for cervical cancer n (%)
194 (90.2)
Has access to a mobile phone n (%)
203 (94.4)
311 **Distance to the local healthcare facility and time to clinic are reported as median (IQR)
312
313 Potential constraints stratified by highest level of educational attainment, marital status and rural
314 versus urban are presented in Table 3. When stratified by education level, women with no college
315 or university were significantly more likely to have to wait >1 hour to be seen at a local clinic
316 versus women with some college or university-level education (p = 0.001), and were significantly
317 less likely to be aware of local clinics offering screening services (p = 0.05). When stratified by
318 marital status, single, widowed, and divorced women were significantly more likely to walk to the
319 clinic versus women who were married or living with their partners (p = 0.03). Married women
320 were more likely to require permission to schedule a doctor appointment versus single women (p
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
15
321 < 0.001). When comparing women living in an urban versus rural area, women living in an urban
322 area spent, on average, a longer duration of time traveling to their local clinic.
323
324
325 Table 3: Individual-level constraints stratified by education level, marital status, and rural
326 versus urban for all women (n = 215)
Prevalence Ratio
Test
Statistic
p-value
Facility-level barrier
No
College/University
(n = 165), %(n)
Some College/
University
(n = 50), %(n)
Distance (km) to local healthcare
facility, mean (SD)
8.7 (11.4)
6.5 (6.5)
1.7
0.09
Time (minutes) to local healthcare
facility, mean (SD)
30.8 (23.2)
27.0 (23.3)
1
0.32
>1 hour wait time at clinic %(n)
64.8 (107)
34.0 (17)
13.7
<0.001
Primary mode of transportation
%(n)
Car (own or rideshare)
24.8 (41)
54.0 (27)
Referent
Bicycle
0.6 (1)
0
NA
Walking
9.7 (16)
4.0 (2)
0.54
0.41
Public transportation
63.6 (105)
40.0 (20)
0.78
0.49
Needs permission to schedule visit %(n)
32.7 (54)
38.0 (19)
0.27
0.6
Knows of clinics with screening %(n)
33.3(55)
50.0 (25)
3.88
0.05
Never been screened for cervical cancer
%(n)
90.3 (149)
90.0 (45)
<0.001
1
Prevalence Ratio
Test
Statistic
p-value
Facility-level barrier
Married/Living
Together (n=162),
n(%)
Single/Divorced/
Widowed (n=53),
n(%)
Distance (km) to local healthcare
facility, mean (SD)
8.4 (11.1)
7.6 (8.2)
0.56
0.58
Time (minutes) to local healthcare
facility, mean (SD)
30.0 (23.3)
30.0 (23.4)
0.002
1
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
16
>1 hour wait time at clinic %(n)
60.5 (98)
49.1 (26)
1.7
0.19
Primary mode of transportation
%(n)
Car (own or rideshare)
32.7 (53)
28.3 (15)
Referent
Bicycle
0
1.9 (1)
NA
Walking
5.6 (9)
17.0 (9)
2.3 (1.2 - 4.3)
0.03
Public transportation
59.9 (97)
52.8 (28)
1.0 (0.58 - 1.8
0.96
Needs permission to schedule visit %(n)
42.6 (69)
7.5 (4)
20.3
<0.001
Knows of clinics with screening %(n)
40.1 (65)
28.3 (15)
1.9
0.17
Never been screened for cervical cancer
%(n)
88.3 (143)
96.2 (51)
2.0
0.15
Prevalence Ratio
Test
Statistic
p-value
Facility-level barrier
Rural (n=70)
Urban (n=145)
Distance (km) to local healthcare
facility, mean (SD)
9.2 (6.9)
7.8 (11.8)
-1.1
0.26
Time (minutes) to local healthcare
facility, mean (SD)
24.2 (18.2)
32.7 (24.9)
2.8
0.005
>1 hour wait time at clinic %(n)
61.4 (43)
55.9 (81)
0.39
0.53
Primary mode of transportation
%(n)
Car (own or rideshare)
32.9 (23)
31.0 (45)
Referent
Bicycle
0
0.7 (1)
NA
Walking
5.7 (4)
9.7 (14)
1.18 (0.87 - 1.59)
0.37
Public transportation
57.1 (40)
58.6 (85)
1.03 (0.83 - 1.27)
0.8
Needs permission to schedule visit %(n)
31.4 (22)
35.2 (51)
0.15
0.7
Knows of clinics with screening %(n)
27.1 (19)
42.1 (61)
3.9
0.05
Never been screened for cervical cancer
%(n)
87.1 (61)
91.7 (133)
0.66
0.42
327
328 Qualitative Results
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
17
329 Based on the results of the quantitative analysis, two main themes related to the practice setting
330 were further explored in these interviews: client-level constraints and health system challenges.
331 The sub-themes that emerged through thematic analysis are presented in Table 4.
332
333 Table 4: Emerging practice setting relevant themes from healthcare worker interviews
Themes
Sub-themes
Financial constraints and cost of treatment
Clients' non-compliance due to shyness and feelings of discomfort
Client level barriers
Fatalistic views about the outcome of cervical cancer screening
Infrastructural inadequacies
Logistical constraints
Health system challenges
Low staff strength
334
335 Healthcare workers reported that women perceived financial constraints to cervical cancer
336 screening either through the direct cost of screening services, or the indirect cost associated with
337 transportation to the clinic. Healthcare workers also noted that women often do not think the
338 financial burden of screening services is worth the benefit it provides in detecting cervical cancer:
339 “Most of them complain about finances. They think that why should I use my money to go and do screening
340 while I can use it for other pressing matters. So, it is not seen as a priority.” (P00 2, laboratory personnel)
341 Healthcare workers also discussed non-compliance with screening due to feelings of discomfort
342 around the VIA procedure. Women experienced shyness for several reasons, including fear of
343 judgment on personal hygiene or discomfort from healthcare workers seeing them naked:
344 “Some women are shy of themselves for their fellow women to see their nakedness…this makes them non-
345 compliant and difficult to handle. At the end of the day, you will not be able to do the screening because they
346 will be reluctant to follow the screening procedures.” (P006, midwife)
347 Fears about the results from the screening process was another theme that emerged from healthcare
348 worker interviews. According to the key informants, their clients feared that the screening would
349 reveal that they had the disease (cervical cancer). This fear about a positive screening result was
350 attributed to fatalistic views held by the clients. That is, healthcare workers thought that clients
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
18
351 feared that once they had been diagnosed as having cervical cancer, it would mean that they were
352 going to die. Consequently, this fear made the clients non-compliant and unwilling to get screened.
353 “Fear of the unknown. You know after educating the women and giving them counseling, they will still not
354 feel comfortable for you to screen them because of fear of the unknown. She doesn’t know what the outcome
355 will be. The fear being diagnosed of cervical cancer and dying from the disease.” (P00 14, midwife)
356 The overarching theme of practice setting constraints was further broken into three sub-themes:
357 infrastructural inadequacies, logical constraints, and lack of staff. Healthcare workers noted that
358 the infrastructural capacity of healthcare facilities was not adequate for providing women's health
359 services. For example, clinics lacked beds necessary for cervical cancer screening, or the clinic
360 set-up did not allow for patient privacy. In clinics with a proper examination room, patients often
361 experienced long wait times due to the availability of only one or a few rooms:
362 “We don’t have a specific room with a special bed. The room that we use is not ideal...it does not support
363 the privacy of the client in any way. Because of that, the clients don’t feel comfortable getting screened.”
364 (P001, midwife)
365 Lastly, healthcare workers noted that inadequate staffing of clinics resulted in longer wait times.
366 This prevents patients from returning for future visits, and patients may also advise against others
367 wanting to use the clinic’s services:
368 “Inadequate staffing is the big issue. We are few, hence, the women wait for a long time to be attended to.
369 They go and do not come back. Sharing such experience with other women also discourage screening
370 uptake.” (P0016, midwife)
371 Discussion
372 This study illustrates the constraints that different subgroups of women in Ghana face when
373 accessing cervical cancer screening services and highlights areas in both the ecological system and
374 practice setting that need to be addressed to increase uptake of screening. In terms of ecological
375 factors affecting uptake, a key constraint is lack of knowledge about facilities offering cervical
376 cancer screening. This finding is consistent with studies conducted in Uganda (34) and Tanzania
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
19
377 (35). The 2016 study in Uganda found that less than half of surveyed participants were aware of
378 cervical cancer screening services (34), and the 2012 study in Tanzania found that women with a
379 higher level of cervical cancer prevention knowledge were significantly more likely to partake in
380 screening services (35). The implication of this finding is that individuals who intend to get
381 screened may miss opportunities for screening due to their lack of awareness about facilities that
382 offer screening. We also found that women with lower educational attainment were less likely to
383 be aware of facilities that provide cervical cancer screening services. This is not surprising as
384 higher educational attainment tends to empower women to access information including health
385 information such as where to access cervical cancer screening services (36,37).
386
387 Contrary to previous studies that have found long travel time for cervical cancer screening among
388 women in rural areas (38,39), we found that long travel times were more likely to be experienced
389 by those living in urban areas. This is possibly due to higher population density, traffic congestion,
390 or longer distances between residences and healthcare facilities in urban areas compared to rural
391 areas. Further research is required to fully comprehend why urban dwelling women in Ghana spend
392 long travel time in accessing cervical cancer screening services. Another key consideration is long
393 wait times, which are a major barrier to accessing cervical cancer screening with more than half
394 of participants (57.7%) waiting over 60 minutes to be seen at the clinic. This result is consistent
395 with a study in which 86% of a sample of 200 Kenyan women reported long wait times as a barrier
396 to cervical cancer screening participation (40). Similar findings were also reported in a qualitative
397 study of 48 women conducted in Accra, Ghana (41).
398
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
20
399 In terms of population characteristics, our study also highlights the fact that women with lower
400 educational attainment were more likely to experience longer wait times at local clinics. A
401 plausible explanation for this finding could be that women with higher education are more likely
402 to access cervical cancer screening services in private healthcare facilities where there is less
403 congestion and faster service delivery. This study did find that women with higher educational
404 attainment received, on average, higher monthly incomes, which increases access to more
405 expensive, private clinics. Women with lower educational attainment may be more likely to access
406 cervical cancer screening services at public healthcare facilities where there is a high volume of
407 patients, long queues, and fewer staff available for screening (42). These factors slow service
408 delivery consistent with our qualitative results from interviews with healthcare workers who
409 asserted that the low staffing in their facilities resulted in longer waiting times that discouraged
410 women from accessing cervical cancer screening services. Another perspective is that lower
411 educational attainment may be a proxy for socioeconomic status and lower information attainment.
412 This is in the sense that women with lower levels of education often face limited access to
413 economic opportunities, resulting in lower income levels and financial resources. This economic
414 disadvantage can directly impact their ability to access healthcare services, including cervical
415 cancer screening, due to financial constraints such as transportation costs or inability to afford
416 screening fees.
417
418 Consistent with another study conducted in Pakistan (36), we found that needing permission to
419 schedule appointments was a barrier to women’s access to cervical cancer screening. The study
420 revealed that married women were more likely to face the challenge of needing permission before
421 they could schedule a visit. This finding may be explained by existing patriarchal and gender norms
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
21
422 that place the male as the decision-maker in the household (43,44). Consequently, the autonomy
423 of women to make decisions regarding their healthcare is significantly reduced, especially among
424 married women. Ghanaian traditional culture also encourages women to ensure that their
425 nakedness is only seen by their husbands or partners. Given that the cervix is examined during
426 cervical cancer screening, married women may feel uncomfortable undergoing this process (12).
427 Those willing to undergo screening would then require their partners to consent and grant
428 permission before they can allow another person to see their ‘nakedness.’ This assertion was
429 corroborated by the accounts of the healthcare workers regarding shyness and discomfort women
430 feel in relation to cervical cancer screening. These results suggest that other forms of screening
431 such as HPV test-based screening through self-collected samples could increase screening uptake
432 by mitigating patient shyness and the need for permission. Self-collection does not require the
433 assistance of healthcare providers, which increases the privacy of the screening procedure.
434 Previous studies have shown that women feel comfortable and confident with this type of
435 screening (45). Further, this type of screening may also increase screening by circumventing the
436 need for clinic-based screening.
437
438 In conclusion, this study finds that a large proportion of Ghanaian women have never been
439 screened for cervical cancer. Both the quantitative and qualitative data highlight major barriers
440 that women face in access to cervical cancer screening. The women’s survey found that long clinic
441 wait times may impact screening uptake. This was especially true for women with lower
442 educational attainment (i.e. those with no college or university education). Additionally, there was
443 an overall lack of knowledge regarding where to obtain screening services, consistent with other
444 studies previous findings. In-depth interviews with healthcare workers confirmed that healthcare
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
22
445 facilities are not adequately staffed, resulting in longer patient wait times. Healthcare workers also
446 noted culturally sensitive issue as a barrier to screening uptake. This expanded upon the
447 quantitative data, which found that over one-third of surveyed women (and over two-thirds of
448 surveyed married women) needed permission prior to scheduling a doctor appointment. Moving
449 forward, the use of self-collected samples for HPV testing could mitigate barriers such as time
450 spent traveling to and waiting at a healthcare clinic. It might also provide a more culturally
451 sensitive tool to address issues of cultural norms because HPV sample self-collection does not
452 require the assistance of healthcare providers. Future studies should assess the feasibility of
453 implementing self-collected HPV samples as a possible method for cervical cancer screening.
454 Declarations
455 Author’s Contributions
456 ABBM: Conceptualization, methodology, data curation, formal analysis, investigation, project
457 administration, writing – original draft, writing – review & editing
458 TOK: Conceptualization, methodology, investigation, data curation, writing – review & editing
459 EA: Methodology, investigation, writing – review & editing
460 JA: Methodology, investigation, writing – review & editing
461 MN: Methodology, data curation, formal analysis, investigation, writing – original draft, writing –
462 review & editing
463 JO: Data curation, formal analysis, writing – original draft, writing – review & editing
464 KOB: Methodology, investigation, writing – review & editing
465 FMEE: Methodology, investigation, writing – review & editing
466 SX: Methodology, writing – review & editing
467 RM: Methodology, writing – review & editing
468 BV: Conceptualization, methodology, data curation, investigation, writing – review & editing
469 SK: Conceptualization, methodology, data curation, formal analysis, investigation, project
470 administration, writing – original draft, writing – review & editing
471 Availability of data
472 The datasets generated and/or analyzed during the current study are not publicly available due to
473 ethical reasons but are available from the corresponding author on reasonable request.
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
23
474 Funding
475 The study was funded by the Institute of Global Cancer Prevention Research (IGCPR). Masonic
476 Cancer Center, University of Minnesota.
477 Conflict of Interest
478 The author(s) declared no potential conflicts of interest with respect to the research, authorship,
479 and/or publication of this article.
480 Acknowledgments
481 We are grateful to all participants who shared their experiences in this study and funder-Institute
482 of Global Cancer Prevention Research (IGCPR). Masonic Cancer Center, University of
483 Minnesota.
484 References
485 1. Bray F, Colombet M, Aitken J, Bardot A, Eser S, Galceran J, et al. Cancer Incidence in Five
486 Continents, Vol. XII (IARC CancerBase No. 19). [Internet]. Lyon: International Agency for
487 Research on Cancer; 2023. Available from: https://ci5.iarc.who.int
488 2. Ghana: Human Papillomavirus and Related Cancers, Fact Sheet 2023. Fact Sheet. 2023;
489 3. Spain: Human Papillomavirus and Related Cancers, Fact Sheet 2023. Fact Sheet. 2023;
490 4. Effah K, Mawusi Wormenor C, Atuguba BH, Gedzah I, Amuah JE, Akakpo PK. A
491 revolution in cervical cancer prevention in Ghana. ecancer [Internet]. 2022 Aug 9 [cited
492 2024 Jan 30];16. Available from: https://ecancer.org/en/journal/editorial/123-a-revolution-in-
493 cervical-cancer-prevention-in-ghana
494 5. Dunyo P, Effah K, Udofia EA. Factors associated with late presentation of cervical cancer
495 cases at a district hospital: a retrospective study. BMC Public Health. 2018 Dec;18(1):1156.
496 6. Cancer Plan Ghana Ministry of Health.pdf [Internet]. [cited 2024 Jan 30]. Available from:
497 https://www.iccp-
498 portal.org/system/files/plans/Cancer%20Plan%20Ghana%20Ministry%20of%20Health.pdf
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
24
499 7. Mensah KB, Mensah ABB. Cancer control in Ghana: A narrative review in global context.
500 Heliyon. 2020 Aug;6(8):e04564.
501 8. 2021-Holistic-Assessment-Report_v1.7.3.pdf [Internet]. [cited 2024 Jan 30]. Available from:
502 https://www.moh.gov.gh/wp-content/uploads/2022/09/2021-Holistic-Assessment-
503 Report_v1.7.3.pdf
504 9. Calys-Tagoe BNL, Aheto JMK, Mensah G, Biritwum RB, Yawson AE. Cervical cancer
505 screening practices among women in Ghana: evidence from wave 2 of the WHO study on
506 global AGEing and adult health. BMC Women’s Health. 2020 Dec;20(1):49.
507 10. Handlogten KS, Molitor RJ, Roeker LE, Narla NP, Bachman MJ, Quayson S, et al. Cervical
508 Cancer Screening in Ghana, West Africa: Prevalence of Abnormal Cytology and Challenges
509 for Expanding Screening. International Journal of Gynecological Pathology. 2014
510 Mar;33(2):197–202.
511 11. Bruni L, Albero G, Mena M, Collado JJ, Gómez D, Muñoz J, et al. Human Papillomavirus
512 and Related Diseases in the World. Summary Report [Internet]. ICO/IARC Information
513 Centre on HPV and Cancer (HPV Information Centre); 2023 Mar. Available from:
514 https://hpvcentre.net/statistics/reports/XWX.pdf
515 12. Binka C, Nyarko SH, Awusabo-Asare K, Doku DT. Barriers to the Uptake of Cervical
516 Cancer Screening and Treatment among Rural Women in Ghana. BioMed Research
517 International. 2019 Nov 3;2019:1–8.
518 13. Tawiah A, Konney TO, Dassah ET, Visser LE, Amo‐Antwi K, Appiah‐Kubi A, et al.
519 Determinants of cervical cancer screening uptake among women with access to free
520 screening: A community‐based study in peri‐urban Ghana. Intl J Gynecology & Obste. 2022
521 Nov;159(2):513–21.
522 14. Appiah-Kubi A, O. Konney T, Amo-Antwi K, Tawiah A, K. Nti M, Ankobea-Kokroe F, et
523 al. Factors associated with late-stage presentation of cervical cancer in Ghana. gmj. 2022
524 May 31;56(2):86–94.
525 15. Omenge Orang’o E, Liu T, Christoffersen-Deb A, Itsura P, Oguda J, Washington S, et al.
526 Use of visual inspection with acetic acid, Pap smear, or high-risk human papillomavirus
527 testing in women living with HIV/AIDS for posttreatment cervical cancer screening: same
528 tests, different priorities. AIDS. 2017 Jan 14;31(2):233–40.
529 16. Orang’o EO, Were E, Rode O, Muthoka K, Byczkowski M, Sartor H, et al. Novel concepts
530 in cervical cancer screening: a comparison of VIA, HPV DNA test and p16INK4a/Ki-67
531 dual stain cytology in Western Kenya. Infect Agents Cancer. 2020 Dec;15(1):57.
532 17. De Vuyst H, Claeys P, Njiru S, Muchiri L, Steyaert S, De Sutter P, et al. Comparison of pap
533 smear, visual inspection with acetic acid, human papillomavirus DNA‐PCR testing and
534 cervicography. Intl J Gynecology & Obste. 2005 May;89(2):120–6.
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
25
535 18. Quao Y, Zhao Y. Experience with careHPV implementation in China. HPV World [Internet].
536 2017 Feb;(1). Available from: https://www.whri.org/wp-
537 content/uploads/2016/11/HPVWORLD_FEBRUARY_2017.pdf
538 19. World Health Organization. WHO guideline for screening and treatment of cervical pre-
539 cancer lesions for cervical cancer prevention, second edition: use of mRNA tests for human
540 papillomavirus (HPV) [Internet]. Geneva; 2021. Available from:
541 https://iris.who.int/bitstream/handle/10665/350652/9789240040434-eng.pdf?sequence=1
542 20. Dartell MA, Rasch V, Iftner T, Kahesa C, Mwaiselage JD, Junge J, et al. Performance of
543 visual inspection with acetic acid and human papillomavirus testing for detection of
544 high‐grade cervical lesions in HIV positive and HIV negative Tanzanian women. Intl Journal
545 of Cancer. 2014 Aug 15;135(4):896–904.
546 21. Katanga JJ, Rasch V, Manongi R, Pembe AB, Mwaiselage JD, Kjaer SK. Concordance in
547 HPV Detection Between Self-Collected and Health Provider–Collected Cervicovaginal
548 Samples Using careHPV in Tanzanian Women. JCO Global Oncology. 2021 Dec;(7):985–
549 91.
550 22. Chambers DA, Glasgow RE, Stange KC. The dynamic sustainability framework: addressing
551 the paradox of sustainment amid ongoing change. Implementation Sci. 2013 Dec;8(1):117.
552 23. Ghana Statistical Service. Ghana - Population and Housing Census 2000 [Internet]. Office of
553 the President; 2013 Feb. Available from:
554 https://www2.statsghana.gov.gh/nada/index.php/catalog/3/study-description
555 24. Nartey Y, Hill PC, Amo-Antwi K, Nyarko KM, Yarney J, Cox B. Cervical Cancer in the
556 Greater Accra and Ashanti Regions of Ghana. JGO. 2017 Dec;3(6):782–90.
557 25. Ma’som M, Bhoo-Pathy N, Nasir NH, Bellinson J, Subramaniam S, Ma Y, et al. Attitudes
558 and factors affecting acceptability of self-administered cervicovaginal sampling for human
559 papillomavirus (HPV) genotyping as an alternative to Pap testing among multiethnic
560 Malaysian women. BMJ Open. 2016 Aug;6(8):e011022.
561 26. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data
562 capture (REDCap)--a metadata-driven methodology and workflow process for providing
563 translational research informatics support. J Biomed Inform. 2009 Apr;42(2):377–81.
564 27. Harris PA, Taylor R, Minor BL, Elliott V, Fernandez M, O’Neal L, et al. The REDCap
565 consortium: Building an international community of software platform partners. J Biomed
566 Inform. 2019 Jul;95:103208.
567 28. R Core Team. R: A language and environment for statistical computing [Internet]. R
568 Foundation for Statistical Computing; 2021. Available from: https://www.R-project.org/
569 29. Mensah KB, Boamah Mensah AB, Yamoah P, Manfo J, Amo R, Wiafe E, et al. Socio-
570 Demographic Factors and Other Predictors of Pap Test Uptake Among Women: A
571 Retrospective Study in Ghana. CMAR. 2023 Jun;Volume 15:489–99.
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
26
572 30. Cunningham MS, Skrastins E, Fitzpatrick R, Jindal P, Oneko O, Yeates K, et al. Cervical
573 cancer screening and HPV vaccine acceptability among rural and urban women in
574 Kilimanjaro Region, Tanzania. BMJ Open. 2015 Mar 10;5(3):e005828–e005828.
575 31. Doody O, Noonan M. Preparing and conducting interviews to collect data. Nurse Res. 2013
576 May;20(5):28–32.
577 32. NVivo [Internet]. Lumivero; 2017. Available from: www.lumivero.com
578 33. Guba EG, Lincoln YS. Effective evaluation. 1st ed. San Francisco: Jossey-Bass Publishers;
579 1981. 423 p. (A Joint publication in the Jossey-Bass higher education and Social and
580 behavioral sciences series).
581 34. Ndejjo R, Mukama T, Musabyimana A, Musoke D. Uptake of Cervical Cancer Screening
582 and Associated Factors among Women in Rural Uganda: A Cross Sectional Study.
583 Tornesello ML, editor. PLoS ONE. 2016 Feb 19;11(2):e0149696.
584 35. Lyimo FS, Beran TN. Demographic, knowledge, attitudinal, and accessibility factors
585 associated with uptake of cervical cancer screening among women in a rural district of
586 Tanzania: Three public policy implications. BMC Public Health. 2012 Dec;12(1):22.
587 36. Habib K, Shafiq M, Afshan G, Qamar F. Impact of Education and Employment on Women
588 Empowerment. European Online Journal of Natural Social Sciences. 8(3):1–13.
589 37. Okyere J, Aboagye RG, Seidu AA, Asare BYA, Mwamba B, Ahinkorah BO. Towards a
590 cervical cancer-free future: women’s healthcare decision making and cervical cancer
591 screening uptake in sub-Saharan Africa. BMJ Open. 2022 Jul;12(7):e058026.
592 38. Allen-Leigh B, Uribe-Zúñiga P, León-Maldonado L, Brown BJ, Lörincz A, Salmeron J, et al.
593 Barriers to HPV self-sampling and cytology among low-income indigenous women in rural
594 areas of a middle-income setting: a qualitative study. BMC Cancer. 2017 Dec;17(1):734.
595 39. Rosser JI, Njoroge B, Huchko MJ. Knowledge about cervical cancer screening and
596 perception of risk among women attending outpatient clinics in rural Kenya. Intl J
597 Gynecology & Obste. 2015 Mar;128(3):211–5.
598 40. Ampofo AG, Adumatta AD, Owusu E, Awuviry-Newton K. A cross-sectional study of
599 barriers to cervical cancer screening uptake in Ghana: An application of the health belief
600 model. PLoS One. 2020;15(4):e0231459.
601 41. Osei EA, Appiah S, Oti-Boadi E, Hammond D, Awuah DB, Menlah A, et al. Experiences of
602 women awaiting cervical CANCER screening results from selected hospitals in Accra,
603 Ghana. BMC Public Health. 2022 Dec;22(1):1467.
604 42. Adongo AA, Azumah FD, Nachinaab JO. A comparative study of quality of health care
605 services of public and private hospitals in Ghana. J Public Health (Berl). 2022
606 Jul;30(7):1809–15.
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
27
607 43. Chen Z, Fiske ST, Lee TL. Ambivalent Sexism and Power-Related Gender-role Ideology in
608 Marriage. Sex Roles. 2009 Jun;60(11–12):765–78.
609 44. Olubodun T, Balogun MR, Odeyemi AK, Odukoya OO, Ogunyemi AO, Kanma-Okafor OJ,
610 et al. Barriers and recommendations for a cervical cancer screening program among women
611 in low-resource settings in Lagos Nigeria: a qualitative study. BMC Public Health. 2022 Oct
612 12;22(1):1906.
613 45. Bakiewicz A, Rasch V, Mwaiselage J, Linde DS. “The best thing is that you are doing it for
614 yourself” – perspectives on acceptability and feasibility of HPV self-sampling among
615 cervical cancer screening clients in Tanzania: a qualitative pilot study. BMC Women’s
616 Health. 2020 Dec;20(1):65.
617
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
for use under a CC0 license.
This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted February 27, 2024. ; https://doi.org/10.1101/2024.02.22.24303192doi: medRxiv preprint
